Selection for feed efficiency traits and correlated genetic responses in feed intake and weight gain of Nellore cattle A. L. Grion, M. E. Z. Mercadante, J. N. S. G. Cyrillo, S. F. M. Bonilha, E. Magnani and R. H. Branco J ANIM SCI 2014, 92:955-965. doi: 10.2527/jas.2013-6682 originally published online February 3, 2014
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.journalofanimalscience.org/content/92/3/955
www.asas.org
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency traits and correlated genetic responses in feed intake and weight gain of Nellore cattle1 A. L. Grion, M. E. Z. Mercadante,2 J. N. S. G. Cyrillo, S. F. M. Bonilha, E. Magnani, and R. H. Branco Instituto de Zootecnia, Centro APTA Bovinos de Corte. Sertãozinho, SP, Brazil, 14160-970
ABSTRACT: The objectives of this study were to estimate genetic parameters for indicator traits of feed efficiency and to recommend traits that would result in better responses to selection for increased weaning weight (weaning weight adjusted to 210 d of age [W210]), ADG, and metabolic BW (BW0.75) and lower DMI. Records of W210 from 8,004 Nellore animals born between 1978 and 2011 and postweaning performance test records from 678 males and females born between 2004 and 2011 were used. The following feed efficiency traits were evaluated: G:F, partial efficiency of growth (PEG), relative growth rate (RGR), Kleiber’s ratio (KR), residual feed intake (RFI), residual weight gain (RWG), and residual intake and gain (RIG). Covariance and variance components were estimated by the restricted maximum likelihood method using multitrait analysis under an animal model. Estimates of genetic gain and correlated responses were obtained considering single-stage
and 2-stage selection. Heritability estimates were 0.22 ± 0.03 (W210), 0.60 ± 0.08 (DMI), 0.42 ± 0.08 (ADG), 0.56 ± 0.06 (BW0.75), 0.19 ± 0.07 (G:F), 0.25 ± 0.09 (PEG), 0.19 ± 0.07 (RGR), 0.22 ± 0.07 (KR), 0.33 ± 0.10 (RFI), 0.13 ± 0.07 (RWG), and 0.19 ± 0.08 (RIG). The genetic correlations of DMI with W210 (0.64 ± 0.10), ADG (0.87 ± 0.06), and BW0.75 (0.84 ± 0.05) were high. The only efficiency traits showing favorable responses to selection for lower DMI were G:F, PEG, RFI, and RIG. However, the use of G:F, PEG, or RFI as a selection criterion results in unfavorable correlated responses in some growth traits. The linear combination of RFI and RWG through RIG is the best selection criterion to obtain favorable responses in postweaning growth and feed intake of Nellore cattle in single-stage selection. Genetic gains in feed efficiency are expected even after preselection for W210 and subsequent feed efficiency testing of the preselected animals.
Key words: Bos indicus, feed intake, genetic correlation, genetic gain, selection objective © 2014 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2014.92:955–965 doi:10.2527/jas2013-6682 INTRODUCTION The efficiency of animal production systems is very important not only from an economic point of view but also from a socio-environmental viewpoint (Hegarty et al., 2007). Several measures that associate inputs (raw material), outputs (products), and time have been proposed over the years, especially the selection of animals that combine higher production with less expenditure and less time (Kleiber, 1936; Koch et al., 1963; Fitzhugh and Taylor, 1971). 1This
work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo, Conselho Nacional de Desenvolvimento Científico e Tecnológico and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior. 2Corresponding author: [email protected] Received May 8, 2013. Accepted January 3, 2014.
The correct use of feed efficiency traits in genetic evaluation programs requires analysis of the additive genetic variability of each trait and of genetic correlations between traits related to inputs (generally corresponding to DMI) and products (generally corresponding to weight gain rate or BW or both). In addition, estimates of genetic gain will permit determination of the most appropriate selection criteria and methods for a given selection objective (Bourdon, 2000). In this respect, an increase in body growth and a decrease in feed intake are the main outcome measures of a breeding program whose objective is to improve feed efficiency using the traits described for this purpose as selection criteria. Few studies have investigated genetic parameters and the effects of selection for efficiency traits in Zebu cattle. Therefore, the objectives of the present study were to estimate genetic parameters for indicator traits of feed efficiency, including G:F, partial efficiency of
955 Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
956
Grion et al.
