Age-related changes in semen quality characteristics and expectations of reproductive longevity in Duroc boars

INTRODUCTION

Semen quality is generally considered to be a proxy measure of boar fertility and represents sire effects in pig production. Thus, during practice of artificial insemination (AI), quality characteristics of collected semen, such as sperm concentration, sperm motility, percentage of morphologically normal sperm and sperm functional competence, for example capacitation, are usually evaluated or kinetically analyzed to ensure its quality before service (q.v. review by Petrunkina et al. 2007). However, these characteristics vary with breed, social environment, season, age, nutrition, and management skills such as stockmanship, sanitary, and facility suitability (cf. Kunavongkrit et al. 2005). Among these variables, ambient temperature may be seen as the most noticeable and general factor that affects semen quality. Numerous references have showed that summer heat not only depresses libido but also reduces spermatogenesis in boars, as sperm concentration and normal sperm percentage are significantly reduced (q.v. Joseph et al. 2005). Among other variables that affecting semen quality, age can be recognized as a prominent and indiscriminate one, besides those of circumstantial and technical determinants of different husbandry practices.

Boars become adolescent at about 9 months old and sperm quality as well as morphometric features of leydig cells and seminiferous tubules mature during pubertal development (Allrich et al. 1983). Ejaculates from adult boars were superior to those aged less than 9 months and the daily sperm output gradually increased with age, being higher in 18-month-old boars than in younger ones (Cameron 1985). In a temperate environment of Canada, total sperm per ejaculate was highest for 32-month-old boars (Swierstra 1973). It has been reported in Canada also, that the maximum semen volume, sperm concentration and doses of semen were obtained from boars 24–29 months old (Kennedy & Wilkins 1984). Contemporary Duroc boars in a tropical environment (Thailand) showed a similar result of semen quality maturation, as their ejaculate volume and sperm number per ejaculate gradually increased at the age of collection from 9 to 33 months (Suriyasomboon et al. 2005).

The maximum quality of boar semen may last for a certain period but the duration has not yet been revealed. Nonetheless, once beyond a certain optimal age, semen quality would inevitably decline at an unknown rate to a level no more acceptable for service. The longevity of a boar's reproductive performance, in terms of least acceptable semen quality for AI, cannot be ascertained from the available literature.

Age and semen quality, soundness and libido are the common reasons for boar culling, in addition to herd genetic improvement which requires replacement of higher indexing boars. In a breeding herd or an AI station, where superior studs are kept, boar longevity has became a great interest since it involves maximizing dose or gene output from the stud to reduce the fixed and variable costs per extension service of semen, especially in the wake of rising feed-grain prices and awareness of energy-saving. However, related references concerning aging and semen quality in boars are rare and insufficient to illustrate a general understanding, not to mention a likely interaction of season with age that involves changes of thermoregulatory ability during aging. A study then is conducted in a commercial Duroc breeding unit for a large cooperation for marketing pig production, to reveal variation of semen quality in Duroc boars of different ages and its interaction with seasons, and to predict reproductive longevity.

MATERIALS AND METHODS

A total of 1441 quality evaluations consisting of 12 months continuous collection of semen (sperm-rich and post-sperm-rich fractions) from 58 boars of pure Duroc breed was used in the present study. Animals varied in age from 10 to 80 month and all started to serve as successful semen donors from 9 month old after training and qualifying. With minimum disease, they were all individually penned in well-ventilated and insulated houses and fed on a ration formulated according to NRC standards. Semen was collected every 5–7 days, but only every fortnight evaluation data were used for analysis. There were a few collections that showed below acceptable quality due to unexpected incidences and not qualified for the use of AI service. These evaluation data are not included in this analysis.

Once the required fraction of ejaculate was collected in a sterile beaker, covered with a triple-layer of gauze to serve as a gel filter, a cylinder immediately measured the volume. The collection was transferred to a nearby laboratory for quality evaluation. Upon arrival, sperm concentration was counted first by a photometer (Minitub SpermaCue Photometer: MINITUB Abfull-und Labortechnik GmbH & Co. KG, Tiefenbach, Germany). Then, an amount of approximately 10 mL semen was placed in a tube and centrifuged (3500 g, for 10 min) to obtain seminal plasma for determination of pH value and osmolarity, measured by a pH meter (780 pH Meter, Metrohm Ltd, Herisan, Switzerland) and an osmometer (M-3250 osmometer; Advanced Instruments, Norwood, MA, USA), respectively. A drop of semen sample was thereafter placed onto a warmed (38°C) slide, overlaid with a cover slip, and examined under a phase contrast microscope to determined sperm mobility and percentage of normal sperm by visual estimation according to Woelders (1991). Abnormalities included deviations of the acrosome head, and/or midpiece morphology, persistent proximal cytoplasmic droplets and coiled tails. Acrosome morphology can be categorized into normal apical ridge, damaged apical ridge, missing apical ridge and loose acrosomal cap (Pursel et al. 1972).

