Novel Approaches to Positively Impact the Early Life Physiology, Endocrinology, and Productivity of Bulls

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

By

Bo R. Harstine, M.S.

Graduate Program in Animal Sciences

The Ohio State University

2016

Dissertation Committee:

Dr. Michael L. Day, Advisor Mel DeJarnette Dr. Christopher Premanandan Dr. Gustavo Schuenemann Dr. Joseph Ottobre

Copyrighted by

Bo Randall Harstine

2016

ABSTRACT

Changes to sire selection, such as the utilization of genomic evaluations, have created a desire to collect semen from superior sires as early as possible. Therefore, a series of experiments was performed in order to determine whether a novel exogenous

FSH treatment hastened puberty and positively impacted postpubertal semen production in bulls.

In the first experiment, angus-cross bulls received either 30 mg NIH-FSH-P1 in a

2% hyaluronic acid solution (FSH-HA, n =11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age. Blood was collected every 7 days to determine concentrations and at 59, 84, 94, 130, and 169 days of age to determine activin A concentrations. FSH concentrations were determined from blood collected preceding treatment every 3.5 days, as well as during three intensive collections commencing at 66,

108, and 157 days of age. Castration was performed at 170 days of age to examine testis weight, volume, diameter of seminiferous tubules, and the number of Sertoli cells per tubule cross section. Concentrations of FSH did not differ from 59 to 91 days of age, but became greater (P < 0.05) in FSH-HA than control bulls from 94 to 167.5 days. For each intensive sampling, FSH concentration was elevated (P < 0.05) in FSH-HA bulls for at least 18 hours post-injection at all ages examined. Activin A concentrations were greater

ii in FSH-HA than control bulls at 84 and 94 days. Testosterone concentrations, testis weight, testis volume, and diameter did not differ between treatments, but FSH-HA bulls had greater (P < 0.05) numbers of Sertoli cells per tubule cross section (45.2 ± 1.4 vs. 41.7 ± 0.9 cells).

In a second experiment, Holstein bulls were utilized to determine the FSH-HA treatment's effect on puberty attainment and mature sperm production. Bulls received either the FSH-HA treatment (FSH-HA, n = 17) or saline (n = 12) every 3.5 days from 62 to 170.5 days of age. Concentrations of FSH did not differ between treatments from 62 to

93.5 days of age, but became greater (P < 0.05) in FSH-HA bulls from 97 to 170.5 days of age. Activin A concentrations measured at 62, 86.5, 107.5, 139, and 170.5 days were greater (P < 0.05) in FSH-HA than control bulls at 86.5 and 107.5 days. FSH-HA bulls reached puberty (ability to produce 50 x 106 cells, 10% motility) sooner (P < 0.05) than control bulls (278 ± 7.7 vs. 303 ± 9.1 days), but there was no differences in mature sperm production measured from 571 to 627 days of age.

Together, these experiments highlight the efficacy of a novel FSH-HA treatment to hasten puberty. We propose that FSH-HA treatment positively effects the hypothalamo-pituitary-gonadal axis of bulls by stimulating the feedback loop involving activin A and FSH production as evidence by increased Sertoli cells. Impacts of this research are important not only to AI organizations, but also for cattle producers who AI their cows, and for consumers who rely on efficient food production.

iii

ACKNOWLEDGEMENTS

Attaining a Ph.D. has been a humbling experience, but reflecting on my work over the last three years makes me realize how fortunate I am. I have made many new personal and professional connections, been given the chance to study and work abroad, and have learned a lot about what drives and motivates me. While few would argue that self-reliance was critical to my success—especially in the later stages of this degree—I must give credit to the many people who helped me along the way.

To Mike Day, thank you for giving me a firm understanding of science and research. There is no doubt that I'll be using skills you taught me during my future endeavors and careers. Thanks to your family for always being welcoming. I pride myself on being your last graduate student (at Ohio State, at least), and please know that your well-respected career has acted as a springboard for myself as well as your other students.

Thank you to my committee members who provided expertise and insight throughout the process. Some, such as Mel DeJarnette and Chris Premanandan, have been honing my research skills for over five years. In addition, I appreciate the guidance and suggestions from Gustavo Schuenemann and Joe Ottobre.

Many of these projects would not have been possible without colleagues at Select

Sires. I associated with many of the personnel at SS even before graduate school, and I'm

iv so excited to begin working alongside some of the same people after completing this degree. Thanks to Clif and Mel for seeing me as a person worth mentoring. Thanks to the calf-campus, collection, and laboratory processing employees for letting me make your work more difficult.

Thank you to the OSU Beef Center crew for helping me with these projects.

Marty and Gregg, thank you for advice and for the care of the animals over the years.

Although I'm finishing graduate school as the sole member of the lab, I have to thank past lab mates for their friendship and the knowledge they provided. Leandro,

Fernanda, Matt, Martin, thank you for being friends and teachers. It is exciting that we may all work closely together in the future. It's comforting to know that we've established work and personal friendships that will last for a long time.

To my family, thank you for being a constant source of support. Mom and Dad,

I'm so grateful for the upbringing you gave Tyler, Colton, and I. You've instilled an incredible sense of determination and decency in all of us. The older I get, the more I appreciate growing up on a dairy farm and the lessons it taught me. Not many of my peers are blessed with such experiences. Katherine, thank you for your love and support!

If you faint in the day of adversity, your strength is small.

Proverbs 24:10

v

Vita

February 7, 1989...... Born – Dundee, Ohio

2011...... B.A. Cellular and Molecular Biology, Washington

& Jefferson College, Washington, Pennsylvania

2013...... M.S. Animal Science, The Ohio State

University, Columbus, Ohio

2013 to present...... OSU-Select Sires-C.E. Marshall Graduate Research

Associate, The Ohio State University, Columbus,

Ohio

Field of Study

Major Field: Animal Sciences

Reproductive Physiology and Endocrinology

vi

Table of Contents

Abstract ...... ii

Acknowledgments...... iv

Vita ...... vi

List of Tables ...... x

List of Figures ...... xi

Chapter 1: Introduction and Statement of the Problem ...... 1

Introduction ...... 1

Statement of the Problem ...... 3

Chapter 2. Review of the Literature ...... 5

2.1. The Events Preceding Puberty in the Bull ...... 5

2.1.1. Putting Bull Puberty in Context ...... 5

2.1.2. Initiation of GnRH Production and Activation of the Hypothalamo-

Pituitary-Gonadal Axis ...... 7

2.1.3. Initiation of LH Pulsatility and Testosterone Production ...... 10

vii

2.1.4. Development of the FSH, Activin, Inhibin, and Pathways

...... 14

2.1.5. Cellular and Structural Changes of the Preceding Puberty and ...... 24

2.2. Overview of Testicular Physiology and Endocrinology in Mature Bulls ...... 31

2.2.1. GnRH, LH, Testosterone, and Leydig Cells ...... 31

2.2.2. FSH, Sertoli Cells, and Spermatogenesis ...... 34

2.2.3. Inhibin, Activin, and Follistatin ...... 37

2.3. Previous Methods to Hasten Puberty in Bulls ...... 39

2.3.1. Direct Endocrine Manipulations to Advance Puberty ...... 40

2.3.2. Dietary Manipulations to Advance Puberty ...... 46

2.4. The Selection and Use of Bulls in the Artificial Insemination Industry ...... 49

2.4.1. Historical Perspective on Bull Acquisition and Use by AI

Companies...... 51

2.4.2. Utilization of AI Sires in the Genomics Era ...... 57

2.4.3. Economic Impacts of the Age of Puberty Attainment in Bulls

Destined for Use in AI ...... 66

2.5. Restatement of the Problem and Rationale for Research ...... 70

Chapter 3: Impact of an Exogenous FSH Treatment Regimen on the Endocrine and

Testicular Development of Prepubertal Beef Bulls...... 73 viii

Introduction ...... 73

Materials and Methods ...... 77

Results ...... 86

Discussion ...... 88

Chapter 4: Impact of a Prepubertal Exogenous FSH Treatment Regimen on the

Endocrinology, Puberty Attainment, and Mature Sperm Production in Holstein Bulls

Destined for Use in the AI Industry ...... 103

Introduction ...... 103

Materials and Methods ...... 107

Results ...... 114

Discussion ...... 116

Chapter 5: General Discussion...... 128

Literature Cited ...... 133

ix

List of Tables

Table 3.1. Testicular measurements, seminiferous tubule diameter, and the numbers of germ and Sertoli cells per tubule cross section (mean ± SE; 170 days of age) for Angus- cross bulls treated with either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n =

11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age...... 96

x

List of Figures

Figure 3.1. Mean (±SE) scrotal circumference measurements from Angus-cross bulls treated with either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline

(control, n = 11) every 3.5 days from 59 to 167.5 days of age...... 97

Figure 3.2. Systemic concentrations of FSH immediately before beef bulls received either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n

= 11) every 3.5 days from 59 to 167.5 days of age...... 98

Figure 3.3A,B,C. Systemic FSH concentrations in beef bull calves after injection of 0.5 ml containing either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA) or saline

(control) at 66 and 69.5 (Figure 3.3A), 108 and 111.5 (Figure 3.3B), and 157 and 160.5

(Figure 3.3C) days of age...... 99

Figure 3.4. Concentrations of activin A in beef bulls which received either 30 mg NIH-

FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age...... 101

Figure 3.5. Systemic testosterone concentrations of beef bulls treated with either 30 mg

NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every

3.5 days from 59 to 167.5 days of age...... 102

xi

Figure 4.1. Mean (±SE) scrotal circumference (SC) of Holstein bulls treated with either

30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days from 62 to 170.5 days of age...... 124

Figure 4.2. Systemic testosterone concentrations of Holstein bulls treated with either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days from 62 to 170.5 days of age...... 125

Figure 4.3. Systemic concentrations of FSH immediately before Holstein bulls received either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n

= 12) every 3.5 days from 62 to 170.5 days of age...... 126

Figure 4.4. Concentrations of activin A in Holstein bulls which received either 30 mg

NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every

3.5 days from 62 to 170.5 days of age...... 127

xii

CHAPTER 1

INTRODUCTION AND STATEMENT OF THE PROBLEM

INTRODUCTION

The world population reached 7.3 billion people in 2015, and despite projected decreases in the rate of growth, researchers estimate there will be 9.7 billion people in

2050 and 11.2 billion by 2100 (United Nations, 2015). Reliable food production will be crucial in sustaining the increasing population. Within agriculture, the cattle industries continue to seek ways to improve the efficiency of production in order to assure milk and meat production can meet consumer demand.

Factors such as improved management, nutrition, and genetics have played a large role in increasing the efficiency of cattle production. For example, U.S. dairy cows increased annual milk production from 8,080 pounds of milk per cow in 1965 to 22,393 pounds per cow in 2015 (USDA, 1966; USDA, 2016). This represents a 277% increase in production in only fifty years. The now common practice of artificial insemination (AI), more so in dairy than beef, has perhaps been the biggest factor contributing to recent genetic advancement in cattle. Once reliable semen extenders and cryogenic freezing procedures had been implemented in the early 1950s (Salisbury et al, 1941; Polge et al,

1

1949; O'Dell and Almquist, 1957; Polge, 1968), AI use increased drastically and is now commonplace in the U.S. industry. A recent USDA Agricultural Resource Management

Survey (ARMS) showed that 80.1% of U.S. dairy farms used AI in 2010, and these farms produced over 90% of the U.S. milk supply for that year (Gillespie et al., 2014). The dissemination of genetics achieved by the use of AI has contributed greatly to genetic improvement and cannot be overlooked. However, the more recent inclusion of genomic technologies has further accelerated the rate of genetic progress and selection intensity.

The adoption of genomic selection into dairy improvement programs in 2009 has impacted fundamental principles of traditional genetic selection by allowing enhanced understanding of individual animals' genomes. Genomic evaluations are often performed on young animals (i.e. prepubertal), meaning that young animals have begun to serve as the genetic foundations of the population. Genomics have positively affected the rate of genetic selection for this reason (Seidel, 2010). For example, the generation interval (GI), or the average age of the parents at the birth of their offspring, has decreased significantly for dairy cattle since the implementation of genomic testing. The GI for sires of bulls has decreased from 7 to 2.5 years of age, and the GI for dams of bulls has decreased from 4 to 2.5 years of age since 2008 (Garcia-Ruiz et al., 2016).

As the rate of genetic improvement in cattle is regularly pushed to new limits, the underlying biology necessary to attain puberty may act as an inextricable barrier to future progress. While genomics allows for the identification of genetically superior animals before sexual maturity (Schefers and Weigel, 2012), the attainment of puberty will remain a crucial biological event preceding the capability of full reproductive capacity

2

(Blood and Studdert, 1999). Furthermore, the semen production of a newly pubertal sire is suboptimal compared to mature production (Amann and DeJarnette, 2012), and this will negatively impact the ability to disseminate a bull's genetics. Hence, as cattle selection moves forward in the "genomic era," scientists will need to find the optimal balance of genetic selection and early-life management in order to determine whether or not it is possible to hasten the physiological processes preceding puberty and further accelerate the rate of genetic progress.

STATEMENT OF THE PROBLEM

The incorporation of new genomic selection tools has quickly changed the traditional paradigms used to select genetically elite sires destined for use in the AI industry. While a genomic evaluation of a sire can be performed as early as the embryonic stage (Humblot et al., 2010), breeders wishing to use the bull's genetics must still wait until the animal reaches puberty so the collection of gametes on a large scale becomes feasible. Moreover, once puberty is attained, a secondary problem exists in that the production capacity of young sires is often not able to meet market demand.

As the reliability and implementation of genomic testing increases, the pressure to hasten puberty attainment and to increase post-pubertal sperm production will continue to increase. The lag in time from the realization of an animal's genetic potential to the time at which they can produce enough gametes to supply the industry demand is detrimental to both genetic progress and financial profitability. Hence, research examining ways to positively advance the sexual maturation and reproductive function of bulls is useful in

3 order to best integrate new scientific advancements in genetics with the actual demands of the market (i.e. farmers).

The main aspect of research examined in this dissertation includes the use of a novel treatment regimen of exogenous (FSH) in bulls during the prepubertal period aimed to accelerate puberty attainment and testicular maturation. The effects of the treatment regimen on endocrine and gonadal development, puberty attainment, and mature sperm production are examined in two experiments using several different groups of bulls. This research has the potential to contribute knowledge useful to not only the AI companies that develop beef and dairy sires, but also to cattle producers and those in food production as a whole.

4

CHAPTER 2

REVIEW OF THE LITERATURE

2.1. The Events Preceding Puberty in the Bull

2.1.1. Putting Bull Puberty in Context

The endocrine events leading to puberty in the bull are dynamic and interwoven with feedback loops between the higher centers of the brain and the gonad. In some fast- maturing species (i.e. rodents), the hypothalamo-pituitary-gonadal (HPG) axis undergoes significant development prenatally. In other more slowly developing species, the majority of HPG axis development occurs postnatally. Thus far, the majority of research in bovine puberty attainment has appropriately focused on the postnatal period, and specifically, the peripubertal period. Puberty has broadly been defined as the "phase of bodily development during which the gonad secretes in amounts sufficient to cause accelerated growth of the genital organs and the appearance of secondary sex characteristics" (Donovan and van der Werff ten Bosch, 1965), and the peripubertal period is defined as a time of significant brain reorganization, ultimately leading to the production of the neuroendorocrine milieu that initiates the onset of fertility (Laroche,

2008).

5 The age of puberty in bulls differs greatly in the literature based on breed and applied experimental treatments. According to Macmillan and Hafs (1968), important development preceding puberty in Holstein bulls commences at 2 months of age and is

"qualitatively completed" by 10 months of age. Coincidentally, later reports generally cite the age of puberty at around 10 months of age. For example, in beef breeds, puberty has been reported at 295 for Angus, 326 days for Herefords, 296 days for Angus x Hereford

(Lunstra and Echternkamp, 1982), and 305 days for Angus x Charolais bulls (Brito,

2007b). Studies examining dairy bulls most often use Holsteins, and reports cite puberty at 288 days (Killian and Amann, 1972), 310 days (Harstine et al., 2015), and between 324 and 369 days (Dance et al., 2015). Another dairy breed, Brown Swiss, were found to attain puberty at 264 days of age (Lunstra and Echternkamp, 1982). Of course, management and the application of experimental treatments may have affected some of the above figures, and these topics will be discussed in further detail later in this review.

These reports simply serve to demonstrate that ages at puberty in bos taurus bulls can differ within breed, and even within experiment, depending on the management applied.

The age of puberty in the majority of the abovementioned studies coincides with a defined literature definition of puberty established by Wolf et al. (1965) as the age when the bull is able to ejaculate 50 million sperm with at least 10% of the cells having motility. However, a secondary definition of puberty has also been used alone or in combination with the Wolf definition and does not require semen collection of the animal. Lunstra et al. (1982) defined puberty in bulls as the age when a 28 centimeter scrotal circumference (SC) is attained, and this age correlates closely to when the animal

6 is physiologically able to ejaculate motile sperm. It is worth noting that bulls often attain a 28 centimeter SC before their ability to ejaculate the 50 million sperm with 10% motility (Wolf et al., 1965). Although a 28 centimeter SC may not functionally represent puberty as well as the Wolf definition, it remains a means of comparison when semen collection is impractical or not feasible.

2.1.2. Initiation of GnRH Signaling and Activation of the Hypothalamo-Pituitary-

Gonadal Axis

Activation of the HPG axis begins during fetal development. In humans, it has been determined that GnRH neurons establish in the during early embryogenesis. GnRH can be detected in the human hypothalamus at 15 weeks post- conception, and follicle stimulating (FSH) and (LH) can be detected even earlier, at 12 to 14 weeks of gestation (Kuiri-Hanninen et al., 2014). GnRH is produced and stored by the arcuate nuclei of the hypothalamus and is secreted in a pulsatile manner from GnRH-releasing neurons. Once released from the hypothalamus,

GnRH will enter the hypothalamo-hypophyseal portal system located in the median eminence (Senger, 1997). GnRH travels via the hypothalamo-hypophyseal portal system to its target tissue, the gonadotropes of the , and the gonadotropes release FSH and LH (Schwanzel-Fukuda and Pfaff, 1989; Campbell et al., 2009). Many of the early reports in bulls cite a general lack of (FSH, LH) fluctuation during the infantile period, classified from birth to approximately 12 weeks old (Pelletier and Lacroix, 1980; Amann, 1983; Amann and Walker, 1983; Amann et al., 1986).

7

However, subsequent studies detected measurable, albeit infrequent, GnRH pulses beginning at 2 weeks of age (Rodriguez and Wise, 1989). It is possible that GnRH is produced in the postnatal animal, but that the release of GnRH is short in duration and low in amplitude. Thus, this minimal production of GnRH may not be able to reach the pituitary or elicit a response from the pituitary (Glanowska et al., 2014).

There are contrasting theories as to how the hypothalamus gains the ability to begin pulsatile release of GnRH in bulls. Several will be discussed here. Original studies that concluded the hypothalamus was quiescent before 10-12 weeks of age in the bull cited a hormonal "block" to the production of GnRH. More studies have been conducted in heifers than in bulls seeking to describe this "gonadostat" theory of hypothalamic

GnRH suppression. Briefly, the gonadostat theory proposes that provides negative feedback to the hypothalamus, preventing the release of GnRH and its downstream product LH (Ramirez and McCann, 1963). Increased LH pulsatility is crucial for the attainment of puberty in both bulls and heifers (Brito et al., 2007a; Day et al., 1984). Indeed, in agreement with other animal models (Foster and Ryan, 1979;

Berardinelli et al., 1984), Day et al. (1987) determined that there is a decline in the number of estradiol receptors in the medial basal hypothalamus in heifers as puberty approached. The decline in estradiol receptor concentrations reduces the number of binding sites for estradiol to exert negative feedback effects on the hypothalamus's production of GnRH. It was similarly hypothesized that estradiol or its precursor, , served as the hormonal block to GnRH in prepubertal bulls

(Amann et al., 1986). For instance, there is measurable production of estradiol in the

8 adrenal glands and neurons outside of the hypothalamus (Henricks et al., 1988; McEwen,

1980).

More recent studies have added complexity to this model, citing that GnRH neurons do not have receptors (Lehman and Karsch, 1993). Rather, kisspeptin and kisspeptin neurons have been implicated as an intermediary "effector" of estradiol feedback. Kisspeptin neurons innervate GnRH neurons (Smith et al., 2007), and kisspeptin itself is a with highly stimulatory effects on the secretion of GnRH and the gonadotropins via increasing the pulsatile rate of GnRH neurons (Kadokawa et al.,

2008; Han et al., 2005). In opposition to kisspeptin's synergetic effects, estradiol acts as an inhibitor to kisspeptin production in the arcuate nucleus (Smith et al., 2007), and estradiol may decrease presence of kisspeptin neurons (Amstalden et al., 2014;

Desroziers et al., 2012). The role of kisspeptins in male puberty attainment is less clear, especially since the anteroventral periventricular nucleus (AVPV) region is sexually dimorphic and contains significantly less kisspeptin neurons than females (Kauffman et al., 2007). Furthermore, in males are thought to greatly affect the size and function of hypothalamic nuclei, including the sexual dimorphic nucleus of the pre-optic area (Dugger et al., 2008) and the suprachiasmatic nucleus (Fernandez-Guasti et al.,

2000). However, it has been observed that male mice with involving kisspeptin signaling pathways do not undergo puberty, are infertile, and have impaired spermatogenesis, all due to insufficient concentrations of gonadotropins (d'Anglemont de

Tassigny, 2010).

9

Our understanding of the neurobiology of GnRH secretion in prepubertal males will undoubtedly increase in the future. It is probable that a similar mechanism of a prepubertal GnRH block, such as the gonadostat theory in females, exists in the male as well. However, more research is needed in males, especially considering several of the key areas of the brain responsible for sexual development are sexually dimorphic.

Regardless of the mechanisms underlying the activation of the HPG axis in bulls, the literature suggests that at birth there is little to no GnRH production. If there is postnatal

GnRH production, it is likely either too low in concentration to elicit response from the pituitary or the pituitary has not yet gained the ability to respond (Glanowska et al.,

2014). Once the hypothalamus gains the ability to secrete GnRH, the anterior pituitary will respond by producing and releasing LH and FSH. If GnRH pulsatility does truly begin as early as two weeks of age (Rodriguez and Wise, 1989), then the anterior pituitary needs several more weeks in order to gain responsiveness and the ability to secrete LH.

2.1.3. Initiation of LH Pulsatility and Testosterone Production

As previously discussed, GnRH released from the hypothalamus travels to anterior pituitary and stimulates the gonadotrope cells to release LH. In turn, circulating

LH controls testicular production and release of testosterone (T; Huhtaniemi et al., 1986).

The hypothalamus and pituitary are closely integrated, and the pituitary must gain GnRH receptors (GnRH-R) in order to gain the stimulation necessary to release LH in a pulsatile manner. As mentioned, early studies cite that there are undetectable concentrations of

10

GnRH before approximately 18 weeks of age (Lacroix and Pelletier, 1979; Amann et al.,

1986), while later studies alternately report early-life GnRH production by the hypothalamus (Miller and Amann, 1986; Rodriguez and Wise, 1989). Interestingly, all of these studies theorize that regardless of hypothalamic GnRH production, the pituitary is unable to respond due to lack of sufficient numbers of GnRH-R. Studies of GnRH-R coincide well with these theories, since GnRH-R in the anterior pituitary increase by over

300% from 6 to 10 weeks of age, and the concentration of LH being stored in the anterior pituitary increases by nearly 70% during the same age interval (Amann et al., 1986).

As the GnRH rate of release increases beginning at 4 to 6 weeks of age, there is a transient rise in overall LH concentration in circulation. This increase in systemic LH is indicative of an increased pulsatility of the anterior pituitary (McCarthy et al., 1979;

Rawlings and Evans, 1995). The increase in LH beginning at 4 to 6 weeks is only transient. Once negative feedback mechanisms of testicular androgens emerge, there is a decrease in the elevated concentrations of LH (Amann, 1983; Bagu et al., 2006).

Postnatal increases in LH are reported as beginning at early as 4 to 5 weeks of age

(Rawlings et al., 2008), with other studies reporting the LH increase approximately one week later, at 6 weeks of age (Amann et al., 1986; Evans et al., 1995; Madgwick et al.,

2008). The transient elevation of LH has been reported to last until 18 weeks (Harstine et al., 2015), 20 weeks (Amann et al., 1986; Evans et al., 1995; Madgwick et al., 2008), and up to 25 weeks of age (Rawlings et al., 2008).

This LH rise early in life is considered essential to the attainment of puberty

(Brito et al., 2007b). The LH produced by the anterior pituitary will reach the Leydig

11 cells of the testes, causing them to release T. Bagu et al. (2006) reported that LH- receptors (LH-R) are static or decrease in number from 17 to 25 weeks of age, but begin a period of rapid increase from 25 to 56 weeks of age. An important biological change in

Leydig cells occurs around 25 weeks of age, where fetal Leydig cells either change to or are replaced by a new population of “mature” Leydig cells (Mendis-Handagama and

Ariyaratne, 2001; Bagu et al., 2006; Shima et al., 2013). The increase in LH corresponds to a period of rapid testicular growth, which begins between 20 to 25 weeks of age

(Amann and Walker, 1983; Bagu et al., 2006). Evidence of LH’s necessary role in prepubertal testicular maturation is provided by studies utilizing Leuprorelin, a GnRH agonist acting on pituitary GnRH receptors, thus reducing LH, FSH, and the subsequent estradiol and T concentrations in both sexes (Wuttke et a., 1996). In humans, Leuprorelin is part of the treatment plan to delay puberty in transgender youth until they begin hormone replacement therapy (Wolfe and Mash, 2008). In bulls, suppression of the LH increase from 10 to 14 weeks of age using Leuprolide acetate delays testicular maturation until at least 50 weeks of age (Chandolia et al., 1997a).

The increase in LH causes the proliferation and final maturation of the testicular

Leydig cells, which will produce T (Amann and Walker, 1983). To briefly describe

Leydig cell synthesis and secretion, Leydig cells are the only cells within the testes where cholesterol can be converted to pregnenolone and then to using the P450scc and 3β-HSD enzymes, respectively (Dong and Hardy, 2004). Progesterone will serve as the precursor for later androgen synthesis. Overall, the four enzymes in

Leydig cells responsible for the of androgens from cholesterol include

12

P450scc, 3β-HSD, P450-17, and 17β-HSD (Hall, 1994). In general, it is believed that LH binds LH-R on Leydig cells, beginning a cascade of events which include cAMP- dependent signaling pathway that causes the transport of cholesterol to the inner mitochondrial membrane. The preliminary steps of T synthesis will begin here before transferring outside of the mitochondria.

Testosterone is at very low or undetectable concentrations at birth, but detectable concentrations are observed in the bull as early as 10 weeks (Harstine et al., 2015), 12 weeks (Lacroix and Pelletier, 1979), and 14 weeks of age (Amann et al., 1986).

Dependent on study, the initiation of T production usually correlates well with the early gonadotropin rise, and this is also displayed in other species such as the ram (Wilson and

Lapwood, 1979). It can be concluded that T concentrations reach a critical level in bulls sometime between 18 – 25 weeks of age considering that there is enough T circulating to negatively feedback on the hypothalamus and pituitary to bring an end to the transient gonadotropin rise at this time (Schanbacher, 1982; Bagu et al., 2006). LH-R on Leydig cells begin to increase in concentration at the time of birth, and approach their highest concentrations soon after birth (Purvis et al., 1977; Hardy et al., 1990; Bagu et al., 2006).

Although concentration of LH-R begins to decrease at approximately 13 weeks of age, they do not lose receptor affinity (KA) for LH ligand at any time from birth to the attainment of puberty (Bagu et al., 2006). The concentration of T will stabilize at around

6 months of age in the bull (Miyamoto et al., 1989), and as the function of Leydig cells plateaus and T production becomes sufficient, the testes will begin a phase of rapid growth at 28 weeks of age. The beginning stages of spermatogenesis may have already

13 commenced by this age (Amann and Walker, 1983). Further regulation and function of

Leydig cell-produced T in mature bulls will be discussed later.

The research involving Leydig cell function and maturation is fairly comprehensive. It took over 100 years from the discovery of Leydig cells by Franz

Leydig until it was determined that they were the main source of androgens in the male

(Wattenberg, 1958). Prenatally, Leydig cells begin as mesenchymal stem cells and will undergo differentiation into progenitor cells located in the testicular interstitium. These progenitor cells will already be present in the interstitium at birth, and these cells will have developed in fetal Leydig cells under the hormonal direction of several molecules, including hormone (Mendis-Handagama and Ariyaratne, 2001). Today, we know that the Leydig cell population undergoes an important biological shift from “fetal” to

“mature” function as the ability to synthesize T is gained (Mendis-Handagama and

Ariyaratne, 2001).

2.1.4. Development of the FSH, Activin, Inhibin, and Follistatin Pathways

Several of the mechanisms and pathways associated with the production and secretion of FSH from the pituitary are similar to that of LH. However, the target cells, the agonists and antagonists affecting secretion, and the importance of FSH to this dissertation's focus are different than LH. Brief descriptions of the hormones and that will be discussed in this section are useful as a preface. In this section, the structure and timing of the presence of each hormone/molecule as it relates to prepuberty will be discussed. The cellular targets relating to each hormone will briefly be mentioned when

14 necessary, but their role in maturation and spermatogenesis will be discussed in detail in the next section.

FSH is a heterodimeric polypeptide hormone that shares an alpha subunit with

LH. Therefore, the beta-subunit gives specificity between the two gonadotropins. The anterior pituitary secretes FSH in response to hypothalamic GnRH, and FSH receptors

(FSH-R) are most notably located on testicular Sertoli cells (Ulloa-Aguirre and Timossi,

1998). Sertoli cells are the somatic cells within the seminiferous tubules essential for spermatogenesis and testicular structure. Sertoli cells facilitate spermatogenesis by

"nurturing" the early stages of germ cells as they mature to spermatozoa during spermatogenesis (Griswold, 1998). Inhibin is a dimeric with two active forms, inhibin A and B. However, inhibin B is the dominant form in most species.

Inhibin A and B isoforms share a common alpha-subunit but differ between their beta- subunits (beta-A or beta-B). Inhibins are produced within Sertoli cells and suppress the pituitary secretion of FSH (O'Connor and Kretser, 2004). Activins, just as inhibins, are members of the transforming growth factor beta (TGFbeta) family of growth factors.

There are three main forms of activin (A, B, and AB), each being composed of dimers of the inhibin beta subunits corresponding to their nomenclature. Therefore, activin A is composed of dimers of inhibin beta-A (βAβA) and so on (activin B, βBβB; activin AB

βAβB; de Kretser et a., 2002). In general, activins have many roles in regulation of cell proliferation throughout the body, but most relevant to this dissertation is that activins stimulate FSH secretion from the pituitary (Ling et al., 1986; Vale et al., 1986). Lastly, follistatin is a glycoprotein hormone with the ability to suppress activin activity by

15 binding it with high affinity, thereby preventing its stimulatory actions of FSH release at the pituitary level (Robertson et al., 1987; Nakamura et al., 1990). FSH, its effects on testicular development, and the feedback mechanisms involving FSH are of significance to this research as bulls were treated prepubertally with exogenous FSH from approximately 2 to 6 months of age, and this treatment had significant effects on the endocrinology and sexual development of the bulls during and after the conclusion of treatment.