growth (PEG), relative growth rate (RGR), Kleiber’s ratio (KR), residual feed intake (RFI), residual weight gain (RWG), and residual intake and gain (RIG) and to recommend traits that would result in the best responses to selection based on the estimated correlated responses of these traits with weaning weight adjusted to 210 d of age (W210), DMI, ADG, and metabolic BW (BW0.75). MATERIAL AND METHODS Records of feed efficiency traits from 785 Nellore animals born between 2004 and 2011 and weaning weight records from 8,004 animals born between 1978 and 2011 were used. The animals belonged to 3 lines of selection started in 1978 at Centro Avançado de Pesquisa Tecnológica do Agronegócio, Bovinos de Corte, Instituto de Zootecnia, Sertãozinho, São Paulo, Brazil. Male animals were selected for weight adjusted to 378 d of age obtained in weight gain tests (Razook et al., 1997) and female animals were selected for weight adjusted to 550 d obtained on pasture. In the control (NeC) Nellore line, animals were selected for the mean of the contemporary group, with a selection differential of about zero. In the selection (NeS) and traditional (NeT) lines, the animals were selected for higher weights, that is, the maximum selection differential for the traits described above (Mercadante and Razook, 2010). The difference between the NeS and NeT lines is that the NeS line, like the NeC line, is a closed experimental line whereas sires from other herds were used in the NeT line, mimicking a conventional commercial herd. The animals were submitted to the same management until they entered the performance test. Performance testing was used to evaluate individual feed intake after weaning. Before the tests, the animals were submitted to a period of adaptation to the diet and facilities for at least 21 d. The parameters of the performance tests are shown in Table 1. In 2005, 2006, and 2009, females were subdivided into 2 test groups performed in sequence due to the limited availability of individual pens. After the adaptation period, 32 randomly chosen females were allocated to individual pens for the measurement of feed intake whereas the other 32 females remained in the collective pens until the end of the test of the first group. Next, the second group was tested. This approach explains the average difference of 3 mo of age and almost 82 kg between the 2 groups at the beginning of the performance tests. After 2010, the 2 groups of females were evaluated simultaneously in different facilities. For males, only 1 group of animals was evaluated in individual pens per year between 2007 and 2011. In 2012, 31 animals were evaluated as done in the previous years and 87 animals were analyzed in collective pens using the GrowSafe system (GrowSafe Systems Ltd., Airdrie, Alberta, Canada).
Table 1. Groups of animals submitted to performance testing Year 2005 2005 2006 2006 2007 2008 2009 2009 2009 2010 2010 2010 2011 2011 2011 2012 2012 2012 2012 Overall
Sex1 F F F F M M M F F M F F M F F M M F F –
Duration, No. of Herd2 d animals NeC and NeS 56 32 NeC and NeS 56 32 NeC and NeS 70 32 NeC and NeS 70 32 NeC and NeS 112 61 NeC and NeS 112 62 NeT 70 60 NeT 85 32 NeT 84 32 NeT 73 60 NeT 84 32 NeT 84 24 NeT 72 62 NeT 84 31 NeT 84 25 NeT 91 31 NeC, NeS 91 87 and NeT NeT 86 26 NeT 86 32 – 83 ± 163 785
Initial Initial weight,3 age,3 d kg 263 ± 18 164 ± 28 341 ± 18 233 ± 34 267 ± 24 155 ± 30 350 ± 21 242 ± 48 268 ± 14 205 ± 31 249 ± 25 194 ± 36 300 ± 21 295 ± 32 272 ± 25 203 ± 28 391 ± 22 292 ± 26 262 ± 22 248 ± 40 294 ± 19 212 ± 28 281 ± 22 208 ± 25 261 ± 23 243 ± 40 294 ± 23 220 ± 33 286 ± 30 212 ± 28 277 ± 19 250 ± 31 270 ± 23 239 ± 41 225 ± 24 223 ± 25 287 ± 40
257 ± 29 256 ± 28 230 ± 49
1F
= female; M = male. = control herd; NeS = selection herd; NeT = traditional herd. 3Mean ± SD. 2NeC
Body weight was recorded at intervals of 14 d after a 12-h fast in 2005 and 2006 and at intervals of 28 d after fasting in 2007 and 2008 for males and from 2009 to 2011 for females. After 2009, males were weighed weekly without fasting, including 3 weight recordings per week on consecutive days in 2009 and 2010, 2 recordings per week on consecutive days in 2011, and 1 recording per week in 2012. The females tested in 2012 were weighed on 2 consecutive days at intervals of 14 d. Therefore, each animal was weighed at least 4 times with prior fasting and at least 11 times without prior fasting. The diets offered over the years differed in terms of the composition and source of feeds but were formulated to have isoenergetic and isoprotein levels (62% total digestible nutrients and 13% CP). In the individual pens, the diet was offered 2 times per day and adjusted daily and individually to permit an amount of leftovers of 5 to 10%, thus guaranteeing ad libitum intake. Weekly samplings of the diet offered and of leftover content were analyzed in composite samples at intervals of 28 d for the determination of DM. In individual pens, daily feed intake was calculated as the difference between the amount of food offered and leftovers. In collective pens, the troughs always contained food and were cleaned weekly. Individual daily intake (as-fed basis) recorded automatically by the GrowSafe system was multiplied by DM of the food offered. Daily feed intake records
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency
957
Table 2. Definition of the indicator traits of feed efficiency Formula1 ADG/DMI
Trait G:F, kg gain/kg DM
Definition Weight gain divided by the amount of feed consumed. Higher values are favorable. Partial efficiency of growth (PEG), kg ADG/(DMI – DMIm) Quantity of weight gain per kilogram feed available for this purpose, that is, subtracting total true gain/kg DM DMI from estimated DMI for maintenance. Higher values are favorable. Relative growth rate (RGR), kg BW/d 100 × [(logBWf – logBWi)/ Growth potential in relation to degree of maturity. days in test] Higher values are favorable. Kleiber’s ratio (KR), g gain/kg BW0.75 100 × (ADG/BW0.75) ADG, in grams, proportional to each kilogram of metabolic weight. Higher values are favorable. Residual feed intake (RFI), kg DM/d DMI – DMIe Difference between observed and estimated DMI based on ADG and BW0.75. Lower values are favorable. Residual weight gain (RWG), kg gain/d ADG – ADGe Difference between observed and estimated ADG based on DMI and BW0.75. Higher values are favorable. Residual intake and gain (RIG) RWG + (–1 × RFI) Simple index including RFI and RWG whose variance is adjusted at 1. Higher values are favorable. 1DMI
m
= DMI for maintenance; BWf = final BW; BWi = initial BW; BW0.75 = metabolic BW; DMIe = estimated DMI; ADGe = estimated ADG.
