The direct evaluation of blood samples from the hypophyseal portal system demonstrated that GnRH pulsatile secretion increased linearly from 2 weeks (3.5 pulses/10 h) to 12 weeks of age (8.9 pulses/10 h) in bull calves. Although GnRH secretion into hypophyseal portal blood was detected at 2 weeks, pulsatile LH secretion was not detected in jugular blood samples before 8 weeks of age. In addition, GnRH pulses were not necessarily accompanied by LH secretion until 8–12 weeks of age, when all GnRH pulses resulted in LH pulses. The increase of pulsatile GnRH release from 2 to 8 weeks of age without a concomitant increase in LH secretion may represent a reduced ability of the pituitary gland to respond to stimulus by GnRH (Rodriguez and Wise, 1989). The period in which GnRH pulses did not stimulate LH secretion corresponded to the period during which there was an increase in pituitary weight, GnRH receptor concentration and LH content (Amann et al., 1986). Moreover, frequent GnRH treatments during the infantile period in calves increased pituitary mRNA for the LH beta subunit (LH-β), LH content and GnRH receptors, with resulting increases in LH pulse frequency and mean concentrations (Rodriguez and Wise, 1991), indicating that increased GnRH pulse frequency results in increased pituitary sensitivity to GnRH. With time, the increased GnRH secretion results in the increased LH pulse frequency observed during the prepubertal period.

During the early gonadotrophin rise that is characteristic of the prepubertal period, there is a transient increase in LH and follicle-stimulating hormone (FSH) concentrations from approximately 2 to 6 months of age (Figs 12.1 and 12.2). Concentrations decrease thereafter and remain at levels only slightly greater than those observed during the infantile period. The main factor responsible for increased gonadotrophin concentrations is the increase in GnRH pulse secretion, as demonstrated by a dramatic increase in LH pulse frequency. The number of LH pulses increased from less than one a day at 1 month of age to approximately 12/day (≥1 pulse/2 h) at approximately 4 months of age. Changes in pulse amplitude during this period were not consistent among reports; amplitude was either reduced, unchanged or augmented. Pulsatile discharges of FSH have been observed in bulls, but are much less evident than those of LH (Amann and Walker, 1983; Amann et al., 1986; Evans et al., 1995, 1996; Aravindakshan et al., 2000).

The crucial role of the early gonadotrophin rise (especially the LH secretion pattern) in regulating sexual development in bulls has been demonstrated in several studies, using a variety of approaches. Prolonged treatment with a GnRH agonist in calves from 1.5 to 3.5 months of age decreased LH and FSH pulse frequency, pulse amplitude and mean concentrations at 3 months of age, delayed the peak mean LH concentration from 5 to 6 months of age, and reduced FSH and testosterone concentrations from 3.5 to 4.5 months of age. These hormonal alterations were associated with delayed puberty and reduced paired testes weight and number of germ cells in tubular cross sections at 11.5 months of age. Conversely, treatment with GnRH every 2 h to mimic pulsatile secretion from 1 to 1.5–2 months of age increased LH pulse frequency and mean concentration during the treatment period, and also resulted in greater scrotal circumference (SC), paired testes weight, seminiferous tubule diameter, and number of germ and Sertoli cells in tubular cross sections at 12 months of age (Chandolia et al., 1997a,b; Madgwick et al., 2008). The LH secretion pattern during the prepubertal period is associated with age at puberty in bulls raised in contemporary groups, suggesting that this is the physiological mechanism by which genetics affects sexual development. Studies have shown that LH pulse frequency was greater at around 2.5 to 5 months of age and that mean LH concentrations increased earlier and reached greater maximum levels in early- than in late-maturing Hereford bulls (age at puberty 9.5 and 11 months, respectively) (Evans et al., 1995; Aravindakshan et al., 2000). Moreover, other studies have indicated that the effects of nutrition on sexual development are mediated through effects on LH secretion patterns (Brito et al., 2007b,c,d).

