AI has been in use worldwide for almost 50 years. As it reaches maturity, the
industry is demanding increased efficiency. Where will this come from?
One of the fundamental weaknesses of any AI system is that the fixed costs are
high, usually amounting to more than 50% of the total cost of a dose of semen.
Good house-keeping and careful negotiation with suppliers will reduce variable
costs, and this is always a worthwhile exercise; but reducing the fixed costs
must be a high priority target for research and development. The only way to have
an impact on the fixed costs per semen dose is to produce more doses per boar
place – effectively to serve more sows per boar. Alongside this objective
one must not lose sight of the importance of ensuring consistently good fertility
results with AI.
The key to enhancing AI efficiency is to find ways of reducing costs whilst never
Most commercial studs using 3.0 x 109 sperm per dose report an average
semen output of 20-22 doses per ejaculate. Collecting each boar approximately
75 times per year, this amounts to a total of 1500 doses per boar per year. Increasing
sperm output per unit time, or reducing the number of sperm required per insemination
dose both could result in increasing the number of doses produced. The average
number of semen doses used per sow per estrus is between 2 and 2.5. A boar’s
annual semen production of 1500 doses will therefore be sufficient to serve 600-750
estrous sows. Clearly, reducing the number of times a sow needs to be served is
also part of the overall efficiency equation.
Increasing sperm output per boar
Semen quality (expressed as the total number of viable sperm produced per unit
time) is influenced by a range of external factors (e.g. environmental conditions,
diet, collection frequency), and internal factors (e.g. health, genetics, age)
(Glossop, 1995). Efforts to enhance sperm production by manipulating one or more
of these factors tend to be carried out on rather an “ad hoc” basis,
with different AI studs operating their own preferred system, the results of which
may well not be repeatable elsewhere.
Dietary supplements and special boar rations designed to enhance sperm production
and boar fertility have also been tried and tested over the years, with variable
results (Wilson, 2000). A recent example of such work is the oral administration
of the fatty acid Docosahexaenoic Acid (DHA) in combination with the anti-oxidants
vitamin E and selenium in the form of Prosperm TM (JSR Clover Ltd). Workers claim
that DHA is an essential component of healthy sperm cells, enhancing membrane
integrity and tail flexibility, as well as increasing output. In field trials,
there was an average improvement of over 100 piglets per 100 sows served with
semen from boars on the supplementation regime, representing an 18:1 return on
investment. In addition, there was a significant effect on sperm production (Penny
et al., 2000). While such observations are interesting and important, and do demonstrate
a cost benefit, it is unlikely that this line of research will ever do more than
increase sperm production by 10-20%.
Reducing sperm numbers per insemination
Most boar studs are working with an insemination dose of 2.5-3.0 billion sperm,
although the early work of Chris Polge (1956) showed us that a minimum of 1.3
billion viable sperm is sufficient. As semen evaluation and preservation techniques
are improved there is potential to reduce the number of sperm required per insemination.
Improving semen assessment may allow reduction of the insemination dose, as well
as promoting standardisation. Reducing the insemination dose to 1 billion cells
would have a significant effect on costs, while enabling superior boars to be
spread even further. Techniques in current use rely on estimations of motility,
photometric measurement of sperm count and microscopical examination of sperm
morphology (Almond et al., 1998). Laser technology is being used to replace both
the microscope and the photometer, providing a simple, accurate system ensuring
Another new technology is computer-aided semen assessment, where image analysis
enables the number of moving sperm cells in a sample to be counted, while analysing
quality of movement. Other work is concentrating on sperm function tests using
differential stains and automated cell counting e.g. the short HOST (hypo osmotic
swelling test), which evaluates both the integrity of the plasma membrane in the
sperm tail and the status of the sperm head membrane (Vazquez et al., 2000).
One major area of new research is focusing on the possibility of reducing the
insemination dose by delivering the semen further into the sow’s reproductive
tract. The extreme version of this concept is the technique of “deep intra-uterine
insemination”. Work from Spain (Gil et al., 2000) has demonstrated that a
dose of 20 million sperm may be sufficient if sperm are deposited deep into the
uterine horns. While deep intrauterine insemination is a technique which may never
be appropriate or acceptable on commercial farms, outside of carefully controlled
conditions, and may even have welfare implications in some countries, this demonstrates
a significant potential in terms of reducing the semen dose. Other groups in Europe
and the USA have also been examining this important and exciting area in a slightly
more conservative way. Insemination simply through the cervix, and just into the
uterine body can also have an impact on the number of sperm required per dose.
There are some concerns about the possible damage caused by the use of extended
catheters inseminating deep into the uterus. Studies on the pathology of the uterus,
along with a consideration of the effect of such treatment on the long-term breeding
potential of a sow are overdue.
Enhancing sperm transport through the uterus may have the same effect as depositing
the semen deeply into it, with acceptable conception rates resulting from lower
sperm dose rate. Recent studies have investigated the role of seminal plasma in
sperm transport, identifying oestrogens as a key factor in this process (Waberski
et al., 1997). The use of synthetic seminal plasma to enhance sperm transport
and fertilization rates is based on the same principals (Lyczynski et al., 2000).