were excluded when there were no leftovers in the individual pens and when the GrowSafe system indicated irregularities in the collective pens, remaining a minimum of 49 feed intake records per animal. Dry matter intake was calculated based on the mean of all valid feed intake values obtained during the test period. Average daily gain of each animal was the coefficient of the linear regression of weights on days of test (DOT): yi = α + β × DOTi + εi, in which yi = weight of the animal in the ith observation, α = intercept of the regression equation corresponding to the initial weight, β = linear regression coefficient corresponding to ADG, DOTi = days of test for the ith observation, and εi = random error associated with each observation. Mean BW0.75 during the test was calculated as BW0.75 = [α + β × (duration of test)/2]0.75, with α and β assuming the values obtained by the equation described above. Twenty animals from the individual pens and 2 other animals from the collective pens of the 785 animals tested were excluded before having RFI, RWG, and RIG estimated by the regressions. The reasons for that were disease, a coefficient of determination less than 0.8 in the model used for the estimation of ADG, and ADG less than 330 g. In addition, animals with high food waste in individual pens and animals whose feed intake was below the reference of the group as indicated by the GrowSafe system were also excluded. Therefore, data from 411 males (56 C, 122 S, and 233 T) and 352 females (41 C, 86 S, and 225 T) were used for genetic evaluation, for a total of 763 animals. Another dataset was created that only contained the data
obtained in individual pens, that is, excluding the data of 85 males (10 C, 46 S, and 29 T) whose feed intake was measured in collective pens, corresponding to a total of 678 animals. Table 2 shows the definition and formula used for the calculation of feed efficiency measures. Dry matter intake for maintenance (DMIm) was estimated according to the NRC (1996) as the ratio between net energy for required maintenance of the animal and net energy required for maintenance of the diet. Considering a mean dietary content of 62.35% total digestible nutrients, corresponding to a net energy for maintenance of approximately 1,383 cal/kg DM, and that 69.3 cal/kg metabolic weight are necessary for maintenance of Zebu cattle, disregarding specific adjustments for sex, physiological condition, and compensatory gain, DMIm was estimated directly from BW0.75 as DMIm = 0.05011 × BW0.75. The models used for the estimation of RFI and RWG were derived from adjustments suggested by Koch et al. (1963) for DMI and ADG, respectively. Estimated DMI (DMIe) used for the calculation of RFI was obtained by the regression equation DMIe = β0 + βT × TG + βTG × TG × ADG + βTP × TG × BW0.75 + ε (i.e., RFI), in which β0 is the intercept, βT, βTG, and βTP are regression coefficients of the classificatory variable test group (TG) and of the interactions between TG and the covariates ADG and BW0.75, respectively, and ε is the residual of the equation (i.e., RFI). Estimated ADG (ADGe) used for the calculation of RWG was obtained by the regression equation
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
958
Grion et al.
