The Role Of Genetics: Lean Growth And Pork Quality

John Webb Cotswold Pig Development Company Limited, UK

Introduction

Over the next five years there can be little doubt that the pig industry will continue to compete on the efficient production of quality lean at minimum cost. Past genetic improvement programmes have been very successful in reducing backfat and improving feed conversion. For the future there are tow main genetic challenges. The first is to maintain the improvement in feed conversion as backfat levels decline. The second is to ensure that the quality of pig meat can keep pace with new developments in processing and retailing, so securing its place on the supermarket shelf.

The rapidly developing technologies of electronics and molecular genetics bring the possibility of faster improvement, and the opportunity for genetic change in traits, particularly meat quality, hitherto very difficult to measure. This second article on the role of genetics examines the strategy for safeguarding the competitive position of the pig industry in the growth of quality lean.

Components of Quality Lean Growth

Heritabilities and heterosis levels for the components of growth and quality are shown in Table 1. Unlike sow productivity lean growth traits have high heritabilities, low heterosis and are easy to measure. On ad libitum feeding, growth rate has a favourable genetic correlation (0.5) with feed conversion, but an unfavourable correlation (0.2) with lean percentage. The quality traits have lower heritabilities, partly due to poor accuracy of measurement, and heterosis levels are not well estimated, but assumed low. Heritabilities may well have been over-estimated in the past due to the presence of the halothane gene.

Table 1: Components of Lean Growth and Pork Quality

Trait Heritability % Heterosis %
Lean growth
Growth rate 30 7
Feed conversion ratio 30 3
Daily feed intake 25 7
Backfat 40 -2
Lean % 40 0
Killing out % 25 0
Pork quality
Meat colour 15
pH45 15
Tenderness 20
Juiciness 20
Taste 20

Efficient Lean Growth

Past selection on a simple performance test of growth rate and ultrasonic backfat from roughly 30 kg to 100 kg has been highly effective. For example in Canada the annual genetic trends over the past 20 years have been 1.4% for backfat and 0.4% for growth rate (Kennedy et al, 1996). It is well recognized that this selection can take one of two routes:

  • increased daily lean growth, with little change in daily fat deposition
  • decreased daily fat deposition through a reduction in daily feed intake, with little change in daily lean growth.

Selection programmes which have placed heavy emphasis on low backfat and feed conversion at the expense of growth rate on ad libitum feeding have taken the second route of a decline in feed intake. This has recently been confirmed in a UK selection study at BBSRC Roslin Institute (Edinburgh) and Wye College (London University) where halothane-free Large White and Landrace lines were selected over eight generations for four objectives:

  • lean tissue growth rate (LTGR) on ad libitum feeding
  • lean tissue growth rate (LTGR) on scale feeding (70% ad lib)
  • lean tissue feed conversion (LTGR) on ad libitum feeding
  • voluntary feed intake (VFI) on ad libitum feeding

As shown in Table 2, selection for LTFC on ad lib feeding led to a strong reduction in daily feed intake and a smaller increase in lean growth rate than selection for LTGR, which actually increased intake. Selection for high VFI produced the expected increase in backfat, but with little change in lean growth rate. With the particular selection index weightings used, the study suggested that LTGR selection on scale feeding might be more successful in improving efficiency under ad lib. In practice selection on ad lib is preferred, since this allows expression of voluntary feed intake, but the index weightings on component traits must be chose to ensure that emphasis falls on efficient and rapid deposition of lean.

Table 2: Responses to 4 generations of section for rate (LTGR) and efficiency (LTFC) of lean growth and voluntary feed intake (VFI) on ad libitum feeding in Edinburgh - Wye lines (table entries are differences high minus low divergent lines)

Trait

LTGR LTFC VFI
Daily live weight gain g +76 +10 +73
Daily feed consumption g +25 +131 +272
Feed conversion ratio -0.24 -0.18 +0.13
Average backfat mm -3.0 -3.1 +3
Lean tissue growth rate g/day +47 +30 +3
Lean tissue feed conversion g/kg +44 +36 -18

From: Cameron and Curran (1994)

Most of the past improvement in feed conversion has arisen from the exchange of lean for fat, where fat deposition requires some 3.5 times more energy than lean. As fat levels decline,t he scope for further genetic improvement of feed conversion by changing body composition is becoming rapidly reduced. Once there is no further fat to be removed, the main route to improved efficiency will be to grow the lean faster and use less of the total feed for maintenance. Hence the main selection objective for the future will be lean tissue growth rate (LTGR), and the challenge will be to identify pigs which can eat more but convert the extra feed to lean rather than fat.