It is speculated that FSH, like LH, undergoes a transient rise in concentrations in a period preceding puberty in bulls. Reports cite an increase in FSH beginning at approximately 4 weeks of age, with the decrease occurring between 25 weeks of age

(Evans et al., 1993; Rawlings and Evans, 1995; Aravindakshan et al., 2000; Bagu et al.,

2006) and 32 weeks of age (Amann and Walker, 1983). Conversely, other studies, such as McCarthy et al. (1979), cite no prepubertal increase in FSH. In the present research, fluctuations in FSH were not observed in control animals (which did not receive exogenous FSH), and this includes a sampling window from approximately 8 to 25 weeks of age. Furthermore, differences in FSH concentrations were not observed in bulls in a similar experiment where serum FSH concentrations were monitored from 5 to 13 weeks of age (unpublished). While these unpublished data did not include sampling prior to 4 weeks of age—and therefore may have missed the initial increase in gonadotropins—it must be noted that measurable decreases in FSH were not observed by end of sampling at

13 and 25 weeks for each experiment, respectively. Granted, these two periods of

16 sampling may have opportunely fell within the ages of the increase and decrease in FSH, and thus represent a window of time when FSH was increased in the animals.

An important differentiation between FSH and LH secretion in bulls centers on the mechanism of secretion. It is known in bulls and heifers that LH is secreted in a pulsatile manner from the pituitary in response to pulsatile release of GnRH from the hypothalamus (Day et al., 1984; Brito et al., 2007c). The anterior pituitary also secretes

FSH in response to hypothalamic GnRH release, but in a nonpulsatile manner (Amann and Walker, 1983; Stumpf et al., 1993). Studies by Akbar et al. (1974) in ewes determined that the half-life of FSH is longer than LH, providing one possible explanation for the lack of detectable pulsatility of FSH in ruminant models. Regarding the timing of the pituitary's ability to produce and secrete FSH, research in both males and females is scarce. However, connections between the pituitary's ability to gain

GnRH-R and secrete LH can be used as translational theories to make inferences for FSH production. While it is likely that nearly undetectable concentrations of GnRH are helping establish the neural networks of the HGP axis before measurable GnRH and LH pulsatility (Rodriguez and Wise, 1989; Glanowska et al., 2014), it is plausible that the pituitary has not gained the ability to respond to GnRH prior to the production of gonadotropins (Amann et al., 1986). Despite the need for further clarifications of prepubertal FSH secretion and function in ruminants, it is known that concentrations of

FSH stabilize prior to puberty and remain stable during maturity while remaining crucial to spermatogenic function (Simoni et al., 1999).

17

The identification of activin in 1968 by two independent groups occurred while isolating proteins in follicular fluid (Ling et al., 1968; Vale et al., 1968). Activin was originally thought to only be produced within the gonadotropes of the pituitary, and thereby exert its action on FSH secretion via paracrine mechanisms (Meunier et al., 1988;

Roberts et al., 1989; Corrigan et al., 1991). However, and important to this dissertation research, activin A production has also been isolated in Sertoli cells (Anderson et al.,

1998).

The timing of activin production and the role it plays in the attainment of puberty in males is difficult to discern due to lack of research. Reports in women cite that there is no change in activin A concentrations before or during puberty (Foster et al., 2000). It is known that activins are present prenatally and play a crucial role in embryonic development since mouse models lacking sufficient activin production have significant skeletal and facial abnormalities. Furthermore, the importance of activin in postnatal reproductive capacity is clear in the same mouse models considering these mice have

"compromised" reproductive performance into adulthood (Matzuk et al., 1995).

According to other sources, activin concentrations are greatest after birth and appear to decrease at a time similar to the decrease in the gonadotropins between 25 and 32 weeks

(Barakat et al., 2008). Activins are reported to have significant positive correlations to

Sertoli cell proliferation early in life (Barakat et al., 2008), so the concurrent decrease in

FSH and activin near the time of puberty attainment implicates both molecules as drivers

Sertoli cell proliferation prior to puberty (Buzzard et al., 2004). The decrease in activin in the time preceding puberty may play an important role in the attainment of

18 spermatogenesis, as it has been hypothesized that a reduction of activin bioactivity in the testes is required prior to the onset of spermatogenesis (Meehan et al., 2000).

Activin's mechanisms of action in the pituitary are generally stimulatory to the production of FSH. However, activin's effect in the testes has been documented as both stimulatory and inhibitory. In prepubertal testicular tissue, activin A stimulates Sertoli cell proliferation while inhibiting germ cell proliferation (Boitani et al., 1995). Other reports disagree, stating that activin induces germ cell proliferation while inhibiting

Sertoli cell function (Mather et al., 1990; de Winter et al., 1993). Clearly, the roles of activin in the male gonad require further elucidation, especially if the role of activin is species-specific. The mechanisms of of activins in the pituitary are under recent investigation. Vale and colleagues, who were one of the first research groups to identify activin, were able to verify that directly bind the same receptor and exert transduction like many other TGF-beta ligands, via a serine/threonine kinase and SMAD effector (Lewis et al., 2000). This indicates that inhibin action is in direct antagonism of activin via binding. To summarize some of the recent findings involving activin signaling in the pituitary, there have been two major activin receptors found in the pituitary (ACVR2A and ACVR2B), of which the type II receptor, ACVR2A, has been found to be the more critical of the two

(Attisano et al., 1992). Upon the binding of activin to activin receptors, SMAD families activate to effect intracellular signaling (Suszko et al., 2003).

It is known in vitro that SMAD transcription factors directly influence pituitary LβT2 cells to induce production of the FSH-beta promoter, thus causing the assembly of FSH

19 molecules within the pituitary (Suszko et al., 2005). Interestingly, there are many less studies examining the activin-induced SMAD pathway production of FSH in vivo. It is known that SMAD3 deficiency in mice manifests with a large reduction in FSH-beta subunit mRNA levels (Coss et al., 2005). Lastly, several recent studies have also described major involvement in FSH synthesis by the forkhead box protein L2 (FOXL2) that is encoded by the FOXL2 gene and is also a transcription factor. SMAD3 and

FOXL2 may very well work in concert or in a cascading pathway to cause the transcription of FSH-beta subunits (Lamba et al., 2010). A comprehensive review of the history of activin-stimulated FSH synthesis, especially in regards to in molecular signaling, can be found a review by Bernard and Tran (2013).

Follistatins are unrelated to the proteins previously discussed here, and they are that bind to activin to neutralize its stimulatory effects on FSH synthesis

(Esch et al., 1987). Specifically, follistatin binds to the C-terminal domain of activin, which is necessary for the binding of activin to the previously described ACVR2A receptor in the pituitary. There are several confirmed areas of follistatin production in the body. Follistatin is produced in the same cells that produce activin, as well as other adjacent cell types (Phillips and de Kretser, 1998). Follistatin production has also been reported in germ cells (Ogawa et al., 1997) and Sertoli cells (Kaipa et al., 1992) within the testes. It is often cited that follistatin concentrations in circulation exceed activin concentrations. Furthermore, it appears that most of the activin in the body is bound to follistatins, and therefore, is biologically inactive. Since most activin is bound, there are

20 theoretically "reservoirs" of activin in several tissues in the body (Sugino et al., 1993;

Sugino et al., 1997).

Information on the timing of follistatin production is scarce, but it is generally hypothesized that follistatin is not present prenatally but begins production sometime prior to maturity. There is no expression of follistatin in the fetal testis (Roberts, 1997), but it is unclear if follistatin is present in the higher brain centers at prenatal or early ages.

The lack of follistatin at early ages is unusual considering activin and inhibin are produced in the early stages of development (Roberts, 1997). Interestingly, it appears that the presence of follistatin early in life may actually be detrimental, as transgenic mice over-expressing follistatin at young ages have Leydig cell hyperplasia, seminiferous tubule degeneration, and problems with spermatogenesis (Guo et al., 1998). The lack of follistatin prenatally and a subsequent presence of follistatin in adult testes suggests that there are dynamic events in the establishment of the paracrine control of activin

(Anderson et al., 1998).

Inhibins are strongly related to activin molecules and share one of the beta- subunits with activin while having a heterologous pairing with a distinct alpha-subunit.

The two main types of inhibin are inhibin A and inhibin B, nomenclatured αβA and αβB

(Mason, 1987). The expression of inhibin A versus B is variable between species. In humans, inhibin B is the circulating form of inhibin, while in sheep, inhibin A is predominant (Illingworth et al., 1996). Interestingly, bulls are known to produce both inhibin A and B in the testes, but only inhibin A is measurable in the circulation (Kaneko et al., 2006). Inhibin is produced by Sertoli cells upon binding FSH ligand to the FSH-R.

21

Once secreted into the circulation, inhibin is able to negatively feedback on the higher centers of the brain to decrease the biosynthesis and secretion of FSH (de Kretser and

Robertson, 1989).

The mechanisms by which inhibin affects the pituitary and causes downregulation of FSH production still require investigation, but thus far, there are two theories describing inhibin’s effect on the pituitary. First, since the beta-subunit of inhibin is shared as the beta-subunit of activin, inhibin formation could technically reduce the formation of activin biosynthesis by not allowing activin beta-subunits to homodimerize.

However, it must be noted that since inhibin is not produced within the pituitary, this would have a small effect on the activin formed in the pituitary, and would more so affect the formation of testicular activin (Gregory and Kaiser, 2004). A second way inhibin negatively affects FSH biosynthesis is through the previously described binding of the activin type II receptor (ACVR2A; Lebrun and Vale, 1997). By binding to the ACVR2A receptor, inhibin could disrupt activin's ability to bind to this receptor and activate the

SMAD/FOXL2 pathway to signal FSH synthesis (Lebrun and Vale, 1997). Interestingly, inhibin binds to the ACVR2A receptor with lower affinity than activin, so other mechanisms of inhibin-induced inhibition of FSH are being investigated, such as the possible existence of a separate inhibin receptor whereby inhibin can exert its effects separately from ACVR2A (Lewis et al., 2000).

In prepubertal bulls, inhibin is detectable at 8 weeks of age (MacDonald et al.,

1991), and concentrations of inhibin begin to inversely correlate to FSH concentrations a few weeks later at 10 weeks of age. The negative correlation of inhibin to FSH

22 concentrations at 10 weeks corresponds chronologically to the age at which Sertoli cells are able to reflexively respond to FSH by producing inhibin (MacDonald et al., 1991). A more intensive study by Kaneko and colleagues (2006) measuring inhibin A and B concentrations from 5 to 50 weeks of age reported that circulating inhibin A is secreted from the infantile to the postpubertal period, with a plateau in concentrations occurring from 10 to 14 weeks of age, with a steady decrease occurring until the end of sampling at

50 weeks. Within the testes, both inhibin A and B concentrations undergo a general trend of increasing in concentration significantly from 5 to 50 weeks of age. The increase in inhibin A production by the testes as puberty approaches is potentially explained by the transition of Sertoli cells from an undifferentiated state to a mature state occurring between 24 and 28 weeks of age (Kaneko et al., 2006; Curtis and Amann, 1981).

The pathways of FSH secretion in concert with the agonist and antagonistic molecules described here are crucial in the context of this dissertation research.

Treatments of exogenous FSH were applied during a prepubertal period based not only on the efficacy of such a treatment in a production setting, but also when the effects of this treatment regimen could have maximum effect on the physiology and endocrinology of bulls. In summary, as the pituitary gains the ability to secrete FSH in response to hypothalamic GnRH, and there are feedback mechanisms developing involving activin, inhibin, and follistatin that will affect the pituitary's ability to produce and secrete FSH.

The data generated from control (untreated) bulls in the projects within this dissertation serve to add to the information already available regarding the prepubertal endocrinology of bulls. In addition, the utilization a novel exogenous FSH treatment regimen in the

23 same experiments has provided insight into the plasticity of the prepubertal endocrinology and physiology of bulls. The ability to manipulate the prepubertal endocrine axis using targeted treatments has meaningful implications for the AI and genetic industries in cattle.

2.1.5. Cellular and Structural Changes of the Gonad Preceding Puberty and

Spermatogenesis

Information regarding the effects of the abovementioned endocrine molecules on the cellular components of the testes was deliberately left for further clarification in this section in an attempt to compartmentalize prepubertal endocrinology from the physiology. The testes undergo drastic developmental changes postnatally before achieving the necessary cellular organization to initiate the first rounds of spermatogenesis, and the reproductive hormones play a large role in the orchestration of this maturation.

Beginning with a brief overview of mature testicular structure to give context to the prepubertal testis development, spermatogenesis occurs within the seminiferous tubules. Within the tubules, two main cell types predominate, one being the spermatogonial stem cells which lie close to the basement membrane of the tubule and which give rise to all subsequent spermatogonia via the process of spermatogenesis. The secondary cell type within the seminiferous tubules, and the only somatic cell, are the

Sertoli cells, which govern spermatogenesis by regulating the biochemical surroundings of the germ cells as the cytoplasm of the Sertoli cell surrounds each developing germ cell

24

(Griswold, 1998). The seminiferous tubules are surrounded by peritubular myoid cells, whose contractile ability will "squeeze" developed spermatogonia and fluids through the tubules (Maekawa et al., 1996). The remaining space of the testis outside of the tubules harbors endothelial cells and , but perhaps most importantly, the steroidogenic

Leydig cells. The development of the cells most important to spermatogenesis, the

Leydig cell, Sertoli cell, and the actual germ cells will be the concentration of discussion in this section.

Leydig cells derive from the mesenchymal stem cells that have migrated to the testis interstitium during embryogenesis. As these differentiate in to progenitor cells, they begin to have Leydig cell characteristics, and continue their development through gestation to become what are classified as fetal Leydig cells. Thyroid hormone is implicated as a main driver of differentiation of mesenchymal cells to fetal Leydig cells

(Mendis-Handagama and Ariyaratne, 2001). The interplay of Leydig cells, LH, and testosterone is dynamic postnatally. As mentioned before, the concentration of LH receptor (LH-R) on Leydig cells increases significantly around the time of birth, and LH-

Rs are probably near their greatest concentrations soon after birth (Purvis et al., 1977;

Hardy et al., 1990; Bagu et al., 2006). Concentrations of LH-Rs decrease from 13 to 21 weeks of age as the numbers of undifferentiated progenitor Leydig cells undergo differentiation to mature Leydig cells (Mendis-Handagama and Ariyaratne, 2001; Bagu et al., 2006). Despite the decrease in LH-Rs prior to puberty, the equilibrium constant of the receptors (KA), or affinity of the LH-R to bind LH, does not decrease. This is likely crucial as the Leydig cells begin to synthesize the enzymes necessary for androgen

25 synthesis that occurs at approximately 20 weeks of age (Bagu et al., 2006; Curtis and

Amann, 1981). Corresponding to the production of androgens, the transient increase of

LH will decrease to a sustained concentration at around 20 weeks of age (Curtis and

Amann, 1981). The initiation of testosterone production coincides with a rapid testicular growth starting at 28 weeks of age, and this growth is caused by the rapid expansion of the populations of Leydig, Sertoli, and germ cells (Bagu et al., 2006). The concentration of T will stabilize at around 6 months of age in the bull (Miyamoto et al., 1989), and the

Leydig cells will continue to serve as the main source of androgens in the male throughout maturity.

The Sertoli cells originate embryonically from the coelomic epithelium, which forms the genital ridges. As the primordial germ cells migrate to the genital ridge from extragonadal origins, the early sex cords are formed (Merchant-Larios, 1976). Early

Sertoli cells express SRY gene, which may facilitate germ cell migration, and is definitely involved in the formation of the testis cord via the downstream induction of anti-mullerian hormone (AMH) production (Barrionuevo et al., 2011). In the postnatal animal, Sertoli cells are generally termed "undifferentiated supporting cells" of the testes before a period of maturation occurs. In bulls, the transition from undifferentiated supporting cells to Sertoli cells occurs from 13 to 25 weeks of age (Bagu et al., 2006).

Differentiation entails distinct morphological changes such as increases in size, attainment of a more columnar shape, and the acquisition of a tripartite nucleus (Sharpe et al., 2003). The tripartite nucleus is characterized by a large nucleolus and two satellite karyosomes, and this unique shape is a visual signature to identify Sertoli cells (Sharpe et

26 al., 2003). Sertoli cells are similar to other epithelial cell-types in their characteristics upon maturity, meaning that they have a polarity to their orientation, with the basal pole containing the nucleus oriented toward the outside of the tubule and the apical pole is oriented toward the tubule lumen (Skinner and Griswold, 2005). Another sign of differentiation of Sertoli cells is the formation of tight junctions, which create the blood- testis barrier and the adluminal compartment of the tubule that will fill with fluid (Sharpe et al., 2003).

As the pituitary begins to elevate production of FSH, FSH-R are established on the Sertoli cells. Concentrations of FSH-R are reported to be unchanging from 5 to 17 weeks of age. A slight decrease in FSH-R with a nadir at 25 weeks of age may be explained by the switching of undifferentiated Sertoli cells to their mature form at this time. It is then believed that Sertoli cell FSH-R increases beginning at 29 weeks of age until stabilized, mature spermatogenesis occurs (Bagu et al., 2006). During the ages examined (5 – 56 weeks) by Bagu et al. (2006), the affinity (KA) of the FSH-R for the ligand FSH was constant regardless of the varying FSH concentrations during the same ages, and it is likely important that the Sertoli cells remain sensitive to FSH during this time for functional spermatogenesis to be attained (Orth, 1984).

The Sertoli cells begin a period of division during the prepubertal period, and it is known that FSH is an important driver to this proliferation (Orth, 1984; Allan et al.

2004). The length of time the Sertoli cells undergo divisions is meaningful, because in bulls the ability of Sertoli cells to divide ceases upon their final maturation preceding puberty at approximately 25 weeks of age when the formation of the blood-testis barrier

27 is established (Gilula et al., 1976; Sharpe et al., 2003). Additionally, the rate and duration of Sertoli cell division, and the final numbers of Sertoli cells, is of additional pertinence because it is known that mature Sertoli cells host a relatively fixed number of germ cells in bulls and other non-seasonally breeding animals (Blanchard and Johnson, 1997; Leal et al., 2004). Accordingly, the final numbers of Sertoli cells correlates to mature sperm production in bulls (Berndtson et al., 1987).

The mechanisms by which FSH stimulates Sertoli cell division has been studied in vitro and in rodent models. It has been determined that binding of FSH to FSH-R does initiate cAMP-dependent protein kinase (PKA) activity in the testes, which is known to be a driver of division in other cell types (Sasaki et al., 2000). Other paracrine modulators, such as beta-endorphin production within the testes, have also been linked to the prepubertal Sertoli cell proliferation (Orth, 1986). The most recent research has also included and IGF-1 signaling as pertinent regulators of Sertoli cell proliferation

(Petetti et al., 2013). For a comprehensive review detailing the signaling pathways involved with the mentioned molecules and their role in Sertoli cell proliferation, see a review by Escott and colleagues (2014).

It is an interesting—albeit extraneous—fact that the lack of Sertoli cell proliferation in adulthood is not conserved across all species, especially those that are seasonally breeding. Djungarian Hamsters have reduced Sertoli cell numbers when exposed to short day length. It is believed that the decrease in the number of viable

Sertoli cells results from decreased gonadotropin production resulting from shorter daily photoperiod (Meachem et al., 2005). The Sertoli cells of these hamsters may be able to

28 revert to an undifferentiated state during the nonbreeding season and exhibit characteristics common to both undifferentiated and mature Sertoli cells (Tarulli et al.,

2006). Exogenous FSH supplementation can restore breeding-season Sertoli number and cause resumption of spermatogenesis in hamsters that have been experimentally exposed to short day length (Lerchl et al., 1993). Stallions of all ages also experience changes in their numbers of Sertoli cells depending on the season (Johnson et al., 1991). However, in-depth studies of hormonal manipulation such as those performed in the hamster have not been completed in stallions. While horses and certain hamster species represent an exception to the norm, the numbers of Sertoli cells within the testes in bulls exhibit no tendencies to differ with season (Curtis and Amann, 1981; Hochereau-de Reviers et al.,

1987).

As the main cellular constituents necessary for spermatogenesis undergo final maturation, spermatogenesis is able to commence. Overall, it is a combination of endocrine and physiological changes that culminate in the bull's ability to produce and ejaculate viable sperm. The early stages of spermatogenesis begin once spermatogonia occupy the spaces along the basement membrane of the seminiferous tubules, and this can be visualized in testis histological sections beginning between 3 to 4 months of age in the bull (Chandolia et al., 1997b). Spermatocytes can be seen by 6 months of age, and elongated spermatids by 8 months of age (Chandolia et al., 1997b; Barth, 2004). Early ejaculations may contain sperm that have visual abnormalities if puberty has not been finalized since the testicular components necessary for sperm maturation (i.e. epididymis) may not be fully developed (Evans et al., 1995). These abnormalities may include the

29 presence of proximal droplets or compromised motility, but there is a marked decrease in sperm abnormalities just prior to puberty (Evans et al., 1995). The decrease in visible sperm abnormalities coincides with the attainment of puberty as testosterone concentrations in the gonad are increasing and are known to be important for the maturation of sperm in the epididymis (Martig and Almquist, 1969).

The testes experience rapid growth beginning between 25 and 28 weeks of age

(Lunstra et al., 1978), and this is most attributable to the changes in individual functions and numbers of somatic and germ cells (Bagu, 2004). Overall, the composition of the testis is changing as the tubules begin to rapidly develop and fill with sperm cell-types of varying maturity. From 12 to 32 weeks of age the percentage of testis comprised of seminiferous tubules increases from 44 to 81% (Curtis and Amann, 1981), and the testes will continue to grow in size before they will be 90% of their final size by 24 months of age (Coulter, 1986). It has been observed that the initial stages of spermatogenesis take longer in immature than fully matured bulls. For example, in prepubertal bulls it takes 84 days for the first preleptotene spermatocytes to develop to spermatozoon (Curtis and

Amann, 1981). This event takes only 41 days in mature bulls (Courot et al., 1970). It is also believed that the delayed initiation of spermatogenesis in young bulls is likely due to the degeneration of the first three to five cellular generations developing from spermatogonia (Courot et al., 1970).

Lastly, the accessory sex glands begin to develop in synchrony with the , and the vesicular glands and begin to increase in size rapidly after 34 weeks of age (Chandolia et al., 1997b). Select Sires, Inc. begins sperm collection attempts in bulls

30 at 40 weeks of age, or at 28 to 30 cm scrotal circumference for some high genomic bulls.

Few bulls will mount a teaser at this age, but as time progresses most will gain the ability to mount and ejaculate into an artificial vagina by 12 months of age, and it is estimated that 95% of bulls have gained the ability to mount and ejaculate sperm of acceptable quality by 12 – 13 months of age (personal correspondence, Mel DeJarnette). Sperm numbers per collection will also continue to increase during the initial year of collection, and most bulls will reach a peak of semen production around 48 months of age (personal correspondence, Don Monke).

2.2. Overview of Testicular Physiology and Endocrinology in Mature Bulls

After describing the prepubertal development of the HGP-axis and the cells responsible for spermatogenesis in the male, this section aims to provide a brief overview of the physiology and endocrinology of the mature bull. Once bulls attain puberty at an average age of 10 months, several more months (or years) may be needed before sustained, maximal sperm production is achieved.

2.2.1. GnRH, LH, Testosterone, and Leydig Cells

As GnRH is produced in regular pulses from the mature hypothalamus it reaches the gonadotropes of the anterior pituitary. As GnRH ligand binds the GnRH-R, the anterior pituitary responds by producing and releasing FSH and LH (Campbell et al.,

2009). In a postpubertal bull, the generation of LH release from the pituitary in response to GnRH is relatively fast, but not instantaneous. Injection of GnRH in 13 month old

31 bulls causes increased LH concentration within the circulation as early as 20 minutes post-administration (Schanbacher and Echternkamp, 1978).

As previously described, the main target tissue for LH in the reproductive axis is the testicular Leydig cells. Upon binding circulating LH to LH-R, Leydig cells synthesize testosterone using cholesterol-derived precursor molecules and steroidogenic enzymes. If a GnRH pulse is detected/administered, it has been determined that the Leydig cells respond with increased testosterone production in as soon as 20 minutes (Schanbacher and Echternkamp, 1978). A more thorough investigation by D'Occhio and Setchell

(1984) using 3-year old bulls similarly concluded that LH rises approximately 20 minutes after the GnRH pulse, but it was further described that the peak in LH following the

GnRH pulse occurred 80 minutes later. This study utilized increasing dosages of GnRH to stimulate the LH response, and the concentration of LH achieved in the circulation was dose-dependent to the amount to GnRH administered. Therefore, the limiting factor of

LH release in mature bulls may be the hypothalamic production of GnRH, not the pituitary's ability to produce and possibly store adequate LH reserves (D'Occhio and

Setchell, 1984). Increase in testosterone in the general circulation in response to LH is not detected until 40 (Schanbacher and Echternkamp, 1978; D'Occhio and Setchell, 1984) to

60 (Malak and Thibier, 1982) minutes after the GnRH pulse. Interestingly, the concentrations of testosterone following the initial GnRH pulse may not respond in a dose-dependent manner like LH, as D'Occhio and Setchell (1984) observed the same testosterone concentrations regardless of the GnRH dose administered or the LH concentrations achieved. The authors concluded from these findings that the existing

32 steroidogenic activity of the testis determines the testosterone production rather than the concentrations of LH received by the LH-R. Once produced, the presence of testosterone in the systemic circulation regulates further production of GnRH and LH via negative feedback mechanisms (Schanbacher, 1982). Testosterone concentrations will increase steadily once Leydig cells are able to produce the hormone until 6 months of age in bulls, and concentration will remain similar from this age through maturity (Miyamoto et al.,

1989).

In mature males, testosterone's most apparent role may not be regulation through feedback loops, but rather the maintenance of spermatogenesis. As circulating testosterone increases before puberty, and as the beginning stages of spermatogenesis commence, intratesticular testosterone concentrations will increase greatly (Miyamoto et al., 1989). The elevated concentration of testosterone within the testis is due to the binding of testosterone to androgen binding protein (ABP), which is produced by mature

Sertoli cells (Dadoune, and Demoulin, 1993). Ranges in the concentrations of intratesticular testosterone have been determined for several species. In the rat, for example, testosterone reaches between 50 to 100 times greater concentration (70 ng/ml) than the surrounding tissue. These high testicular testosterone concentrations are known to be crucial for the commencement and maintenance of spermatogenesis (Senger, 1997).

Specifically, through histological observation, elevated concentrations of testosterone are necessary to prevent the degradation of spermatocytes as they mature within the seminiferous tubules (Walker and Cheng, 2005). After puberty attainment and the commencement of spermatogenesis, the efficiency of sperm production continues to

33 increase through 1 year of age in bulls until it reaches that of a mature bull, which has been reported as 12 million sperm/gram parenchyma/day (Amann and Almquist, 1976).

2.2.2. FSH, Sertoli Cells, and Spermatogenesis

The other gonadotropin under the influence of GnRH is FSH. In similar fashion to

LH, FSH will be produced by the gonadotropes of the anterior pituitary in response to the ligand GnRH binding the GnRH-R as described in detail in the previous chapter. Again, it is important to note that FSH does not seem to be secreted in a pulsatile manner as is

GnRH and LH, and circulating concentrations of FSH seem to remain relatively stable in males (Dunkel et al., 1992; Genazzani et al., 1994).

Once in the circulation, FSH has target receptors on the Sertoli cells of the testes.

FSH is a necessary factor for spermatogenesis to occur, and FSH is required for the viability of Sertoli cells (Bagu et al., 2006). As briefly touched upon before, a major product of the Sertoli cells necessary for sustaining high intratesticular testosterone concentrations is ABP (Dadoune and Demoulin, 1993). Testosterone production by

Leydig cells will affect the secretion of FSH just as it does LH. The testosterone (and estrogen) production by the adult testis is an important regular of gonadotropin secretion in the higher centers of the brain.

The Sertoli cell population plays a crucial role in the adult as the facilitator of spermatogenesis within the seminiferous tubules by regulating the biochemical surroundings of germ cells (Griswold, 1998). In most species, including the bull, the number of Sertoli cells is unchanging (Gilula et al., 1976; Sharpe et al., 2003). The

34 mature Sertoli cell has a specialized shape and attributes that allow it to foster spermatogenesis. In addition to structural support for germ cells, Sertoli cells also phagocytize germ cells flagged as degenerative as well as the residual bodies of sperm cells at the completion of their maturation (Russell and Griswold, 1993). Differentiated

Sertoli cells are described as having a columnar shape. The nuclei are generally located near the basal surface (perimeter of the tubule), and Sertoli cells have elongated, thin mitochondria (Johnson, 1991).

A crucial component of the Sertoli cells that allows spermatogenesis to occur is the presence of the blood-testis barrier (BTB). The BTB is composed of tight junctions located between neighboring Sertoli cells. The BTB compartmentalizes the seminiferous tubule and differing development stages of germ cells. Specifically, the BTB segregates the early, diploid germ cell types (spermatogonia and early preleptotene spermatocytes) within the basal compartment of the tubule from the further-matured, haploid spermatocytes and spermatids in the adluminal compartment (Setchell and Waites, 1975;

Waites, 1977). This, of course, means that the Sertoli cells must also facilitate the

"crossing" of preleptotene primary spermatocytes across the BTB to gain access to the adluminal compartment (Waites, 1977). The immunological barrier established by the

BTB is crucial because the recognizes haploid germ cell types

(spermatocytes and spermatids) as non-self and would stimulate an autoimmune response within the testes were it not for the BTB (Johnson et al., 2008). The molecular mechanisms responsible for allowing preleptotene primary spermatocytes to cross the

BTB are still under investigation, but it is known that there is a general disassembly and

35 assembly of the tight junctions which must occur (Wong and Cheng, 2005). Namely, two have been implicated in the restructuring of the BTB tight junctions, transforming growth factor-beta3 (TGF-beta3) and -alpha

(TNFalpha). These cytokines elicit mitogen-activated protein kinases within Sertoli cells, allowing for the control of the tight junctions (Wong and Cheng, 2005).

The germ cells within the seminiferous tubules undergo spermatogenesis in waves under the direction of the hormonal milieu established by the Sertoli cells. The spermatogonia are located near the base of the tubule and in the basal portion of the

Sertoli cells. Spermatocytes are in the middle sections, and spermatids are closest to the lumen of the tubule in the branching apical portions of the Sertoli cells. As the spermatids are spermiated, or released, into the lumen of the tubule they are called spermatozoa

(Amann, 1970). To briefly describe some of the maturational steps of spermatogenesis, the spermatogonium nearest the base of the seminiferous tubule undergo mitosis to either replenish the population, or a committed spermatogonia is formed that will undergo further differentiation. It is the committed spermatogonia that will continue through spermatogenesis to form the spermatocytes (Johnson et al., 2008). Much of the cytoplasm of the developing sperm will be resorbed by the Sertoli cell to decrease the sperm cytoplasmic volume, and at the end of maturation there will be a small amount of cytoplasm remaining called the cytoplasmic droplet which is attached to the sperm where the head and midpiece connect (Johnson et al., 1978).