ADGe = β0 + βT × TG + βTG × TG × DMI + βTP × TG × BW0.75 + ε (i.e., RWG), in which β0 is the intercept, βT, βTG, and βTP are regression coefficients of the classificatory variable TG and of the interactions between TG and the covariates DMI and BW0.75, respectively, and ε is the residual of the equation (i.e., RWG). As suggested by MacNeil et al. (2011), W210 was included in the analyses to improve the genetic parameter estimates due to the large number of animals in the herd with this type of information. Regarding to the genetic correlations between traits, the use of a trait with a large number of records adds information to traits with a small number of animals analyzed, as is the case of performance test measures, feed intake, and feed efficiency. Weaning weight adjusted to 210 d of age is calculated based on weight gain from birth to weaning using the following equation: W210 = [(WW – BW)/AW] × 210 + FBW, in which WW is the weaning weight, FBW is the first BW or the birth weight, and AW is the age of the animal at weaning. The genetic parameter estimates were obtained by the restricted maximum likelihood method using multitrait analysis under an animal model that included 4 (W210, DMI, ADG, and BW0.75) or 5 traits (the previous traits plus an indicator trait of feed efficiency: G:F, PEG, RGR, KR, RFI, RWG, or RIG). The WOMBAT software was used for the analyses (Meyer, 2007). The model can be represented in matrix form as y1 X 1b1 Z1a1 V1m1 W1c1 e1 y X b Z a 0 0 e , 2 2 2 2 2 2 y3 = X 3b3 + Z 3a 3 + 0 + 0 + e3 y4 X 4 b4 Z 4a 4 0 0 e4 y5 X 5b5 Z 5a 5 0 0 e5
in which yi is the vector of observations for index i ranging from 1 to 5, respectively, for W210, DMI, ADG, BW0.75, and an indicator trait of feed efficiency, βi is the vector of fixed effects associated with the incidence matrix Xi, ai is the vector of random direct additive genetic effects associated with the incidence matrix Zi, mi is the vector of maternal additive genetic effects associated with the incidence matrix Vi, ci is the vector of maternal permanent environmental effects associated with the incidence matrix Wi, and ei is the vector of residual effects. The usual assumptions are
y1 X 1b1 yi X i bi a1 0 E a1 = 0 ; m 0 1 c1 0 e 0 1 ei 0 As2a1 Asai a1 a1 0 Á s A s2a1 a1 ai a1 0 0 0 m VAR 1 = c1 0 0 0 e1 0 0 0 0 0 0 ei 0 0 0
0
0
0
0 0
0 0
0 0
As2m1
0
0
0
Is
0
0
Ise21
0 0
0 0
Isei e1
2 c1
0
0 0 0 0 0 0 0 0 0 Isei e1 Ise21 0
in which A is the relationship matrix and I is an identity matrix. For W210, the models included the fixed effects of contemporary group (year of birth, sex, and herd) and month of birth and included age of dam (linear and quadratic effects) and age of animal at weaning (linear effect) as covariates. For the other traits, the fixed effects of contemporary group (test group) and linear effects of the covariates age of dam and age of animal in the middle of the test were included in the models. All 763 animals tested for performance and efficiency were born to 67 sires (12 with their own performance records) and 458 dams (63 with their own performance records), which provided a relationship matrix with 1,946 animals. The 678 animals tested in individual pens were born to 58 sires (9 with their own performance records) and 408 dams (52 with their own performance records), which provided a relationship matrix with 1,802 animals. For both datasets, the inclusion of 8,004 animals with W210 records, born to 305 sires and 1,969 dams, increased the relationship matrix to 8,357 animals. The expected response to direct selection was estimated for the traits of interest (W210, DMI, ADG, and BW0.75) using the following equation (Falconer and Mackay, 1996): ΔGY = rtiY × iY × σaY, in which ΔGY is the genetic gain in trait Y (W210, DMI, ADG, and BW0.75) per generation, rtiY is the accuracy of
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
959
Selection for feed efficiency
genetic prediction of Y (square root of heritability in the case of selection based on own performance), iY is the intensity of selection for trait Y, and σaY is the genetic variation (SD of the additive genetic effect) in trait Y. The correlated responses in W210, DMI, ADG, and BW0.75, considering selection for feed efficiency traits, were estimated by the following equation (Falconer and Mackay, 1996): ΔGY|X = rgXY × rtiX × iX × σaY, in which ΔGY|X is the genetic gain (ΔG) in trait Y (W210, DMI, ADG, and BW0.75) per generation given selection for X (feed efficiency trait), rgXY is the genetic correlation between X and Y, rtiX is the accuracy of genetic prediction of X, iX is the intensity of selection for trait X, and σaY is the SD of the additive genetic effect on trait Y. Estimates of ΔG were calculated considering a selection intensity of 1.2, corresponding to selection of 10% of males and 60% of females in a single stage or as cumulative genetic gain in 2 stages. The first stage was simulated with selection of 50% of males and 80% of females (average selection intensity of 0.57) for W210 and the second stage with selection of 20% of males and 75% of females (average selection intensity of 0.91) for feed efficiency traits. In the case of 2-stage selection, all variances and covariances in the calculation of genetic gain based on selection for efficiency measures (second stage) were adjusted by previous selection for W210 using the following equation (Cochran, 1951; Cunningham, 1975): σxy* = σxy – [(σxz × σyz)/σzz] × iz × (iz – tz), in which σxy* is the covariance between traits X and Y adjusted by previous selection for Z, σxy is the covariance estimated before selection for Z, σxz is the covariance between X and Z, σyz is the covariance between Y and Z, σzz is the variance in Z, and iz and tz are the intensity of selection and truncation point of selection for Z, respectively. RESULTS AND DISCUSSION The phenotypic mean and SD of W210 was 185 ± 30.1 kg. The mean age of the animals at weaning and at the beginning of the efficiency test, after the adaptation period, was 199 ± 20 and 287 ± 41 d, respectively. The mean duration of the efficiency tests was 83 ± 16 d. Therefore, the mean age of the animals in the middle of the test was 329 ± 39 d. All animals were 7 to 15 mo old at the time of the test. This age range permits an adequate evaluation of the postweaning growth phase of Nellore beef cattle raised under tropical conditions.