Feed Intake Measurements

Theoretically a measurement of individual feed intake on ad lib performance test can add 15 to 20% to the rate of genetic improvement in efficient lean growth. For some years, the only way to measure this was by individual penning, with the risk that the isolated social environment could lead to behavioural changes which might negate any benefits under commercial conditions. This has been overcome by electronic feeding stations with transponder ear tags, which record individual intake for pigs fed ad-lib and penned in groups.

One such system (Hunday FIRE) has been used in Cotswold sire line nucleus populations for six years. As each meal is individually recorded, the system provides detailed measurements of feeding pattern and behaviour. In co-sponsored research at SAC Edinburgh, there was no indication that feed intakes differed from normal commercial levels, or, more important, that selection in the stations would increase aggressive feeding behaviour (Nielsen, 1995). However, feeding pattern showed wise variation in terms of number of means per day and rate of eating, and at higher stocking rates pigs readily adapted by eating fewer larger and faster meals with little impact on lean growth.

Recent genetic analysis has shown that many of the feeding pattern traits are quite highly inherited (Table 3), and through their correlations with lean growth, might actually be used to speed up genetic improvement. In the Cotswold nucleus feed intake records from 45 to 95 kg increased improvement by around 18%. Further addition of feeding pattern traits raised this figure to 24$ (A.D. Hall, personal communication).

Table 3: Estimated feeding pattern parameters from Hunday FIRE feeders in a Cotswold UK sire line.

Trait Mean Heritability Genetic correlation with:
Average backfat Feed Conversion Growth rate
Daily feed intake kg 2.06 0.21 0.78 0.65 0.61
Feed intake per visit kg 0.20 0.27 0.35 -0.12 0.49
Visit per day 10.1 0.34 -0.15 0.31 -0.29
Time per visit min 6.0 0.11 0.17 -0.27 0.33
Feeding rate kg/min 0.14 0.00 -0.08 0.25 -0.39

From: Hall et al (1997)

Maximising Lean Tissue Growth Rate

Taking a simple view, this pig is over-fat because its feed intake is not in balance with its genetic potential for LTGR. Fat is then a consequence of a genetic mis-match between the two traits. In the ideal pig of the future, the animal would eat to realise its potential LTGR and a pre-determined level of fat. In practice the modern pig may eat too little to fully realise the potential LTGR prior to 40 kg, and eat too much after 70 kg (Figure 1).

The electronic feeding stations now offer an individual feed intake curve for each pig. The opportunity therefore already exists to select pigs to ear more in early life and less later on, changing the shape of the feed intake curve (Figure 2). However this requires a knowledge of genetic potential for lean growth throughout the finishing period. To find this out Cotswold has recently completed a serial dissection study with Bristol University, slaughtering pigs from FIRE feeders at 20, 30, 60, 80, 100 and 120 kg live. The resulting lean growth patterns brought two surprises. First, muscle growth rates of over 400 g/day were reached before 40 kg, suggesting that selection to increase early intake may indeed be worthwhile (Figure 3). Second, the relationship between muscle weight and live weight was almost linear, making prediction of LTGR very easy (Figure 4).

Carcase Value

The financial value of the carcase to the processor depends on the quantity, quality and distribution of lean in the carcase, and may very probably differ from the fat or lean measurements on which payment is based. Probes of the Fat-O-Meater and Hennessy type estimate only lean percentages, and the accuracy of the prediction equations used varies notoriously across lines differing in muscle depth. Genetic improvement programmes require more accurate predictions of financial value from ultrasonic measurements in the live pig. In one Cotswold trial the accuracy (R2) of ultrasonic fat and muscle depths in predicting value was as low as 0.58 and 0.25 respectively, but the best equation using all measurements gave an accuracy of 0.81. The means to select and indeed pay producers on financial value therefore already exists.