All the analyses were completed within 2 h of each collection and measurements were routinely compared and confirmed once a month by a computer-assisted sperm morphometry (Hamilton-Thorne analyzer, Version 12.2; Hamilton-Thorne Biosciences, Beverly, MA, USA). The evaluation data was divided into two groups for comparison, those grouped as hot season-included data gathered from May to October (n = 919, monthly mean temperature from 22.9°C to 29.9°C with average 86.6% RH), and cool season data consisted of the other 6 months (n = 522, monthly mean temperature from 14.0°C to 22.7°C with average 81.5% RH). Statistical software of SAS (2002) was used to test significance of the effects and interaction of season and age by GLM procedure rather than a mixed model, simply because data sets were only partially repeated measurements, although dependent. Adopting quadratic regression procedures generated best fittings.

RESULTS

Best fittings of total ejaculated volume collected against animal age in cool and hot seasons generated respective quadratic regressions (Fig. 1). It is noted that initially (10-month-old), two season-regressions differed little but gradually parted. In the cool season, volume reached a theoretical maximum point (M_c) of 242 mL at 46.2 months, while in the hot season, volume increased relatively less and attained a maximal point (M_h) of 218 mL at 36.0 months of age. In evidence, the hot season depressed maximum production of semen volume and accelerated aging effects.

Figure 1 also showed that hot and cool season regressions declined after reaching maximums at different rates that rendered quadratic regressions increasingly distant along with increasing age. The graduated separating of two regressions indicates different age responses to heat stress as young boars of around 1 year old could endure the hot season and provide an equal amount of semen volume as in the cool season. However, the endurance gradually diminished along with increasing age.

The hot maximum volume value (M_h) shown in Figure 1, horizontally extrapolated to M_hc-1 (23.0 months) and M_hc-2 (68.8 months) points to cool regression. This means the cool season kept boars of any age between 23.0- and 68.8-month-old, a 45.8-month duration, performing better than or equally well as 36-month-old boars exhibiting their best in the hot season. In addition, the average volume in 10-month-old boars in cool and hot seasons (185 mL) can be seen as a minimum requirement for acceptable quality. This essential level horizontally intercepts at 62.8 months in hot season regression (R_h point) and at 81.0 month in cool season regression (R_c point) and, respectively, yields the longevity in hot and cool seasons.

The changes of total sperm counts per ejaculation from boars of different ages in different seasons are shown in Figure 2. Similar quadratic plottings and horizontal extrapolations were also applied to mobile sperm number and total normal sperm number per ejaculation and results are showed in Figure 3 and in Figure 4, respectively. Curves of quadratic regression in 2-4 all share a similar start value, showing a similar pattern, but differ in regression coefficient with significant interaction (P < 0.01) and resulting in different peak values and estimated longevity.

The peak performance age and longevity of Duroc boars as semen donors, decided by quadratic plotting and extrapolations shown in 1-4 are summarized in Table 1. It is observed that, in theory, hot season-kept boars only have 1 month duration of peak performance at ages around 33 months (average from 4 estimates from different parameters). On the other hand, peak performance of cool season-kept boars could last for 48 months from 12 month old onward (based on normal sperm number). The least reproductive longevity of Duroc boars, as normal sperm donors, should be 51 months of age in a subtropical environment, and it may extend to 70 months old if summer heat stress can be avoided.

DISCUSSION

Changes in semen quality can occur from host factor effects such as age. Domestic pigs of conventional breeds do not attain mature weight within the first 3 years of life (Solanes & Stern 2001). Thus, it is reasonable to believe that boars would exhibit maximum quality of semen when they have reached full maturity. The literature reveals that maximum semen quality was obtained from boars 24–29 months old in a temperate area (Swierstra 1973; Kennedy & Wilkins 1984) and 33 months old in a tropical environment (Suriyasomboon et al. 2005). The present study, conducted in a semitropical region, generated a similar result, as measured quality characteristics of semen reached its maximum at 30–36 months old (hot season), or 38–46 months old (cool season). All the above-mentioned results, despite being approximately a quarter-century apart and obtained from different environment regions, are all fairly consistent in showing that the quality characteristics of semen were highest for boars about 3-year-old and of mature weight.