Reduced gonadotrophin secretion marks the end of the prepubertal period and the beginning of the pubertal period. The rapidly increasing testosterone secretion and, possibly, increased hypothalamic sensitivity to negative feedback from androgens are probably responsible for the decrease in LH secretion, whereas inhibin produced by Sertoli cells may act on the gonadotrophs to limit FSH secretion, because immunization with inhibin antiserum resulted in a marked increase in FSH concentrations in prepubertal bulls (Kaneko et al., 1993; Rawlings and Evans, 1995). Testosterone pulse frequency did not increase after the peri-pubertal period and remained at approximately 4.5 to 6.8 pulses/24 h from 6 to 10 months of age. However, pulse amplitude increased during the pubertal period, with consequent increase in testosterone mean concentrations until approximately 12 months of age. Elevated testosterone secretion is essential for increasing the efficiency of spermatogenesis that eventually leads to the appearance of sperm in the ejaculate (Rawlings et al., 1972, 1978; McCarthy et al., 1979a,b; Rodriguez and Wise, 1991).

Circulating concentrations of insulin-like growth factor I (IGF-I) in calves increased continuously and only reached a plateau (or decreased slightly) after sexual development was mostly completed after 12–14 months of age; increasing circulating concentrations of IGF-binding protein 3 and decreasing concentrations of IGF-binding protein 2 were also observed during sexual development (Renaville et al., 1993, 1996, 2000; Brito et al., 2007a,b,c,d). The concomitant decrease in circulating GH concentrations and increase in IGF-I concentrations during sexual development in bulls indicates that there are either drastic changes in liver sensitivity to GH or that other sources are responsible for IGF-I production. A possible IGF-I source might be the testes, as Leydig cells are capable of secreting this hormone in other species. Observations that intact bulls tended to have greater IGF-I concentrations than castrated steers at 12 months of age further support the hypothesis that the testes might contribute substantially to circulating IGF-I concentrations during the peri-pubertal and pubertal periods in bulls (Lee et al., 1991). Close temporal associations observed in a series of nutrition studies strongly suggest that circulating IGF-I might be involved in regulating the GnRH pulse generator and the magnitude and the duration of the early gonadotrophin rise in beef bulls (Brito et al., 2007b,c,d).

A possible effect of IGF-I on testicular steroidogenesis in bulls has also been suggested. Leydig and Sertoli cells both produce IGF-I, indicating the existence of paracrine/autocrine mechanisms of testicular regulation involving IGF-I (Bellve and Zheng, 1989; SpiteriGrech and Nieschlag, 1992). It is assumed that most of the IGF-I in the testes is produced locally, and that circulating IGF-I may play a secondary role in regulating testicular development and function. However, the temporal patterns and strong associations among circulating IGF-I concentrations, testicular size and testosterone secretion observed in bulls receiving different nutrition argue for a primary role for this hormone (Brito et al., 2007b,c,d). This primary role of increased circulating IGF-I during the pubertal period may be to promote the increase in testosterone concentrations by regulating Leydig cell multiplication, differentiation and maturation. Because testosterone has been shown to upregulate IGF-I production and IGF-I receptor expression by Leydig and Sertoli cells (Cailleau et al., 1990), the establishment of a positive feedback loop between IGF-I secretion and testosterone production may be important for sexual development.

Circulating leptin and insulin concentrations also increased during the pubertal period in bulls. Notwithstanding, developmental and nutritional differences in LH pulse frequency were not related to differences in leptin or insulin concentrations in beef bulls (Brito et al., 2007b,d). Other studies have also demonstrated that leptin did not stimulate in vitro GnRH secretion from hypothalamic explants or gonadotrophin secretion from adenohypophyseal cells collected from bulls and steers maintained at an adequate level of nutrition (Amstalden et al., 2005). These results indicate that the role of these hormones in regulating GnRH secretion, if any, might be purely permissive in bulls.