INSERT FIGURE 1

Two-dimensional (real-time) scanners are now in routine use for Cotswold sire line populations. Software is now available to provide an automatic computer analysis of the ultrasonic image of the cross-section of the longissimus dorsi (Liu and Stouffer, 1995). Eventually this could provide muscle: fat ratios in the streak as well as intramuscular fat content, although recent preliminary trials in Canada gave poor success (Sather et al, 1996). More expensive techniques such as X-ray, CT scanning offer more detailed measurement of muscle properties such as water holding capacity, but require anaesthesia and are prohibitively expensive for all but research. TOBEC (total body electrical conductivity) methods being developed in the US measure total lean content (Akridge, 1992), but with little extra accuracy and at far greater cost than ultrasonics.

Population Structure

Population structures such as the group nucleus to improve little size make use of very cheap measurements (litter size, growth rate, backfat) in a large population (see Webb, 1997). Electronic measures of individual feed intake are much more expensive and could not economically be applied to all the progeny from 1000 sows. The solution adopted for Cotswold sire lines is an open nucleus (Figure 5), in which expensive measurements of intake and real-time ultrasonics are confined to a single core nucleus farm of 300 sows.

Genes pass from the core nucleus via AI to an array of four multipliers producing purebred terminal sires for sale. Boars on the multiplier units undergo a less expensive ad lib group-fed performance test measuring only growth rate and backfat. With AI sires in common, the genetic merit of boars at nucleus and multiplier levels can then be directly compared using BLUP. When by chance boars of exceptionally high merit occur at multiplication, their genes can be returned to the nucleus via AI. Although, with lean growth traits of high heritability, the nucleus of 300 sows is more than enough for a competitive rate of improvement, the additional 1200 sows at multiplication greatly reduces the rate of inbreeding, giving longer term security.

Nucleus populations of course represent a sizeable, albeit essential, overhead cost on the production base. Unfortunately more accurate BLUP selection results in faster inbreeding, requiring larger populations. However, new methods are being developed which balance inbreeding and selection by combining BLUP estimates of merit with individual inbreeding coefficients to give a single criterion of sections (Grundy et al, 1996). Hopefully, this will allow some reduction in the size of nucleus populations.

Halothane gene

Since 1991, HAL-1843 DNA test, available on licence from the Toronto Innovations Foundation, has offered a means of eliminating this gene. Its effects in increasing lean yield along with liability to stress deaths and PSS have been very well documented. The debate centres on the case for continuing to produce heterozygous (Nn) terminal boars for mating to homozygous negative females (NN) to give a slaughter generation containing 50% Nn : 50% NN. Within litters it is clear that Nn do have a 1-3% advantage in lean yield. However, as illustrated by a recent French study in Table 4, evidence continues to accumulate that the gene (n) is also additive rather than recessive for meat quality, with Nn inferior to NN.

There are several reasons for favouring complete elimination:

  • new controlled atmosphere packaging systems to prolong the shelf-life of fresh pork require the meat to retain its natural content of water to the fullest extent.
  • as lean % and LTCR increase be selection, it is very likely that the adverse effects of the gene on quality and stress susceptibility will also increase.
  • as well as the high cost, there are welfare concerns over maintaining stress-susceptible (nn) nucleus populations.
  • continued presence of the halothane gene may obscure important longer term relationships between meat quantity and quality.

After some 15000 DNA tests, Cotswold is in the final stages of elimination from all lines. In one very lean composite line, the frequency of the gene has been reduced from 90% to zero over ten years, first by halothane testing, then by blood typing for neighbouring marker genes (PHI, P)2, 6-PGD), and finally by DNA testing (Figure 6). During this time the lean content lost with the halothane gene has been almost completely restored to within 0.5% lean by selection on the many other genes with smaller effects on yield. Meat colour however has been improved in the pure line by 40%. The line is currently used to produce a crossbred terminal sire bearing the HAL-1843 nm trademark guaranteeing freedom from the gene.