Table 3. Phenotypic means and SD of growth, intake, and feed efficiency traits in Nellore cattle Trait1 DMI, kg DM ADG, kg BW0.75, kg G:F, kg gain/kg DM PEG, kg gain/kg DM RGR, kg BW/d KR, g gain/kg BW0.75 RFI, kg DM RWG, kg gain RIG, index
Individual pens (n = 678) Collective pens (n = 85) 6.63 ± 1.08 6.83 ± 1.21 0.930 ± 0.216 1.205 ± 0.201 65.5 ± 9.66 0.141 ± 0.0289 0.284 ± 0.0644 0.362 ± 0.0936 14.3 ± 3.30 0.00205 ± 0.354 0.000228 ± 0.0921 –0.00214 ± 1.76
70.8 ± 8.31 0.178 ± 0.0256 0.391 ± 0.106 0.417 ± 0.0586 17.0 ± 2.17 –4.47 × 10–11 ± 0.817 –9.41 × 10–12 ± 0.138 –1.65 × 10–11 ± 1.63
1BW0.75 = metabolic BW; PEG = partial efficiency of growth; RGR = relative growth rate; KR = Kleiber’s ratio; RFI = residual feed intake; RWG = residual weight gain; RIG = residual intake and gain.
The phenotypic means and SD of the feed efficiency traits (Table 3) agree with those reported for Nellore cattle raised in Brazil (Santana et al., 2012) but were generally lower than those described in studies on taurine breeds in which the diets are more energetic (Berry and Crowley, 2012). The means and SD obtained for individual pens (n = 678) differed from those obtained for collective pens (n = 85) since the former contain males and females of different ages evaluated in different years whereas the latter is a dataset of only 1 performance test with males. Heritabilities (Table 4) differed when estimated with or without information from collective pens. The greatest differences were observed for DMI and the feed efficiency traits that are phenotypically more correlated with DMI, such as PEG, RFI, and RIG. These differences suggest that feed intake in an individual pen is a distinct trait when compared to voluntary intake in the group. The evaluation of these 2 traits as a single trait may explain the residual variance inflation observed in DMI, PEG, and mainly RFI. This change in RFI may have also altered the variance component estimates for RIG. The evaluation of animals housed in individual pens is useful since it permits the measurement of voluntary intake in the absence of external interferences; however, it does not represent the natural pattern of feed intake of pasture-fed cattle. According to Crews et al. (2010), feed intake data obtained in individual pens are inadequate for genetic evaluation. In general, the heritability estimates for DMI, ADG, and BW0.75 were of high magnitude (>0.42) whereas the heritabilities for feed efficiency traits were moderate to low (0.42). ADG was poorly correlated with animal weight (0.8% in the growth traits of Nellore cattle during the postweaning period. A trait comprising RFI and RWG, such as RIG, may yield results similar to those reported by Rolfe et al. (2011) who used economic and empirical indices and observed
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency
963
Figure 1. Genetic change in weaning weight adjusted to 210 d of age (W210), DMI, ADG, and metabolic BW (BW0.75) when selecting for W210 followed by selection for W210, DMI, ADG, BW0.75, G:F, partial efficiency of growth (PEG), relative growth rate (RGR), Kleiber’s ratio (KR), residual feed intake (RFI), residual weight gain (RWG), or residual intake and gain (RIG).
that an increase in gain is even more important than a reduction in intake. The authors recommended an index consisting of RFI and ADG since it provides the best economic results. Archer et al. (2004) evaluated different cost scenarios of feed efficiency testing. Despite the relatively high cost of measuring feed intake, the authors concluded that the inclusion of indicator traits of feed efficiency is profitable, especially if preselection is performed to reduce costs by decreasing the number of animals evaluated in postweaning feed efficiency tests. Furthermore, in view of the need to select 10% of males as replacement sires and 60% of females as replacement cows, selection can be performed in 2 stages: in the first stage, 50% of males and 80% of females are selected at weaning for W210; in the second stage, 20% of males and 75% of females are selected after evaluation of feed efficiency. Figure 1 shows the cumulative genetic gains as a result of selection for weight at weaning in the first stage and for the other traits in the second stage. Positive genetic gains in W210 are expected in all cases, even if selection to reduce DMI is performed in the second stage. However, only selection for PEG and RFI provides favorable genetic gains in both indicator growth traits and DMI. Selection for PEG in the second stage would reduce DMI
by 0.77% of the mean in the next generation and would provide gains of 1.32% in W210 and of 0.26% in ADG. In contrast, selection for RFI in the second stage would provide gains of 2.11% in W210 and of 0.46% in ADG, in addition to reducing DMI by 0.49%. Taken together, the results highlight the importance of PEG and RFI because they permit the best responses to selection for reduced DMI proportionally to high genetic gains in growth traits in 2-stage selection schemes, including preselection during the postweaning period. Conclusions The additive genetic variability observed in feed efficiency traits of Nellore cattle can provide moderate genetic gains through selection. Among the indicator traits of feed efficiency, the combination of residuals of the prediction equations of intake and gain, represented by RIG, should result in the best responses to selection designed to reduce the use of feed and to increase growth traits. This finding suggests that the use of selection indices combining more than 1 efficiency trait or even measures of direct interest (DMI, ADG, and weights) is the most effective solution for the improvement of Nellore cattle when the
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
964
Grion et al.