Advance in age beyond maturity is inevitably to be associated with changes in cellular or physiological functions. In healthy men, semen volume and sperm motility start to decrease continuously from 22 years of age (Chen et al. 2003; Eskenazi et al. 2003). Age-related alternations are due to many attributes, from morphological changes such as narrowing and sclerosis of the testicular tubular lumen, decrease in spermatogenic activity, increased degeneration of germ cells and decreased numbers of functions of leydig cells (Bishop 1970; Johnson 1986), to DNA damage, chromatin integrity, gene mutations and aneuploidies in sperm (Wyrobek et al. 2006). All these alternations and attributes observed in aging man may be commonly seen in other aging mammals, including boars.

The age-related declining rate of each quality characteristic measured in the presented study can be simply indicated by duration period from maximum point down to minimum acceptable (boar replacement or longevity) point. In the hot season, the period (M_h to R_h) is 26.8 month for total ejaculate volume, longer than that of any other sperm parameters measured (23.9, 23.6, 21.1 months for total sperm number, mobile sperm number and normal sperm number, respectively). In the cool season, the duration period of ejaculate volume (M_c to R_c) is also the longest when compared to that of sperm parameters. These results indicate that the semen volume declines at a rate slower (longer period) than sperm parameters of Duroc boars in either season. It means that spermatogenic function is more vulnerable to aging effects than accessory glands and chiefly seminal vesicles, since it is the place where the bulk of semen volume is derived. In men, Kidd et al. (2001) reviewed and concluded that, decreases in sperm motility to 37% and decreases in semen volume to 22% were likely. The relatively lower decrease in semen volume than in values of sperm parameters observed in either aging Duroc boars or men, suggests that the age-related different declining rate of testicle and accessory glands functions may be a common aging process in mammals.

Age also influences susceptibility of pigs to heat stress through either physical or physiological functions and, in consequence, affects semen quality. A better ability in thermoregulation of boars aged 10 months than those of full maturity (36 month) may mainly be due to body surface ratio difference. Applying an equation of body surface area (cm²) equal to 734 kg^0.656 in female pigs (Kelley et al. 1973) to boars, then, a 10 month old boar of 150 kg has a body surface ratio of 131 cm²/kg, which is about 30% more than 103 cm²/kg shown by mature boars of 300 kg. In addition, younger boars have less subcutaneous fat and less insulation, facilitating heat loss. The combined feature of higher surface area ratio and less insulation seems to provide a sufficient means for 10 month Duroc boars to endure subtropical heat and render semen quality differing little between cool and hot seasons.

The total contribution in quantity of a Duroc boar in his entire reproductive life can be calculated from a mean characteristics value per ejaculation times service duration (from 10 months old to replacement age point), and further multiplied by ejaculation frequency per month (assuming 10). For instance, the total collectable volume in a cool season would be 230 mL/ejaculation × 71 months (10 to R_c) × 10 ejaculation/month = 163.3 L in theory. In an actual environment with seasons alternating, the real collectable volume should be 226 mL/ejaculation (mean of cool and hot) × 53 month × 10 ejaculation/month = 119.8 L. The ratio of cool to actual equals 1.36, meaning an extra amount of 36% in semen volume could be obtained from a Duroc boar in his entire reproductive life if heat stress could be avoid. By a similar calculation, the quantity ratio of cool to actual in sperm parameters were 1.40 for sperm number, 1.51 for mobile sperm number and 1.80 for normal sperm number. Therefore, in view of normal sperm production, to maintain one Duroc boar in a temperate region (bellow 22°C) would be theoretically equivalent to keeping 1.8 boars in a subtropical environment.

It may be argued that boars beyond a certain age (e.g. 60 months old) may produce semen with normal ex-vivo quality and yet became less fertile once inseminated to sows. This was not verified by results obtained from following litter-recording systems in our study. There were four different boars (total n = 58) aged from 60 to 80 months involved in the investigation and a total of 112 collections obtained from them had been evaluated and inseminated. No following evidence to indicate conception rate and litter size of these 112 inseminations were inferior to those of younger boars. This means ex-vivo measures currently used for boar semen quality evaluation are applicable to boars up to 80 months old. It also implies that boar reproductive longevity is likely to be one-third of its life span, which may reach 20–25 years. (Donglas 1972)

Pig production shall be sustainable with environmental friendliness and social acceptance (ISO14045, 1999; Yang 2007). Life Cycle Assessment (ISO14045, 1999) has been used to evaluate different pig feed choices (Strid et al. 2005) and different pig production systems (Stern et al. 2005) to understand likely changes in economic and environmental impacts. Such an assessment should also be applicable to evaluate different scenarios for boar-raising and regression data generated from the present study may be essential in this.