Treatment of the sow with PGF2µ either by injection or in the semen dose
at the time of insemination has been shown also to enhance this process (Pena
et al., 1998), as well as possibly having an impact on the timing of ovulation.
This area needs further study, particularly with reference to reducing the number
of sperm inseminated. It may be of even greater value where the sperm to be used
are particularly precious, valuable, scarce (e.g. a particular genotype, or sexed),
or in some way damaged (e.g. aged, or frozen/thawed) where motility and viability
may be impaired.
Intracytoplasmic sperm injection (ICSI), perhaps the ultimate in maximizing the
use of a single ejaculate, is really an adjunct to other embryo manipulations.
A single sperm is injected through the zona pellucida of an oocyte, thereby effecting
fertilization. It is preceded by retrieval of eggs from a donor sow, and is followed
by other in vitro manipulations before the fertilized eggs are introduced into
the recipient (Cameron, 1998). It offers a complex technique that is extremely
efficient in the use of sperm, and would be of particular value where sperm are
very precious e.g. from a subfertile or extremely rare sire, or possible sperm
which have been sexed. The technique is in use in human infertility cases, but
is not yet in widespread use in farm livestock due to the high costs and technical
Reducing number of insemination doses per
Timing is the single most important factor in successful application of AI. Consideration
of the timing of events leading to successful fertilization in the sow demonstrates
the importance of ensuring that viable sperm are present within the reproductive
tract in advance of ovulation. Ova remain viable in the oviduct for 4-8 hours
after ovulation, and exposure to sperm after this time results in reduced fertilization
rate and subsequent litter size (Hunter and Dziuk, 1968). One of the most exciting
areas of research now in progress is that of real-time ultrasonography of ovarian
activity, which is able to pinpoint the timing of ovulation and the factors which
influence this important event (Soede et al., 1995, Waberski et al., 1994). Single
fixed-time AI should be the long-term goal here, although the commercial AI sector
needs to consider exactly how this can be marketed whilst maintaining margin.
It is helpful to consider stud productivity in terms of the number of doses of
semen used per boar place per unit time, as this takes into account ordering the
correct number of doses and storing them properly on the farm as well as simply
maximizing output from the stud itself. While this is of more relevance when a
stud is producing semen for in-house use, it is a worthwhile exercise to monitor
semen wastage on any breeding unit, regardless of semen source.
Semen extenders expand the volume of the ejaculate and preserve its viability
for 3-7 days. Understanding of sperm membrane physiology will help to develop
longer-life extenders, which will in turn reduce semen wastage. Prolonging the
viability of semen for up to 8 days would be of great value as semen could be
delivered once a week, reducing transportation costs and wastage (Revell and Glossop,
1989). Another interesting area is the development of low temperature extenders
(e.g. 4 0 C) that would simplify semen shipping and storage. An understanding
of sperm membrane stability and the physiological requirements of these cells
are essential if a longer-life semen is to become a reality. The newer science
of micro-encapsulation also is being applied to sperm. This technique seeks to
entrap sperm in gel spheres that slowly dissolve to release them, and real progress
is now being made in this area (Vigo et al., (2000). Imagine inseminating a sow
once at the start of estrus, knowing that populations of live sperm will gradually
be released into the uterus throughout the time when ovulation may be occurring!
The cattle AI industry relies upon the use of frozen semen, whereas pig AI is
based almost entirely upon fresh semen, as fertility achieved with frozen pig
semen has been disappointing. Pig sperm are adversely affected by cooling and
freezing, and the cryoprotectants used to reduce damage to bull sperm during the
cooling and freezing processes do not have the same desirable effects with boar
sperm. The most promising techniques available involve freezing the semen in straws
or sachets, using egg yolk and glycerol as protectants against damage during the
cooling, freezing and thawing processes (Glossop, 1998). The optimum cooling and
freezing rates for pig sperm, identification of the best ingredients to protect
sperm from damage during the process, and development of the ideal shape for a
pig semen straw are all under review. Frozen semen would offer benefits to everyone
- simplifying international exchange of genes, and providing all producers with
a perfect contingency plan. Current freezing systems are expensive in terms of
time and sperm (a frozen insemination dose is double that of fresh semen) - but
the benefits are clear. This work may have to wait until the insemination dose
has been reduced in order to become commercially viable.
Producing tailor-made end products
The ability to predetermine the sex of piglets offers benefits throughout the
Separation of semen into populations enriched for sperm carrying either the X
or Y chromosome is one means of doing this, although embryo-sexing techniques
are also being developed. Any technique for sexing sperm must be rapid, and practical
in a commercial AI laboratory, while not damaging sperm in any way. Fluorescence
Activated Cell Sorting (Johnson et al., 2000) is an accurate though rather slow
technique (5 million sperm sorted per hour), and the fluorescent DNA stains used
in the process may cause damage to the genetic component of the sperm - a problem
that must be taken seriously. Other techniques are attempting to utilize immunological
differences between X and Y-bearing sperm to separate out the two different populations
en masse, removing the problems of sorting cells individually.