Postnatal testicular development and establishment of spermatogenesis

Although the overall pattern of testicular growth is similar in all breeds, the characteristics of the growth curve are greatly affected by genetics. In general, the rapid growth phase is shorter and testicular growth plateaus sooner in bulls from breeds that mature faster (reach puberty earlier) than in bulls from late-maturing breeds, resulting in marked differences in the slope of the curve. In addition, the asymptotic value of the curve, i.e. adult testicular size, also differs considerably among breeds (Coulter and Keller, 1982; Coulter et al., 1987b; Gregory et al., 1991; Pratt et al., 1991; Johnson et al., 1995; Bell et al., 1996; Lunstra and Cundiff, 2003). These same differences can be observed within breed between early- and late-maturing bulls, which emphasizes the effects of genetics on bull testicular growth (Evans et al., 1995).

Testicular growth is accompanied by marked histological changes. The testicular intertubular cell population is composed of mesenchymal-like cells, fibroblasts, Leydig cells, peri-tubular cells and mononuclear cells. At around 1 to 2 months of age, mesenchymal-like cells made up the majority of the cells in the testicular interstitial tissue. These pluripotent cells proliferated by frequent mitoses and were the precursors of Leydig cells, contractile peri-tubular cells and fibroblasts. Approximately 20–30% of all intertubular cells at all ages were mononuclear cells, including lymphocytes, plasma cells, monocytes, macrophages and light inter-calated cells (monocyte-derived, Leydig cell-associated typical cells of the bovine testis). The thickness of the tubular basal lamina was approximately 3 μm at 4 months of age, but decreased continuously to 1.2 μm at 5 months of age. Mesenchymal-like cells transformed into peri-tubular cells with elongated nuclei at 4 months of age, while these, in turn, transformed into contractile myofibroblasts at 6 months of age (Wrobel et al., 1988).

Leydig cells are present in the testes from birth to adulthood. At approximately 1 month of age, most (70%) intertubular cells were mesenchymal-like cells with a high mitotic rate. Typical Leydig cells constituted about 6% of all intertubular cells and a number of these cells were found in an advanced degenerative state, probably as remnants of the fetal Leydig cell population. Degenerating fetal and newly formed Leydig cells coexisted until 2 months of age, but only Leydig cells formed postnatally were observed thereafter. At 2 months of age, Leydig cells accounted for approximately 20% of all intertubular cells and comprised 10% of the total testicular volume. At approximately 4 months of age, the mesenchymal-like cells ceased proliferation and transformed into contractile peri-tubular cells or Leydig cells, thus decreasing the proportion of mesenchymal-like cells to approximately 20%. At 5 months of age, undifferentiated mesenchymal-like cells were rare and the Leydig cell population increased by mitotic proliferation. Newly differentiating, intact and degenerating Leydig cells were observed in close proximity from 4 to 7 months of age. The Leydig cell population found in the adult bull was present at approximately 7 months of age, as mitosis after this age was rare (Wrobel, 1990).

Leydig cell mass increased from 0.15 g/testis at 1 month to 5.8 g/testis at 7–8 months of age. After this, Leydig cell mass increased slowly but continuously to reach about 10 g in the young adult testis at 24 months of age. Nuclear and whole Leydig cell volumes increased from 1 to 4 months of age, but then remained unchanged up to 12 months of age. However, from approximately 12 to 24 months of age there was a considerable increase in nuclear and whole Leydig cell volumes. Leydig cell numbers per testis increased from 1 to 7 months of age (0.42 to 6 billion, respectively) and remained unchanged after that. Therefore, the increase in Leydig cell mass after 7 months of age was a result of hypertrophy and not hyperplasia. Leydig cell mitochondrial mass increased from 1 to 4 months of age and remained relatively constant up to 10 months of age, but then more than doubled from that age until approximately 24 months of age (Wrobel, 1990). Functional maturation of Leydig cells was observed to involve the expression of steroidogenic enzymes and the production of testosterone, although the age-dependent expression of steroidogenic enzymes did not parallel changes in the numbers of Leydig cells in the testis (Waites et al., 1985).