objective is postweaning feed efficiency. Genetic gains in feed efficiency are expected even after preselection of animals at weaning and subsequent feed efficiency testing of the preselected portion. LITERATURE CITED Archer, J. A., P. F. Arthur, R. M. Herd, P. F. Parnell, and W. S. Pitchford. 1997. Optimum postweaning test for measurement of growth rate, feed intake, and feed efficiency in British breed cattle. J. Anim. Sci. 75:2024–2032. Archer, J. A., S. A. Barwick, and H. U. Graser. 2004. Economic evaluation of beef cattle breeding schemes incorporating performance testing of young bulls for feed intake. Aust. J. Exp. Agric. 44:393–404. Arthur, P. F., J. A. Archer, D. J. Johnston, R. M. Herd, E. C. Richardson, and P. F. Parnell. 2001a. Genetic and phenotypic variance and covariance components for feed intake, feed efficiency, and other postweaning traits in Angus cattle. J. Anim. Sci. 79:2805–2811. Arthur, P. F., G. Renand, and D. Krauss. 2001b. Genetic and phenotypic relationships among measures of growth and feed efficiency in young Charolais bulls. Livest. Prod. Sci. 68:131–139. Barwick, S. A., M. L. Wolcott, D. J. Johnston, H. M. Burrow, and M. T. Sullivan. 2009. Genetics of steer daily and residual feed intake in two tropical beef genotypes, and relationships among intake, body composition, growth and other post-weaning measures. Anim. Prod. Sci. 49:351–366. Bergh, L., M. M. Scholtz, and G. J. Erasmus. 1992. Identification and assessment of the best animal: The Kleiber ratio (growth/ metabolic mass) as a selection criterion for beef cattle. In: Proc. Aust. Assoc. Anim. Breed. Genet., vol. 10, Rockhampton, Australia. p. 338–340. Berry, D. P., and J. J. Crowley. 2012. Residual intake and body weight gain: A new measure of efficiency in growing cattle. J. Anim. Sci. 90:109–115. Bouquet, A., M. N. Fouilloux, G. Renand, and F. Phocas. 2010. Genetic parameters for growth, muscularity, feed efficiency and carcass traits of young bulls. Livest. Sci. 129:38–48. Bourdon, R. M. 2000. Understanding animal breeding. 2nd ed. Prentice Hall, Upper Saddle River, NJ. Cochran, W. G. 1951. Improvement by means of selection. In: Proc. 2nd Berkeley Symp. Math. Stat. and Prob., Berkeley, CA. p. 449–470. Crews, D. H., Jr., M. E. Dikeman, S. L. Northcutt, D. J. Garrick, T. T. Marston, M. D. MacNeil, L. W. Olson, J. C. Paschal, G. H. Rouse, R. Weaber, T. L. Wheeler, S. D. Shackelford, R. E. Williams, and D. E. Wilson. 2010. Animal evaluation. In: L. V. Cundiff, L. D. Van Vleck, and W. D. Hohenboken, editors, Guidelines for uniform beef improvement programs. Beef Improvement Federation, Raleigh, NC. p. 16–55. Crowley, J. J., R. D. Evans, N. Mc Hugh, D. A. Kenny, M. McGee, D. H. Crews Jr., and D. Berry. 2011. Genetic relationship between feed efficiency in growing males and beef cow performance. J. Anim. Sci. 89:3372–3381. Crowley, J. J., M. McGee, D. A. Kenny, D. H. Crews Jr., R. D. Evans, and D. P. Berry. 2010. Phenotypic and genetic parameters for different measures of feed efficiency in different breeds of Irish performance-tested beef bulls. J. Anim. Sci. 88:885–894. Cunningham, E. P. 1975. Multi-stage index selection. Theor. Appl. Genet. 46:55–61.