Table 4: Influence of halothane genotype on meat quality in 224 F2 Pietrain x Large White pigs

Trait NN Nn nn
pH90 6.36 6.13 5.65
Lactate umol/g 67 76 94
Glycogen umol/g 36 28 18
Colour L* 51.9 53.9 58.4
Cohesiveness 9.6 10.0 10.9
Toughness 8.5 9.2 10.7
Easiness to swallow 10.7 10.1 9.3
From Larzul et al (1996)

Pork Quality

Once the halothane gene has been removed, there are a number of critical questions for the future:

  1. How far will quality deteriorate as a consequence of continued selection for lean growth?
  2. What is the scope for further improvement of quality?
  3. How much of the selection effort within lines should be diverted to quality traits?

In a recent study of halothane negative (NN) Dutch Yorkshires, there were no favourable genetic correlations of lean growth with water holding capacity or meat colour (De Vries et al, 1994). However, each 1% genetic increase in lean content was predicted to reduce intramuscular fat by around 0.07%. A threshold at which low intramuscular fat becomes a limiting factor on quality could therefore be reached within 5 to 10 years. If so, in vivo prediction of intramuscular fat becomes a priority. Dutch reports of a marker gene accounting for half the genetic variation in intramuscular fat are still under investigation.

Duroc crosses certainly improve tenderness, but at the expense of lean yield and feed conversion. Less clear is the extent to which this is due to higher intramuscular fat content. UK MLC trials (1991) suggest that at least 50% Duroc may be needed for a worthwhile improvement, requiring a very high premium to compensate for the loss of lean growth. Meishan crosses appear to show less advantage in eating quality than their Duroc counterparts, even though the Meishans probably have more intramuscular fat. The RN-gene discovered by the French reduces curing yield but improves eating quality, and appears confined to the Hampshire breed.

Until quality can be more easily measured in the live pig, selection effort is probably best concentrated on lean growth. Meanwhile research is urgently needed on the basic biology and measurement of quality. Recent studies suggest that fast lean growth may improve quality through increased post mortem activity of proteolytic enzymes. Another approach might be to select for a more favourable distribution of fibre types in the muscle by using marker genes.

Risks of lean growth selection

As faster methods of genetic change take populations further from their foundation performance, there must be a risk of adverse effects from selection for lean growth. In UK the unfavourable genetic correlation between backfat and leg condition (0.2) was recognised early on, and today some 20 to 40% of selection effort in breeding companies is still devoted to physical condition. As fat levels fall, the greatest risk must be to reproduction, particularly sire line nucleus herds. For example, the Edinburgh-Wye selection lines show signs of reductions in conception rate, litter size and litter weight for lines with heavy emphasis on either efficiency or low feed intake (Table 5). Fortunately the preferred objective of lean growth on ad lib so far shows no adverse changes.

Table 5: Changes in reproduction of gilts after 7 generations of selection in Edinburgh Large White lines (table entries are differences high minus low divergent lines).

Selection objective Test regime Conception rate % Number born Litter weight kg
Lean feed conversion ad lib +2 -0.3 -2.3
Lean growth ad lib +4 +0.8 +0.4
Lean growth scale -19 -0.7 +0.9
Daily feed intake ad lib -7 -0.5 -0/9

From: Kerr and Cameron (1996)

Potentially even more worrying is the suggestion from Iowa State University that genetically improved lines could suffer a proportionally greater drop in performance when exposed to disease. This has yet to be confirmed elsewhere, but is another good reason to promote research on fundamental understanding of immune responsiveness.

Choice of Genetics Package

How then does the commercial producer choose the right source of breeding stock? Random sample testing is not the way forward since it always suffers from disease, a non-representative test environment, a long time delay, and worst of all the bias which accompanies small samples. The important indicators are:

  • the size of nucleus populations and therefore the selection differentials
  • the investment in performance testing and policing the selection process
  • balance in performance traits between sire and dam lines
  • attention to heterosis and physical soundness
  • price and level of service
  • practical advice on feeding and management, together with performance targets
  • commitment to research for the future
  • flexibility to change quickly with the market, and the resources to quickly exploit new technology

With modern recording schemes, producers should monitor carefully the consequences of any change including genotype on the farm. Genetic-nutritional computer models are available to assist in identifying the best combination of genotype and feeding regime for any given situation.

New technology

What new technology is likely to accelerate genetic change? Gene transfer is in its infancy, and transgenic pigs containing extra growth genes have experienced well-publicised side effects on reproduction and locomotion. Transfer of sry and related genes controlling maleness on the Y- chromosome might conceivably offer a means of producing a single gender to increase both efficiency and uniformity. Some form of genetic manipulation might help in reducing protein turnover in order to cut down nitrogen excretion. For feed efficiency, genetic manipulation might be better targeted to fodder crops rather than animals.