At approximately 1 month of age, undifferentiated Sertoli cells were characterized by round or oval nuclei with numerous flakes of heterochromatin dispersed throughout the nucleoplasm. Lateral cell membranes of neighbouring presumptive Sertoli cells contained few interdigitations and no special junctional complexes. Between 4 and 5 months of age, opposing cell membranes of adjacent undifferentiated Sertoli cells started to develop extended junctional complexes above the spermatogonia and in the basal portion of the tubules. The nuclei of these cells became more elongated and irregular, a vacuolar nucleolus characteristic of the mature cell developed, and the clumps of heterochromatin present at younger ages disappeared. Most Sertoli cells had completed their morphological differentiation and attained adult structure by 6 to 7 months of age. Junctional complexes consisting of many serially arranged points or lines of fusion involving neighbouring Sertoli cell membranes could be observed. These junctions formed a functional blood–testis barrier and divided the tubular epithelium into a basal compartment containing spermatogonia and an adluminal compartment (connected to the lumen) containing germ cells at later stages of spermatogenesis (Abdel-Raouf, 1960; Sinowatz and Amselgruber, 1986; Wrobel, 2000). In Holstein bulls, the number of adult-type Sertoli cells increased dramatically from 202 to 8862 million cells/testis between 5 and 8 months of age (Curtis and Amann, 1981).

The accelerated testicular growth observed after 6 months of age in bulls occurs when circulating gonadotrophin concentrations are decreasing, which points to the existence of important GnRH-independent mechanisms regulating testicular development. The period of accelerated testicular growth coincides with increasing circulating IGF-I and leptin concentrations, and strong associations between these hormones and testicular size have been observed in growing beef bulls (Brito et al., 2007a,d), indicating that metabolic hormones may be involved in regulating GnRH-independent testicular development. As there was no association between circulating metabolic hormones and gonadotrophin concentrations, the possible effects of metabolic hormones on testicular growth were apparently direct and independent of the hypothalamus and pituitary. Although IGF-I and leptin concentrations were associated with testicular size, there were no associations between these hormones and seminiferous tubule diameter and area, seminiferous epithelium area, or volume occupied by the seminiferous tubules (L. Brito, unpublished). These observations suggest that increased circulating IGF-I and leptin concentrations were associated with increased length of the seminiferous tubules and probably with overall increases in the total number of testicular cells. Considering the cellular events in the testis during the pubertal period, the temporal patterns of metabolic hormone concentrations in bulls indicated that circulating IGF-I and leptin could be involved in regulating Leydig cell multiplication and maturation, Sertoli cell maturation and germ cell multiplication during the period of accelerated GnRH-independent testicular growth.

Puberty

After spermatogenesis is established, there was a gradual increase in the number of testicular germ cells supported by each Sertoli cell and an increase in the efficiency of the spermatogenesis, i.e. an increase in the number of more advanced germ cells resulting from the division of precursor cells. The yields of different germ cell divisions, low during the onset of spermatogenesis, increased progressively to the adult level (Macmillan and Hafs, 1968; Curtis and Amann, 1981). The eventual appearance of sperm in the ejaculate was the result of the increasing sperm production efficiency after initiation of spermatogenesis. In general terms, puberty is defined as the process of changes by which a bull becomes capable of reproducing. This process involves the development of the gonads and secondary sexual organs, and the development of the ability to breed. For research purposes however, puberty in bulls is usually defined as an event instead of a process. Most researchers define attainment of puberty by the production of an ejaculate containing ≥50 million sperm with ≥10% motile sperm (Wolf et al., 1965). The interval between the first observation of sperm in the ejaculate and puberty as defined by these criteria was approximately 30 to 40 days (Lunstra et al., 1978; Jiménez-Severiano, 2002).