Welfare considerations of new breeding
As new techniques become available which enhance the productivity and profitability
of the swine industry, we must take seriously our responsibilities to the health
and welfare of the animals involved. In the UK, the welfare debate has been in
progress for 10 years or more. In
1995, the Banner Committee Report raised questions regarding the welfare of practices
such as AI and embryo transfer in a farm animals. AI in pigs was included in this
review, which in particular highlighted the importance of staff training and veterinary
involvement on units practicing such breeding technologies (Glossop, 2000). Stockmanship
and animal handling can vary widely and there are, clearly, welfare implications
when AI is carried out by badly trained, unsympathetic, disinterested staff. Provided
that staff have received appropriate training in the details of reproductive anatomy
and physiology, as well as in basic hygiene, these potential problems should be
minimized. Excessive numbers of insemination during estrus, dirty technique and
late insemination can all contribute to increased incidence of uterine infections
and vaginal discharge. It is, of course, easier to AI a sow, which is going off
heat, than to persuade her to stand to the boar. As further developments in breeding
technologies emerge it is important to consider their welfare implications and
make sure that adequate training is given to all those involved in their application.
*Almond,G, Britt,J, Flowers,B, Glossop,C, Levis, D, Morrow,M, and See,T (1998).
Swine AI Book, 2nd Edition, Ed. Ruth Cronje, North Carolina State University.
*Cameron, RDA (1998). Porcine reproduction now and in the future. Proceedings
of the 15 th
International Pig Veterinary Society Congress,, Birmingham, England, 209.
*Gil, J, Tortades, JM and Alevia, A (2000). Post cervical insemination. Proceedings
16th International Pig Veterinary Society Congress, Melbourne, p399.
* Glossop, CE (1995). Investigation into boar infertility (2). The Pig Journal,
*Glossop, CE (1998). AI in pigs: production of quality-assured, healthy semen.
20, (4), 182-188
*Glossop, CE (2000). Animal welfare and the artificial insemination industry.
Preservation IV, Beltsville, Maryland, August 2000. Pages 207-211.
*Hunter, RHF and Dziuk, PJ (1968). Sperm penetration of pig eggs in relation to
the timing of
ovulation and insemination. Journal of Reproduction and Fertility, 15, 199-208.
*Johnson, LA, Dobrinsky, JR, Guthrie, HD and Welch, GR (2000). Sex preselection
flow cytometric sorting of x- and y- chromosome bearing sperm to produce offspring.
Semen Preservation IV, Beltsville, Maryland, August 2000., 107.
*Lyczynski, A, Soczywko, T, Martin Rillo, S and de Alba Romero, C (2000). The
Predil-Mra Synthetic seminal plasma used the inseminate sows and gilts on their
efficiency. Boar Semen Preservation IV, Beltsville, Maryland, August 2000., 250.
*Pena, FJ, Dominguez, JC, Alegre, B and Pelaez, J (1998). Effect of vulvo-mucosal
of PGF2? at insemination on subsequent fertility and litter size in pigs under
Animal Reproduction Science, 52, 63-69.
*Penny, PC, Maldjian, A and Noble, RC (2000). An enhancement of boar fertility
reproductive performance. Proceedings of the 16th International Pig Veterinary
Congress, Melbourne, p404.
*Polge, C (1956). Artificial insemination in the pig. Veterinary Record, January
28 th 1956,
*Revell,SG and Glossop,CE, (1989). A long-life ambient temperature diluent for
Animal Production 48, 579-584
*Soede, NM, Wetzels, CCH, Zondag, W, de Koning, MAI and Kemp, B (1995). Effects
time of insemination relative to ovulation, as determined by ultrasonography,
rate and accessory sperm count in sows. Journal of Reproduction and Fertility,
*Vasquez, JM, Martinez, EA, Roca, J, Lucas, X and Gil, MA (2000). Hypoosmotic
test as a predictor of membrane integrity in boar spermatozoa. Boar Semen Preservation
Beltsville, Maryland, August 2000., 263.
*Vigo, D, Faustini, M, Munari, E, Pace, M, Russo, V, Torre, ML, Maggi, L, Conte,
Maffeo, G (2000). Boar sperm encapsulation in barium alginate membranes: mechanical
resistance of capsules and sperm kinetics. Proceedings of the 16th International
Veterinary Society Congress, Melbourne, p405.
*Waberski, D, Weitze, KF, Gleumes, T, Schwartz, M, Willmen, T, and Petzoldt, R
Effect of time of insemination relative to ovulation on fertility with liquid
and frozen semen.
Theriogenology, 42, 831-840.
*Waberski, D, Claassen, R, Hahn, T, Jungblut, PW, Parvizi, N, Kallweit, E, and
(1997). LH profile and advancement of ovulation after transcervical infusion of
plasma at different stages of estrus in gilts. Journal of Reproduction and Fertility,
*Wilson, ME, (2000). Nutritional effects on boar semen production. Boar Semen
IV, Beltsville, Maryland, August 2000., 193.