36

2.2.3. Inhibin, Activin, and Follistatin

The mature Sertoli cells will produce inhibin upon binding FSH to FSH-R, and circulating concentrations of inhibin will relate inversely to FSH concentration in the mature animal (de Kretser and Robertson, 1989). It was first observed that inhibin negatively feeds back on FSH production during experiments where bovine follicular fluid was injected into bulls between 1 to 2 years of age. Inhibin was isolated as the molecule in the follicular fluid that caused decreases in FSH in these bulls (McGowan et al., 1988). In mature males, the concentration of circulating inhibin correlates to the number of Sertoli cells within the testes, and it has therefore been proposed to utilize inhibin concentrations as an estimation of the number of Sertoli cells when invasive means to collect data are infeasible (Sharpe et al., 1999). Furthermore, Johnson et al.

(1984) suggested further predictive power of using inhibin to approximate sperm production because the number of Sertoli cells correlates to mature sperm output in bulls, and the number of Sertoli cells dictates systemic inhibin concentrations. FSH is regarded as the main stimulatory protein for inhibin secretion. The up-regulation of the alpha- subunit of inhibin in Sertoli cells is likely the main mode of action (Le Gac and de

Kretser, 1982).

The role of activins in the mature males is less apparent than the importance of activin in the developing male. The source of activins in the mature male has been identified mainly within the testes, including within the Sertoli cells (de Winter et al.,

1993; de Kretser et al, 2002), as well as Leydig and peritubular myoid cells (de Winter et al., 1994; Buzzard et al., 2003). Activin has been described as necessary for the

37 maintenance of spermatogenesis. Activin may facilitate the act of spermatogenesis by acting in an autocrine or paracrine manner and having stimulatory effects of Sertoli cell reaggregation, or maintenance of the structure of the seminiferous tubules established by

Sertoli cells (Mather et al., 1990). Additionally, activin is necessary for germ cell proliferation (Boitani et al., 1995; Mather et al., 1990).

Follistatins' role in the adult male is a continuation of function seen in the immature male. By binding activin with high affinity, follistatin is able to act in a local, paracrine fashion against the actions of activin, including germ cell proliferation (Mather et al., 1993). A better understanding of the source of follistatin has been achieved in mature males because of its expression patterns in the ejaculate. The concentrations of follistatin in ram seminal plasma is 100 times greater than in the circulation (Tilbrook et al., 1996). It is known that follistatin is produced within the testes (Guo et al., 1998), but it appears that the follistatin produced within the testes does not contribute greatly to the circulating concentrations (Tilbrook et al., 1996). However, vasectomy in men does not result in a significant decrease in follistatin in the seminal plasma, indicating a large portion of the follistatin in the body is produced elsewhere. An alternate source of follistatin has been identified in the prostate gland (Thomas et al., 1997). Regardless of the source of follistatin, the high concentrations present in the ejaculate indicate that it is localized within the testes. The close proximity of follistatin to the major sources of activin production is not a surprise, as the mechanisms of follistatin are widely believed to be mainly paracrine in action. Therefore, follistatin is probably acting as a controller of activin activity within the testes.

38

2.3. Previous Methods to Hasten Puberty in Bulls

Although a main focus of this dissertation research is the economics underlying puberty attainment in bulls destined for use within the AI industry, this could be considered a relatively new and niche concern when the interests of the cattle production industries are weighed as a whole. The time and money spent growing calves to maturity, regardless of breed, represents a loss to industry, as immature animals are not generating income until they are ready for milk or meat production. In dairy, replacement heifers must be grown to appropriate body sizes before breeding, and transitioning to the milking herd generally occurs at around 2 years of age. In the beef industry, pre-slaughter animals must first be fed or grazed by the cow-calf producer, and then intensively managed by a grower/finisher before reaching appropriate slaughter weights. AI companies rearing young bulls face the same challenges as the beef and dairy producers. Immature, growing bulls fail to provide income until they are able to produce semen as saleable product. To minimize monetary losses and to accelerate profit attainment, cattle producers have started to explore ways to accelerate growth. For example, one of the many tools used in the beef industry includes the use of FDA-approved estradiol-based implants in beef cattle. Most of these implants successfully increase the rate of weight gain and improve feed efficiency (Bagley et al., 1989).

Since the implementation of genomic testing in the cattle industries, producers utilizing AI demand semen from genetically superior bulls as early as they become available. Companies rearing these AI sires are seeking to minimize non-productive periods of a bull's life and maximize the rate of genetic progress by developing bulls on

39 accelerated growth programs and by beginning collections as soon as possible. The following sections provide summaries of some of the experimental methods performed to attempt to manipulate and better understand prepubertal endocrinology of the bull. The number of studies performed in bull calves is limited, so applicable studies in rodents, humans, and other livestock species will be included when pertinent. Also, significant focus will be given to FSH as a target for manipulations based on the similarity to this dissertation's research. Many of the studies in bulls have similar experimental designs between different research groups. Therefore, some studies have been omitted from this review to prevent unnecessary repetitiveness. In general, experiments have sought to manipulate prepubertal endocrinology either through treatments that directly affect hormones via exogenous supplementation or immunizations, or through dietary treatments, which have been known to affect the endocrinology of growing animals.

2.3.1. Direct Endocrine Manipulations to Advance Puberty

Many of the studies determining the prepubertal, peripubertal, and mature endocrinology of the bull were conducted in the 1960s and 1970s. The first attempts seeking to alter the physiology and age at puberty by applying prepubertal treatments of exogenous gonadotropins began in the mid-1970s. However, these attempts utilizing

GnRH were largely unsuccessful at altering the age at puberty (Mongkunpunya et al.,

1975; Haynes et al.,1977; Schanbacher et al., 1982). It was retrospectively surmised that these attempts might have failed due to the timing of treatment being too late (occurring

40 after the prepubertal events had already initiated), the durations of treatments being too short, or the frequency of treatments being too low (Amann, 1983).

Later attempts became more successful, if not only in significantly altering the endocrinology and physiology of animals prior to puberty. Madgwick and colleagues

(2008) treated prepubertal bulls with 120 ng/kg GnRH twice daily from 4 to 8 weeks of age. These bulls experienced an LH pulse after each injection, and the bulls receiving

GnRH had more rapid testicular growth from 22 to 44 weeks of age and a hastening of puberty by 4 weeks relative to untreated bulls in terms of their ability to ejaculate 50 million sperm with 10% motility. The reduction in the age at puberty in this experiment was attributed directly to the advancement of the transient increase in LH that is normally associated as occurring at 6 weeks of age (Madgwick et al., 2008). Chandolia et al.

(1997c) treated bulls with 200 ng GnRH intravenously every 2 hours from 4 to 6 weeks of age and reported that increasing the systemic concentrations of GnRH attributed to increases in LH and FSH concentrations. Although unmeasured, it was concluded that these endocrine changes would culminate into enhanced testicular development in terms of increased numbers of Sertoli and germ cells (Chandolia, 1997c). Studies have been performed in rats providing insight into the effect prepubertal GnRH concentrations have on Sertoli cell numbers. When GnRH antagonists are administered to rats neonatally

(when development is similar to the postnatal, prepubertal bull), FSH concentrations are decreased and there is a 45 to 52% decrease in the number of Sertoli cells and a 48% decrease in testis mass in adulthood compared to untreated rats (Sharpe et al., 1999;

Atanassova et al., 1999).

41

Studies administering FSH prepubertally indicate very similar hypotheses and objectives as those expressed in this dissertation, mainly that increasing FSH prepubertally will cause increased proliferation of Sertoli cells (Orth, 1984). In a main experiment influencing the methods of this dissertation's research, Bagu et al. (2004) treated bull calves from 4 to 8 weeks of age with 10 mg FSH every other day. The treatments successfully elevated FSH concentrations post-administration, hastened puberty as determined by a 28 cm scrotal circumference (39.3 vs 44.8 weeks of age), and increased the numbers of Sertoli cells, elongated spermatids, and spermatocytes when measured at 56 weeks of age. Meyers et al. (1983) administered 5 mg FSH twice daily for

10 days in 4 month old bulls. The study indicated increased LH concentrations from FSH administration, and testicular weight was increased by 38%. However, the increase in LH concurrent with the administration of FSH causes questioning of the purity of the FSH and (or) the specificity of the LH assay, as the molecules are similar in structure and FSH administration has not been reported to be associated with LH production.

Rodents and humans have also been utilized as experimental models for exploring the effects of prepubertal FSH manipulation. Boys diagnosed with hypogonadotropic do not produce FSH. The recommended treatment regimen established to allow them to reach puberty includes administration of recombinant FSH (r-hFSH;

Gonal-f, Serono) every 3 weeks beginning between 2 months to 2.8 years preceding the expected age at puberty. Once an appropriate age to undergo puberty is reached, puberty is induced by human chorionic gonadotropin (hCG; Profasi HP) administration several times per week, and the administration of hCG can continue after puberty to maintain

42 fertility (Delemarre-Van de Waal, 1993; Raivio et al., 2007). In rats, treatment with recombinant FSH during the neonatal period has marked positive effects on testis development, with a treated animal having 49% percent more Sertoli cells in adulthood and a 24% increase in testicular mass (Meachem et al., 1996).

The induction of hyper- and hypothyroidism has provided insight into puberty attainment, especially in rodents and boars. The and thyroxine (T3 and T4) are important for prenatal gonadal development along with a host of other growth-related processes. Regarding puberty, inactivating the thyroid (induced hypothyroidism) in neonatal mice results in an extended time of Sertoli cell proliferation, enlargement of the testes, and an increase in spermatogenic capacity (Joyce et al., 1993).

Elaborating on this experiment's results, the window of Sertoli cell proliferation was extended from 15 to 25 days of age, which led to a subsequent increase in the numbers of

Sertoli cells at maturity (Joyce et al., 1993). In the rat, the induction of neonatal hypothyroidism causes increases of 82 to 157% in the final number of Sertoli cells (Hess et al., 1993). The negative effects of hyperthyroidism have also been reported in rat models. Induced neonatal hyperthyroidism can cause a 50% reduction in the final numbers of Sertoli cells (van Haaster et al., 1993). Furthermore, T3's effect on androgen receptor expression on Sertoli cells has been examined. In in vitro Sertoli cell cultures, addition of T3 caused significant increases in androgen-receptor mRNA expression. The authors hypothesized that in vivo treatment with T3 would simulate hyperthyroidism and could hasten puberty, as the expression of androgen receptors in Sertoli cells is a major

43 marker of puberty in rats and would signify the time when Sertoli cells lose their proliferative capabilities (Arambepola et al., 1998).

In the boar, transiently induced hyperthyroidism during the neonatal period causes a decline in the proliferation of Sertoli cells, most likely because of a hastened age at puberty (McCoard et al., 2003). Boars may be an imperfect model of Sertoli cell proliferation, however, given that FSH concentrations are not correlated to Sertoli cell proliferation prenatally or postnatally (McCoard et al., 2003), and FSH concentrations during the prepubertal period does not influence mature testis size like in other animals

(Ford et al., 2001). Considering other livestock species, Fallah-Rad et al. (2001) found that inducing a hyperthyroid state in ram lambs from 6 to 12 weeks of age caused decreased concentrations of FSH during treatment, increased seminiferous tubule diameter at 36 weeks of age, and hastened puberty attainment. This seems contrary to results seen in other species where hyperthyroidism generally causes negative impacts on reproductive status. Reports linking T3 and T4 to puberty attainment in larger ruminants is scarce, especially in bulls. In Buffalo heifers, reports indicate that T3 gradually increases postnatally to reach peak concentrations prior to puberty, when high concentration of T3 are necessary for weight gain and enhanced protein synthesis (Ingole et al., 2012). No differences have been reported in T3 or T4 prior to puberty in buffalo bull calves (Sharma et al., 1985).

Lastly in this hormonal manipulation review section, immunizations have been utilized in several species to study the immunological removal of various reproductive hormones. In rats, McLachlan et al. (1995) effectively shut down testicular function in

44 adult rats utilizing GnRH immunizations to study the role of the gonadotropins and downstream hormones in spermatogenesis. In these GnRH immunized males, FSH, LH, testosterone, and inhibin production was minimized, and these "gonadotropin-deficient" rats experienced severe spermatogenic regression by 12 weeks after GnRH immunization. Next, recombinant FSH was administered for 7, 14, or 21 days after the cessation of spermatogenesis. After 7 days of FSH treatment, testis weight had increased by 43% and a resumption of spermatogenesis was observed. This study serves to demonstrate the applicability of immunizations to study gonadotropin function in the gonads. Another study in rats immunized against FSH at various early postnatal ages and examined testis histology soon after treatment. At all ages of FSH immunization, examination of the testes 2 to 4 days later revealed decreased Sertoli cell proliferation and increased germ cell . Together, these two studies in rats showcase FSH's necessary role in Sertoli cell proliferation and successful spermatogenesis (Meachem et al., 2005).

Immunological applications have been utilized in bull studies to a surprising extent, albeit consistently by a few different research groups. The main target molecule under manipulation immunologically in bulls is inhibin. Kaneko et al. (1993) first verified that inhibin immunizations at 6 months of age increased FSH concentrations for up to 168 hours post-injection. Inhibin immunizations did not affect LH or testosterone secretion. Later, Kaneko et al. (2001) immunoneutralized inhibin in four groups of bulls at either 7, 21, 60, or 120 days of age. FSH increased in plasma at all ages examined, and the magnitude of the FSH rise was greater as bulls aged. LH was unaffected by inhibin

45 immunization. These studies helped clarify that the inhibin feedback from Sertoli cells regulates the pituitary's ability to secrete FSH. Other studies validated that inhibin immunizations could have long-term effects on the testes. Martin et al. (1991) immunized bulls against inhibin at 14 weeks of age followed by several boosters, and at 34 weeks of age bulls had greater serum concentration of FSH and testosterone. In a subsequent manuscript from this experiment, it was determined that these immunized bulls had increased numbers of elongated spermatids per gram parenchyma later in life (Lunstra et al., 1993). Lastly, Bame et al. (1999) immunized against inhibin at 2 months of age with boosters given until 13 months of age. While no histology was performed, it was noted that immunized bulls had increased scrotal circumference and serum FSH concentrations throughout the study. The use of inhibin immunizations represents an alternative method to alter FSH concentrations. However, individual animal response within experiments seems variable (Bame et al., 1999), and in the context of designing the experiments presented in this dissertation, more direct methods to increase FSH were targeted by using the direct administration of exogenous FSH.

2.3.2. Dietary Manipulations to Advance Puberty

In a production setting, the times during which a calf's nutrition could be altered depends upon the breed and purpose. For example, the producer plays a larger role in supplying nutrition to the postnatal dairy replacement heifer than the beef calf since dairy heifers are generally removed from their mothers and fed milk replacer. Still, it is known

46 that growth in beef calves suckling their dams plays a large role in their sexual maturation. Increased pre-weaning body weight gains results in earlier attainment of puberty in beef heifers (Wiltbank et al., 1966). Furthermore, weaning weights of beef heifers correlates to age at puberty (Arije and Wiltbank, 1971; Greer et al., 1983; Roberts et al., 2007). In a series of experiments by Gasser et al. (2006a, b, c, d), early weaning and the feeding of high energy diets post-weaning caused beef heifers to precociously reach puberty 85 days earlier. Early puberty attainment has large implications for beef cow productivity considering that advancing puberty can allow for additional estrous cycles before the onset of the first breeding season, and this results in a greater probability for early conception and optimal lifetime productivity. Mechanistically, increased nutrition causes hastened puberty in heifers by accelerating the start of LH pulsatility (Gasser et al., 2006a, b, c, d). In accordance with this theory, fasting prepubertal heifers for 48 hours reduces the mean frequency of LH pulses (Amstalden et al., 2000).

Much of the literature linking nutrition and puberty in dairy focuses on the rearing of heifers to reach acceptable weights to perform breeding at a time such that she has her first calf at approximately two years of age. It is known that intensive nutrition before weaning (before approximately 40 days of age) can result in increased growth, earlier attainment of puberty, and therefore younger age at first calving when age at breeding is based on weight. Importantly, in this study enhanced early life nutrition did not negatively increase calving difficulties or decrease milk production later in life (Davis

Rincker et al., 2011). Other studies also report decreased age at puberty and first calving

47 using intensive management, but also a decrease in first-lactation milk yield (Peri et al.,

1993; Hoffman et al., 1996). A significant amount of research has been performed linking enhanced growth in dairy heifers and the undesirable effect of udder fat deposition, which has been one reason intensive management may have negative correlations to milk production later in life as reported by some studies (Lohakare et al., 2012).

In bulls, limiting prepubertal nutrition delays puberty (Flipse and Almquist, 1961;

Pruitt et al., 1986; Dance et al., 2015). Other studies which also limited nutrition in bulls reported smaller testes (VanDemark and Mauger, 1964), decreased SC (Pruitt et al.,

1986), lower intratesticular testosterone concentrations (Mann et al., 1967), and lower sperm numbers in ejaculates of lower volume (VanDemark et al., 1964; Dance et al.,

2015). Early studies utilizing high-energy diets revealed little about their effect on the

HGP-axis. In Brahman bulls, high-energy diets prior to puberty led to earlier puberty, and this was manifested most in an increased SC. However, there were no apparent changes in the secretion of GnRH of LH in bulls fed high-energy diets (Nolan et al., 1990). More recently, several experiments have provided insight as to how high-energy diets in

Holstein bulls positively affect the HPG-axis. Brito et al. (2007b) fed bulls high energy diets beginning at 10 weeks of age and reported larger testes, greater scrotal circumference, and greater daily sperm production compared to bulls fed control rations.

Likewise, Dance et al. (2015) and Harstine et al. (2015) performed similar experiments utilizing high-energy diets in Holstein bulls. Both studies cited that high-energy diets cause an earlier increase in pulsatility of LH. This translates to increased systemic concentrations of LH and testosterone during the pubertal period and an increased scrotal

48 circumference. The hastening of the initial LH pulsatility in these experiments seems mechanistically very similar to the model of precocious puberty in beef heifers developed by Gasser.

The large number of studies seeking to positively affect puberty attainment in cattle showcases the malleability of the prepubertal . This dissertation’s research follows several of the same principles as these experiments by seeking to positively manipulate the prepubertal endocrinology by introducing an exogenous gonadotropin. It was examined how these administered treatments affected endocrinology, testicular development, puberty attainment, and mature sperm production.

Fortunately, these projects do seek to provide answers to unknown questions, such as how exogenous FSH administration mechanistically affects the cells of the testes and how any positive effects of these treatments could be utilized in an industry setting.

2.4. The Selection and Use of Bulls in the Artificial Insemination Industry

The history of the use of AI spans many decades and across several countries, and the implementation of AI has undoubtedly been one of the greatest advancements to cattle production genetics and the cattle industries in general. Many scientists and industry personnel argue that the recent incorporation of genomic evaluations might be considered one of the largest, comprehensive changes to the cattle breeding industries since the widespread use of AI began in the 1950s.

Sperm cells were first visualized by Leeuwenhoek as early as 1678

(Leeuwenhoek, 1683), and the first published reports of AI (in dogs) did not surface until

49 one hundred years later (Spallanzani, 1784). Research utilizing AI in cattle grew in prevalence in the 1900s. Milovanov pioneered the use of an artificial vagina (AV) to collect bull semen in 1938, and the first large-scale cattle AI experiments were performed by Sørensen in 1936. Coincidentally, Sørensen was also said to have established the first bull stud cooperative in conjunction with this research endeavor (Sørensen, 1940).

Growth of AI in America began early in the 1940s and relied purely on the use of fresh semen. The first American AI co-ops began in 1938 in both New Jersey and New York, and the New York Artificial Breeders Cooperative still operates today under the Genex,

Inc. name (Foote, 2001). After the ability to cryopreserve sperm was revolutionized using glycerol in the 1950s, many of the AI cooperatives consolidated into the few large companies present today (Polge et al., 1949; O'Dell and Almquist, 1957; Polge, 1968).

For thorough historical narratives of the cattle AI industry, see reviews by Foote (2001) and Funk (2006). Both of these reviews give excellent highlights of the history of the AI industries and the recent changes caused by globalization.

With the sequencing of the bos taurus genome in 2009 (The Bovine Genome

Sequencing and Analysis Consortium, 2009) and the genetic evaluations that followed, fundamental shifts have begun in the strategies of cattle breeding. Undoubtedly, overviews of the industry written in the coming years will seek to assess and describe the major changes caused by the inclusion of genomics in the industry in the 2000s.

Although methods of sire selection are changing, product differentiation between AI companies is minimal, and each remains in business mainly by selling desirable sire genetics that produce profitable results for producers. As surmised by DeJarnette and

50 colleagues (2004), "it is not the responsibility of the AI industry to dictate to the producer what type of cattle they should be breeding nor to narrowly define what types of genetic packages will be offered for sale."

2.4.1. Historical Perspective on Bull Acquisition and Use by AI Companies

Determining Bull Genetic Worth and Profitability in the Traditional System

From the beginning of the commercial AI industry until the last decade, the genetic worth of a bull was determined by his daughters' performance, better known as progeny testing. Progeny testing has served as a standard for comparing genetic merit of sires, and although the current emphasis on genomics may seem to overshadow progeny testing data based on "newness" and great consumer adoption, the necessity for progeny testing and its continued role in verifying genomic evaluations cannot be understated. To briefly summarize progeny testing, a potential AI sire will have semen collected and distributed to designated "test" herds as soon as possible, generally soon after one year of age. After sufficient numbers of cows are bred, the sire and his contemporaries usually begin a "sire-in-waiting" period until his daughters can be evaluated upon entering the milking herd. Once the bull's daughters are milking, an evaluation of their type and production can be used to confirm and increase the accuracy of the bull's previously determined predicted transmitting abilities (PTAs). In a traditional system of sire selection, many producers choose to wait for these progeny tested, "proven" sires to use in their breeding programs.

51

The use of progeny testing has been the most reliable system designed for sire selection since the widespread use of AI. Although there are drawbacks to waiting until sires become proven, the progeny testing system must be given credit for improving the production and efficiency of cattle as highlighted in this dissertation's Introduction.

Indeed, selection intensity amongst bulls has increased over time by using progeny testing. For example, the percentage of progeny tested bulls actually returning to active service declined from the 1960s to the 1990s (Norman et al., 2003). In similar fashion to genomics, the utilization of progeny testing has been edited by the AI companies to best impact the effectiveness and the rate of genetic improvement it provides. For example, AI companies have made measurable strides in sampling bulls at consistently young ages.

Since 1960, the mean age for Holstein bulls at semen release for sampling by the industry at 16 months of age did not change, but the standard deviation in age decreased from ± 4 months to ± 2.4 months by the 1990s. Additionally, the mean age of bulls at the time their progeny-test daughters calved decreased from 56 ± 6 to 52 ± 3 months from 1960 to 1990

(Norman et al., 2003). More recent reports estimate the average age for bulls' first proofs as 55 months (Amann and DeJarnette, 2012). Regardless, an inefficiency in the traditional system lies in the fact that sires must wait until daughters are producing before their genetic merit is confirmed. This represents a huge investment for AI companies. In

2006 ABS Global claimed that the total cost of progeny testing a Holstein sire was estimated to be approximately $30,000. At this time of this claim, approximately 12% of bulls were utilized in a proven bull line-up, meaning that nearly $250,000 was invested per graduate after not utilizing the less desirable bulls (Funk, 2006).

52

In contrast to the often fast-paced transitions in sire line-ups today, the reliability of the "proofs" generated from progeny testing were apparent as some sires attain celebrity-like status as indicated by incredible sales of their semen over many years. To put sire popularity and longevity in context, approximately 50 Holsteins have succeeded in producing and selling over 1 million units of semen as of October 2014 (Hurtgen,

2014). The youngest sire to exceed 1 million units in sales did so in 2014 at less than 9 years of age. Additionally, it is believed seven Holstein sires have sold over 1.5 million straws, and this includes one bull that has successfully sold more than 2 million doses of semen (Geiger, 2012). This latter bull, Jenny-Lou Mrshl Toystory-ET, reportedly produced and sold 2.415 million doses of semen before he was unable to produce at 13 years of age according to his owners, Genex.

Acquisition, Rearing, and Use of Bulls in the Traditional System

When a bull was selected to enter a traditional sire sampling program in the past, the AI company would purchase the bull from a purebred producer deemed to have elite genetics. Once a breeder was identified, a contract was formed to ensure a particular mating between a chosen dam and predetermined sire. Contracts not only designate a mating, but often contain leasing agreements instead of outright acquisition, and often dictate the order in which particular stud companies can choose calves if multiple are born. To fulfill a contract mating, multiple ovulations and embryo transfer (MOET) is often used. Attributes of MOET include increasing the reproductive rate, increasing genetic selectivity, and decreasing the generation interval of valuable cattle (Smith, 1988;

53

Teepker and Keller, 1989). In a contract mating, the dam of the bull calf will be superovulated using a regimented hormone treatment protocol. The time of and estrus is closely calculated or watched, and the cow is artificially inseminated to the sire of choice at the appropriate time. Collection of the embryos has been thoroughly researched and is most successfully performed when the embryos are 6-8 days-old

(Seidel, 1981). Next, the embryos are evaluated or 'graded,' and those deemed viable will be introduced into one of the uterine horns of a 'recipient' cow that is in the appropriate stage of her estrous cycle or that has had her estrous cycle manipulated such that she will be able to sustain a pregnancy. A second way to accomplish the same goal is to collect oocytes directly from a superovulated cow, perform in vitro fertilization (IVF), allow embryos to mature in vitro for a short period, and then implant the embryos into recipient cows. This process is called in vitro embryo production (IVP) and is a feasible alternative to the aforementioned MOET system (Hasler et al., 1995). The AI center obviously seeks a male from the resultant calves in either scenario, and the more to choose from, the better. After these processes, the AI center will allow the bulls to grow for some months to allow them the possibility of physically differentiating themselves from one another.

Once the time to select an individual arises, the stud company will consider overall physical soundness, health, growth, and testis size when making their decision. Only after careful qualitative and quantitative assessment will the company be able to select the prime specimen out of several 'littermates' (Amann and DeJarnette, 2012).

Traditionally, the age at which bulls are moved to the AI company varied from 4 to 16 months of age. However, the historical average for dairy breeds tends to be before

54 the animal is 12 months of age, and the average for beef breeds generally occurs when the animal is more than 12 months of age (Monke, 2006). Bulls from different farms do not enter the stud rearing-facilities as a uniform group due to differences in the manner in which they have been reared. This variation was an accepted fact within the AI industry for many years, and this problem is exacerbated when the use of a contract in is effect since bulls purchased in this manner are 'pre-paid,' and there is not large incentive to provide specialized care prior to transport. As a result, it is not uncommon for contracted bulls to be maintained on pasture or with diets that meet maintenance needs, but which do not sustain optimal growth before they are moved to collection facilities. The variability in growth resulting from acquiring bulls at approximately 12 months of age is contrary to literature stating a steady rate of growth from birth to 12 months of age is optimal.

According to protocols on raising dairy bulls before semen production, large breed dairy bulls should weigh 400-450 pounds at 6 months of age, and they should preferably weight 800 to 900 pounds by 12 months of age (Monke, 2002). A gradual, consistent rate of growth, rather than rapid or fluctuating growth, is most desirable. Overall, it is recommended that dairy bulls grow approximately 2 to 2.5 pounds per day during their first year, and that the daily dry matter intake (DMI) be about 3% of body weight during this time in order to ensure that a bull does not undergo delayed puberty (Monke, 2002).

Under-nutrition early in life can postpone sexual maturation. Malnutrition in newborn to 12 month old bulls can lead to delayed or impaired androgenic function, which may have later impacts on testicular development and sperm production (Cupps,

1991). Conversely, over-nutrition can have negative effects preceding puberty as well.

55

Rapid growth from 6 to 12 months of age may result in excessive fat deposition in the body and neck of the scrotum that can negatively affect sperm production at later times in the bull's life (Monke, 2006). The assumption that all bulls entering the semen collection facilities are vastly different in their development from one another may be an exaggeration, but extremes in both directions can be observed, especially before AI companies were fully aware of the abovementioned implications that nutritional differences can have on the future productive life of bulls. Today, AI companies manage nutrition of their bulls more carefully because they are owned from a younger age, as will be described later in this review.

Once arriving at the AI center, a bull is subjected to a brief quarantine period, but the commencement of semen collection will occur as soon as possible to accomplish the initial steps of progeny testing. It typically takes several attempts to train a young bull to interact with a teaser animal and to finally ejaculate into an AV. It is also assumed that collections commence under ideal circumstances when the young bull does not have phimosis, subnormal testicular or epididymal development, a delay in puberty, or combinations of these factors which will delay or hinder semen production. After the waiting period associated with the progeny testing process, bulls coming back onto collection as "graduates" in the active lineup will be collected at a maximum of every other day. As semen production nears mature quantities, bulls on frequent collection schedules can produce upwards of 40 billion sperm per week, equaling 2 trillion sperm cells per year. When packaged at 15 million total sperm per straw, this translates into well over 100,000 straws per year (Funk, 2006). Even when considering conservative

56 approximations of sperm numbers per straw and conception rates, popular bulls can sire tens of thousands of calves annually. Toystory, the bull previously mentioned, was estimated to have sired over 500,000 calves, according to Genex.

The logistics underlying progeny testing have been fine-tuned since the 1950s, and the traditional system of sire selection served as a main working business model for the AI companies. However, the rate of genetic improvement and the decreasing of the generation interval may have approached its peak using progeny testing. The emergence of genomics has introduced a useful new tool to add accuracy and actual predictive abilities regarding the genetic merit of young sires, and it has been recognized by both AI companies and producers that changes to cattle breeding practices are imminent.

2.4.2. Utilization of AI Sires in the Genomics Era

The use of genomics has changed the industry by adding powerful tools to better select bulls for AI programs. Interestingly, the acceptance of genomically tested, unproven sires by the industry is greater than some AI companies predicted. A 2010 issue of Nature Biotechnology featured an article highlighting the incorporation of genomic evaluations into the dairy industry. In the article, Roy Wilson, Genex's technology development manager, said that in a 2008 planning meeting Genex expected roughly

15% of the company's business would switch to the use of unproven sires in the following years. They drastically underestimated, considering that 40 to 45% of Genex's

December 2009 sales were from sires with no milking daughters (Strauss, 2010). This occurrence seems to be similarly reported amongst other AI companies, as well. Today, it

57 is estimated that more than half of all AI matings in the United States are now made to genomically tested young sires (Hutchinson, et al., 2014).

Even though genomics have increased in popularity tremendously, proofs from progeny testing provide higher reliability of a sire's transmitting abilities, and this information is integrated with genomic data, often post hoc, to enhance genomic predictions. Some of the traits for economic indexes in Holsteins, such as net merit, have as much as 70% reliability in genomic proofs. Conversely, some traits still have relatively low reliability as predicted by genomic proofs. Genomic predictions still pale in comparison to proofs of progeny-tested bulls, where reliabilities in many traits approach and surpass 85 to 90% (Van Raden et al., 2009; Cassel, 2010). The companies which market cattle genetics recognize that a "paradigm shift" is occurring (Amann and

DeJarnette, 2012), and the genomic era of cattle selection shows no signs of slowing in its implementation and utility as more breeds are included in genomic analyses and the current breeds have additional animal records added to their expanding databases.