The epididymis continued to grow until at least 6 years of age in Holstein bulls, with epididymal weight increasing from 9 g at 8 months of age to 15, 23, 27 and 38 g at 12, 18, 25–48 and 73–96 months of age, respectively (Almquist and Amann, 1961; Killian and Amann, 1972). The tube-like vesicular glands in newborn calves increased in length and became lobulated during development. The weight of the vesicular glands increased until approximately 4 years of age in Holstein bulls – from 13 g at 8 months of age to 26, 35, 54 and 78 g at 12, 18, 25–48 and 73–96 months of age, respectively (Almquist and Amann, 1961; Killian and Amann, 1972). The sigmoid flexure of the penis began to develop at about 3 months of age. Penis length increased by up to five times by the onset of puberty, and length continued to increase until sexual maturity (Coulter, 1986). The penis in Friesian bulls 13 to 19 months old measured 73 to 89 cm (Ashdown et al., 1979a); in contrast, the penis in Holstein bulls ≥25 months old measured 95 to 106 cm (Almquist and Amann, 1961). First protrusion of the penis during mounting was observed at approximately 8 months of age, with complete separation of the penis and sheath observed at approximately 8.5 months of age in Angus, Charolais and Hereford bulls (Wolf et al., 1965; Barber and Almquist, 1975). Complete sheath–penile detachment evaluated during electroejaculation (EE) was observed at around the same time of puberty, whereas first completed service – when evaluated during libido testing – was only observed approximately a month after puberty, indicating that bulls started producing sperm before complete sheath–penile detachment (Lunstra et al., 1978).

Effects of nutrition on sexual development

The effects of nutrition on the sexual development and reproductive function of bulls were mediated through the hypothalamic-pituitary-testicular axis. Nutrition affects the GnRH pulse generator in the hypothalamus, because differences in LH pulse secretion in bulls receiving different nutrition can be observed even in the absence of differences in pituitary LH secretion capability, as determined by GnRH challenge (Brito et al., 2007c,d). Interestingly, when low nutrition was imposed on bulls by limiting the amount of nutrients in a ration fed ad libitum, only reduced LH pulse frequency was observed; but when nutrition was restricted by restricting food availability, reduced LH pulse frequency, mean and peak concentrations and secretion were observed after GnRH challenge, (Brito et al., 2007b). These results seem to indicate that the inhibitory effects of limited nutrient availability on LH secretion appear to be exerted on the hypothalamus only, while the combination of limited availability of nutrients combined with hunger sensation experienced by bulls with restricted intake affected both hypothalamic and pituitary function, producing a much more severe inhibition of LH secretion. The effect of nutrition on Leydig cell number and/or function in bulls receiving different diets was demonstrated by differences in testosterone secretion after GnRH challenge, even in the absence of differences in LH secretion after the challenge (Brito et al., 2007c,d).

Differences in yearling SC due to age of the dam in beef bulls could also be interpreted as an indication that nutrition during the pre-weaning period affects sexual development, although possible in utero effects cannot be completely ruled out. SC in beef bulls increases as age of the dam increases up to 5 to 9 years of age and then decreases as dams get older. Adjustment factors of 0.7–1.4 cm, 0.2–1.0 cm, 0.1–1.0 cm and 0.3–0.75 cm for yearling SC have been suggested for bulls raised by 2-, 3-, 4- and ≥10-year old dams, respectively (Bourdon and Brinks, 1986; Nelsen et al., 1986; Lunstra et al., 1988; Kriese et al., 1991; Evans et al., 1999; Crews and Porteous, 2003). In these studies, the inclusion of weight as a covariate in the models describing SC resulted in decreased effects of age of the dam, indicating that the effect of dam age on testicular growth seems to be primarily the result of dam age effects on the bull’s body weight, probably related to differences in milk production. This theory is also supported by reports that, as observed in bulls receiving low nutrition, LH secretion after GnRH challenge was greater from 3.5 to 6 months of age in bulls raised by multiparous rather than by primiparous females (Bagu et al., 2010c).