Del Claro, A. C., M. E. Z. Mercadante, and J. A. I. I. V. Silva. 2012. Meta-análise de parâmetros genéticos relacionados ao consumo alimentar residual e a suas características componentes em bovinos. (In Portuguese.) Pesqui. Agropecu. Bras. 47:302–310. Falconer, D. S., and T. F. C. Mackay. 1996. Introduction to quantitative genetics. 4th ed. Addison Wesley Longman, Harlow, UK. Fan, L. Q., D. R. C. Bailey, and N. H. Shannon. 1995. Genetic parameter estimation of postweaning gain, feed intake, and feed efficiency for Hereford and Angus bulls fed two different diets. J. Anim. Sci. 73:365–372. Fitzhugh, H. A., Jr., and C. S. Taylor. 1971. Genetic analysis of degree of maturity. J. Anim. Sci. 33:717–725. Giannotti, J. D. G., I. U. Packer, and M. E. Z. Mercadante. 2005. Meta-análise para estimativas de herdabilidade para características de crescimento em bovinos de corte. (In Portuguese.) R. Bras. Zootec. 34:1173–1180. Gunsett, F. C. 1984. Linear index selection to improve traits defined as ratio. J. Anim. Sci. 59:1185–1193. Hegarty, R. S., J. P. Goopy, R. M. Herd, and B. McCorkell. 2007. Cattle selected for lower residual feed intake have reduced daily methane production. J. Anim. Sci. 85:1479–1486. Herd, R. M., and S. C. Bishop. 2000. Genetic variation in residual feed intake and its association with other production traits in British Hereford cattle. Livest. Prod. Sci. 63:111–119. Hoque, M. A., P. F. Arthur, K. Hiramoto, and T. Oikawa. 2006. Genetic relationship between different measures of feed efficiency and its component traits in Japanese Black (Wagyu) bulls. Livest. Sci. 99:111–118. Kleiber, M. 1936. Problems involved in breeding for efficiency of food utilization. J. Anim. Sci. 1936b:247-258. Koch, R. M., L. A. Swiger, D. Chambers, and K. E. Gregory. 1963. Efficiency of feed use in beef cattle. J. Anim. Sci. 22:486–494. Lancaster, P. A., G. E. Carstens, D. H. Crews Jr., T. H. Welsh Jr., T. D. A. Forbes, D. W. Forrest, L. O. Tedeschi, R. D. Randel, and F. M. Rouquette. 2009. Phenotypic and genetic relationships of residual feed intake with performance and ultrasound carcass traits in Brangus heifers. J. Anim. Sci. 87:3887–3896. MacNeil, M. D., N. Lopez-Villalobos, and S. L. Northcutt. 2011. A prototype national cattle evaluation for feed intake and efficiency of Angus cattle. J. Anim. Sci. 89:3917–3923. Mercadante, M. E. Z., and A. G. Razook. 2010. Programa de melhoramento genético do Instituto de Zootecnia. In: A. V. Pires, editor, Bovinocultura de Corte. Vol. 2, 1st ed. (In Portuguese.) Fundação de Estudos Agrários Luiz de Queiroz, Piracicaba, Brazil. p. 847–860. Meyer, K. 2007. WOMBAT – A tool for mixed model analyses in quantitative genetics by restricted maximum likelihood. J. Zhejiang Univ. Sci. B 8:815–821. Nkrumah, J. D., J. A. Basarab, Z. Wang, C. Li, M. A. Price, E. K. Okine, D. H. Crews Jr., and S. S. Moore. 2007. Genetic and phenotypic relationships of feed intake and measures of efficiency with growth and carcass merit of beef cattle. J. Anim. Sci. 85:2711–2720. NRC. 1996. Nutrient requirements of beef cattle. 7th ed. National Academy Press, Washington, DC. Razook, A. G., L. A. Figueiredo, J. N. S. G. Cyrillo, L. J. Pacola, L. M. Bonilha Neto, J. B. F. Trovo, A. C. Ruggieri, and M. E. Z. Mercadante. 1997. Prova de Ganho de Peso: Normas adotadas pela estação experimental de Zootecnia de Sertãozinho. Boletim Técnico, n. 40. (In Portuguese.) Instituto de Zootecnia, Sertãozinho, Brazil. Robinson, D. L., and V. H. Oddy. 2004. Genetic parameters for feed efficiency, fatness, muscle area and feeding behavior of feedlot finished beef cattle. Livest. Prod. Sci. 90:255–270.
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency Rolfe, K. M., W. M. Snelling, M. K. Nielsen, H. C. Freetly, C. L. Ferrel, and T. G. Jenkins. 2011. Genetic and phenotypic parameter estimates for feed intake and others traits in growing beef cattle, and opportunities for selection. J. Anim. Sci. 89:3452–3459.
965
Santana, M. H. A., P. Rossi Jr., R. Almeida, and D. C. Cucco. 2012. Feed efficiency and its correlations with carcass traits measured by ultrasound in Nellore bulls. Livest. Sci. 145:252–257. Schenkel, F. S., S. P. Miller, and J. W. Wilton. 2004. Genetic parameters and breed differences for feed efficiency, growth, and body composition traits of young beef bulls. Can. J. Anim. Sci. 84:177–185.