An Overview of Genomics and its Current Use in the AI Industries

A brief overview of the how genomics were developed and how this technology is translated into commercial settings is useful. As scientists learned more about genetics, it was realized that it is possible to predict the genetic value of both plants and animals using genome-wide maps that mark useful traits (Humblot et al., 2010). In cattle, the combined efforts of over 300 scientists in 25 countries resulted in the mapping of the bos taurus genome sequenced from a Hereford cow. The bovine genome contains

58 approximately 22,000 genes, including 14,000 that are common amongst all mammalian species (The Bovine Genome Sequencing and Analysis Consortium, 2009). The genetics of bos taurus cattle breeds encompass surprisingly little genetic variety. Genetic variation is induced at the time of fertilization as large chromosomal regions randomly assort from paternal and maternal genomes, and this random assortment makes offspring unique from even genetic siblings. Besides these large, endogenous variations between genes of individuals, even small, single base pair differences can effect phenotypes of the animal.

The difference in one base pair between genes serving the same function is referred to as a single nucleotide polymorphism (SNP). SNPs occur randomly throughout the genome, and on average there is a SNP present every 700 base pairs in bos taurus cattle (The

Bovine HapMap Consortium, 2009). Inferences can be made based on large numbers of individuals that a difference in a particular SNP is associated with a certain phenotype.

SNPs can be considered a marker for an allele, or variation of a gene, and by comparing enough resultant phenotypes it can be accurately predicted whether a specific gene is present or not in an animal. In essence, SNPs are the genetic markers used in the cattle industries to detect differences between individuals (Seidel, 2010). The mathematical calculations used to compare individuals via SNP comparisons are enormous, but advances in statistical modeling and the growing number of data points from continued sampling makes comparison possible. The development of the statistical packages and programs required to create regressions for predictions has flourished, and many researchers are looking for more accurate ways to predict phenotypes based on genomic data such through the use of Bayesian methods (Perez and de los Campos, 2014).

59

Additionally, after the success of genomic prediction in Holsteins, similar programs were initially applied in other species such as beef cattle (Pollak et al., 2012), sheep

(Duchemin et al., 2012), pigs (Ibanez-Escriche et al., 2014), and poultry (Preisinger,

2012).

Many sources identified SNPs within the bovine genome soon after it was published, with early investigators publishing well over 2 million relevant SNPs

(Margulies et al., 2007; Matukumalli et al., 2009; The Bovine Genome Sequence and

Analysis Consortium, 2009; The Bovine HapMap Consortium, 2009). These SNPs were utilized by commercial entities in order to create genotyping microarray technology. The number of SNPs required to be analyzed to provide biological useful predictions is still under investigation. For several years, the recommended "SNP chip" for use in cattle was the Illumina BovineSNP50 BeadChip (Illumina, San Diego, CA). It was deemed to provide the most accurate predictions for several phenotypic performance parameters in dairy cattle while reading approximately 50,000 pre-selected SNPs, hence the "50" in its title (VanRaden et al., 2009). At the time, over 5300 Holstein bulls and their progeny data were used to develop the predictive regressions of the BovineSNP50 BeadChip

(VanRaden et al., 2009). Today, Holstein Association USA, the premier breed organization for purebred Holstein owners, advertises and recommends either a 9K, 77K, and 800K SNP tests to producers. The 9K SNP test is cited as identifying common genetic diseases such as Deficiency of Uridine Monophosphate Synthase (DUMPS), giving basic reproductive information such as freemartin status, identifying whether animals carry the identified lethal haplotypes (VanRaden et al., 2012), and will provide

60 an official GTPI value (signifies genomic information was used in the calculation of the predicted transmitting abilities (PTAs)). The 77K chip indicates better reliability over both parent averages than the 9K chip, and the 800K chip advertises more reliability yet.

Overall, the projected reliabilities for the 9K to 800K SNP tests are between 72 and 74% as indicated by the Holstein Association USA, and this represents significant increases in the reliability calculated from solely parent averages, which is cited as only 42%

(Holstein Association USA, 2016). It is also important to note that the centralized entity analyzing, compiling, and reporting genomic SNP data has changed recently. Initially, the United States Department of Agriculture –Agricultural Research Service –Animal

Genomics and Improvement Laboratory (USDA-ARS-AGIL) was critical in developing and controlling cattle genome information. However, there has been a transition of these services to the Council on Dairy Cattle Breeding (CDCB). A non-funded cooperative agreement (NFCA) was initiated in 2013 between USDA and CDCB that committed to have the transition complete by December of 2015 (CDCB, 2015).

Today, nearly all bulls with the potential to be used in the AI industry are genotyped at a young age. Generally, the AI companies rely on external growers to house bull dams and infant bulls prior to weaning ages. Often, the early attainment of genomic values can aid in culling decisions before infantile bulls are even brought to rearing facilities owned by the AI company. While less post-pubertal bulls may be housed during sire-in-waiting periods, AI companies are now having to house large quantities of bulls at young ages. Accordingly, large changes in the facilities and demographics of bulls have occurred since the 2008 inclusion of genomics in the industry.

61

Changing Demographics of Bulls in AI and Operational Changes to AI Centers

The ability to more accurately predict genetic merit of bulls by using genomic evaluations has substantially impacted management and acquisition of sires. Today, AI companies will rarely allow producers or growers to raise bulls for any length of time prior to movement to an AI company's rearing facilities. Bulls can be selected as targets for potential active-lineups at birth or even before by genotyping embryos (Humblot,

2010). Genotyping embryos has become more common, as it hastens the rate of genetic progress (Seidel, 2010), and pregnancy and implantation rates are not negatively affected by embryonic genotyping (Ponsart et al., 2008; Humblot, 2010). In the future, phenotype of bulls may simply be used to remove physically undesirable animals from a pool of potential AI stud candidates.

The consumer demand for sires that are younger than ever before will affect the turnover rate of sires within the AI companies. In the genomic era it may be a rare occurrence to see bulls older than 5 years of age in collecting facilities. Young desirable sires will be marketed as soon as possible and will dominate the facilities and active lineups (Amann and DeJarnette, 2012). The use of genomically-selected bulls may quickly make the genetics of traditionally raised bulls obsolete, and the scenario could exist where bulls raised on a traditional system may have to compete with their genomically-selected sons, as their sons will not go through a 'sire in waiting' period and will possess current and more desirable genetics (Boichard, 2010). If is not uncommon

62 for AI companies to house grandsons or even great grandsons of sires currently in the active lineups due to the decreased generation interval observed today.

As AI companies transition to marketing a larger proportion of their bulls based on genomics, their facilities and traditional housing methods will need be revamped to accommodate all ages of sires. It is now commonplace for AI companies to be in ownership of young calves that are not weaned, and these large numbers of immature animals will put stress on the infrastructure of housing. Additionally, separate quarantine barns will become crucial, as the movement of bulls will happen more frequently and strict quarantine protocols are customary to ensure biosecurity. In the genomic era, all bulls will probably be housed in individual production stalls by 10 months of age, as this age represents the target for initiation of semen collection for young bulls today for many companies. Sires will remain in individual collection stalls for the remainder of their productive lives, granted they are commercially popular, which is in contrast to the traditional system where bulls assigned as sires in waiting are not collected and are group housed while waiting for the results of their progeny tests. Overall, there will be an increased number of young, extremely valuable young calves replacing the space originally needed for sires-in-waiting, and there may be a decrease in space needed for mature, actively-producing sires because the turnover rate will be greatly accelerated due to genomics.

The reproductive efficiency of this changing young population of bulls will be different from older, more mature population of bulls of the past. Bulls are introduced to collection facilities beginning at 10 months of age, but the bull's scrotal circumference

63

(SC) is probably the best estimate of his production potential. Bulls should have a SC of

30 to 32 centimeters before attempting semen collection (DeJarnette and Monke, personal correspondence). There are problems associated with semen collected from sires that are not yet fully mature. Semen from young animals may have too high a proportion of seminal fluid in their collections, and the total number of sperm cells they produce will always be lower than that of mature bulls (Monke, personal correspondence). Also, bulls that have recently attained puberty may fail to produce sperm of acceptable motility or enough sperm of acceptable quality to warrant packaging and cryopreservation (Amann and DeJarnette, 2012). Unfortunately, it may take several years after the commencement of semen collections before bulls for reach full production capacity, and this will conflict with the fact that as much semen as possible is needed from a high-genomic sire as soon as he begins collections.

Lastly, there will be a transition and reallocation of funds within the industry in the genomic era. One significant change will be in the acquisition costs of bulls. Bulls purchased based on pedigree information are a large investment to AI companies, but more so due to the large numbers of animals that are purchased, not necessarily because the acquisition cost of any one bull is exorbitant. Stated directly, the cost of each sire has increased significantly in recent years due to genomics. Breeders have quickly realized the worth of a genomically-superior bull and the acquisition costs of sires have increased a reported 4-5 times since the incorporation of genomics into the selection process. It is also becoming frequently common for breeders to choose to lease their bulls to AI companies rather than sell them outright (DeJarnette, personal correspondence). In

64 anticipation of rising sire costs, several AI companies have begun owning and breeding females of high genetic merit with the goal of executing an internal breeding program to balance against the rising costs of a bull on the open market.

Other notable financial changes that can be associated with genomics include the cost of labor for the AI companies due to the increasing need for more employees specializing in raising calves from birth. Also, collection schedules of greater intensity for popular bulls will translate into the need for more workers and facilities, or perhaps the addition of extra days (i.e. weekends) for collection, processing, and freezing (Amann and DeJarnette, 2012). Intellectual property, such as the development of the Illumina

BovineSNP50 BeadChip had cost the major stud companies millions of dollars for use, and there may be similar technological advances in the future requiring the input of resources from stud companies and farmers alike (Seidel, 2010). Finally, the cost of a straw of semen for a producer may change in the future, but to what degree is uncertain.

Farmers will probably pay a premium to utilize the most recent, in-vogue sires as they are released into the active lineups, but this is no different than what occurs in the traditional marketing scheme. Genomically superior sires' semen may sell for more than today's average semen price, but it has been argued that AI centers expect to house and maintain less bulls overall, perhaps evening out or controlling any drastic increases in semen price.

65

2.4.3. Economic Impacts of the Age of Puberty Attainment in Bulls Destined for Use in

Artificial Insemination

The demand for young sires is continuing to grow as the reliability of a bull's genetic transmitting ability as indicated by genomic testing becomes more accurate.

Furthermore, the ability to calculate the transmitting ability of a sire as a young calf creates a disparity between the time at which his genetics are marketable (i.e. immediately) and when the sire can physiologically provide his genetics to the market.

Hence, a novel push to collect genetics from young animals exists in order to advance the production abilities of the population as well as to generate income by selling superior genetics. AI companies would be interested in advancing gamete collection from young sires not only to provide their customers with the best genetics, but also to gain advantages in their sire lineup compared to other competitive AI companies.

Hastening puberty could contribute to increased profits for AI companies in several ways. Foremost, the production of the first viable sperm could be utilized by the company in order to create desired matings for future sires. Once the in-house need for early collections are met, subsequent collections of a genomically superior bull would be available for commercial sale, garnering profits at an earlier time and theoretically for a longer duration of time before the bull's consumer demand falls. To elaborate on the first concept, AI companies generally use desirable genetics to create the next generation of sires. Often, sires of high genomic value are mated to heifers of similarly high genetic value in an attempt to amplify attributes of the parents into a new generation of superior offspring. In a sense, a genetic "race" has begun to own and market sires of the highest

66 genetic merit. This occurrence extends even beyond the established AI companies, and several new niche companies have emerged since 2008 that solely aim to market a small number of elite genomic sires.

When assessing the female side of genetic progress, heifers could be considered more precocious than sires. Oocytes can be collected from very young heifers before they attain puberty using technologies such as gonadotropin treatments and ovum pick-up

(OPU) procedures (Duby et al., 1996). Although the developmental capacity and the potential to develop live births from oocytes collected from prepubertal heifers is lower than in that of older, cycling females, the possibility still exists in heifers ranging from 2 to 4 months of age (Revel et al., 1995; Armstrong et al., 1992; Kajihara et al., 1991).

Granted, in the studies highlighted, oocytes were collected either via laparoscopic follicular aspiration or after slaughter, and often included the use of ovulation-inducing gonadotropin treatments. Regardless, these proof-of-concept studies verify that oocytes in young heifers are viable, albeit much less so than oocytes of postpubertal females (Revel et al., 1995). In a non-experimental setting, it is not unusual to attempt OPU with IVF and

ET in heifers closer to 6 months of age without slaughtering the animal, since the donor animal is likely to be valuable because of her genetics. Bulls are similar to heifers in that prepubertal collection of gametes would include invasive procedures. Accordingly, most

AI companies would likely be unwilling to perform testicular biopsy to attain sperm for round spermatid injection (ROSI), intracytoplasmic sperm injection (ICSI), or IVF to create a pregnancy. The cons of risking testicular damage in a potentially valuable sires

67 outweigh the pros of decreasing the generation interval in this case. Therefore, obtaining sperm via natural ejaculation is the most desirable option for AI companies.

Alternatively, hastened puberty could generate profit for AI companies via the earlier release of a sire in the active lineup and through the commercial sale of fresh or frozen units of semen. As previously described, bulls produced via contract mating were historically not received at the AI facility until 6 to 12 months of age. Accordingly, it was commonplace to acquire the first collections from sires to begin progeny testing when the animals were around one year of age. As operational changes have occurred with AI companies, calves are often owned from a much younger age, and collection attempts are beginning at younger and younger ages. If a sire has genomic potential, it is now common for some AI companies to attempt initial collections as early as eight months of age. The decision to attempt collections earlier may be warranted based on potential demand of the sire, but is largely dictated by his maturational characteristics such as an acceptable scrotal circumference (Monke, personal correspondence).

The income generated from collections of bulls with hastened puberty could be significant when compared to normally maturing counterparts, but the parameters for collections and processing are different for young sires compared to mature sires. For example, it is common for many companies to package more sperm cells per straw in young sires. This is likely for several reasons, including the increased prevalence of morphological abnormalities, decreased numbers of motile cells, and unknown fertility potential of untested young sires. One production manager at a U.S. AI company reported that it is standard procedure to package 35 million sperm per straw for young sires versus

68 the average of 15 to 20 million cells per straw for mature sires. For this company, more than doubling cell numbers in young sires must have been determined to be sufficient to overcome potential compensable traits of sperm and equalize fertility potential between young and mature sires (Sullivan and Elliot, 1968; Saacke, 2008). Additionally, an AI company will have to decide what quantity of total cell numbers are acceptable to warrant further processing. Based on the increased cell numbers per straw in young bulls, it may not be economically profitable to process collections garnering less than several dozen straws because of processing, labor, and distribution costs. All of these factors are under intense scrutiny as AI companies transition to owning many more young sires.

Apart from the logistics of processing and collecting young bulls, there is still income to be generated from young sires. Scanning the price lists of young sires from several major AI companies reveals that there is an increased profit margin per straw in young bulls versus the average mature bulls. While the cost of mature Holstein sires average $12 to $20 per straw, most newly released young sires sell for $40 to $50 (and up) per straw. In this dissertation's experiments, the initiation of sperm production was hastened by exogenous FSH administration protocols by approximately 1 month over control animals (comparisons of pubertal and mature collections are discussed at length in their respective Chapters of this dissertation). In a theoretical situation where a young sire was producing 50 straws per collection on twice weekly collections (400 straws per month) at $50 per straw, this would generate $20,000 gross income per month. Several treated sires in this dissertation were producing over 1.75 billion cells per collection at just over 300 days of age, meaning that they would qualify for processing thresholds

69 aimed at producing 50 straws with 35 million sperm per straw. The obtainment of saleable collections at 300 days (10 months) is a markedly different scenario than the standard protocol of initiating the attempts for collections at this same age in some AI companies, especially since saleable collections are not expected to be obtained until several weeks after these attempts begin. In a conservative economic model estimating

$20,000 per month in sales from each young sire, a large stud company with many sires could experience measurable increases in profit if the commencement of collections were earlier. Furthermore, exogenous FSH treatments in this dissertation's research enhanced testicular morphology at the end of treatment at 6 months of age in terms increasing of the number of Sertoli cells. There is the potential that these treatments develop the testes in such a manner that overall sperm production could be elevated not only at the time of puberty attainment, but permanently for the productive life of the sire. Any increases in sperm production would positively contribute to the profit generated as a sire overcame processing thresholds set by a company. A reduction in the age of sperm production in young sires as well as an increase in semen production after puberty would not only benefit the AI company monetarily, but would also benefit the producers by allowing access to elite genetics at an earlier time.

2.5. Restatement of the Problem and Rationale for Research

Many experiments have validated our understanding of prepubertal bull growth, endocrinology, and testicular development. Many experiments seeking to enhance sexual development and to positively augment the endocrinology of bulls so that they are

70 profitable to AI companies have also been conducted. However, the prior advancements in this area of research have come under new pressure due to novel changes of sire selection and marketing in the form of genomic predictions of sire transmitting ability.

A current limiting factor to genetic progress in the cattle industries that has not been easily overcome is that the most popular, genetically superior sires cannot produce enough semen at an early enough age to meet the demand of farmers. An increase in semen supply earlier in the life of a genomically prominent sire would allow greater accessibility to the best genetics available, thus preventing producers from breeding females to sires of lesser genetic merit. Based on these legitimate dilemmas, a reliable system for increasing and maximizing semen production from young bulls would increase profits for AI companies and give producers better access to genetically superior sires.

It is known that several types of treatments early in life have the potential to positively affect sperm production in the mature bull. Treatment with exogenous FSH preceding puberty, during the short period of Sertoli cell proliferation, can cause increases in the final number of Sertoli cells (Bagu et al., 2004; Harstine, unpublished), and this could have large impacts on mature sperm production (Berndtson et al., 1987).

However, the time at which FSH administration is most effective in impacting Sertoli cell proliferation, as well as the modes of action, are not well established. The need for more studies examining bull testicular development and factors causing puberty in bulls was conveyed by Amann and Schanbacher in 1983, and this need still exists today.

71

The studies within this dissertation were designed with the objective of resolving several of the aforementioned gaps in knowledge. Several earlier studies conducted by our laboratory validated the use of an exogenous FSH treatment regimen early in life to enhance prepubertal bull endocrinology and testicular development. The comprehensive nature of the current experiments seeks to build upon these previous findings in several ways. The duration of treatment for these experiments was strategically chosen to be from 2 to 6 months of age, not only because this is an age range when development can be significantly impacted, but also because the timing of treatment is actually feasible for implementation by a working AI company. We seek to gain understanding of the effect of this novel exogenous FSH treatment regimen on bull endocrinology and testicular development, the mechanisms underlying the potential changes to bull testicular physiology, the age of puberty attainment, and the mature semen production abilities.

Advances in this area of research could contribute not only to the AI companies that development and market genetics, but also to cattle producers and the growing population of food consumers as a whole.

72

CHAPTER 3

IMPACT OF AN EXOGENOUS FSH TREATMENT REGIMEN ON THE

ENDOCRINOLOGY AND TESTICULAR DEVELOPMENT OF PREPUBERTAL

BEEF BULLS

INTRODUCTION

Within the cattle breeding and genetics industries, the recent inclusion of genomic evaluations into AI sire proofs has placed increased emphasis on the timeliness of sire selection and marketing (Amann and DeJarnette, 2012). Proofs generated from progeny testing of daughters are becoming supplementary to initial genomic estimates, and these genomic tests are usually performed at a young age or can even be performed during the embryonic stage (Humblot et a., 2010). Accordingly, there is a new initiative to collect bulls destined for use in the AI industry as early as possible. In addition, young bulls often do not have the ability to produce enough semen to meet market demand to the extent older sires do. Due to the recent, changing demands of young bulls in the AI industry, novel ways to advance puberty and positively influence testicular development of bulls are warranted.

73

Dynamic changes in the endocrinology and physiology of young bulls culminate in the attainment of puberty. For example, concentrations of follicle-stimulating hormone

(FSH) have been reported to undergo a transient elevation in bulls from approximately 4 to 25 weeks of age (Amann et al., 1986; Bagu et al., 2006). However, FSH concentrations have also been reported to remain stable and unchanging in the prepubertal bull

(McCarthy et al., 1979; Harstine et al., unpublished). Sertoli cells, which are responsible for nurturing sperm during spermatogenesis, proliferate in response to FSH in the prepubertal animal, but once Sertoli cells reach a mature state at 25 weeks of age they lose the ability to divide and proliferate (Orth, 1984; Sharpe et al., 2003). Since each mature Sertoli cell hosts a fixed number of germ cells, the timing and rate of Sertoli cell proliferation preceding puberty is of interest because the final number of Sertoli cells correlates to the sperm production capacity of the adult animal (Berndtson et al., 1987).

Therefore, a mature bull with more Sertoli cells per testis has the potential to produce more sperm per unit of time than a bull possessing lesser numbers of Sertoli cells (Orth,

1984).

Methods to increase prepubertal Sertoli cell numbers represent a promising target of research, and various methods have been utilized to manipulate the prepubertal endocrinology of bulls in an attempt to enhance testicular physiology. For example, immunizations against inhibin remove the suppressive effects of inhibin on FSH secretion (McGowan et al., 1988). Inhibin immunizations are marginally successful in increasing prepubertal FSH concentrations, and the level of response seems variable between experiments and even between animals within experiment (Kaneko et al., 1993;

74

Kaneko et al., 2001; Martin et al., 1991; Lunstra et al., 1993; Bame et al., 1999). The development of a method to directly and efficiently increase systemic FSH in the prepubertal bull would provide a means to circumvent challenges associated with indirect immunizations. In this regard, Bagu et al. (2004) administered 10 mg NIH-FSH-S1 every other day to 4-week old bull calves until 8 weeks of age and observed a threefold increase in FSH concentrations 105 minutes after administration at 4 weeks of age. This plateau in

FSH returned to control levels within 24 hours, likely due to the short half-life of FSH in the circulation. As the bulls grew in body size during the experiment, the same treatment applied at 8 weeks of age caused a peak in FSH concentration 120 minutes after administration that returned to control levels within 9 hours. The treatments successfully elevated FSH concentrations after each administration, hastened puberty as determined by a 28 cm scrotal circumference (39.3 vs 44.8 weeks of age), and increased the numbers of Sertoli cells, elongated spermatids, and spermatocytes in the testis when measured at

56 weeks of age using histology. Due to the short duration of FSH elevation resulting from this treatment, as well as need to handle bulls every other day, we aimed to develop and test the effects of a novel time-released FSH treatment on the endocrinology and testicular development of bulls.

The utilization of a slow-release FSH treatment has been proposed before to superovulate cows. Mixing commercially available FSH with either 2% (Bó and

Mapletoft, 2012) or 1% (Tribulo et al., 2012) hyaluronic acid (HA) reduces the number of FSH treatments needed for superovulation of beef cattle from eight treatments at 12 hour intervals to a single dose or to a double dose given at 48 hour intervals. In the

75 present experiment, the FSH (Folltropin-V; Bioniche Animal Health, Ontario, Canada) used was pituitary-derived porcine FSH suspended in a solution containing 2% HA as a vehicle as has previously been used in cattle (Bó et al., 1991). The ability of HA to extend the release of drug at its site of absorption is due to its mucoadhesive properties

(Surini et al., 2003).

It was desired that the treatment utilized in this experiment not only positively affected the physiology and endocrinology of the bulls, but also that the timing of this treatment regimen was feasible for implementation by an operating AI company. First, a dosage was chosen based on prior research by our laboratory indicating that 30 mg of

NIH-FSH-P1 (Folltropin-V) in a 2% HA solution administered twice weekly from 39 to

91 days of age could positively affect testicular development. Next, this treatment was administered from 59 to 167.5 days of age, which represents a plausible age range during which an AI company would be in possession of their AI sires. It was hypothesized that treatment of beef bulls with 30 mg NIH-FSH-P1 in a 2% HA solution from 59 to 167.5 day of age would cause increases in FSH following treatment and increase the numbers of Sertoli cells within the testis at the end of the treatment period. Therefore, the objective of the present experiment was to test the effect of this exogenous slow-release

FSH treatment from 59 to 167.5 days of age on the endocrinology and testicular histology of bulls. In terms of this dissertation's research, this experiment serves as a precursor to a second experiment where the time of puberty attainment and mature sperm production are assessed. Together, these two experiments seek to better understand the endocrinology and physiology resulting from this treatment regimen in prepubertal bulls.

76

The mechanisms underlying observed changes in endocrinology and physiology will also be explored. Additionally, the second experiment in this series will provide insight as to whether the application of this treatment actually hastens puberty and positively affects sperm production at puberty attainment and in maturity.

MATERIALS AND METHODS

Animals and handling were conducted in accordance with procedures approved by The Ohio State University Agricultural Animal Care and Use Committee

(#2014A00000047).

Animals and Treatments

Twenty-two age-matched Angus-cross bull calves born at The Ohio State

University's Beef and Sheep Center from multiparous cows were selected for this experiment. Bulls were allowed ad libitum access to pasture, water, and suckling of dams until 53 ± 3.8 days of age, at which time they underwent weaning. Weaning was accomplished using a fence-line technique where calves were placed in a separate pasture adjacent to the cows (Price et al., 2003). Calves were kept in this pasture with access to shelter, water, ad libitum grass hay, and were provided with a receiving ration consisting of 50% corn, 15% alfalfa, 15% soy hull pellets, and 20% vitamin/mineral supplement with 16% protein overall. This receiving diet was fed once daily to achieve a 1.1 kg/d average daily gain (ADG). At 140 days of age, all calves were switched to a growing ration consisting of 60% crimped corn, 30% soy hull pellets, and 10% vitamin/mineral

77 supplement. This growing ration contained 0.25 mg/kg rumensin, and calves achieved approximately 1.4 kg/d ADG while receiving this growing ration. The bulls received the growing ration through the age of castration at 170 days of age.

Bulls were randomized and evenly allocated in two treatments groups based on birth date and pedigree, and beginning at 59 ± 3.8 days of age, the bulls were injected i.m. with either 30 mg NIH-FSH-P1 (Folltropin-V; Bioniche Animal Health, Athens,

GA) in a 2% hyaluronic acid solution (FSH-HA, n = 11) or saline (control, n = 11) every

3.5 days until 167.5 ± 3.8 days of age. Hyaluronic acid (HA) was research grade dried sodium hyaluronate with a molecular weight ranging from 601 – 850 KDa (Lifecore

Biomedical, Chaska, MN). The FSH-HA treatment was formulated by mixing Folltropin-

V to 60 mg/ml in a 2% HA solution and delivering 0.5 ml every 3.5 days to achieve the delivery of 30 mg NIH-FSH-P1 per injection. Control animals received 0.5 ml saline every 3.5 days. Injections were given i.m. in the neck, being sure to deliver the entire bolus of the treatment to one area. Side of the neck of the treatment administration was recorded and switched the next treatment day to ensure previously placed boluses were not disturbed.

Body Weight and Scrotal Circumference Measurements

Body weights (BW) were recorded every 30 days from 59 to 167.5 days while bulls were examined on a cattle squeeze chute equipped with a digital scale. Scrotal circumference (SC) measurements were recorded for all bulls monthly at 76.5, 104.5,

136, and 167.5 days of age. SCs were obtained with a metal scrotal tape (Sullivan

78

Supply, Inc., Dunlap, IA) by the same technician, being sure to measure the circumference of the testes at their widest point. With the bull properly restrained, the scrotum was grasped at the neck with one hand and the were pulled gently into the bottom of the scrotum. The scrotal tape was placed at the site of the greatest circumference, snugged, and the measurement recorded.

Blood Sample Collection

A blood sample to assess hormone concentrations consisted of a maximum of 10 ml of blood collected from the jugular vein of each animal using 1 inch, 18 gauge needle and syringe. When both plasma and serum were needed, the 10 ml of blood were divided evenly into two respective Vacutainer tubes intended for either serum or plasma isolation

(Becton Dickinson and Company, Franklin Lakes, NJ). A 5 ml sample was collected when only serum was needed for analyses. For serum isolation, glass Vacutainer tubes were placed on ice until storage at 4°C, allowed to clot for 48 hours at 4°C, and then centrifuged at 7,735 x g for 20 minutes. Serum was harvested and frozen at -20°C until analyzed for circulating concentrations of FSH or activin A. For plasma, the sample was placed into K2-EDTA Vacutainer tubes, centrifuged at 7,735 x g for 20 minutes, plasma harvested, and samples frozen at -20°C until analyzed for concentration of testosterone.

Regarding the timing of sample collection, serum to measure FSH was regularly collected immediately before each treatment administration every 3.5 days from 59 to

167.5 days of age. Additionally, three periods of "intensive" blood collection to examine the blood hormone profiles of FSH following treatment commenced at 66, 108, and 157

79 days of age. During an intensive blood sampling period, which lasted for a total of 7 days and encompassed two treatment administrations, blood was collected immediately before treatment (0 hour) and every 6 hours thereafter for a 24 hour period, and at 36, and 60 hours post-treatment. At the next treatment 3.5 days (84 hours) later, blood was again collected prior to treatment (0 hour) and every 6 hours for 24 hours and at 36 and 60 hours post treatment. For testosterone analysis, blood samples for plasma isolation were collected every 7 days immediately prior to treatment from 59 to 164 days of age. Serum samples were also used to determine activin A concentrations.

Hormone Analyses

Follicle-Stimulating Hormone

Concentration of FSH in serum was determined in duplicate for all samples with a double radioimmunoassay (RIA) previously validated for use in measuring bovine FSH (Burke et al., 2003). Briefly, purified ovine FSH (LER1976-A2) was iodinated using the chloramine-T method, primary antibody was rabbit-anti-ovine FSH

(JAD #17-7.6.9), and secondary antibody was donkey-anti-rabbit and normal rabbit serum. For samples collected every 3.5 days, all samples were analyzed using the same iodinated FSH in three assays run simultaneously, being sure that an individual bull's samples were kept within a singular assay. The average intra-assay coefficient of variation (CV) was 7.8% and the average inter-assay CV (three assays) was 11.9% for a

"low" standard (2.69 ng/ml) and 9.42% for a "high" standard (5.1 ng/ml). The average sensitivity was 1.1 ng/ml.

80

The samples collected during the three "intensive" blood collections were run in a separate group of four RIAs utilizing the same methods and as described above. The average intra-assay CV was 5.3% and the average inter-assay CV (four assays) for the low standard (2.79 ng/ml) was 15.5% and for the high standard (4.87 ng/ml) was 13.4%. The average sensitivity was 1.4 ng/ml.

Activin A

An enzyme-linked immunosorbent assay (ELISA) kit (Ansh Labs, Webster, TX) was used to determine activin A in serum. The kit was reported to measure "total" activin

A, which includes all unbound and follistatin-bound activin A. Cross reactivity with inhibin A, inhibin B, and activin AB is 2.8% as reported by the manufacturer. All samples were processed in duplicate and according to manufacturer's recommendations.

The optical density of the plates was measured within 15 minutes of assay completion using a spectrophotometric plate reader at 450 nm (BioTek, Winooski, VT). Activin A concentrations of unknowns were calculated by plotting the data on a log versus log scale using a cubic regression curve-fit. The lowest limit of detection reported was 0.065 ng/ml, and the measurable range of detection was 0.1 – 10 ng/ml. The average intra-assay and inter-assay CV was 7.2% and 8.8%, respectively.