Though low nutrition during the pubertal period has adverse effects on reproductive function, the potential beneficial effects of high nutrition after weaning are questionable, at best. Effects of energy on sexual development were not consistent in a study with Simmental and Hereford bulls fed diets with low, medium or high energy content (approximately 14, 18 and 23 Mcal/bull daily, respectively) from 7 to 14 months of age. Dietary energy affected sexual development in Simmental bulls, but not in Hereford bulls. Simmental bulls in the high-energy group were heavier and had greater SC and testosterone concentrations than bulls in the low-energy diet group (in general, the medium energy group had intermediate values). However, increased dietary energy did not hasten age at puberty. The only semen trait affected by dietary energy was semen volume, which was depressed in Simmental bulls in the medium-energy group. Serving capacity was greater for Hereford bulls in the high-energy diet; in contrast, medium- and high-energy diets were associated with a decrease in the number of services between two testing periods in Simmental bulls. There was a trend for lower sperm motility and a lower proportion of normal sperm in Simmental bulls fed the low-energy diet (Pruitt and Corah, 1985; Pruitt et al., 1986).

In Holstein bulls producing semen for AI, high energy intake was associated with visual evidence of weakness of the feet and legs and increased reaction time after 3 years of age (Flipse and Almquist, 1961). Under field conditions, post-weaning high-energy diets are frequently associated with impaired reproductive function in bulls, which is most likely related to altered testicular thermoregulation as a result of excessive fat deposition above and around the testes in the scrotum. In one report, sperm motility decreased and the proportion of sperm defects increased with age in Hereford bulls fed to gain >1.75 kg/day, and significantly differed from bulls fed to gain approximately 1 kg/day (control). Even after the high nutrition diet was changed to a control diet, bulls previously receiving high nutrition continued to have lower semen quality. There was greater deposition of fat around the testicular vascular cone in the scrotal neck in bulls in the high nutrition group, and the difference between body and testes temperature was reduced in this group compared with that of bulls in the control group. This difference was still present after the high-energy diet was changed and the bulls had lost a considerable amount of weight, indicating that fat that has accumulated in the scrotum is more difficult to lose than other body fat (Skinner, 1981).

This summary of the literature supports the intuitive assumption that low nutrition has adverse effects on sexual development and reproductive function regardless of a bull’s age. Then again, most research seems to indicate that high nutrition is only beneficial during the first 6 months of life, which presents a challenge to bull producers. Beef bull calves are often nursed until 6 to 8 months of age, with very little attention being paid to their nutrition, whereas nutrition offered to dairy bull calves is often suboptimal. Efforts to obtain maximum weight gain during the first months after birth by offering highly nutritious diets and adopting management practices such as creep feeding should be compensated by reduced age at puberty and greater sperm production capacity in adult bulls. It is also clear that, although high nutrition diets after 6 months of age might be associated with greater SC, this effect is likely to be caused by fat accumulation in the scrotum and is not actually the result of greater testicular size. Moreover, sperm production, semen quality and serving capacity are all compromised in bulls receiving excessive nutrition after this age. Adjusting diets accordingly to maximize growth, but prevent over-conditioning after the peri-pubertal period, is advisable.

Testicular Thermoregulation

Testes in bulls are maintained at 4–5.5^oC below body temperature and this lower temperature is essential for normal spermatogenesis (Kastelic et al., 1995, 1997). Adequate testicular temperature is maintained by complex physiological mechanisms that involve the scrotum, the testicular vascular cones and the testes proper. The temperature is determined primarily by the heat conveyed to the testes by the arterial blood, and the temperature of the arterial blood reaching the testes is determined by the amount of heat lost in the testicular vascular cone. This, in turn, is largely dependent on the temperature of the venous blood leaving the testes. Because the testicular veins run on the surface of the testes just below the tunica albuginea, the temperature of the testicular venous blood is greatly affected by the temperature of the scrotum. Therefore, the scrotum, which is equipped with thermoreceptors, is responsible for actively regulating testicular temperature by controlling vascular flow, sweat gland activity and muscle contractility (Waites, 1970; Setchell, 1978; Sealfon and Zorgniotti, 1991).