A major goal has to be a better biological understanding of the animal and its behaviour. For example the recently discovered ob mutation in mice causes obesity by over-eating. The cause is a deficiency of a satiety factor leptin, which when injected causes fat mice to become thin. This could be a first step in understanding the control of feed intake at molecular level, offering the prospect of correcting the pig's major deficiency of over-consumption.

Conclusions

Until new methods are available for measuring pork quality in the live pig, genetic improvement of finishing performance should focus on reducing the cost per kilo of lean. As fat levels fall, this will be achieved by shifting the selection emphasis to lean tissue growth rate (LTGR). Electronic measures of feed intake on performance test can speed up genetic improvement without adverse changes in behaviours, and may offer a means of better balancing voluntary feed intake and LTGR throughout the finishing period. Real-time ultrasound offers the most cost-effective prospect for improvement of carcase value.

To maximise pork quality, the halothane gene should be completely eliminated. Research should be stepped up on the fundamental biology affecting muscle quality and growth. Meanwhile there seems no limit in sight to the further improvement of LTGR, with live weight gains of up to 2 kg per day from 30 to 100 kg biologically attainable. As for sow productivity, it is clear that genetic improvement of lean growth can continue to make a major impact on the competitive position of the pig industry for many years. To make best use of higher genetic potential, geneticists, nutritionists and veterinarians will need to work closely together to ensure that the animal and its environment are perfectly matched for maximum efficiency.

References

Akridge, J.T., Brorsen, B.W., Whipker, L.D., Forrest, J.C., Kuei, C.H. and Schinckel, A.P. (1992)
Evaluation of alternative techniques to determine pork carcase value. Journal of Animal Science 70 (1): 18-28
Cameron, N.D. and Curran, M.K. (1994)
election for components of efficient lean growth rate in pigs 4. Genetic and phenotypic parameter estimates and correlated responses in performance test traits with ad libitum feeding. Animal Production 59: 281-291
De Vries, A.G., van der Wal, PG., Long, T., Eikelenboom, G. And Merks, J.W.M. (1994)
Genetic parameters of pork quality and production traits in Yorkshire populations. Livestock Production Science 40: 277-289
Grundy, B., Villanueva, B. and Woolliams, J.A. (1996)
Utilising genetic contributions to maximise longterm response to selection. BSAS Winter Meeting, paper 115 (abstract)
Jones, F.A. and Wiseman, J. (1997)
Relationship between carcase muscle weight and live weight in two diverse pig lines. BSAS Winter Meeting (Abstract)
Kennedy, B.W., Quinton, V.M. and Smith, C. (1996)
Genetic changes in Canadian performance-tested pigs for fat depth and growth rate. Canadian Journal of Animal Science 76: 41-48
Kerr, J.C. and Cameron, N.D. (1996)
Responses in gilt traits measured during performance test, at mating and at farrowing with selection for components of efficient lean growth rate. Animal Science 63: 235-241
Larzul, C., Rousset - Akrin, S., Le Roy, P., Gogue, J., Talmant, A., Vernin, P., Touraille, C., Morin G. And Sellier, P. (1996)
Effect of halothane genotype on texture characteristics of pig meat. Journees de la Recherche Porcine en France 28: 39-44
Lui, Y. And Stouffer, J.R. (1995)
Pork carcase evaluation with an automated and computerised ultrasonic system. Journal of Animal Science 73 (1): 29-39
Meat and Livestock Commission (1991)
Stotfold Pig Development Unit Second Trial Results. Milton Keynes, UK
Nielsen., B. L., Lawrence, A.B. and Whittemore, C.T. (1996)
Feeding behaviour of growing pigs using single or multi-space feeders. Applied Animal Science 47: 235-246
Sather, A.P., Bailey, D.R.C. and Jones, S.D.M. (1996)
Real-time ultrasound image analysis for the estimation of carcase yield and pork quality. Canadian Journal of Animal Science 76: 55-62
Webb, A.J. (1997)
The role of genetics: sow productivity. Manitoba Swine Seminar