Testosterone

Concentrations of testosterone in plasma were analyzed in duplicate using a commercial RIA kit (Coat-a-Count, Siemens Medical Solutions Diagnostics, Los

81

Angeles, CA) previously validated in our laboratory for bovine samples (Burke et al.,

2005). All samples were run in one assay, and the intra-assay CV was 5.1% and the average intra-assay CV for the low pool (0.02 ng/ml) was 25.9%, medium pool (1.9 ng/ml) was 2.29%, and high pool (8.9 ng/ml) was 14.9%. The sensitivity of the assay was

0.05 ng/ml.

Castration and Testicular Measurements

The bulls were castrated according to the OSU Beef and Sheep Center Standard

Operating Protocol at 170 ± 3.8 days of age and testes were collected. Local anesthetic

(Lidocaine HCl 2%, 5 ml per testis; Pro Labs Ltd., MO, USA) was injected into the scrotal neck after bulls were restrained in a squeeze chute. A single opening was cut into the bottom of the scrotum and each pulled out of the scrotum and away from the body cavity. Testicles were pulled downward, exposing the spermatogenic cord while pushing back the connective tissue. The cord was cut above the testicle using an emasculator with crimper, and bulls were treated with spray-on antibacterial medication and insect repellent in the area post-operatively.

Immediately following castration, one testis was utilized to measure volume and weight, and the other used for fixation. Testes were designated for either measuring and fixation by alternating between bull within treatment. Testis volume was measured by submersion in a graduated cylinder filled with saline after the epididymis and the pampiniform plexus were removed, leaving only the testis and tunica albuginea. A weight was recorded for this same testis using a digital scale.

82

Testicular Histology and Immunohistochemistry

Using the testis designated for fixation, sections approximately 5 mm thick representing the proximal, medial, and distal third of each testis were collected for histological analysis. Sections were fixed in 10% formalin for 12 hours, placed in 70% ethanol for at least 24 hours, and then imbedded in paraffin for microtome sectioning. A sample representing each of the three collected testis sections was stained using a standard hematoxylin and eosin stain.

Using the hematoxylin and eosin stained sections, the medial section was used to measure the seminiferous tubule diameters. Seminiferous tubule diameter was measured for ten randomly selected round tubules per section using an ocular micrometer (Zeiss,

Fisher Scientific, Pittsburgh, PA). Only the middle section was utilized to generate an average seminferous tubule diameter for each bull because previous research in our lab has indicated that the tubule composition does not differ between the proximal, medial, and distal third of the testis (Harstine et al., 2015).

The number of Sertoli and germ cells contained within a seminiferous tubule monolayer cross section was determined for each bull using histological sections from the middle of the testis. Tissue sections were subjected to immunohistochemistry for the localization of the transcription factor GATA-4, which is a marker of Sertoli cell nuclei

(McCoard, et al., 2001). An affinity purified goat polyclonal antibody (C-20, Santa Cruz

Biotechnology, Santa Cruz, CA) developed against a peptide mapping at the carboxy- terminus of mouse origin GATA-4 was used at a 1:100 dilution. The epitope is highly

83 conserved between species and has been used in cattle (Jimenez-Severiano et al., 2005).

Secondary antibody was a rabbit-anti-goat, and slides were counterstained with hematoxylin, dehydrated, cleared, and mounted. Specificity of the GATA-4 marker was verified using histological sections from mature bulls (18 months of age) utilized in a prior experiment (Harstine et al., 2015). The same procedures described above were applied to these sections, but with a preliminary incubation with GATA-4 Blocking

Peptide (Santa-Cruz Biotechnology, Santa Cruz, CA) in 10X excess of the primary antibody. Negative controls were created by incubating with buffer solution and no primary antibody to check for nonspecific binding of the secondary antibody. It was determined that GATA-4 was selectively staining only Sertoli cell nuclei within the seminiferous tubules using these procedures based on cell location and visual identification of the cells within the tubules. All procedures were performed by The Ohio

State University's Comparative Pathology and Mouse Phenotyping Shared Resource.

After GATA-4 staining, the middle section of each testis was examined at 400X magnification, and ten round seminiferous tubules were randomly selected. The number of stained Sertoli cell nuclei and unstained germ cell (spermatogonia) nuclei present in a circular monolayer within the round tubule were recorded.

Statistical Methods

Body weight (BW) and scrotal circumference (SC) of the bulls were analyzed by

ANOVA using the MIXED procedure of SAS (version 9.3, 2010) with repeated measures

(age) analysis and an autoregressive (AR(1)) covariate structure included in the model.

84

Each model examined the effect of treatment, age, and their interaction on the respective dependent variable, and if either the main effects or their interactions were significant according to Wilks' lambda, differences between least squares means were reported using

Tukey's HSD test.

The concentrations of FSH collected every 3.5 days prior to treatment were analyzed by ANOVA using the MIXED procedure of SAS with repeated measures (age) analysis and initial concentration of FSH (d59) used as a covariate in the model. An autoregressive (AR(1)) covariate structure was applied to the data based on the lowest

Bayesian information criteria (BIC) in a comparison of all covariate structures. FSH concentrations obtained from the three intensive samplings commencing at 66, 108, and

157 days of age were all analyzed using the same statistical model, where the effect of treatment, age, their interaction was tested. The covariate for these analyses was the FSH concentration immediately preceding treatment administration. FSH concentrations were analyzed by multivariate ANOVA using the MIXED procedure of SAS. Repeated measures (age) analysis with a compound symmetry (CS) covariate structure was applied to the data after comparison of all applicable covariate structures (lowest BIC).

Activin A and testosterone concentrations were analyzed by ANOVA using the

MIXED procedure of SAS. For testosterone, repeated measures (age) analysis was included in the model and an autoregressive (AR(1)) covariate structure was utilized. For activin A concentrations, repeated measures (age) analysis was also included in the model, but with a heterogeneous compound symmetry (CSH) covariance structure based on the most appropriate fit for the data (lowest BIC).

85

Testicular morphometry data of testis weight, testis volume, seminiferous tubule diameter, and the number of germ and Sertoli cells per tubule cross section were analyzed with one-way ANOVA using PROC GLM procedure of SAS. When initially analyzing testis weight and volume, side (L or R) was included in the model, but this variable was removed after backward stepwise elimination of independent variables based on a Wald statistic of greater than P < 0.1.

All results are reported in terms of mean ± SE.

RESULTS

Neither the weight nor SC of the bulls differed between treatments at any of the ages measured during the experiment and BW and SC increased during the experiment across treatments (Age, P < 0.05). BW increased from 93.8 ± 2.4 kg at 59 days of age to

224.5 ± 4.9 kg at 167.5 days of age, and SC increased from 14.2 ± 0.2 cm at 76.5 days of age to 19.2 ± 0.3 cm at 167.5 days of age (Figure 3.1).

Concentrations of FSH in serum collected every 3.5 days preceding treatment administration did not differ between control and FSH-HA bulls from 59 to 90.5 days of age. However, at 94 days of age FSH-HA bulls had greater (P < 0.05) concentrations of

FSH than control animals, and this difference was sustained until the end of the sampling period at 167.5 days of age (Trt*Age, P < 0.05; Figure 3.2). When comparing initial FSH concentrations on d 59 to subsequent concentrations within the FSH-HA treatment, FSH concentrations were increased (P < 0.05) over day 59 concentration beginning at 97.5 days of age, and were greater throughout the remainder of the sampling period (Figure

86

3.2). In the control treatment, concentrations of FSH remained relatively unchanged from the initial sampling at d 59. However, control bulls did have lower (P < 0.05) concentrations of FSH on day 73 than the initial sampling at day 59. All other values were not different from the initial day 59 concentration within the control treatment.

The intensive blood samplings that commenced at 66, 108, and 157 days of age to examine FSH hormone profiles were analyzed independently, but the same statistical procedures were used for each sampling window. An important consideration for comparison of these data is that concentration of FSH preceding the first of 2 FSH/saline treatments in a sampling window was included as a covariate. For the first intensive sampling that included treatment administration at 66 and 69.5 days of age, peripheral

FSH concentrations were greater (Trt*Age, P < 0.05) in the FSH-HA than control treatment 6, 12, and 18 hours after treatment at 66 days of age, and the concentration of

FSH was greater (Trt*Age, P < 0.05) in FSH-HA than control bulls 6, 12, 18, and 24 hours after treatment at 69.5 days of age (Figure 3.3A). At the next sampling encompassing treatment administration at 108 and 111.5 days of age, FSH concentrations were greater (Trt*Age, P < 0.05) in FSH-HA than control bulls 6, 12, 18, and 24 hours post-treatment at both 108 and 111.5 days of age (Figure 3.3B). During the last intensive sampling at 157 and 160.5 days of age, FSH concentrations were greater (Trt*Age, P <

0.05) in FSH-HA than control bulls 6, 12, 18, 24, and 36 hours after treatment at 157 days, and after treatment at 160.5 days FSH was greater in FSH-HA than control bulls at

6, 12, and 18 hours post-treatment (Figure 3.3C).

87

The ages at which activin A concentrations were examined were chosen based on the timing of the FSH differences between treatments observed during the regular blood samplings every 3.5 days (Figure 3.2). Foremost, the start (d59) and end (d167.5) of treatment were selected. Next, the day which FSH concentrations first differed (P < 0.05) between treatments (d94), as well as midway time points (d83.5, d129) between this age and the start and end of treatment were chosen. Activin A concentration did not differ in control animals at any age examined (Figure 3.4). However, activin A concentrations

(Trt*Age, P < 0.05) in FSH-HA bulls were greater (P < 0.05) than control bulls at 83.5 and 94 days of age. Similarly, within the FSH-HA treatment, concentrations at d83.5 and d94 were greater than concentrations at d59, d129, and d167.5 (Figure 3.4).

Testosterone concentrations measured in plasma increased with age (P < 0.05) but did not differ between treatments at any ages examined (Trt*Age, P = 0.99; Figure 3.5).

Evaluation of the testes after castration at 170 days of age revealed no differences in testis weight, testis volume, diameter of seminiferous tubules, or number of germ cells between treatments (Table 3.1). However, bulls in the FSH-HA treatment had a greater (P

< 0.05) number of Sertoli cells per seminiferous tubule cross section (FSH-HA, 45.2 ±

1.4 cells; control, 41.6 ± 0.9 cells; Table 3.1).

DISCUSSION

The FSH-HA treatment given every 3.5 days increased systemic FSH concentrations in the FSH-HA bulls compared to control bulls at 94 days of age in the present experiment. Within the FSH-HA treatment, FSH concentrations became greater

88 than initial day 59 concentrations on day 97.5 and remained greater until the end of sampling at 167.5 days of age. When examining the intensive blood samplings used to determine FSH profiles after FSH-HA treatment at 66, 108, and 157 days of age, concentrations of FSH were elevated in FSH-HA bulls a minimum of 18 hours post- treatment in each of the ages analyzed. The increases in FSH concentrations observed in

FSH-HA bulls are believed to initiate several interesting endocrine and physiological changes. For example, FSH-HA bulls had greater numbers of Sertoli cells per seminiferous tubule cross section than control bulls at 170 days of age. Also, concentrations of activin A were greater in FSH-HA than control bulls at 83.5 and 94 days of age. Overall, the experimental hypotheses were correct in stating that the applied

FSH-HA treatment regimen would cause increases endogenous FSH secretion and cause increases in Sertoli cells at the time of castration. The possibilities of how FSH-HA treatment potentiates these effects is discussed in detail in the following section.

The concentration of FSH has been reported to undergo transient increases from approximately 4 to 25 weeks of age (28 to 175 days; Amann et al., 1986; Bagu et al.,

2006). In the present experiment, blood sampling did not occur at early enough ages to verify whether FSH concentrations increased early in life, and sampling may not have occurred at a late enough age to determine whether FSH concentrations in control bulls would have decreased as described in these reports. The results of FSH sampling in this experiment agree with other reports, such as McCarthy et al., 1979, which report that

FSH concentrations remain unchanging in the prepubertal bull. An earlier starting age and later ending age of sampling would be necessary to link FSH concentrations

89 observed in this experiment to other literature. Nonetheless, the increases in FSH seen in

FSH-HA bull beginning at 94 days of age are noteworthy, as these concentrations continue to rise until a peak at 143 days of age, where the FSH-HA bulls have an approximate 230% increase in FSH over control bulls at the same age (FSH-HA, 7.82 ±

1.1 ng/ml versus control, 3.43 ± 0.4 ng/ml).

Further examining the FSH increase observed in FSH-HA bulls at 94 days of age, the results of this study correlate well to a previous study conducted by our laboratory that utilized the same FSH-HA treatment in cross-bred beef bulls. In this experiment, the bulls were treated from 35 to 91 days of age compared to the 59 to 167.5 days of age in the current experiment. Coincidentally, the same number of treatments were given before an endogenous increase in FSH was observed in FSH-HA bulls in these two experiments

(11 treatments). These findings suggest that the prepubertal endocrine system is responding in the same manner regardless of whether treatments begin at 35 or 59 days of age. The start date chosen for this experiment was of significance, however, because it represents a plausible date of implementation for such a treatment by an AI company, whereas it is unusual for AI companies to have regular access to their bulls at 35 days of age.

The weeklong intensive blood collections initiated at 66, 108, and 157 days of age provide insight into the hormone profiles generated in FSH-HA animals in the time following treatment. Overall, the profiles observed on the initial day of sampling for a given period and the profile of the subsequent (3.5 later) sampling (70.5, 111.5, and 160.5 day of age, respectively) correlate well with one another in the sense that an FSH increase

90 was observed at 6 hours post-injection in all instances. In all ages examined, an FSH-HA injection resulted in increased FSH concentrations for at least 18 hours post-treatment. In some instances, the increase in FSH over the pre-treatment FSH concentration was sustained for up to 24 hours after the FSH-HA treatment. These ranges in time represent significant increases over prior research utilizing exogenous FSH injections. Bagu et al.

(2004) reported that administration of 10 mg NIH-FSH-S1 in saline at 57 ± 6 days of age resulted in a peak in FSH concentration 2 hours after treatment that remained greater than untreated bulls for 8.25 hours. The range in time of the FSH increase caused by the FSH-

HA treatment (30 mg NIH-FSH-P1 in 2% HA) in the present study is likely contributable to the use of HA as a vehicle for administration. As mentioned before, the ability of HA to extend the release of drug at its site of absorption is likely due to its mucoadhesive properties (Surini et al., 2003). Of importance, the hormone profiles generated during in the first intensive sampling at day 66 and 70.5 demonstrate that FSH concentrations increased by FSH-HA return to basal concentrations before the next treatment administration. Considering that the collections of blood for FSH analysis from 59 to

167.5 days of age occurred prior to each treatment every 3.5 days, it can be concluded that the increases in FSH observed at 94 days of age is attributable to endogenous production of FSH, not from residual FSH from prior FSH-HA treatments.

The differences in activin A concentrations between treatments may elucidate a mechanism underlying the increase in endogenous FSH seen in FSH-HA bulls.

Concentrations of activin A were increased in FSH-HA bulls initially at 83.5 days of age.

This age precedes the first observed increase of FSH in FSH-HA observed at 94 days of

91 age. Furthermore, the concentration of activin A was greatest at 94 days of age in FSH-

HA bulls. Activins in bulls have been reported to be at their highest concentrations after birth, with concentrations decreasing at a similar time to the decrease of the gonadotropins between 25 and 32 weeks of age (Barakat et al., 2008). The last date of sampling in this experiment at 167.5 days (24 weeks) of age comes before the reported decrease in activins. Therefore, it is unknown whether activin concentrations in control bulls would decrease if blood sampling continued past the age range examined. In a prior experiment where the same FSH-HA treatment regimen was implemented from 35 to 91 days of age (Harstine et al., unpublished), FSH-HA bulls also had increased FSH concentrations 11 treatments after the initiation of treatment, but differences in activin A were not observed in FSH-HA bulls. A less frequent sampling and a differing age range of bulls in this experiment may be the reason differences in activin A were not observed.

The increase in activin A in FSH-HA bulls in the present experiment likely explains the increased endogenous production of FSH in FSH-HA bulls at 94 days of age.

Activin A production does occur in Sertoli cells (Anderson et al., 1998; Buzzard et al.,

2004), and activin A is known to stimulate FSH secretion from the pituitary (Ling et al.,

1986; Vale et al., 1986). Mechanistically, the binding of activin to ACVR2A receptors in the pituitary causes upregulation of SMAD transcription factors, and this pathway can cause upregulation of FSH-beta promoter within pituitary gonadotrope cells in vitro

(Suszko et a., 2003; Suszko et al., 2005). Furthermore, Smad3 deficiency in mice causes a large reduction in FSH-beta subunit mRNA levels (Coss et al., 2005). Overall, we propose that exogenous FSH-HA treatment initiates a positive feedback loop starting

92 with increased activin A production by a larger number of Sertoli cells, and this in turn stimulates endogenous secretion of FSH from the pituitary.

Another primary and highly relevant finding of this experiment was increased number of Sertoli cells in FSH-HA bulls at the time of castration at 170 days of age.

FSH-HA bulls had approximately 4 more Sertoli cells per cross section than control bulls

(45.2 ± 1.4 vs. 41.6 ± 0.9 cells). It can be inferred that the increase in Sertoli cells per tubule cross section in the FSH-HA treatment would translate to larger numbers of Sertoli cells on a per-testis basis since there were no differences between treatments in the weight or volume of the testes, or the diameter of the seminiferous tubules. These results correspond well to the results observed in the previously mentioned study by our laboratory where cross-bred beef bulls were treated with the same FSH-HA treatment from 35 to 91 days of age. In that experiment, bulls treated with FSH-HA a similar numerical increase in the numbers of Sertoli cells per seminiferous tubule cross section when compared to control bulls at the conclusion of the treatment administration (93 days of age, 33.35 ± 0.9 vs. 28.27 ± 0.9 cells). When comparing these two experiments, the percentage increase in Sertoli cells in FSH-HA bulls was greater in the earlier study where bulls were treated and castrated at earlier ages. It is unknown whether the stimulatory effects of FSH-HA treatment on Sertoli cell proliferation diminishes as animals grow older, if these differences reflect the varying age intervals of treatment administration, or both.

The increase in Sertoli cell numbers in these experiments is likely attributed to either the exogenous FSH provided by the FSH-HA treatment, the increase in the

93 endogenous FSH production, or both. It is known that FSH has a synergistic and proliferative effect on Sertoli cells before puberty (Orth et al., 1984; Bagu et al., 2004).

Increases in the number of Sertoli cells in the present experiment are promising because there is a positive correlation between the number of Sertoli cells within the adult testis and total sperm production (Berndtson et al., 1987). Experiments are currently underway to examine whether the FSH-HA treatment regimen's positive effect on Sertoli cell numbers translates into increased sperm production of post-pubertal and mature bulls.

Interestingly, the number of germ cells per seminiferous tubule cross section were not different between treatments in this experiment. As it is believed that the number of germ cells per Sertoli cell is fixed in bulls and other non-seasonally breeding animals

(Blanchard and Johnson, 1997; Leal et al., 2004), it is unclear whether the population of germ cells within the testes of FSH-HA bulls will increase proportionally to the extent of the Sertoli cell increases to maintain a set ratio as indicated in the literature. Histological assessment of the testes at 170 days of age did reveal that they were immature in the sense that only Sertoli cells and spermatogonium were visualized within the seminiferous tubules. It has been reported that spermatogonia will begin to occupy the spaces along the basement membrane between 3 to 4 months of age (Chandolia et al., 1997a).

Additionally, it has been reported that spermatocytes can be seen by 6 months of age and elongated spermatids by 8 months of age (Chandolia et al., 1997a; Barth, 2004). Bulls in this study were developmentally lagging in comparison to these studies, as spermatogonia had not yet migrated to the basement membrane of the tubule.

94

Overall, the results of this experiment illustrate the physiological and endocrine changes that can be induced by treatment with an exogenous FSH in prepubertal bulls.

Our experimental hypotheses were correct in stating that the applied FSH-HA treatment regimen would cause increases endogenous FSH secretion and cause increases in Sertoli cells at the time of castration. Potential mechanisms underlying these changes have been proposed in this manuscript. This study yields positive results that warrant follow-up studies examining how the observed changes may impact bull puberty attainment and mature sperm production. Accordingly, such experiments have been conducted concurrently in bulls housed within an AI center, and the results of these studies are paired in this dissertation.

95

Table 3.1. Testicular measurements, seminiferous tubule diameter, and the numbers of germ and Sertoli cells per tubule cross section (mean ± SE; 170 days of age) for Angus- cross bulls treated with either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age.

Treatment

Item FSH-HA Control P-Value

Testis Weight, g 46.6 ± 2.9 52.9 ± 3.9 0.21

Testis Volume, ml 45.2 ± 2.2 50.0 ± 4.2 0.32

Semiferous Tubule Diameter, μm 122.8 ± 2.9 125.4 ± 2.3 0.6

Germ cell per round tubule cross-section 4.5 ± 0.9 3.9 ± 0.6 0.58

Sertoli cells per round tubule cross-section 45.2 ± 1.4 41.6 ± 0.9 ˂ 0.05

96

21.0 20.0 Control 19.0 FSH-HA 18.0 17.0 16.0 15.0

Scrotal Circumference (cm) Circumference Scrotal 14.0 13.0 12.0 76.5 104.5 136 167.5 Average age (days)

Figure. 3.1. Mean (±SE) scrotal circumference measurements from Angus-cross bulls treated with either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age. Treatments did not differ at any age examined.

97

9.0

8.0 FSH-HA 7.0 Control * 6.0 5.0

4.0 FSH (ng/mL) FSH 3.0 2.0 1.0

Age (d)

Figure 3.2. Systemic concentrations of FSH immediately before bulls received either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age. Concentrations of FSH (Trt*Age, P > 0.05) were increased (P < 0.05) in the FSH-HA over control treatment beginning at 94 days of age (denoted by asterisk). Within the FSH-HA treatment, FSH concentrations were increased (P < 0.05) over day 59 concentration beginning at 97.5 days of age, and this difference was maintained through the remainder of the sampling period. In the control treatment, concentrations of FSH remained relatively unchanged from the initial sampling at d 59 although at a single time point ((d 73) control bulls had lower (P < 0.05) concentrations of FSH than the d 59.

98

Figure 3.3.

A.

7.0 Control 6.5 FSH-HA 6.0 5.5 * * * * 5.0 * * * 4.5

FSH (ng/ml) FSH 4.0 3.5 3.0 2.5 2.0 65 66 67 68 69 70 71 72 73 Age (d)

Figure 3.3A,B,C. Systemic FSH concentrations in bull calves after injection of 0.5 ml containing either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA) or saline (control) at 66 and 69.5 (Figure 3.3A), 108 and 111.5 (Figure 3.3B), and 157 and 160.5 (Figure 3.3C) days of age. There was a treatment by age interaction (P < 0.05) for each intensive sampling period. Within any sampling period, an asterisk (*) signifies that FSH- HA bulls had greater (P < 0.05) concentrations of FSH than contol bulls at that time point; after adjustment for initial FSH concentration.

continued

99

Figure 3.3. continued

B.

7.0 * * * * * * * 6.5 * 6.0 5.5 5.0 4.5

FSH (ng/ml) FSH 4.0 3.5 3.0 2.5 2.0 107 108 109 110 111 112 113 114 115 Age (d)

C.

7.0 * * 6.5 * * * * * * 6.0 5.5 5.0 4.5

FSH (ng/ml) FSH 4.0 3.5 3.0 2.5 2.0 156 157 158 159 160 161 162 163 164 Age (d)

100

0.6 bc Control 0.5 FSH-HA

0.4 b A (ng/ml) A - 0.3 a a a a a a a a Activin 0.2

0.1

0.0 59 83.5 94 129 167.5 Age (d)

Figure 3.4. Concentrations of activin A in bulls which received either 30 mg NIH-FSH- P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age. Concentrations of FSH (Trt*Age, P > 0.05) were increased (P < 0.05) in the FSH-HA over control treatment at 83.5 and 94 days of age. Similarly, within the FSH-HA treatment (Trt, P < 0.05), concentrations at 83.5 and 94 days of age were greater (P < 0.05) than concentrations at 59, 129, and 167.5 days of age. Concentration of activin A in the control treatment was not different at any of the ages examined.

101

6.0 5.0 4.0 Control 3.0 FSH-HA 2.0 1.0

Testosterone (ng/ml) Testosterone 0.0

Age (d)

Figure 3.5. Systemic testosterone concentrations of bulls treated with either 30 mg NIH- FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 11) or saline (control, n = 11) every 3.5 days from 59 to 167.5 days of age. Concentrations were not different between treatments at any age examined (Trt*Age, P = 0.99), but concentrations did increase for all bulls as they aged (Age, P < .05).

102

CHAPTER 4

IMPACT OF A PREPUBERTAL EXOGENOUS FSH TREATMENT REGIMEN

ON THE ENDOCRINOLOGY, PUBERTY ATTAINMENT, AND MATURE

SPERM PRODUCTION IN HOLSTEIN BULLS DESTINED FOR USE IN THE

ARTIFICIAL INSEMINATION INDUSTRY

INTRODUCTION

The methods for testing and selecting bulls for use by artificial insemination (AI) organizations has changed in recent years due to the inclusion of genomic evaluations.

Progeny testing had been the main system for sire selection since the use of AI in cattle became widespread in the 1950s. However, as the reliabilities of the transmitting abilities for some traits approach and surpass 90% in genomic evaluations, progeny testing has, in a sense, become supplemental to genomic evaluations (Van Raden et al., 2009; Cassel,

2010). Genomic evaluations of an animal can be conducted at a very young age (Humblot et al., 2010). Because the genetic worth of a bull can be determined well before he is able to produce sperm, there is a new emphasis on accelerating puberty in sires owned by AI organizations. Additionally, the production ability of a young sire is much less than that of a mature sire, and novel ways to maximize testicular development prior to the

103 attainment of puberty would maximize productivity of bulls destined for use in the industry.

Several areas of research have focused on positively manipulating the prepubertal endocrinology and physiology of the bull. Most of this research focuses on positively affecting the prepubertal rise in gonadotropins (FSH and LH) that occurs approximately between 4 to 25 weeks of age (Amann et al., 1986; Bagu et al., 2006). Diet can have a large effect on puberty attainment in bulls, mainly by hastening and increasing LH pulsatility (Brito et al., 2007a,b,c; Dance et al., 2015; Harstine et al., 2015).

Immunizations against hormones that limit gonadotropin production, such as inhibin, have also been reported and have varying success at accelerating puberty and increasing sperm production later in life (Martin et al., 1991; Kaneko et al., 1993; Bame et al., 1999;

Kaneko et al., 2001). Direct supplementation of gonadotropins has also been examined.

For example, administration of GnRH early in life (from 4 to 8 weeks of age) was determined to reduce the age at puberty and cause positive development in the testes

(Chandolia et al., 1997; Madgwick et al., 2008). Other more recent studies have administered FSH directly and reported similar results. Bagu et al. (2004) treated bull calves from 4 to 8 weeks of age with 10 mg FSH-NIH-S1 every other day. This treatment successfully elevated FSH concentrations post-administration, hastened puberty as determined by a 28 cm scrotal circumference (39.3 vs 44.8 weeks of age), and increased the numbers of Sertoli cells, elongated spermatids, and spermatocytes when measured in histology samples at 56 weeks of age.

104

The concentration of FSH during the prepubertal period has become of interest to our laboratory because of its ability to not only accelerate puberty, but also to positively affect testicular development in a way that could translate into increased mature sperm production. FSH causes Sertoli cells to proliferate during the prepubertal period, but they cease to divide once the blood-testis barrier is established and they attain a mature state preceding puberty (Orth, 1984; Sharpe et al., 2003). Since each mature Sertoli cell hosts a fixed number of germ cells, the timing and rate of Sertoli cell proliferation before puberty has become a topic of focus since the final number of Sertoli cells correlates to the sperm production capacity of the mature bull (Blanchard and Johnson, 1997; Berndtson et al.,

1987).

The basis for the research in this study comes from past projects in our laboratory that have demonstrated the effectiveness of a novel timed-release FSH treatment regimen to positively affect the prepubertal endocrinology and testicular development of bulls. In a past experiment, it was determined that 30 mg of NIH-FSH-P1 (Folltropin-V) in a 2%

HA solution administered twice weekly from 39 to 91 days of age positively affected testicular development and prepubertal endocrinology as evident by increased endogenous FSH production. This novel exogenous FSH treatment has been termed

"FSH-HA." In this dissertation research, two experiments were conducted concurrently in order to examine the effects of this FSH-HA treatment during a different age range.

Although many AI companies are acquiring bulls at younger ages today than in the past, it is more appropriate to begin treatments at a later date than in our prior research.

105

Accordingly, these current projects seek to apply experimental treatments at ages when

AI companies are more likely to be in possession of their bulls.

In this dissertation's other experiment, beef bulls were administered FSH-HA every 3.5 days from 59 to 167.5 days of age. Bulls receiving FSH-HA had increased endogenous FSH production over control animals beginning at 94 days of age. After an examination of activin A concentrations, a mechanism was proposed whereby FSH-HA increases testicular production of activin A by stimulating Sertoli cells prior to (83.5 days of age) the increase in systemic FSH. Overall, we proposed that FSH-HA perpetuates a positive feedback loop between the testes and the upper centers of the brain

(hypothalamus and pituitary). At the conclusion of treatment, FSH-HA bulls had greater number of Sertoli cells per seminiferous tubule cross section, which hinted at the possibility of increased capacity for sperm production. However, this hypothesis was unable to be tested because bulls were castrated in order to examine testicular physiology.

This current experiment seeks to expand upon this knowledge by determining whether

FSH-HA treatment administered from 62 to 170.5 days of age to Holstein bulls results in hastened puberty and increased mature sperm production. We hypothesized that treatment of Holstein bulls every 3.5 days with FSH-HA from 62 to 170.5 days of age would result in increased endogenous FSH concentrations, increased activin A concentrations, hastened age at puberty, and increased mature sperm production.

Together, these two experiments seek to provide a holistic understanding of the endocrinology and physiology resulting from this treatment regimen in prepubertal bulls.

Furthermore, the mechanisms underlying the observed changes in physiology, whether or

106 not this FSH-HA treatment results in the hastening of puberty, and whether mature sperm production is impacted was also examined.

MATERIALS AND METHODS

Animals and handling were conducted in accordance with procedures approved by The Ohio State University Agricultural Animal Care and Use Committee.