Increased testicular temperature results in reduced vasoconstrictor tone of the smooth muscles of the skin arteries, which increases blood flow to the scrotum and heat loss by irradiation. Increased testicular temperature also results in increased scrotal sweat production and heat loss by evaporation; scrotal sweat gland density and sweat production are greater than in the skin of other parts of the body (Blazquez et al., 1988). The cremaster muscle and the dartos tunic become completely relaxed when testicular temperature rises, thus maintaining the testes as far away as possible from the abdomen, maximizing the contact of the testes with the scrotum and increasing the scrotal surface area for heat irradiation. Scrotal surface temperature is lower than both body and intra-testicular temperatures, and a positive temperature gradient is also observed between the proximal and distal parts of the scrotum (Plate 32). Scrotal surface temperatures of approximately 30–31^oC at the proximal portion and 28–29^oC at the distal portion have been reported in beef bulls (Cook et al., 1994; Kastelic et al., 1995, 1996a, 1997).

Anatomical changes of the testicular vascular cone as the bull ages and starts to produce sperm are indications that the efficiency of the countercurrent mechanism needs to improve in order to couple with increasing testicular metabolism. In Angus and Angus × Charolais bulls, the testicular vascular cone diameter measured by ultrasonography increased until approximately 13.5 months of age, or until 1 to 8 weeks before the SC reached a plateau (Brito et al., 2012c). In crossbred beef bulls, the length and diameter of the testicular artery in the vascular cone increased from 6 to 12 months of age (1.8 m and 1.9 mm versus 3.1 m and 3.5 mm, respectively), but did not increase significantly thereafter (Cook et al., 1994). In 15-month-old Angus bulls, the testicular artery length and volume in the vascular cone were 1.6 m and 6 ml, respectively (Brito et al., 2004a). Other studies have reported that the testicular vascular cone length was approximately 10 to 15 cm and that the testicular artery length varied from 1.2 to 4.5 m in adult beef bulls of several breeds (Kirby, 1953; Kirby and Harrison, 1954; Hofmann, 1960).

In addition to the lengthening of the testicular artery, the distance between the arterial and venous blood in the testicular vascular cone also decreased with age. This occurred via thinning of the arterial wall (from 317 to 195 μm thick at 6 and 36 months of age, respectively) and reduction of the distance between the artery and the closest veins (Cook et al., 1994). Another interesting observation was that the artery wall thickness and distance between the artery and the veins also decreased from the proximal to the distal portions of the testicular vascular cone (Hees et al., 1984; Cook et al., 1994; Brito et al., 2004a). The reduction of the distance between the warmer arterial blood and the cooler venous blood may facilitate heat transfer and compensate for the gradual reduction in the arterial–venous temperature gradient that occurs as blood flows through the vascular cone. Evaluation of intra-artery temperature on the dorsal portion of the testis indicated that the arterial blood was cooled down by 3–4.5°C after passing through the testicular vascular cone in 15–18 month-old beef bulls (Kastelic et al., 1997; Brito et al., 2004a).

The testicular artery reaches the dorsal pole of the testis and runs under the tunic albuginea, in close relation to the epididymal body, until it ramifies at the ventral pole. After ramification, several arterial branches run dorsally under the tunic until they penetrate into the testicular parenchyma. The temperature of the arterial blood decreases only slightly between the testicular dorsal pole and the ventral pole, but decreases significantly between the ventral pole and the point of penetration. This arrangement results in a negative sub-tunic temperature gradient between the dorsal and ventral portions of the testis. The combination of the opposing ‘top-to-bottom’ temperature gradients of the scrotum (positive gradient) and the testicular surface (negative gradient) results in a constant temperature throughout the entire testicular parenchyma. Intra-testicular temperature in beef bulls 15–18 months old was approximately 33.5–35^oC (Kastelic et al., 1996a, 1997; Brito et al., 2004a).

Sperm Production and Semen Quality

No differences were found in testicular development, spermatogenesis efficiency or extragonadal sperm reserves at 7 years of age between Holstein bulls that had been submitted to semen collection frequencies of either once or six times a week from 1 year of age using artificial vaginas, indicating that ejaculation at high frequency was not detrimental to either testicular development or sperm physiology. More importantly, the lack of adverse effects on sperm production and semen quality supported the long-term use of a high rate of ejaculation to maximize sperm harvest from bulls in AI centres (Almquist, 1982). With increased weekly ejaculation frequency there was a decrease in ejaculate volume, though sperm concentration was not affected (Almquist and Cunningham, 1967; Almquist et al., 1976). Although ejaculate volume either remained unchanged or decreased with multiple ejaculations on the same day, there was a gradual decrease in sperm concentration and total sperm per ejaculate. In Angus bulls from which seven consecutive ejaculates were obtained, total sperm number decreased continuously from the first to the third ejaculate, but did not change significantly thereafter. The first, first two and first four ejaculates contained 31, 55 and 77%, respectively, of the total sperm output obtained in the seven ejaculates (Foster et al., 1970). A small but significant increase in the proportion of motile sperm in consecutive ejaculates was observed in some studies (Almquist and Cunningham, 1967; Foster et al., 1970; Almquist et al., 1976). In 10–18 month-old Holstein bulls, volume and concentration of the first and second ejaculates were 4 and 3.3 ml and 1.14 and 0.63 billion/ml, respectively (Diarra et al., 1997), whereas in Holstein bulls of various ages, volume, sperm concentration and total sperm in the first and second ejaculates were 8.2 and 7.3 ml, 1.5 and 0.95 billion/ml and 12.0 and 6.8 billion, respectively (Everett and Bean, 1982). Ejaculate volume, sperm concentration and total sperm in dairy and beef B. taurus bulls in Brazil were 6.9 to 8.2 ml, 1.2 billion/ml and 8.2 to 9.1 billion, respectively (Brito et al., 2002a,b).

A good understanding of the dynamics of spermatogenesis is crucial for understanding abnormal sperm production. Total duration of spermatogenesis is 61 days in bulls. After spermiation, sperm are transported through the epididymis in a process that takes approximately another 10 days. Thus, sperm present in the ejaculate began being produced 71 days earlier and semen quality is a reflection of those events in the previous 2 months that influenced spermatogenesis, sperm maturation and transport. Different testicular germ cell lines have different sensitivities to different insults. For example, spermatocytes and spermatids are particularly sensitive to increased testicular temperature, but spermatogonia are more resistant (Setchell, 1998). Therefore, particular changes in semen quality are manifested after an interval that varies according to the developmental stage of the germ cells at the time of the insult, and the time required for the damaged cells to be released into the seminiferous tubules and transported through the epididymis. If exposure to the insult is limited, a consistent sequence of appearance of sperm defects in the ejaculate is expected, whereas if the exposure is prolonged, a variety of sperm defects may be simultaneously present in the ejaculate. Semen quality improves as the spermatogonia that resisted the insult enter mitosis and meiosis and restart generating normal sperm.

The mildest form of testicular degeneration produces no grossly detectable changes in the testicular tissue and is manifested exclusively by increased production of abnormal sperm. Elevated testicular temperature and endocrine disruption are probably the most common causes of mild testicular degeneration. Normal spermatogenesis in bulls depends upon maintenance of optimal testicular temperature, with increased testicular temperature having detrimental effects on sperm production and semen quality. Metabolic rate and oxygen demand increase as a result of augmented testicular temperature. However, the long and extremely coiled testicular artery limits the blood supply to the testes. As blood flow to the testes either does not increase at all, or does not increase sufficiently to match the increased metabolic rate of the heated tissue, testicular hypoxia develops (Setchell, 1998). Causes of increased testicular temperature include increased whole body temperature (high ambient temperature, pyrexia), increased local temperature (scrotal trauma or dermatitis, orchitis, periorchitis, epididymitis), decreased local heat irradiation (hydrocele, scrotal oedema, fat accumulation around the spermatic cords), and alteration of normal testicular mobility (tunic adhesions, inguinal and scrotal hernias).