Animals and Treatments

A total of 29 Holstein bulls were utilized for this experiment. Initially, the bulls were all potential candidates to be considered for use by Select Sires, Incorporated (Plain

City, Ohio). Therefore, the majority of calves were born at a separate calf rearing facility as a result of embryo transfer (ET) and handled in the same manner prior to the experiment. All experimental methods were performed after the calves' arrival at Select

Sires' rearing facilities. Two different groups of calves underwent the same experimental treatment regimen, but the average birth date of the two groups was staggered approximately three months apart. In general, calves assigned to receive the FSH-HA treatment regimen were genetic culls based on genomic testing that occurred prior to arrival at Select Sires, and oftentimes untreated (control) bulls were siblings of treated bulls. The first group (SS1) contained 16 bulls (control, n = 8; FSH-HA, n =8), and the second group (SS2) contained 13 bulls (control, n = 4; FSH-HA, n = 9). Calves arrived at

Select Sires' housing facilities at an average age of 59 ± 1.7 days of age and were still receiving twice daily feedings of milk replacer (Amplifier Max; 1.9 L/feeding, 22% protein, 20% fat) and receiving free choice calf starter (4 kg/d maximum; 22% protein,

107

2.5% fat). Calves were weaned at 69 ± 1.7 days of age and transitioned to a ration consisting of grower pellets (5 kg/d; 18% protein, 2% fat) and ad libitum access to hay and remained on this diet until receiving a maintenance ration beginning at 278 ± 1.7 days of age consisting of corn silage TMR formulated to provide 1.5 Mcal/kg NEm and

0.9 Mcal/kg NEg when fed at 2.4% of mature BW.

Beginning at 62 ± 1.7 days of age, bulls were injected i.m. with either 30 mg

NIH-FSH-P1 (Folltropin-V; Bioniche Animal Health, Athens, GA) in a 2% hyaluronic acid solution (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days until 170.5 ±

1.7 days of age. Hyaluronic acid (HA) was research grade dried sodium hyaluronate with a molecular weight ranging from 601 – 850 KDa (Lifecore Biomedical, Chaska, MN).

The FSH-HA treatment was formulated by mixing Folltropin-V to 60 mg/ml in a 2% HA solution and delivering 0.5 ml every 3.5 days to achieve the delivery of 30 mg NIH-FSH-

P1 per injection. Control animals received 0.5 ml saline every 3.5 days. Injections were given i.m. in the neck, being sure to deliver the entire bolus of the treatment to one area.

Side of the neck of the treatment administration was recorded and switched the next treatment day to ensure previously placed boluses were not disturbed.

Body Weight and Scrotal Circumference Measurements

Body weights (BW) were recorded every 30 days from 62 to 170.5 days while bulls were examined on a cattle squeeze chute equipped with a digital scale. Scrotal circumference (SC) measurements were recorded for all bulls monthly from 76 to 358 days of age. SCs were obtained with a metal scrotal tape (Sullivan Supply, Inc., Dunlap,

108

IA) by the same technician, being sure to measure the circumference of the testes at their widest point. With the bull properly restrained, the scrotum was grasped at the neck with one hand and the testicles were pulled gently into the bottom of the scrotum. The scrotal tape was placed at the site of the greatest circumference, snugged, and the measurement recorded.

Blood Sample Collection

A blood sample to assess hormone concentrations consisted of a maximum of 10 ml of blood collected from the jugular vein of each animal using a 1 inch, 18 gauge needle and syringe. When both plasma and serum were needed, the 10 ml of blood was divided evenly into two respective Vacutainer tubes intended for either serum or plasma isolation (Becton Dickinson and Company, Franklin Lakes, NJ). A 5 ml sample was collected when only serum was needed for analyses. For serum isolation, glass

Vacutainer tubes were placed on ice until storage at 4°C, allowed to clot for 48 hours at

4°C, and then centrifuged at 7,735 x g for 20 minutes. Serum was harvested and frozen at

-20°C until analyzed for circulating concentrations of FSH or activin A. For plasma, the sample was placed into K2-EDTA Vacutainer tubes, centrifuged at 7,735 x g for 20 minutes, plasma harvested, and samples frozen at -20°C until analyzation for concentration of testosterone.

Regarding the timing of sample collection, serum to measure FSH was regularly collected immediately before each treatment administration every 3.5 days from 62 to

170.5 days of age. For testosterone analysis, blood samples for plasma isolation were

109 collected every 7 days immediately prior to treatment from 62 to 167 days of age. Serum used to determine activin A concentrations came from the same samples used to measure

FSH concentrations on the respective day of analysis.

Hormone Analyses

Follicle-Stimulating Hormone

Concentration of FSH in serum was determined in duplicate for all samples with a double antibody radioimmunoassay (RIA) previously validated for use in measuring bovine FSH (Burke et al., 2003). Briefly, purified ovine FSH (LER1976-A2) was iodinated using the chloramine-T method, primary antibody was rabbit-anti-ovine FSH

(JAD #17-7.6.9), and secondary antibody was donkey-anti-rabbit and normal rabbit serum. All samples were analyzed using the same batch of Iodine-125, but two iodinations were necessary to run all samples. Samples from SS1 (three assays) and SS2

(two assays) were run simultaneously within their respective group, being sure that an individual bull's samples were kept within a singular assay. The sensitivity and coefficient of variations (CV) did not vary significantly between assays performed utilizing the two iodinations, so intra-assay CV, inter-assay CV, and sensitivity estimates include all five assays performed. The average intra-assay coefficient of variation (CV) was 8.7% and the average inter-assay CV was 16.3% for a "low" standard (2.23 ng/ml) and 16.9% for a "high" standard (4.72 ng/ml). The average sensitivity was 0.93 ng/ml.

110

Activin A

An enzyme-linked immunosorbent assay (ELISA) kit (Ansh Labs, Webster, TX) was used to determine activin A in serum. The kit was reported to measure "total" activin

A, which includes all unbound and follistatin-bound activin A. Cross reactivity with inhibin A, inhibin B, and activin AB is 2.8% as reported by the manufacturer. All samples were processed in duplicate and according to manufacturer's recommendations.

The optical density of the plates was measured within 15 minutes of assay completion using a spectrophotometric plate reader at 450 nm (BioTek, Winooski, VT). Activin A concentrations of unknown samples were calculated by plotting the data on a log versus log scale using a cubic regression curve-fit. The lowest limit of detection reported was

0.065 ng/ml, and the measurable range of detection was 0.1 – 10 ng/ml. The average intra-assay and inter-assay CV was 4.9% and 5.2%, respectively.

Testosterone

Concentrations of testosterone in plasma were analyzed in duplicate using a commercial RIA kit (Coat-a-Count, Siemens Medical Solutions Diagnostics, Los

Angeles, CA) previously validated in our laboratory for bovine samples (Burke et al.,

2005). All samples were run in one assay. The average intra-assay CV for unknown samples was 5.1%, and the average intra-assay CV for the low pool (0.01 ng/ml) was

16.5%, medium pool (1.65 ng/ml) was 8.3%, and high pool (10.32 ng/ml) was 8.5%. The sensitivity of the assay was 0.03 ng/ml.

111

Puberty Assessment

Puberty was defined using the parameters established by Wolf et al. (1965) as when the bull is first able to ejaculate 50 million spermatozoa with 10% motility.

Beginning at 244 days of age, trained technicians at Select Sires, Inc. attempted to collect semen from each bull every 7 days using a teaser animal and the artificial vagina (AV) method. When collection was successful, semen was evaluated by the same technician for the percent of motile sperm and concentration using a NucleoCounter SP-100

(ChemoMetec, CA) to determine total sperm numbers. The age at which a bull ejaculated

>50 million spermatozoa with at least 10% motile sperm was defined as the age at puberty for that bull. Bulls were removed from semen collection upon attainment of puberty, with the last bull attaining puberty at 385 days of age (overall average age at puberty was 289 days).

Mature Sperm Production

One control bull was culled prior to the mature sperm collections. Data excludes this bull (control, n = 11; FSH-HA, n = 17). Beginning at 564 days of age, mature semen production was assessed via thrice weekly semen collections using industry standard procedures (teaser animals, false mounts, and collection of two consecutive ejaculates each day with an AV). Data from the first week of collections (3 collections) was not utilized to eliminate biases based on bull adaptation and potential variation due to potential differences in epididymal storage capacities between bulls. Therefore, the first collection included in data analysis was obtained at 571 ± 1.7 days of age, and collections

112 to assess mature production ended at 627 days of age. Total daily sperm production of each animal was considered the total amount of sperm produced in two ejaculates during one collection period.

Statistical Methods

For all data analysis, the replicate of group (SS1 and SS2) was initially included in the model. However, in all variables examined, the variable of group was removed after backward stepwise elimination of independent variables based on a Wald statistic of greater than P < 0.1.

Body weight (BW) and scrotal circumference (SC) of the bulls were analyzed by

ANOVA using the MIXED procedure of SAS (version 9.3, 2010) with repeated measures

(age) analysis and an autoregressive (AR(1)) covariance structure included in the model.

Each model examined the effect of treatment, age, and their interaction on the respective dependent variable, and if either the main effects or their interactions were significant according to Wilks' lambda, differences between least squares means were reported using

Tukey's HSD test.

The concentrations of FSH collected every 3.5 days prior to treatment were analyzed by ANOVA using the MIXED procedure of SAS with repeated measures (age) analysis and initial concentration of FSH (d62) used as a covariate in the model. An autoregressive (AR(1)) covariance structure was applied to the data based on the lowest

Bayesian information criteria (BIC) in a comparison of all covariate structures. Activin A and testosterone concentrations were analyzed by ANOVA using the MIXED procedure

113 of SAS. For testosterone, repeated measures (age) analysis was included in the model and an autoregressive (AR(1)) covariance structure was utilized. For activin A concentrations, repeated measures (age) analysis was also included in the model, but with a spatial power

(SP(POW)) covariance structure based on the most appropriate fit for the data (lowest

BIC).

The age at puberty attainment was analyzed with one-way ANOVA using PROC

GLM procedure of SAS. The mature collections were analyzed using PROC MIXED procedure of SAS, with repeated measures (age) analysis included with an autoregressive

(AR(1)) covariance structure. Additionally, the average daily sperm production during the mature assessment period was compared using one-way ANOVA by PROC GLM of

SAS.

All results are reported in terms of mean ± SE.

RESULTS

There were no differences in body weight between the control and FSH-HA bulls at any of the ages measured. Although FSH-HA bulls had numerically greater SC at all dates measured after the initiation of treatment administration, there were no differences

(Trt*Age, P > 0.1) in SC between FSH-HA and control bulls at any of the ages examined

(Figure 4.1).

Testosterone concentrations measured in plasma increased with age (Age, P <

0.05) but did not differ between treatments at any ages examined (Trt*Age, P > 0.1;

Figure 4.2). In general, bulls did have increases in testosterone concentrations during the

114 age range examined (Age, P < 0.05), having an average concentration of 0.13 ± 0.03 ng/ml at 62 days of age and a concentration of 1.0 ± 0.2 ng/ml at 167 days of age (Figure

4.2).

Concentrations of FSH analyzed from serum collected every 3.5 days prior to treatment administration did not differ between control and FSH-HA bulls from 62 to

93.5 days of age. However, at 97 days of age FSH-HA bulls had greater (P < 0.05) concentrations of FSH than control animals, and this difference was sustained until the end of sampling at 170.5 days of age (Trt*Age, P < 0.05; Figure 4.3). Within the control treatment (Trt, P < 0.05) concentrations of FSH did not change during sampling. Within the FSH-HA bulls, FSH concentrations were increased (P < 0.05) over initial day 62 sampling beginning at 93.5 days of age. Also in FSH-HA bulls, all samples taken after

93.5 days of age were greater (P < 0.05) than day 93.5 FSH concentrations except on day

114.5 (P = 0.06), day 121.5 (P = 0.2), day 125 (P =0.2), and day 170.5 (P = 0.06; Figure

4.3).

The ages at which activin A was analyzed were chosen based on the timing of the

FSH increase in FSH-HA bulls. Foremost, the start (day 62) and end (day 170.5) day of treatment administration were examined. Next, two sampling points (day 86.5 and 107.5) that surround the day of increased FSH concentrations at 97 days of age were chosen.

Lastly, activin A was analyzed at day 139 to represent a midpoint between sampling dates 107.5 and 170.5. Activin A concentrations did not differ in control animals at any age examined (Figure 4.4). However, activin A concentrations (Trt*Age, P < 0.05) in

FSH-HA bulls were greater (P < 0.05) than control bulls at 86.5 and 107.5 days of age.

115

Within the FSH-HA treatment, concentrations of activin A at 86.5 days of age were greater than on day 62, 107.5, 139, and 170.5 (Figure 4.4).

The FSH-HA bulls reached puberty (ability to produce 50 million cells with 10% motility) at a younger age (P < 0.05) than control bulls, with FSH-HA bulls attaining puberty at 278 ± 7.7 days of age and control bulls at 303 ± 9.1 days of age. Regarding the mature collections from 571 to 627 days of age, there were no differences in sperm production between control and FSH-HA bulls on any day (Trt*Age, P = 0.7).

Additionally, there were no differences in the average daily sperm production between control and FSH-HA bulls (Trt, P = 0.5; 5.84 ± 0.11 billion cells).

DISCUSSION

Exogenous FSH administered in the form of the FSH-HA treatment every 3.5 days from 62 to 170.5 days of age resulted in an increase in endogenous FSH concentrations beginning at 97 days of age. This increase in FSH concentrations in FSH-

HA bulls are likely related to increased activin A concentrations observed in FSH-HA bulls when compared to control bulls at 86.5 and 107.5 days of age. These endocrine changes translated into hastened puberty in FSH-HA bulls, with FSH-HA bulls attaining puberty approximately 25 days sooner than control bulls (278 ± 7.7 vs. 303 ± 9.1 days of age). However, when mature sperm production was assessed from 571 to 627 days of age, FSH-HA and control treatments did not differ.

When examining the chronology of prepubertal FSH secretion in bulls, previous reports have been variable. While some reports suggest no substantial changes in FSH

116 concentrations during the prepubertal period (McCarthy et al., 1979; Harstine et al., unpublished), others report a transient increase that begins between 4 to 6 weeks of age and decreases back to basal concentrations between 25 to 32 weeks of age (Amann and

Walker, 1983; Amann et a., 1986; Evans et al., 1993; Rawlings and Evans, 1995;

Aravindakshan et al., 2000; Bagu et al., 2006). Although FSH concentrations in control bulls in this experiment did not change, it cannot be concluded that this observation applies to the entire prepubertal period since treatment administration and blood sampling occurred from 62 to 170.5 days of age (9 to 24 weeks of age), and it is possible that control bulls were experiencing a transient plateau of FSH concentrations. Blood sampling prior to 62 days of age and after 170.5 days of age would have been needed to clarify whether the concentrations of FSH in control bulls experienced transient elevation in FSH as described originally by Amann and Walker (1983) and others. The increase in

FSH observed in the FSH-HA bulls over control bulls in this experiment either represents an increase over an unchanging, basal concentration of FSH as described by McCarthy et al. (1979) and Harstine et al. (2016, under review) or an additional increase above the transient FSH elevation described by Amann and Walker (1983) and others.

The increase in endogenous FSH concentrations in FSH-HA bulls at 97 days of age corresponds to the results of other experiments by our laboratory that utilized the same FSH-HA treatment regimen. The closest source of comparison lies in the concurrent study conducted to this experiment that utilized beef bulls. Bulls were treated with the same FSH-HA treatment from 59 to 167.5 days of age, and FSH-HA bulls had greater FSH concentrations than control bulls beginning at 94 days of age. Interestingly,

117 the number of treatments before an increase in endogenous FSH concentrations in FSH-

HA bulls was eleven in both experiments (35 days of treatment). This highlights the ability of this treatment to affect the endocrinology of both beef and Holstein bulls.

Additionally, it is worth noting the analyses of "intensive" bleeds in the beef bull study provided insight into the differentiation between exogenous versus endogenous FSH cited in this experiment's results. It was demonstrated that while increases in measurable systemic FSH occur after each FSH-HA injection, concentrations return to basal (pre- treatment) concentrations prior the next FSH-HA injection. Hence, the increase in FSH in

FSH-HA bulls observed at 97 days of age is referred to as endogenous and not considered to be from unmetabolized or compounding exogenous FSH from FSH-HA treatment.

We believe that the increase in FSH concentrations in FSH-HA bulls at 97 days of age is due to the effect of exogenous FSH on activin A production. Activins in bulls have been reported to be at their highest concentrations after birth, with concentrations decreasing at a similar time to the decrease of the gonadotropins between 25 and 32 weeks of age (Barakat et al., 2008). Unchanging concentrations of activin A in control animals may correspond to this report. However, sampling occurring after the last date

(170.5 days) of blood collection in this experiment would be necessary to determine whether activin concentrations decreased in control bulls. Activin A concentrations did not differ between control and FSH-HA bulls at 62, 139, or 170.5 days of age. However, there were differences present in activin A concentrations between control and FSH-HA bulls at 86.5 and 107.5 days of age, and these differences may provide insight into a mechanism whereby increased activin A causes increased endogenous FSH secretion.

118

The largest difference in activin A concentrations in FSH-HA bulls versus control bulls occurred at 86.5 days of age (0.36 ± 0.06 versus 0.97 ± 0.03 ng/ml). The sampling at 86.5 days of age occurred 10.5 days prior to the first observed FSH increase in FSH-HA bulls.

Additionally, the increase in activin A in FSH-HA bulls was maintained through the next sampling at 107.5 days of age (10.5 days after the FSH increase). Activin A production does occur in Sertoli cells (Anderson et al., 1998; Buzzard et al., 2004), and activin A is known to stimulate FSH secretion from the pituitary (Ling et al., 1986; Vale et al., 1986).

Mechanistically, the binding of activin to ACVR2A receptors in the pituitary causes upregulation of SMAD transcription factors, and this pathway can cause upregulation of

FSH-beta promoter within pituitary gonadotrope cells in vitro (Suszko et a., 2003; Suszko et al., 2005). Furthermore, Smad3 deficiency in mice causes a large reduction in FSH- beta subunit mRNA levels (Coss et al., 2005). Overall, we propose that exogenous FSH-

HA treatment initiates a positive feedback loop starting with increased activin A production by Sertoli cells, and this in turn stimulates endogenous secretion of FSH from the pituitary. This possible mechanism is further supported by unpublished data generated within our laboratory demonstrating that FSH-HA treatment result in increased numbers of Sertoli cells. It is conceivable that a larger population of Sertoli cells could produce increased concentrations of activin.

Bulls receiving FSH-HA attained puberty at 278 ± 7.7 days of age and control bulls at 303 ± 9.1 days of age. These ages are similar or slightly earlier than other reports which utilized Holstein bulls, including Killian and Amann (1972; 288 days), Harstine et al. (2015; 310 days), and Dance et al. (2015; 324 to 369 days). The hastening of puberty

119 for FSH-HA bulls in this study was likely caused by accelerated testicular development and the positive effects of the FSH-HA treatment on their endocrinology. Our other concurrent study and a prior study currently under review demonstrated that this treatment regimen results in greater numbers of Sertoli cells per testis at the conclusion of treatment. It is unknown whether the increases in Sertoli cells at 170.5 days of age represent a hastening of maturation or an increase in Sertoli cell numbers that would be sustained into adulthood since the bulls in this experiment were not castrated. If a hastening of maturation was occurring, meaning that bulls eventually attain the same numbers of Sertoli cells regardless of treatment, then it is likely that increases in mature sperm production would not be observed in FSH-HA bulls. Indeed, there were no statistical differences in the daily sperm production during the mature collection assessments that took place in this experiment. As previously described, the timing of treatment administration in this experiment occurred within the window of time when

FSH concentrations are reported to be elevated in the bull from approximately 4 to 25 weeks of age (Amann et al., 1986). Treatments took place from 62 to 170.5 days of age

(approximately 9 to 24 weeks of age), and increases in FSH concentrations observed in the FSH-HA bulls were therefore either increases above already elevated FSH or an increase above an unchanging basal (control) concentration of FSH such as reported in

McCarthy et al. (1979). Therefore, the FSH-HA treatment is not hastening or extending the proposed increase in FSH reported by Amann et al. (1986). Accordingly, we propose that FSH-HA is likely accelerating puberty via the hastening of testicular maturation.

120

Although statistical differences in mature sperm production were not observed, numerical differences did exist between treatments (control, 5.58 ± 0.64 billion cells/day versus FSH-HA, 6.03 ± 0.38 billion cells/day). Perhaps a larger sample population would have yielded both numerical and statistical differences based on the results observed in this study. Regardless, the production of 400 million more sperm per day is not insignificant in terms of increase in saleable product. Depending on packaging concentration per straw, FSH-HA bulls could theoretically produce an extra 16 to 27 straws of semen per day compared to control animals. Scrotal circumference was numerically greater in FSH-HA than control bulls by 1 cm at the last day of measurement at 358 days of age (Figure 4.1), and while comparisons of SC were not taken during the mature sperm collection period, if this difference in testis size remained then perhaps

FSH-HA bulls had larger testes in maturity. Increased SC could be due to differences in testicular physiology induced by FSH-HA treatments such as an increased number of

Sertoli cells such as was observed in the beef bulls from this dissertation's other study and previous experiment under review for publication. To prove this hypothesis in a study such as this, castration and histological examination would have been necessary in order to categorize differences in Sertoli cell or germ cell numbers. This was not possible for the genetically valuable and marketable control animals used in this study.

The observed hastening of puberty and numerical increases in daily sperm production are compelling findings of this study due to the economic impacts they could have for AI companies and the cattle industries in general. While theoretical monetary figures will not be suggested at this time, the general impacts of such results are worth

121 discussing. The advantages of a 25 day decrease in the ability of an AI company to obtain genetic material from a sire are twofold. Foremost, it is common for the first collections from a genetically superior bull to be used to create pregnancies which may result in the next crop of sires. Although seemingly inconsequential, a 25 day decrease in the attainment of puberty represents an important advantage in the competitive genetic "race" that exists amongst companies that market sires and the obvious impact on generation interval. Secondly, the productive life of bull begins once he is able to produce enough semen of acceptable quality that it can be sold. A hastening of the first collections from a bull would likely result in earlier marketing of his genetics to the industry. Regarding the numerical differences in mature sperm production realized in FSH-HA bulls, an increase of 400 million sperm per day could have large impacts in the dissemination of a bull's genetics. When applied across the many dozens of bulls owned by the average major AI organization, this increase could manifest in large increases in efficiency of overall semen production, as well as the ability to better serve customers by increasing the availability of the bulls that possess the best genetics. It is important to reiterate that only numerical, not statistical, differences were observed for this data. Perhaps if a larger sample size were used, statistical differences would have been observed. Regardless, the economic aspects of these finding are an important aspect of discussion for this

Dissertation.

In summary, the results of this experiment demonstrate the positive effects that a novel FSH-HA treatment can have in Holstein bulls when treated every 3.5 days with

FSH-HA from 62 to 170.5 days of age. Overall, in this experiment, treatment of Holstein

122 bulls with 30 mg NIH-FSH-P1 (FSH-HA) every 3.5 days from 62 to 170.5 days resulted in increased concentrations of endogenous FSH at 97 days of age. This increase in FSH secretion at 97 days of age was attributed to an increased production of activin A detected at 86.5 days of age. These endocrine changes manifested themselves as an earlier attainment of puberty and a numerical (albeit not statistical) increase in daily mature sperm production. The impacts of this research are considerable for not only the AI organizations marketing sires, but also for the dairymen and beef producers who rely on a supply of superior genetics to produce a saleable product. In coordination with the other experiment in this dissertation, the results of these studies provide a better holistic understanding of how the FSH-HA treatment regimen positively affects the prepubertal endocrinology and physiology of bulls. Additionally, this experiment has yielded a better understanding of how these endocrine and physiological changes induced by FSH-HA treatment create lasting positive changes in terms of puberty attainment and mature sperm production capacity.

123

40.0 Control 35.0 FSH-HA 30.0

25.0

20.0

15.0 Scrotal Circumference (cm) Circumference Scrotal 10.0 76 104 132 163 209 244 290 323 358 Age (d)

Figure 4.1. Mean (±SE) scrotal circumference (SC) of Holstein bulls treated with either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days from 62 to 170.5 days of age. Values did not differ between treatments across ages.

124

2.0 Control 1.6 FSH-HA 1.2

0.8

Testosterone (ng/mL) Testosterone 0.4

0.0

Age (d)

Figure 4.2. Systemic testosterone concentrations of bulls treated with either 30 mg NIH- FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days from 62 to 170.5 days of age. Concentrations were not different between treatments across age (Trt*Age, P > 0.1), but concentrations did increase with age (Age, P < .05).

125

5.0 control 4.5 * FSH-HA 4.0 3.5 3.0

2.5 FSH (ng/ml) FSH 2.0 1.5 1.0

Age (days)

Figure 4.3. Systemic concentrations of FSH immediately before bulls received either 30 mg NIH-FSH-P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days from 62 to 170.5 days of age. Concentrations of FSH (Trt*Age, P < 0.05) were increased (P < 0.05) in the FSH-HA over control treatment beginning at 97 days of age (denoted by asterisk). Within the control treatment (Trt, P < 0.05) concentrations of FSH did not change during the sampling period. Within the FSH-HA bulls (Trt, P < 0.05), FSH concentrations were increased (P < 0.05) over initial day 62 sampling beginning at 93.5 days of age. Within FSH-HA bulls, all samples taken after 93.5 days of age were greater (P < 0.05) than day 93.5 FSH concentrations except on day 114.5 (P = 0.06), day 121.5 (P = 0.2), day 125 (P =0.2), and day 170.5 (P = 0.06).

126

1.4 c control 1.2 b FSH-HA 1

0.8 ab 0.6 ab ab ab Activin A (ng/ml) A Activin 0.4 a ab ab a 0.2

0 62 86.5 107.5 139 170.5 Age (d)

Figure 4.4. Concentrations of activin A in bulls which received either 30 mg NIH-FSH- P1 in 2% hyaluronic acid (FSH-HA, n = 17) or saline (control, n = 12) every 3.5 days from 62 to 170.5 days of age. Concentrations of FSH (Trt*Age, P < 0.05) were increased (P < 0.05) in the FSH-HA over control treatment at 86.5 and 107.5 days of age. Within the FSH-HA treatment (Trt, P < 0.05), the concentration at 86.5 days of age were greater (P < 0.05) than concentrations measured at 62, 107.5, 139, and 170.5 days of age. Concentration of activin A in the control treatment was not different at any of the ages examined. Values not sharing subscripts differ (P < 0.05) from one another.

127

CHAPTER 5

GENERAL DISCUSSION

The projects encompassed in this doctoral research could be considered extensions of research completed for my previous Master's thesis. Many of the comments in the General Discussion of the M.S. thesis still hold true for the current research:

"An overall goal of this thesis research was to examine the efficacy of technologies that may advance puberty and enhance sperm production in bulls. The benefits of such technologies are twofold. Foremost, advancements in either of these objectives represents monetary benefits to both AI companies and cattle producers.

Successful implementation of these technologies means AI companies will have more saleable product from a bull at an earlier time, and producers may be less limited in the product they wish to use. However, a second and possibly more valuable benefit results from the genetic advancements the industry may experience by increasing use of young, genomically-superior bulls. Increasing the availability of the best genetics from young bulls will allow for expedited improvement of genetic traits that are useful in today's cattle."

In the thesis research that formed a foundation for these current projects, an exogenous FSH treatment was developed utilizing 30 mg porcine FSH (Folltropin-V)

128 suspended in a 2% hyaluronic acid solution to prolong or slow the delivery of the hormone. This treatment was termed FSH-HA, and due to the "time-release" aspect of the treatment, administration occurred every 3.5 days. This FSH-HA treatment was administered to bulls from 35 to 91 days of age and showed promise in its ability to affect sperm production positively by increasing the number of Sertoli cells per testis at 93 days of age. The cause of this increase in Sertoli cells was attributed to the FSH-HA treatment's ability to increase endogenous FSH concentrations beginning at 70 days of age. A major output of this study was the verification of the usefulness of this FSH-HA treatment, but it was also explained that the mechanisms underlying the increase in endogenous FSH needed to be elucidated in subsequent experiments. Additionally, it was deemed important to determine whether administration of FSH-HA treatment would positively impact puberty attainment and mature sperm production.

Accordingly, the main objectives of this dissertation research were to administer the same FSH-HA treatment at an age range feasible for incorporation in a production setting, seek to better understand the endocrinology and physiology that is associated with FSH-HA treatment, and to determine how FSH-HA affects the age at puberty attainment and mature sperm production.

In the first experiment, beef bulls were administered FSH-HA every 3.5 days from 59 to 167.5 days of age in order to examine the endocrine profiles of key hormones as well as to examine the testicular histology that resulted from treatment. In both the first and second experiments, this age range (~60 to 170 days of age) was deemed most appropriate for treatment administration since many AI companies acquire their bulls

129 when they are approximately 2 months old if they are not born on-site, and the timing of the treatment occurs when the endocrinology and physiology of bulls is still extremely plastic and able to the manipulated. In this first experiment, the FSH-HA bulls had increased endogenous FSH concentrations beginning at 94 days of age and increased numbers of Sertoli cells per testis when castration occurred at 170 days of age.

Importantly, further analyses of blood determined that FSH-HA bulls had increased activin A concentrations at 83.5 and 94 days of age, and these ages are likely related to the increase in FSH observed at 94 days of age. We proposed that exogenous FSH provided by the FSH-HA treatment provides a positive stimulus to the pituitary-testis axis by increasing the numbers of Sertoli cells which in turn produce activin A, thus signaling the pituitary to further produce and secrete endogenous FSH. While potential negative inputs to this positive feedback loop (such as inhibin) were not examined in this current project, it is clear that the FSH-HA treatment is effectively able to overcome such negative factors as evidence by the increase in Sertoli cell numbers and endogenous FSH production.

While the abovementioned experiment sheds insight into the endocrinology underlying the physiological effects of FSH-HA, a second experiment was performed with the objectives of determining if FSH-HA treatment has positive impacts on puberty attainment and mature sperm production. Holstein bulls that were essentially genetic culls of Select Sires, Inc. received the FSH-HA treatment from 62 to 170.5 days of age. In similar fashion to all previous reports, treatment every 3.5 days with FSH-HA resulted in increased endogenous FSH concentrations at 97 days of age and increased activin A

130 concentration at sampling points of 86.5 and 107.5 days of age. Importantly, FSH-HA bulls attained puberty 25 days earlier than control bulls. While FSH-HA bulls made slight numerical increases in sperm production in maturity (571 to 627 days of age), the data were not statistically significant. This project showcases the FSH-HA treatment's ability to hasten puberty, and although 25 days may seem inconsequential, this represents large gains in advancing genetic distribution and profitability for the AI companies that own these bulls.

Many extensive studies documenting the prepubertal endocrinology of bulls have been published. Hopefully, the findings of this current research contribute new knowledge to this area. Specifically, the literature available on activin A concentrations and potential links to prepubertal FSH concentrations and testis development are lacking.

In this sense, these reports may provide basic information to the field as well as an in- depth description of how prepubertal endocrinology and physiology of the bull can be positively manipulated. This type of research will remain valid until drastic changes are made to the commercial aspect of semen production. These changes could include concepts such as in vitro spermatogenesis or working with the induced differentiation of embryonic stem cells to create pregnancies that hasten the normal generation interval.

A logical next step to this dissertation research is to see if the FSH-HA treatment regimen could be incorporated in a production setting. This may include approval by legislative entities as well as a need to better understand the public's perception of such technologies. As stated in this dissertation's Introduction, the world population reached

7.3 billion people in 2015 and is expected to reach 9.7 billion by 2050. Increased

131 efficiency of food production is as important now as it has ever been. Within the cattle industries, technologies such as those researched in this dissertation will continue to be crucial in order to meet a growing population's demand for food.

132

LITERATURE CITED

Akbar, A. M. T., T. M. Nett, and G. D. Niswender. 1974. Metabolic clearance and secretion rates of gonadotropins at different stages of estrous cycle in ewes. Endo. 94:1318-1324.

Allan, C. M., A. Garcia, J. Spaliviero, F. P. Zhang, M. Jimenez, I. Huhtaniemi, and D. J. Handelsman. 2004. Complete Sertoli cell proliferation induced by follicle-stimulating hormone (FSH) independently of luteinizing hormone activity: evidence from genetic models of isolated FSH action. 145(4):1587-1593.

Amann, R. P. 1970. Sperm production rates. In: Johnson, A. D., W. R. Gomes, and N. L. Van Demark. (Eds.), The Testis, vol 1. Academic Press Inc., New York, NY. pp. 433- 482.

Amann, R. P. and J. O. Almquist. 1976. Bull management to maximize sperm output. Proc. 6th Tech. Conf. on Artificial Insemination and Reprod. p 1.

Amann, R. P. 1983. Endocrine changes associated with onset of spermatogenesis in Holstein bulls. J. Dairy. Sci. 66:2606-2622.

Amann, R. P. and O. J. Walker. 1983. Changes in the pituitary-gonadal axis associated with puberty in Holstein bulls. J. Anim. Sci. 57:422-433.

Amann, R. P., M. E. Wise, J. D. Glass and T. M. Nett. 1986. Prepubertal changes in the hypothalamic-pituitary axis of Holstein bulls. Biology of Repro. 34:71-80.

Amann, R. P. and J. M. DeJarnette. 2012. Impact of genomic selection of AI dairy sires on their likely utilization and methods to estimate fertility: A paradigm shift. Therio. 77:795-817.

Amstalden, M., M. R. Garcia, S. W. Williams, R. L. Stanko, S. E. Nizielski, C. D. Morrison, D. H. Keisler, and G. L. Williams. 2000. gene expression, circulating leptin, and luteinizing hormone pulsatility are acutely responsive to short-term fasting in prepubertal heifers: relationships to circulating insulin and insulin-like growth factor I. Biol. Reprod. 63(1):127-133.

133

Amstalden, M., R. C. Cardoso, B. R. C. Alves, and G. L. Williams. 2014. Reproduction Symposium: Hypothalamic neuropeptides and the nutritional programming of puberty in heifers. J. Anim. Sci. 92:3211-3222.

Anderson, R. A., L. W. Evans, D. S. Irvine, M. A. McIntyre, N. P. Groome, and S. C. Riley. 1998. Follistatin and activin A production by the male reproductive tract. Human Reprod. 13(12):3319-3325.

Arambepola, N., D. Bunick, and P. Cooke. 1998. Thyroid Hormone and Follicle- Stimulating Hormone Regulate Mullerian-Inhibiting Substances Messenger Ribonucleic Acid Expression in Cultured Neonatal Rat Sertoli Cells. 139, 11:4489-4495.

Aravindakshan, J .P., A. Honaramooz, P. M. Bartlewski, A. P. Beard, R. A. Pierson, and N. C. Rawlings. 2000. Gonadotropin secretion in prepubertal bull calves born in spring and autumn. J. Reprod. and Fert. 120:159-167.

Arije G. F. and J. N. Wiltbank. 1971. Age and weight at puberty in Hereford heifers. J. Anim. Sci. 33:401-406.

Armstrong, D. T., P. Holm, B. Irvin, B. A. Petersen, R. B. Stubbings, D. McLean, G. Stevens, and R. F. Seamark. 1992. Pregnancies and live birth from in vitro fertilization of calf oocytes collected by laproscopic follicular aspiration. Therio. 38:667-678.

Atanassova, A-M., C. McKinnell, M. Walker, K. Turner, J. Fisher, M. Morley, M. Millar, N. Groome, and R. Sharpe. 1999. Permanent effects of neonatal estrogen exposure in rats on reproductive hormone levels, Sertoli cell number, and the efficiency of spermatogenesis in adulthood. Endo. 140:5364-5373.

Attisano, L., J. L. Wrana, S. Cheifetz, and J. Massague. 1992. Novel activin receptors: distinct genes and alternative mRNA splicing generate a repertoire of serine/threonine kinase receptors. Cell. 68:97-108.

Bagley, C. P., D. G. Morrison, J .I. Feazel, and A. M. Saxton. 1989. Growth and sexual characteristics of suckling beef calves as influenced by age at castration and growth implants. J. Anim. Sci. 67(5):1258-1264.

Bagu, E. T., S. Madgwick, R. Duggavathi, P. Bartlewski, D. Barrett, S. Huchkowsky, S. Cook, and N. Rawlings. 2004. Effects of treatments with LH or FSH from 4 to 8 weeks of age on the attainment of puberty in bull calves. Therio. 62:861-873.

Bagu, E. T., S. Cook, C. L. Gratton, and N. C. Rawlings. 2006. Postnatal changes in testicular gonadotropin receptors, serum gonadotropin, and testosterone concentrations and functional development of the testes in bulls. Reprod. 132:403-411.

134

Bame, J., J. C. Dalton, S. Degelos, T. Good, J. Ireland, F. Jimenez-Krassel, T. Sweeny, R. G. Saacke, and J. J. Ireland. 1999. Effect of Long-Term Immunization against Inhibin on Sperm Output in Bulls. Bio. Repro. 60:1360-1366.

Barakat, B., A. E. O'Conner, E. Gold, D. M. de Kretser, and K. L. Loveland. 2008. Inhibin, activin, follistatin and serum FSH levels and testicular production are highly modulated during the first spermatogenic wave in mice. Reprod. 136:345-359.

Barrionuevo, F., M. Burgos, and R. Jimenez. 2011. Origin and function of embryonic Sertoli cells. Biomol. Concepts. Vol. 2. pp. 537-547.

Barth, A. D. 2004. Pubertal development of Bos taurus beef bulls. Proc. WBC Congress, Quebec, Canada.

Berardinelli, J. G., J. J. Ford, R. K. Christenson and L. L. Anderson. 1984. Luteinizing hormone secretion in ovariectomized gilts: effects of age, reproductive state and estrogen replacement. J. Anim. Sci. 58:165- 73.

Bernard, D. J. and S. Tran. 2013. Mechanisms of activin-stimulated FSH synthesis: The story of a pig and a FOX. Biol. Reprod. 88(3):78, 1-10.

Berndtson, W., G. Igboeli, and W. Parker. 1987. The Number of Sertoli cells in mature Holstein bulls and their relationship to quantitative aspects of spermatogenesis. Biol. Reprod. 37(1):60-67.

Blanchard, T and L. Johnson. 1997. Increased germ cell degeneration and reduced germ cell number: Sertoli cell ratio in stallions with low sperm production. Therio. 47(4):665- 677.

Blood, D. C. and V. P. Studdert. 1999. Saunders Comprehensive Veterinary Dictionary. Second Edition. W.B. Saunders, London, p. 942.

Boichard, D., V. Ducrocq, S. Fritz, and J. Colleau. 2010. Where is Dairy Cattle Breeding Going? A Vision of the Future. Interbull Bulletin. 41:63-67.

Boitani, C., M. Stefanini, and A. R. Morena. 1995. Activin stimulates Sertoli cell proliferation in a defined period of rat testis development. Endo. 136(12):5438-5444.

Brito, L.F.C., A.D. Barth, N.C. Rawlings, R.E. Wilde, D.H. Crews, Y.R. Boisclair, R.A. Ehrhardt, and J.P. Kastelic. 2007a. Effect of feed restriction during calfhood on serum concentrations of metabolic hormones, gonadotropins, testosterone, and on sexual development in bulls. Reprod. 134:171-181.

135

Brito, L.F.C., A.D. Barth, N.C. Rawlings, R.E. Wilde, D.H. Crews, P.S. Mir, and J.P. Kastelic. 2007b. Effect of nutrition during calfhood and peripubertal period on serum metabolic hormones, gonadotropins and testosterone concentrations, and on sexual development in bulls. Domest. Anim. Endo. 33:1-18.

Brito, L.F.C. A.D. Barth, N.C. Rawlings, R.E. Wilde, D.H. Crews, P.S. Mir, and J.P. Kastelic. 2007c. Effect of improved nutrition during calfhood on serum metabolic hormones, gonadotropins, and testosterone concentrations, and on testicular development in bulls. Domest. Anim. Endo. 33:460-469.

Burke, C. R., M. L. Mussard, C. L. Gasser, D. E. Grum, and M. L. Day. 2003. Estradiol benzoate delays new follicular wave emergence in a dose-dependent manner after ablation of the dominant in cattle. Therio. 60:647-658.

Burke, C. R., H. Cardenas, M. L. Mussard, and M. L. Day. 2005. Histological and steroidogenic changes in dominant follicles during oestradiol-induced atresia in heifers. Reprod. 129:611-620.

Buzzard, J. J., P. G. Farnworth, D. M de Kretser, A. E. O'Connor, N. G. Wreford, and J. R. Morrison. 2003. Proliferative phase of Sertoli cells display a developmentally regulated response to activin in vitro. Endo. 144(2):474-483.

Buzzard, J. J., K. L. Loveland, M. K. O'Bryan, A. E. O'Conner, M. Bakker, T. Hayashi, N. G. Wreford, J. R. Morrison, and D. M. Kretser. 2004. Changes in circulating and testicular levels of inhibin A and B and Activin A during postnatal development in the rat. Endo. 145(7):3532-3541.

Campbell, R. E., G. Gaidamaka, S. K. Han and A. E. Herbison. 2009. Dendro-dendritic bundling and shared synapses between gonadotropin-releasing hormone neurons. Proceedings Natl Acad. Sci. U.S.A. 106(26):10835-10840.

Cassell, B. Genetic improvements using young sires with genomic evaluations. Virginia Coop. Extension. Publications and Educational Resources. Pub. 404-090. https://pubs.ext.vt.edu/404/404-090/404-090.html

CDCB. 2015. CDCB – Activity Report. 2015 Industry Meeting. Accessed 8 Sept. 2016. https://www.cdcb.us/News/Joao%20Durr.pdf

Chandolia, R. K., A. C. O. Evans, and N. C. Rawlings. 1997a. Effect of inhibition of increased gonadotropin secretion before 20 wk of age in bull calves on testicular development. J. Reprod. Fert. 109:65-71.

136

Chandolia, R. K., A. Honaramooz, B. C. Omeke, R. Pierson, A. P. Beard, and N.C. Rawlings. 1997b. Assessment of development of the testes and accessory sex glands by ultrasonography in bull calves and associated endocrine changes. Therio. 48:119-132.

Chandolia, R. K. A. Honaramooz, P. M. Bartlewski, A. P. Beard, and N. C. Rawlings. 1997c. Effects of treatment with LH releasing hormone before the early rise in LH secretion on endocrine and reproductive development in bull calves. J. Reprod. Fert. 111:41-50.

Courot, M., M. T. Hochereau de Reviers, and R. Ortavant. 1970. Spermatogenesis. In: The Testes. Vol. 1. Academic Press, New York. pp. 339-431.

Corrigan, A. Z., L. M. Bilezikjian, R. S. Carroll, L. N. Bald, C. H. Schmelzer, B. M. Fendly, A. J. Mason, W. W. Chin, R. H. Schwall, and W. Vale. 1991. Evidence of an autocrine role of activin B within rat anterior pituitary cultures. Endo. 128:1682-1684.

Coss, D., V. G. Thackray, C. X. Deng, and P. L. Mellon. 2005. Activin regulates luteinizing hormone beta-subunit gene expression through Smad-binding and homeobox elements. Mol. Endo. 19:2610-2623.

Coulter, G.H. 1986. Puberty and postpubertal development of beef bulls. In Current Therapy In Theriogenology. Edited by Morrow, D.A., W.B. Saunders, Philadelphia. 142- 148.

Cupps, P. T. 1991. Reproduction in Domestic Animals. 4th ed. Academic Press, Inc., San Diego, p. 448.

Curtis, S. K. and R. P. Amann. 1981. Testicular development and establishment of spermatogenesis in Holstein bulls. J. Anim. Sci. 53:1645-1657. d'Anglemont de Tassigny, X. 2010. The Role of Kisspeptin Signaling in Reproduction. Physiology. 25(4):207-217. doi: 10.1152/physiol.00009.2010

Dadoune, J. P. and A. Demoulin. 1993. "Structure and functions of the testis" in Reproduction in Mammals and Man. C. Thibault, M.C. Levasseur and R.H.F. Hunter. Ellipses, Paris.

Dance, A., J. Thundathil, R. Wilde, P. Blondin and J. Kastelic. 2015. Enhanced early-life nutrition promotes hormone production and reproductive development in Holstein bulls. J. Dairy Sci. 98:1-12.

Davis Rincker, L. E., M. J. Vadehaar, C. A. Wolf, J. S. Liesman, L. T. Chapin, and M. S. Weber Nielsen. 2011. Effect of intensified feeding of heifer calves on growth, pubertal age, calving age, milk yield, and economics. J. Dairy Sci. 94(7):3554-3567. 137

Day, M. L., K. Imakawa, M. Garcia-Winder, D. D. Zalesky, B. D. Schanbacher, R. J. Kittok, and J. E. Kinder. 1984. Endocrine mechanisms of puberty in heifers: estradiol negative feedback regulation of luteinizing hormone secretion. Biol. Reprod. 31(2):332- 341.

Day, M. L., K. Imakawa, P. L. Wolfe, R. J. Kittok, and J. E. Kinder. 1987. Endocrine mechanisms of puberty in heifers. Role of hypothalamo-pituitary estradiol receptors in the negative feedback of estradiol on luteinizing hormone secretion. Biol. Reprod. 37(5):1054-1065.

DeJarnette, J.M., C.E. Marshall, R.W. Lenz, D.R. Monke, W.H. Ayars and C.G. Sattler. 2004. Sustaining the fertility of artificially inseminated dairy cattle: The role of the artificial insemination industry. J. Dairy Sci. 87:(E.Suppl.):E93-E104.

Delemarre-Van de Waal, H. A. 1993. Induction of testicular growth and spermatogenesis by pulsatile, intravenous administration of gonadotrophin-releasing hormone in patients with hypogonadotrophic hypogonadism. Clin. Endo. (Oxford). 38(5):473-480.

Desroziers, E., J. D. Mikkelsen, A. Duittoz, and I. Franceschini. 2012. Kisspeptin- immunoreactivity changes in a sex- and hypothalamic-region-specific manner across rat postnatal development. J. Neuroendo. 24:1154–1165. de Kretser, D. M. and D. M. Robertson. 1989. The isolation and physiology of inhibin and related proteins. Biol. Reprod. 40:1899-1903. de Kretser, D. M., M. P. Hedger, K. L. Loveland, and D. J. Phillips. 2002. Inhibins, activins and follistatin in reproduction. Hum. Reprod. Update. 8(6):529-541. de Winter, J. P., H. M. Vanderstichele, M. A. Timmerman, L. J. Blok, A. P. Themmen, and F. H. de Jong. 1993. Activin is produced by rat Sertoli cells in vitro and can act as an autocrine regulator of Sertoli cell function. Endo. 132:975-982. de Winter, J. P., H. M. Vanderstichele, G. Verhoeven, M. A. Timmerman, J. G. Wesseling, and F. H. de Jong. 1994. Peritubular myoid cells from immature rat testes secrete Activin A and express activin receptor type II in vitro. Endo. 135:759-767.

D'Occhio, M. J. and B. P. Setchell. 1984. Pituitary and testicular responses in sexually mature bulls after intravenous injections of graded doses of LH-releasing hormone. J. Endo. 103:371-376.

Dong, Q., and M. P. Hardy. 2004. Leydig Cell Function in Man. In: Winters, S., eds. Male Hypogonadism: Basic, Clinical, and Therapeutic Principles. pp. 23-43.

138

Donovan, B. T. and J. J. van der Werff ten Bosch. 1965. Physiology of Puberty. Edward Arnold Ltd., London.

Duby, R. T., P. Damiani, C. R. Looney, R. A. Fissore, and J. M. Robl. 1996. Prepubertal calves as oocyte donors: Promises and problems. Therio. 45(1):121-130.

Duchemin S. I., C. Colombani, A. Legarra, G. Baloche, H. Larroque, J. M. Astruc, F. Barillet, C. Robert-Granié, and E. Manfredi. 2012.Genomic selection in the French Lacaune dairy sheep breed. J. Dairy Sci. 95:2723–2733.

Dugger, B. N., J. A. Morris, C. L. Jordan, and S. M. Breedlove. 2008. Gonadal steroids regulate neural plasticity in the sexually dimorphic nucleus of the preoptic area of adult male and female rats. Neuroendo. 88:17-24.

Dunkel, L., H. Alfthan, U. H. Stenman, G. Selstam, S. Rosberg, and K Albertsson- Wikland. 1992. Developmental changes in 24-hour profiles of luteinizing hormone and follicle-stimulating hormone from prepubertal to midstages of puberty boys. J. Clin. Endo. & Metab. 74:890-897.

Esch, F. S., S. Shimasaki, M. Mercado, K. Cooksey, N. Ling, S. Ying, N. Ueno, and R. Guillemin. 1987. Structural characterization of follistatin: a novel follicle-stimulating hormone release-inhibiting polypeptide from the gonad. Mol. Endo. 1(11):849-855.

Escott, G. M., L. A. da Rosa, and E. da Silviera Loss. 2014. Mechanisms of Hormonal Regulation of Sertoli Cell Development and Proliferation: A Key Process for Spermatogenesis. Current. Mol. Pharm. 7:96-108.

Evans, A. C . O., W. D. Currie, and N. C. Rawlings. 1993. Opiodergic regulation of gonadotropin secretion in the early prepubertal calf. J. Reprod. and Fert. 99:45-51.

Evans, A. C. O, F. J. Davies, L. F. Nasser, P. Bowman, and N.C. Rawlings. 1995. Differences in early patterns of gonadotropin secretion between and early and late maturing bulls, and changes in semen characteristics at puberty. Therio. 43:569-578.

Fallah-Rad, A. H., M. L. Connor, and R. P. Del Vecchio. 2001. Effect of transient early hyperthyroidism on onset of puberty in Suffolk ram lambs. Reprod. 121(4):639-646.

Fernandez-Guasti, A., F. P. Kruijver, M. Fodor, and D. F. Swaab. 2000. Sex differences in the distribution of androgen receptors in the human hypothalamus. J. Comp. Neurol. 425:422-435.

Flipse, R. J. and J. O. Almquist. 1961. Effect of total digestible nutrient intake from birth to four years of age on growth and reproductive performance of dairy bulls. J. Dairy Sci. 44:905. 139

Ford, J. J., T. H. Wise, D. D. Lunstra, and G.A. Rohrer. 2001. Interrelationships of porcine X and Y chromosomes with pituitary gonadotropins and testicular size. Biol. Reprod. 65:906-912.

Foster, D. L. and K. D. Ryan. 1979. Endocrine mechanisms governing transition into adulthood: a marked decrease in inhibitory feedback action of estradiol on tonic secretion of luteinizing hormone in the lamb during puberty. Endo. 105:896-904.

Foster, C. M., D. J. Phillips, T. Wyman, L. W. Evans, N. P. Groome, and V. Padmanabhan. 2000. Changes in serum inhibin, activin and follistatin concentrations during puberty in girls. Human. Reprod. 15(5):1052-1057.

Funk, D. A. 2006. Major advances in globalization and consolidation of the artificial insemination industry. J. Dairy Sci. 89:1362-1368.

Garcia-Ruiz, A., J. B. Cole, P. M. VanRaden, G. R. Wiggans, F. J. Ruiz-Lopez, and C. P. Van Tassell. 2016. Changes in genetic selection differentials and generation intervals in US Holstein dairy cattle as a result of genomic selection. Proc. Nat. Acad. of Sci. 113, 28:E3995-E4004. doi: 10.1073/pnas.1519061113

Gasser, C. L., E. J. Behlke, D. E. Grum, and M. L. Day. 2006a. Effect of timing of feeding a high-concentrate diet on growth and attainment of puberty in early-weaned heifers. J. Anim. Sci. 84:3118–3122.

Gasser, C. L., G. A. Bridges, M. L. Mussard, D. M. Dauch, D. E. Grum, J. E. Kinder and M. L. Day. 2006b. Induction of precocious puberty in heifers III: Hastened reduction of estradiol negative feedback on secretion of luteinizing hormone. J. Anim. Sci. 84:2050– 2056.

Gasser, C. L., C. R. Burke, M. L. Mussard, E. J. Behlke, D. E. Grum, J.E. Kinder and M. L. Day. 2006c. Induction of precocious puberty in heifers II: Advanced ovarian follicular development. J. Anim. Sci. 84:2042–2049.

Gasser, C. L., D. E. Grum, M. L. Mussard, F. L. Fluharty, J. E. Kinder and M. L. Day. 2006d. Induction of precocious puberty in heifers I: Enhanced secretion of luteinizing hormone. J. Anim. Sci. 84:2035–2041.

Geiger, C.A. 2012. "Genex TOYSTORY Bull Produces World Record Setting 2 Millionth Unit of Semen." Hoard's Dairyman. 4 May 2012. Web. Accessed 30 August 2016.

140

Geiger, C.A. 2012. "Genex TOYSTORY Bull Produces World Record Setting 2 Millionth Unit of Semen." Hoard's Dairyman. 4 May 2012. Web. Accessed 30 August 2016.

Genazzani, A. D., F. Petraglia, M. Gastaldi, C. Volpogni, O. Gamba, F. Massolo and A. R. Genazzani. 1994. Evidence suggesting an additional control mechanism regulating episodic secretion of luteinizing hormone and follicle-stimulating hormone in pre- pubertal and post-menopausal women. Human Reprod. 9(10):1807-1812.

Gillespie, J., R. Nehring, and I. Sitienei. 2014. The adoption of technologies, management practices, and production systems in U.S. milk production. Agri. and Food Econ. 2:17. doi:10.1186/s40100-014-0017-y

Gilula, N. B., D. W. Fawcett, and A. Aoki. 1976. The Sertoli cell occluding junctions and gap junctions in mature and developing mammalian testis. Dev. Biol. 50(1):142-168.

Glanowska, K. M., L. L. Burger, and S. M. Moenter. 2014. Developement of Gonadotropin-Releasing Hormone Secretion and Pituitary Response. J. Neuro. 34(45):15060-15069.

Greer, R. C., R. W. Whitman, R. B. Staigmiller, and D. C. Anderson. 1983. Estimating the impact of management decisions on the occurrence of puberty in beef heifers. J. Anim. Sci. 56: 30 – 39.

Gregory, S. J. and U. B. Kaiser. 2004. Regulation of Gonadotropins by Inhibin and Activin. Seminars Reprod. Med. 22(3):253-267.

Griswold, M. D. 1998. The central role of Sertoli cells in spermatogenesis. Semin. Cell Dev. Biol. 9(4):411-416.

Guo, Q., R. Kumar, T. Woodruff, L. A. Hadsell, F. J. DeMayo, and M. M. Matzuk. 1998. Overexpression of mouse follistatin causes reproductive defects in transgenic mice. Mol. Endo. 12(1):96-106.

Hall, P. F. 1994. Testicular steroid synthesis. In: Knobil E., Neill, J. D., eds. The Physiology of Reproduction. Raven, New York. pp. 1335-1362.

Han, S. K., M. L. Gottsch, K. J. Lee, S. M. Popa, J. T. Smith, S. K. Jakawich, D. K. Clifton, R. A. Steiner, and A. E. Herbison. 2005. Activation of gonadotropin-releasing hormone neurons by kisspeptin as a neuroendocrine switch for the onset of puberty. J. Neurosci. 25:11349–11356.

Hardy, M. P., W. R. Kelce, G. R. Klinefelter, and L. L. Ewing. 1990. Differentiation of Leydig Cells precursors in vitro: a role for androgen. J. Endo. 127:488-490. 141

Harstine, B. R., M. Maquivar, L. A. Helser, M. D. Utt, C. Premanandan, J. M. DeJarnette and M. L. Day. 2015. Effects of dietary energy on sexual maturation and sperm production in Holstein bulls. J. Anim. Sci. 93(6):2759-2766. doi: 10.2527/jas.2015-8952.

Hasler, J. F., W. B. Henderson, P. F. Hurtgen, Z. Q. Jin, A. D. McCauley, S. A. Mower, B. Neely, L. S. Shuey, J. E. Strokes, and S. A. Trimmer. 1995. Production, freezing and transfer of bovine IVF embryos and subsequent calving results. Therio. 43:141-152.

Haynes, N. B., H. D. Hafs, and J. G. Manns. 1977. Effect of chronic administration of gonadotropin releasing hormone and thytotrophin releasing hormone to pubertal bulls on plasma luteinizing hormone, and testosterone concentrations, the number of epididymal sperm and body weight. J. Endo. 73(2):227-234.

Henricks, D., J. Hoover, T. Gimenez and L. Grimes. 1988. A study of the source of estradiol-17 beta in the bull. Horm. Met. Research. 20:494-497.

Hess, R. A., P. Cooke, D. Bunick, and J. Kirby. 1993. Adult testicular enlargement induced by neonatal hypothyroidism is accompanied by increased Sertoli cell and germ cell numbers. J. Endo. 132 (6):2607.

Hochereau-de Reviers, M. T., C. Monet-Kuntz, and M. Courot. 1987. Spermatogensis and Sertoli cell numbers and function in rams and bulls. J. Reprod. Fertil. Suppl. 34:101- 114.

Hoffman, P. C., N. M. Brehm, S. G. Price, and A. Prill-Adams. 1996. Effect of accelerated postpubertal growth and early calving on lactation performance of primiparous Holstein heifers. J. Dairy Sci. 79(11):2024-2031.

Holstein Association USA. 2016. Holstein Association USA Genomic Testing Services. Accessed 8 Sept. 2016. http://www.holsteinusa.com/programs_services/genomics_6k_snp.html

Huhtaniemi, I .T., N. Nevo, A. Amsterdam, and Z. Naor. 1986. Effect of postnatal treatment with a gonadotropin-releasing hormone antagonist on sexual maturation of the rat. Biol. Reprod. 35:501-507.

Humblot, P., D. LeBourhis, S. Fritz, J. J. Colleau, C. Gonzalez, C. G. Joly, A. Malafosse, Y. Heyman, Y. Amigues, M. Tissier, and C. Ponsart. 2010. Review: Reproductive Technologies and Genomic Selection in Cattle. Vet. Med. Internatl. Volume 2010: 8 pages.

Hurtgen, P. 2014. "Holstein million-unit fraternity hits 50." Hoard's Dairyman. 29 Oct. 2014. Web. Accessed 30 August 2016. http://www.hoards.com/blog_millionaire-sires 142

Hutchinson, J. L., J. B. Cole, and D. M. Bickhart. 2014. Short communication: Use of young bulls in the United States. J. Dairy Sci. 97(5):3213-3220.

Ibáñez-Escriche, N., S. Forni, J. L. Noguera, and L. Varona. 2014.Genomic information in pig breeding: science meets industry needs. Liv. Sci. 166:94–100.

Illingworth, P. J., N. P. Groome, W. Byrd, W. E. Rainey, A. S. McNeilly, J. P. Mather, and W. J. Bremner. 1996. Inhibin-B: a likely candidate for the physiologically important form of inhibin in men. J. Clin. Endo. Metab. 81:1321-1325.

Ingole, S. D., B.T. Deshmukh, A. S. Nagvekar, and S. V. Bharucha. 2012. Serum profile of thyroid hormones form birth to puberty in buffalo calves and heifers. J. Buffalo Sci. 1:39-49.

Jimenez-Severiano, H., M. L. Mussard, L. A. Fitzpatrick, M. J. D'Occhio, J. J. Ford, D. D. Lunstra, and J. E. Kinder. 2005. Testicular development of Zebu bulls after chronic treatment with a gonadotropin-releasing hormone agonist. J. Anim. Sci. 83(9):2111-2122.

Johnson, L., R. P. Amann, and B. W. Pickett. 1978. Scanning electron and light microscopy of the equine seminiferous tubule. Fertil. Steril. 29:208-215.

Johnson, L., R. S. Zane, C. S. Petty, and W. B. Neaves. 1984. Quantification of the human Sertoli cell population: its distribution, relation to germ cell number and age- related decline. Biol. Reprod. 31:785-795.

Johnson, L. 1991. Spermatogenesis. In: Cupps, P. T. (Ed.), Reproduction in Domestic Animals, fourth ed. Academic Press Inc., San Diego, CA. pp. 173-219.

Johnson, L., D. D. Varner, M. E. Tatum, and W. L. Scrutchfield. 1991. Season but not age affects Sertoli cell number in adult stallions. Biol Reprod. 45:404-410.

Johnson, L., D. L. Thompson, and D. D. Varner. 2008. Role of Sertoli cell number and function on regulation of spermatogenesis. Anim. Reprod. Sci. 105:23-51.

Joyce, K.L., J. Porcelli and P. Cooke. 1993. Neonatal goitrogen treatment increases adult testis size and sperm production in the mouse. J. Andrology. 14,6:448-455.

Kadokawa, H., M. Matsui, K. Hayashi, N. Matsunaga, C. Kawashima, T. Shimizu, K. Kida, and A. Miyamoto. 2008. Peripheral administration of kisspeptin-10 increases plasma concentrations of GH as well as LH in prepubertal Holstein heifers. Endo. 196:331–334.

143

Kaipa, A, T. L. Penttila, S. Shimaski, N. Ling, M. Parvinen, and J. Toppari. 1992. Expression of inhibin betaA and betaB, follistatin and activin-A messenger ribonucleic acids in the rat seminiferous epithelium. Endo. 131:2703-2710.

Kaneko, H., M. Yoshida, Y. Hara, K. Taya, K. Araki, G. Watanabe, S. Sasamoto, and Y. Hasegawa. 1993. Involvement of inhibin in the regulation of FSH secretion in prepubertal bulls. J. Endo. 137:15-19.

Kaneko, H., J. Noguchi, K. Kikuchi, S. Akagi, A. Shimada, K. Taya, and G. Wantanabe. 2001. Production and Endocrine Role of Inhibin During the Early Development of Bull Calves. Bio. Repro. 65:209-215.

Kaneko, H., M. Matsuzaki, J. Noguchi, K. Kikuchi, K. Ohnuma, and M. Ozawa. 2006. Changes in circulating and testicular levels of inhibin A and B during postnatal development in bulls. J. Reprod. Dev. 52(6):741-749.

Kauffman, A. S., M. L. Gottsch, J. Roja, A. C. Byquist, A. Crown, D. K. Clifton, G. E. Hoffman, R. A. Steiner and M. Sempere. 2007. Sexual differentiation of Kiss1 gene expression in the brain of the rat. Endo. 148:1774–1783.

Kijihara, Y., E. G. Blakewood, M. W. Myers, N. Kometani, K. Goto, and R. A. Godke. 1991. and fertilization of follicular oocytes obtained from calves. Therio. 35:220 (Abstract)

Killian, G. J. and R. P. Amann. 1972. Reproductive Capacity of Dairy Bulls. IX. Changes in Reproductive Organ Weights and Semen Characteristics of Holstein Bulls During the First Thirty Weeks After Puberty. J. Dairy. Sci. 55(11):1631-1635.

Kuiri-Hanninen, T., U. Sankilampi, and L. Dunkel. 2014. Activation of the hypothalamic-pituitary-gonadal axis in infancy: Minipuberty. Horm. Res. Paediatr. 82:73-80.

Lacroix, A. and J. Pelletier. 1979. LH and testosterone release in developing bulls following LH-RH treatment. Effect of gonadectomy and chronic testosterone propionate pre-treatment. ACTA Endocrinologica. 91:719-729.

Lamba, P., Y. Wang, S. Tran, T. Ouspenskaia, V. Libasci, T. E. Hebert, G. J. Miller, and D. J. Bernard. 2010. Activin A regulates porcine follicle-stimulating hormone beta- subunit transcription via cooperative actions of SMADs and FOXL2. Endo. 151:5456- 5467.

144

Laroche, J. 2008. Neuroendocrine effects of peripubertal stress exposure in the female mouse (Order No. 3315522). Available from ProQuest Dissertations & Theses A&I; ProQuest Dissertations & Theses Global. (304566378). Retrieved from http://proxy.lib.ohiostate.edu/login?url=http://search.proquest.com/docview/304566378?a ccountid=9783

Leal, M., S. Becker-Silva, H. Chiarini-Garcia, and L. Franca. 2004. Sertoli cell efficiency and daily sperm production in goats (Capra hircus). Anim. Reprod. 1,1:122-128.

Lebrun, J. J. and W. W. Vale. 1997. Activin and inhibin have antagonistic effects on ligand-dependent heteromerization of the type I and II activin receptors and human erythroid differentiation. Mol. Cell Biol. 17:1682-1691.

Leeuwenhoek, A. 1683. De natis e semine genitali animalculis. Royal Soc. (London) Philos. Trans. 12:1040-1043.

Le Gac, F. and D. M. de Kretser. 1982. Inhibin production by Sertoli cell cultures. Mol. Cell. Endo. 28:487-498.

Lehman, M. N. and F. J. Karsch. 1993. Do gonadotropin-releasing hormone, tyrosine hydroxylase-, and beta-endorphin-immunoreactive neurons contain estrogen receptors? A double-label immunocytochemical study in the Suffolk ewe. Endocrinology 133:887– 895.

Lerchl, A., S. Sotiriadou, H. M. Behre, J. Pierce, G. F. Weinbauer, S. Kliesch, and E. Nieschlag. 1993. Restoration of spermatogenesis by follicle-stimulation hormone despite low intratesticular testosterone in photoinhibited hypogonadotropic Djungarian hamsters (Phodopus sungorus) Biol Reprod. 49:1108-1116.

Lewis, K. A., P. C. Gray, A. L. Blount, L. A. MacConnel, E. Wiater, L. M. Bilezikjian, and W. Vale. 2000. Betaglycan binds inhibin and can mediate functional antagonism of activin signalling. Nature. 404:411-414.

Ling, N., S. Y. Ying, N. Ueno, S. Shimasaki, F. Esch, M. Hotta, and R. Guillemin. 1986. Pituitary FSH is released by a heterodimer of the b subunits of the two forms of inhibin. Nature. 321: 779-782.

Lohakare, J. D., K. H. Sudekum, and A. K. Pattanaik. 2012. Nutrition-induced Changes of Growth from Birth to First Calving and Its Impact on Mammary Development and First-lactation Milk Yield in Dairy Heifers: A Review. Asian-Australas J. Anim. Sci. 25(9):1338-1350.

145

Lunstra, D., J. Ford, and S. Echterncamp. 1978. Puberty in Bulls: hormone concentration, growth, testicular development, sperm production and sexual aggressiveness in bulls of different breeds. J. Anim. Sci. 46:1054-1062.

Lunstra, D. and S. Echternkamp. 1982. Puberty in Beef Bulls: Acrosome Morphology and Semen Quality in Bulls of Different Breeds. J. Anim. Sci. 55:638-648.

Lunstra, D.D., T.L. Martin, G.L. Williams and J.J. Ireland. 1993. Immunization against inhibin increases sperm production in young beef bulls. Beef Research Program: Roman L. Hruska U.S. Meat Animal Research Center. Paper 144:93-95.

MacDonald, R. D., D. R. Deaver, and B. D. Schanbacher. 1991. Prepubertal changes in plasma FSH and inhibin in Holstein bull calves: responses to castration and(or) estradiol. J. Anim. Sci. 69(1):276-282.

Macmillan, K. I. and H. D. Hafs. 1968. Pituitary and hypothalamic endocrine changes associated with reproductive development of Holstein bulls. J. Anim. Sci. 27:1614-1617.

Madgwick, S., E. T. Bagu, R. Duggavathi, P. M. Bartlewski, D. M. W. Barrett, S. Huchkowsky, S. J. Cook, A. P. Beard, and N. C. Rawlings. 2008. Effects of treatment with GnRH from 4 to 8 weeks of age on the attainment of sexual maturity in bull calves. Anim. Reprod. Sci. 104:177-188.

Maekawa, M., K. Kamimura, and T. Nagao. 1996. Peritubular myoid cells in the testis: their structure and function. Arch. Histol. Cytol. 59(1):1-13.

Malak, A. G. and M. Thibier. 1982. Plasma LH and testosterone responses to synthetic gonadotropin-releasing hormone (GnRH) or dexamethasome-GnRH combined treatment and their relationship to semen output in bulls. J. Reprod. Fertil. 64:107-113.

Mann, T., L.E.A. Rowson, R.V. Short and J.D. Skinner. 1967. The relationship between nutrition and androgenic activity in pubescent twin calves and the effect of orchitis. J. Endo. 38:455-462.

146

Marguiles, E.H., G.M. Cooper, G. Asimenos, D.J. Thomas, C.N. Dewey, A Siepel, E. Birney, D. Keefe, A.S. Schwartz, M. Hou, J. Taylor, S. Nikolaev, J.I. Montoya-Burgos, A. Loytynoia, S. Whelan, F. Pardi, R. Massingham, J.B. Brown, P. Bickel, I. Holmes, J.C. Mullikin, A. Ureta-Vidal, B. Paten, E.A. Stone, K.R. Rosenbloom, W.J. Kent, G.G. Bouffard, X. Gaun, N.F. Hansen, J.R. Idol, W. Maduro, B. Maskeri, J.C. McDowell, M. Park, P.J. Thomas, A.C. Young, R.W. Blakesley, D.M. Muzny, E. Sodergren. D.A. Wheeler, K.C. Worley, H. Jiang, G.M. Weinstock, R.A. Gibbs, R. Graves, R. Fulton, E.R. Mardis, R.K. Wilson, M. Clamp, J. Cuff, S. Gnerre, D.B. Jaffe, J.L. Chang, K. Lindblad-Toh, E.S. Lander, A. Hinrichs, H. Trumbower, H. Clawson, A. Zweig, R.M. Kuhn, G. Barber, R. Harte, D. Karolchik, M.A. Field, R.A. Moore, C.A. Matthewson, J.E. Schein, M.A. Marra, S.E. Antonarakis, S. Batzoglou, N. Goldman, R. Hardison, D. Haussler, W. Miller, L. Pachter, E.D. Green, and A. Sidow. 2007. Analyses of deep mammalian sequence alignments and constraint predictions for 1% of the human genome. Gemone Res. Jun;17(6):760-774.

Martig, R. C. and J. O. Almquist. 1969. Reproductive capacity of beef bulls. III. Postpubertal changes in fertility and sperm morphology at different ejaculate frequencies. J. Anim. Sci. 28:375.

Martin, T., D. Lunstra and J. Ireland. 1991. Immunoneutralization of Inhibin Modifies Hormone Secretion and Sperm Production in Bulls. Bio. Repro. 45:73-77.

Mason, A. J. 1987. Human inhibin and activin: structure and recombinant expression in mammalian cells. In: Burger, H.G., D. de Kretser, J. Findlay and M. Igarashi. eds. Inhibin: non-steroidal regulation of follicle-stimulating hormone secretion. New York: Ravel Press:42-47.

Mather, J. P., K. M. Attie, T. K. Woodruff, G.C. Rice, and D. M. Phillips. 1990. Activin stimulates spermatogonial proliferation in germ-Sertoli cell co-cultures from immature rat testes. Endo. 127:3206-3214.

Mather, J. P., P. E. Roberts, and L. A. Krummen. 1993. Follistatin modulates activin activity in a cell- and tissue-specific manner. Endo. 132:2732-2734.

Matukumalli, L.K., C.T. Lawley, R.D. Schnabel, J.F. Taylor, M.F. Allan, J. O'Connell, T.S. Sonstegard, C.P Van Tassell, M.P Heaton, T.P.L. Smith and S.S. Moore. 2009. Development and Characterization of a High Density SNP Genotyping Assay for Cattle. PLoS ONE 4(4): e5350. doi:10.1371/journal.pone.0005350

Matzuk, M. M., T.R. Kumar, and A. Bradley. 1995. Different phenotypes for mice deficient in either activins of activin receptor type II. Nature. 374(6520):356-360.

McCarthy, M. S., H. D. Hafs, and E. M. Convey. 1979. Serum hormone patterns associated with growth and sexual development in bulls. J. Anim. Sci. 49:1012-1020. 147

McCoard, S. A., D. D. Lunstra, T. H. Wise, and J. J. Ford. 2001. Specific staining of Sertoli cell nuclei and evaluation of Sertoli cell number and proliferative activity in Meishan and White Composite boars during the neonatal period. Biol. Rerpod. 64(2):689-695.

McCoard, S. A., T. H. Wise, and J.J. Ford. 2003. Endocrine and molecular influences on testicular development in Meishan and White composite boars. J. Endo. 178:405-416.

McEwen, B. S. 1980. Gonadal Steroids: Humoral modulators of nerve cell function. Mol. Cell Endo. 18:151-164.

McGowan, R. D., G. H. Kiracofe, and D. J. Bolt. 1988. Depression of follicle-stimulating hormone in bulls injected with follicular fluid. Therio. 29:979-986.

McLachlan, R. I., N. G. Wreford, D. M. de Kretser, and D. M. Robertson. 1995. The effects of recombinant follicle-stimulating hormone on the restoration of spermatogenesis in the gonadotropin-releasing hormone-immunized adult rat. Endo. 136(9):4035-4043.

Meachem, S., R. McLachlan, D. de Kretser, D. Robertson, and N. Wreford. 1996. Neonatal exposure of rate to recombinant follicle-stimulating hormone increases adult Sertoli and spermatogenic cell numbers. Biol. Repro. 54:36-44.

Meachem, S., J. Aiolkowski, J. Ague, M. Skinner, and K. Loveland. 2005. Developmentally distinct in vivo effects of FSH on proliferation and apoptosis during testis maturation. J. Endo. 186:429-446.

Meehan, T., S. Schlatt, M. K. O'Bryan, D. M. Kretser, and K. L. Loveland. 2000. Regulation of Germ Cell and Sertoli Cell Development by Activin, Follistatin, and FSH. Devel. Biol. 220:225-237.

Mendis-Handagama, S. L. M. C. and H. B. S. Ariyaratne. 2001. Differentiation of the adult Leydig cell population in the postnatal testis. Biol. Reprod. 65:660-671.

Merchant-Larios, H. 1976. The onset of testicular differentiation in the rat: an ultrasound study. Am. J. Anat. 145: 319-329.

Meunier, H., C. Rivier, R. M. Evans, and W. Vale. 1988. Gonadal and extragonadal expression of inhibin alpha, beta-A, and beta-B subunits in various tissues predicts diverse fuctions. Proc. Natl. Acad. Sci. USA. 85:247-251.

Meyers, S. and L. Swanson. 1983. The Effect of Exogenous Follicle Stimulating Hormone Treatment on Anterior Pituitary and Testicular Function in the Prepubertal Bull. J. Andrology. 4,6:371-377. 148

Miller, C. J. and R. P. Amann. 1986. Effects of pulsatile injection of GnRH into 6- to 14- wk-old Holstein bulls. J. Anim. Sci. 62:1332-1339.

Milovanov, V.K. 1938. The Artificial Insemination of Farm Animals. Transcribed by A. Birron and Z.S. Cole. S. Monson. Tech. Services, U.S. Dept. Commerce, Washington, D.C.

Miyamoto, A., M. Umezu, S. Ishii, T. Furusawa, J. Masaki, Y. Hasegawa, and M. Ohta. 1989. Serum Inhibin, FSH, LH, and Tesotsterone Levels and Testicular Inhibin Content in Beef Bulls from Birth to Puberty. Anim. Reprod. Sci. 20:165-178.

Mongkonpunya, K., H. D. Hafs, E. M. Convey, and H.A. Tucker. 1975. Serum LH and testosterone and sperm numbers in pubertal bulls chronically treated with gonadotropin releasing hormone. J. Anim. Sci. 41:160-165.

Monke, D. R. 2002. Bull Management: Artificial Insemination Centres. In H. Rodginski, J. W. Fuquay, and P. F. Fox. Encyclopedia of Dairy Sciences. Academic Press, London, p. 198-204.

Monke, D. R. 2006. Managing Young Sires: Prepubertal Care. Proceedings 21st Tech. Conference on Artificial Insemination & Reprod., NAAB, p 77-81.

Nakamura, T., K. Takio, Y. Eto, K. Shibai, and H. Sugino. 1990. Activin binding protein form rat is follistatin. Science. 247:836-838.

Nolan, C. J., D. A. Neuendorff, R. W. Godfrey, P. G. Harms, T. H. Welsh, Jr., N. H. Arthur, and R.D. Randel. 1990. Influence of dietary energy intake on prepubertal development of Brahman bulls. J. Anim. Sci. 68:1087-1096.

Norman, H. D., R. L. Powell, J. R. Wright, and C.G. Sattler. 2003. Timeliness and effectiveness of progeny testing through artificial insemination. J. Dairy Sci. 86:1513- 1525.

O'Connor, A. E., and D. M. Kretser. 2004. Inhibins in normal male physiology. Semin. Reprod. Med. 22(3):177-185.

O'Dell, W.T. and J.O. Almquist. 1957. Freezing bovine semen. I. Techniques for freezing bovine spermatozoa in milk diluents. J. Dairy Sci. 40:1534-1541.

Ogawa, K., O. Hashimoto, K. Kurohmaru, T. Mizutani, H. Sugino, and Y. Hayashi. 1997. Follistatin-like immunoreactivity in the cytoplasm and nucleus of spermatogenic cells in the rat. Eur. J. Endo. 137:523-529.

149

Orth, J. M. 1984. The Role of Follicle-Stimulating Hormone in Controlling Sertoli Cell Proliferation in Testes of Fetal Rats. Endocrinology. 115(4): 1248-55.

Orth, J. M. 1986. FSH-induced Sertoli cell proliferation in the developing rat is modified by beta-endorphin produced in the testis. Endo. 119(4):1876-1878.

Pelletier, J., M. and A. Lacroix. 1980. Chronobiologie des evenements endocriniens precedant la puberte chez le vau et l'agneau males. In: Ortavant R, Reinberg A (eds.), Ryhthmes et Reproduction. Paris: Masson, pp. 81-90.

Perez, P and G. de los Campos. 2014. Genome-wide regression & prediction with the BGLR statistical package. Genetics. 204(1):1-14.

Peri, I., A. Gertler, I. Bruckental, and H. Barash. 1993. The effect of manipulation in energy allowance during the rearing period of heifers on hormone concentrations and milk production in first lactation cows. J. Dairy Sci. 76(3):742-751.

Pitetti, J. L., P. Calvel, C. Zimmermann, B. Conne, M. D. Papaioannou, F. Aubrey, C. R. Cederroth, F. Urner, B. Fumel, M. Crausaz, M. Docquier, P. L. Herrera, F. Pralong, M. Germond, F. Guillou, B. Jegou, and S. Nef. 2013. An essential role for insulin and IGF1 Receptors in Regulating Sertoli Cell Proliferation, Testis Size, and FSH Action in Mice. Mol. Endo. 27:814-827.

Phillips, D. J. and D. M. de Kretser. 1998. Foolistatin: a multifunctional regulatory protein. Front. Neuroendo. 19:287-322.

Polge, C., A. U. Smith, and A. S. Parkes. 1949. Revival of spermatozoa after vitrificationand dehydration at low temperature. Nature. 164:666.

Polge, C. 1968. Frozen semen and the A.I. Programme in Great Britain. Proc. 2nd Tech. Conf. Artif. Insem. Reprod. NAAB, Columbia, MO. p. 46-51.

Pollak E.J., G. L. Bennett, W. M. Snelling, R. M. Thallman, and L.A. Kuehn. 2012. Genomics and the global beef cattle industry. Anim. Prod. Sci. 52:92–99.

Ponsart, C., L. Salas Cortes, N. Jeanguyot, and P. Humblot. 2008. Le sexage des embryons a le vent en poupe. Bulletin Technique de l'Insemination Artificielle. 129:30- 31.

Preisinger, R. 2012. Genome-wide selection in poultry. Anim. Prod. Sci. 52:121–125.

Price, E. O., J. E. Harris, R. E. Borgwardt, M. L. Sween, and J. M. Connor. 2003. Fenceline contact of beef calves with their dams at weaning reduced the negative effects of separation on behavior and growth rate. J. Anim. Sci. 81:116-121. 150

Pruitt, R. J., L. R. Corah, J. S. Stevenson, and G. H. Kiracofe. 1986. Effect of energy intake after weaning on sexual development of beef bulls. II. Age at first mating, age at puberty, testosterone and scrotal circumference. J. Anim. Sci. 63:579.

Purvis, K., R. Calandra, O. Naess, A. Attramadal, P. A. Torjesen, and V. Hansson. 1977. Do androgens increase Leydig cell sensitivity to luteinising hormone. Nature. 265:169- 170.

Raivio, T., A. Wikstrom, and L. Dunkel. 2007. Treatment of gonadotropin-deficient boys with recombinant human FSH: long-term observation and outcome. Europ. J. Endo. 156:105-111.

Ramirez, D.V. and S. M. McCann.M. 1963. Comparison of the regulation of luteinizing hormone (LH) secretion in immature and adult rats. Endocrinology 72:452-64.

Rawlings, N. C. and A. C. O. Evans. 1995. Androgen negative feedback during the early rise in LH secretion in bull calves. J. Endocrin. 145:243-249.

Rawlings, N. C., A. C. O. Evans, R. K. Chandolia,s and E.T. Bagu. 2008. Sexual Maturation in the Bull. Reprod. Domestic Anim. 43 (Suppl 2): 295-301.

Revel, F, P. Mermillod, N. Peynot, J. P. Renard, and Y. Heyman. 1995. Low developmental capacity of in vitro matured and fertilized oocytes from calves compared with that of cows. J. Reprod. and Fer. 103:115-120.

Roberts, V., H. Meunier, P. E. Sawchenko, and W. Vale. 1989. Differential production and regulation of inhibin subunits in rat testicular cell types. Endo. 125:2350-2359.

Roberts, V. J. 1997. Tissue-specific expression of inhibin/activin subunit and follistatin mRNAs in mid- to late-gestational age human fetal testis and epididymis. Endocrine. 6:85-90.

Roberts, A .J., E. E. Grings, M. D. MacNeil, L. Waterman, J. Alexander, and T.W. Geary. 2007. Reproductive performance of heifers offered ad libitum or restricted access to feed for a 140-d period after weaning. Journal of Animal Science. 85 (Suppl. 2):169.

Robertson, D. M., R. Klein, F. L. de Vos, R. I. McLachlan, R. E. H. Wettenhall, M. T. W. Hearn, H. G. Burger, and D. M. de Kretser. 1987. The isolation of polypeptides with FSH suppressing activity from bovine follicular fluid which are structurally different to inhibin. Biochem. Biophys. Res. Commun. 149: 744-749.

151

Rodriguez, R. P. and M. E. Wise. 1989. Ontogeny of pulsatile secretion of gonadotropin- relasing hormone in the bull calf during infantile and pubertal development. Endo. 124:248-256.

Russell, L. D. and M. D. Griswold. 1993. The Sertoli Cell. Cache River Press. Clearwater, FL.

Salisbury, G. W., H. K. Fuller, and E. L. Willett. 1941. Preservation of bovine spermatozoa in yolk-citrate diluent and field results from its use. J. Dairy Sci. 24:905- 910.

Saacke, R. G. 2008. Sperm morphology: Its relevance to compensable and uncompensable traits in semen. Therio. 70:473-478.

Sasaki, M., M. Yamamoto, K. Arishima, and Y. Equchi. 2000. Effect of follicle- stimulating hormone on Sertoli cell division in cultures of fetal rat testes. Biol. Neonate. 78(1):48-52.

Schanbacher, B. D. and S. E. Echternkamp. 1978. Testicular steroid secretion in response to GnRH-mediated LH and FSH release in bulls. J. Anim. Sci. 47:514-520.

Schanbacher, B. D. 1982. Hormonal interrelationships between hypothalamus, pituitary and testis of rams and bulls. J. Anim. Sci. Suppl. 55:56-67.

Schanbacher, B. D., M. J. D'Occhio, and J. E. Kinder. 1982. Initiation of spermatogenesis and testicular growth in oestradiol-17β-implanted bull calves with pulsatile infusion of luteinizing hormone releasing hormone. J. Endo. 93(2):183-192.

Schefers, J. M. and K. A. Weigel. 2012. Genomic selection in dairy cattle: Integration of DNA testing into breeding programs. Animal Frontiers. Vol. 2. No. 1:4-9.

Schwanzel-Fukuda, M. and D. W. Pfaff. 1989. Origin of luteinizing-hormone-releasing hormone neurons. Nature. 338:161-164.

Seidel, G. E. Jr. 1981. Superovulation and embryo transfer in cattle. Science. 211:351- 358.

Seidel, G. E. Jr. 2010. Brief introduction to whole-genome selection in cattle using single nucleotide polymorphisms. Reprod. Fert. and Develop. 22:138-144.

Senger, P. L. 1997. Pathways to Pregnancy and Parturition. Second Edition. Current Conceptions, Inc., Moscow, p.68.

152

Setchell, B. P. and G. M. H. Waites. 1975. The blood testis barrier. In: Hamilton, D. W. and R. O. Greep. (Eds.), Handbook of Physiology, vol. V. American Physiological Society, Washington, DC. pp. 143-172.

Sharma, I. J., S. P. Agarwal, V. K. Agarwal, and P. K. Dwaraknath. 1985. Serum thyroid hormone levels in male buffalo calves as related to age and sexual development. Therio. 24(5):509-517.

Sharpe, R. M., K. J. Turner, C. McKinnell, N. P. Groome, N. Atavassova, M. R. Miller, D. L. Buchanan, and P.S. Cooke. 1999. Inhibin B levels in plasma of the male rat from birth to adulthood: effect of experimental manipulation of Sertoli cell number. J. Androl. 20:94-101.

Sharpe, R., C. McKinnell, C. Kivlin and J. Fisher. 2003. Proliferation and functional maturation of Sertoli cells, and their relevance to disorders of testis function in adulthood. Repro. 125:769-784.

Shima, Y., K. Miyabayashi, S. Haraguchi, T. Arakawa, H. Otake, T. Baba, S. Matsuzaki, Y. Shishido, H. Akiyama, T. Tachibana, K. Tsutsui, and K. Morohashi. 2013. Contribution of Leydig and Sertoli cells to testosterone production in mouse fetal testes. Mol. Endo. 27(1):63-73.

Skinner, M. K. and M. D. Griswold. 2005. Sertoli cell biology. San Diego, Ca: Elsevier Academic Press.

Simoni, M., G.F. Weinbauer, J. Gromoll, and E. Nieschlag. 1999. Role of FSH in male gonadal function. Annals Endo. 60(2):102-106.

Smith, C. 1988. Applications of embryo transfer in animal breeding. Therio. 29:203-212.

Smith, J. T., C. M. Clay, A. Caraty, and I. J. Clarke. 2007. KiSS-1 messenger ribonucleic acid expression in the hypothalamus of the ewe is regulated by sex steroids and season. Endo. 148:1150–1157.

Sørensen, E. 1940. Insemination with gelatinized semen in paraffined cellophane tubes [in Danish]. Medlernsbl. Danske Dyrlaegeforen. 23:166-169.

Spallanzani, L. 1784. Dissertations relative to the natural history of animals and vegetables. Transcribed by Beddoes in Dissertations Relative to the Natural History of Animals and Vegetables. Vol. 2:195-199.

Strauss, S. 2010. Biotech breeding goes bovine. Nature Biotechnology. 28:540-543.

153

Stumpf, T. T., M. W. Wolfe, M. S. Roberson, R. J. Kittok, and J. E. Kinder. 1993. Season of the year influences concentration and pattern of gonadotropins and testosterone in circulation of the bovine male. Biol. Reprod. 49:1089-1095.

Sugino, K., N. Kurosawa, T. Nakamura, K. Takio, S. Shimasaki, N. Ling, K. Titani, and H. Sugino. 1993. Molecular hetergeneity of follistatin, and activin binding protein. J. Biol. Chem. 68:15579-15587.

Sugino, H., K. Sugino, O. Hashimoto, H. Shoji, and T. Nakamura. 1997. Follistatin and its role as an activin-binding protein. J. Med. Invest. 44:1-14.

Sullivan, J. J. and F. I. Elliot. 1968. Bull fertility as affected by an interaction between motile spermatozoa concentration and fertility level in artificial insemination. VI Int. Cong. Anim. Reprod. Artif. Insem. 2:1307.

Suszko, M. I., D. J. Lo, H. Suh, S. A. Camper, and T. K. Woodruff. 2003. Regulation of the rat follicle-stimulating hormone beta-subunit promoter by activin. Mol. Endo. 17:318- 332.

Suszko, M. I., D. M. Balkin, Y. Chen, and T. K. Woodruff. 2005. Smad3 mediates activin-induced transcription of follicle-stimulating hormone beta-subunit gene. Mol. Endo. 19:1849-1858.

Tarulli, G., P. Stanton, A. Lerchl, and S.J. Meachem. 2006. Adult Sertoli Cells Are Not Terminally Differentiated in the Djungarian Hamster: Effect of FSH on Proliferation and Junction Protein Orgaization. Biol Reprod. 74: 798-806.

Teepker, G. and D. S. Keller. 1989. Selection of sires originating from a nucleus breedingunit for use in a commercial dairy population. Canadian J. Anim. Sci. 69:595- 604.

The Bovine Genome Sequencing and Analysis Consortium. 2009. The genome sequence of taurine cattle: a window into ruminant biology and evolution. Science. 324:522-528.

The Bovine HapMap Consortium. 2009. Genome-wide survey of SNP variation uncovers the genetic structure of cattle breeds. Science. 324:528-532.

Thomas, T. Z., H. Wang, P. Niclasen, M. K. O'Bryan, L. W. Evans, N. P. Groome, J. Pedersen, and G. P. Risbridger. 1997. Expression and localization of activin subunits and follistatins in tissues from men with high grade prostate . J. Clin. Endo. Metab. 82(11):3851-3858.

154

Tilbrook, A. J., D. M. de Kretser, F. R. Dunshea, R. Klein, D. M. Robertson, I. J. Clarke, and S. Maddocks. 1996. The testis is not the major source of circulating follistatin in the ram. J. Endo. 149(1):55-63.

Ulloa-Aguirre, A., and C. Timossi. 1998. Structure-function relationship of follicle- stimulating hormone and its receptor. Hum. Reprod. Update. 4(3):260-283.

United Nations, Department of Economic and Social Affairs, Population Division. 2015. World Population Prospects: The 2015 Revision, Key Findings and Advance Tables. Working Paper No. ESA/P/WP. 241.

USDA. 1966. Milk Production. Statistical Reporting Service, Crop Reporting Board. Washington, D.C.

USDA-National Agricultural Statistics Service (NASS). 2016. Milk Production. Washington, D.C. p. 4

Vale, W., J. Vaughan, R. McClintock, A. Corrigan, W. Woo, D. Karr, and J. Spiess. 1986. Purification and characterization of and FSH releasing protein form porcine follicular fluid. Nature. 321:776-779.

VanDemark, N. L., G. R. Fritz, and R.E. Mauger. 1964. Effect of energy intake on reproductive performance of dairy bulls. II. Semen production and replenishment. J. Dairy Sci. 47:898.

VanDemark, N. L. and R. E. Mauger. 1964. Effect of energy intake of reproductive performance of dairy bulls. I. Growth, reproductive organs and puberty. J. Dairy Sci. 47:798. van Haaster, L., F. DeJong, R. Docter, and D. de Rooij. 1993. High neonatal triiodothyronine levels reduce the period of Sertoli cell proliferation and accelerate tubule lumen formation in the rat testis, and increase serum inhibin levels. Endo. 133:755-760.

VanRaden, P. M., M. E. Toole, and J. B. Cole. 2009. Can you believe those genomic evaluations for young bulls? J. Dairy Sci. 92(E-Suppl. 1):314(abstr. 279).

VanRaden, P. M., D. J. Null, T. S. Sonstegard, H. A. Adams, C. P. Van Tassel, and K. M. Olson. 2012. Fine mapping and discovery of recessive mutations that cause abortions in dairy cattle. J. Dairy Sci. 95(Suppl. 2):ii-iii(abstr. LB6).

Waites, G. M. H. 1997. Fluid secretion. In: Johnson, A. D. and W. R. Gomes. (Eds.), The Testis, vol. IV. Academic Press Inc., New York, NY. pp. 91-123.

155

Walker, W. H. and J. Cheng. 2005. FSH and testosterone signaling in Sertoli cells. Repro. 130: 15-28.

Wattenberg, L. W. 1958. Microscopic histochemical demonstration of steroid-3β-ol dehydrogenase in tissue sections. J. Histochem. Cytochem. 6:225-232.

Wilson, P. R. and K. R. Lapwood. 1979. Studies of reproductive development in Romney rams. I. Basal levels and plasma profiles of LH, testosterone, and prolactin. Biol. Reprod. 20:965-970.

Wiltbank, J. N., K. E. Gregory, L. A. Swiger, J. E. Ingalls, J. A. Rothlisberger, and R. M. Koch. 1966. Effects of heterosis on age and weight at puberty in beef heifers. Journal of Animal Science. 25: 744 - 751.

Wolf, F., J. Almquist and E. Hale. 1965. Prepubertal Behavior and Puberal Characteristics of Beef Bulls on High Nutrient Allowance. J. Anim. Sci. 24:761-765.

Wolfe, D. A. and E. J. Mash. 2008. Behavioral and Emotional Disorders in Adolescents: Nature, Assessment, and Treatment. Guilford Press. p. 556.

Wong, C. H. and C. Y. Cheng. 2005. The blood-testis barrier: its biology, regulation, and physiological role in spermatogenesis. Current Top. Dev. Biol. 71:263-296.

Wuttke, W., H. Jarry, C. Feleder, J. Moguilevsky, S. Leonhardt, J. Y. Seong, and K. Kim. 1996. The neurochemistry of the GnRH pulse generator. Acta Neurobio. Exp. 56(3):707- 713.

Ying, S.Y., Ueno, N., Shimasaki, S., Esch, F., Hotta, M. and Guillemin, R. (1986) Pituitary FSH is released by a heterodimer of the b subunits of the two forms of inhibin. Nature, 321, 779-782.

156