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
References
This article cites 27 articles, 14 of which you can access for free at: http://www.journalofanimalscience.org/content/92/3/955#BIBL
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.journalofanimalscience.org/content/92/3/955
www.asas.org
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency traits and correlated genetic responses in feed intake and weight gain of Nellore cattle1 A. L. Grion, M. E. Z. Mercadante,2 J. N. S. G. Cyrillo, S. F. M. Bonilha, E. Magnani, and R. H. Branco Instituto de Zootecnia, Centro APTA Bovinos de Corte. Sertãozinho, SP, Brazil, 14160-970
ABSTRACT: The objectives of this study were to estimate genetic parameters for indicator traits of feed efficiency and to recommend traits that would result in better responses to selection for increased weaning weight (weaning weight adjusted to 210 d of age [W210]), ADG, and metabolic BW (BW0.75) and lower DMI. Records of W210 from 8,004 Nellore animals born between 1978 and 2011 and postweaning performance test records from 678 males and females born between 2004 and 2011 were used. The following feed efficiency traits were evaluated: G:F, partial efficiency of growth (PEG), relative growth rate (RGR), Kleiber’s ratio (KR), residual feed intake (RFI), residual weight gain (RWG), and residual intake and gain (RIG). Covariance and variance components were estimated by the restricted maximum likelihood method using multitrait analysis under an animal model. Estimates of genetic gain and correlated responses were obtained considering single-stage
and 2-stage selection. Heritability estimates were 0.22 ± 0.03 (W210), 0.60 ± 0.08 (DMI), 0.42 ± 0.08 (ADG), 0.56 ± 0.06 (BW0.75), 0.19 ± 0.07 (G:F), 0.25 ± 0.09 (PEG), 0.19 ± 0.07 (RGR), 0.22 ± 0.07 (KR), 0.33 ± 0.10 (RFI), 0.13 ± 0.07 (RWG), and 0.19 ± 0.08 (RIG). The genetic correlations of DMI with W210 (0.64 ± 0.10), ADG (0.87 ± 0.06), and BW0.75 (0.84 ± 0.05) were high. The only efficiency traits showing favorable responses to selection for lower DMI were G:F, PEG, RFI, and RIG. However, the use of G:F, PEG, or RFI as a selection criterion results in unfavorable correlated responses in some growth traits. The linear combination of RFI and RWG through RIG is the best selection criterion to obtain favorable responses in postweaning growth and feed intake of Nellore cattle in single-stage selection. Genetic gains in feed efficiency are expected even after preselection for W210 and subsequent feed efficiency testing of the preselected animals.
Key words: Bos indicus, feed intake, genetic correlation, genetic gain, selection objective © 2014 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2014.92:955–965 doi:10.2527/jas2013-6682 INTRODUCTION The efficiency of animal production systems is very important not only from an economic point of view but also from a socio-environmental viewpoint (Hegarty et al., 2007). Several measures that associate inputs (raw material), outputs (products), and time have been proposed over the years, especially the selection of animals that combine higher production with less expenditure and less time (Kleiber, 1936; Koch et al., 1963; Fitzhugh and Taylor, 1971). 1This
work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo, Conselho Nacional de Desenvolvimento Científico e Tecnológico and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior. 2Corresponding author: [email protected] Received May 8, 2013. Accepted January 3, 2014.
The correct use of feed efficiency traits in genetic evaluation programs requires analysis of the additive genetic variability of each trait and of genetic correlations between traits related to inputs (generally corresponding to DMI) and products (generally corresponding to weight gain rate or BW or both). In addition, estimates of genetic gain will permit determination of the most appropriate selection criteria and methods for a given selection objective (Bourdon, 2000). In this respect, an increase in body growth and a decrease in feed intake are the main outcome measures of a breeding program whose objective is to improve feed efficiency using the traits described for this purpose as selection criteria. Few studies have investigated genetic parameters and the effects of selection for efficiency traits in Zebu cattle. Therefore, the objectives of the present study were to estimate genetic parameters for indicator traits of feed efficiency, including G:F, partial efficiency of
955 Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
956
Grion et al.
growth (PEG), relative growth rate (RGR), Kleiber’s ratio (KR), residual feed intake (RFI), residual weight gain (RWG), and residual intake and gain (RIG) and to recommend traits that would result in the best responses to selection based on the estimated correlated responses of these traits with weaning weight adjusted to 210 d of age (W210), DMI, ADG, and metabolic BW (BW0.75). MATERIAL AND METHODS Records of feed efficiency traits from 785 Nellore animals born between 2004 and 2011 and weaning weight records from 8,004 animals born between 1978 and 2011 were used. The animals belonged to 3 lines of selection started in 1978 at Centro Avançado de Pesquisa Tecnológica do Agronegócio, Bovinos de Corte, Instituto de Zootecnia, Sertãozinho, São Paulo, Brazil. Male animals were selected for weight adjusted to 378 d of age obtained in weight gain tests (Razook et al., 1997) and female animals were selected for weight adjusted to 550 d obtained on pasture. In the control (NeC) Nellore line, animals were selected for the mean of the contemporary group, with a selection differential of about zero. In the selection (NeS) and traditional (NeT) lines, the animals were selected for higher weights, that is, the maximum selection differential for the traits described above (Mercadante and Razook, 2010). The difference between the NeS and NeT lines is that the NeS line, like the NeC line, is a closed experimental line whereas sires from other herds were used in the NeT line, mimicking a conventional commercial herd. The animals were submitted to the same management until they entered the performance test. Performance testing was used to evaluate individual feed intake after weaning. Before the tests, the animals were submitted to a period of adaptation to the diet and facilities for at least 21 d. The parameters of the performance tests are shown in Table 1. In 2005, 2006, and 2009, females were subdivided into 2 test groups performed in sequence due to the limited availability of individual pens. After the adaptation period, 32 randomly chosen females were allocated to individual pens for the measurement of feed intake whereas the other 32 females remained in the collective pens until the end of the test of the first group. Next, the second group was tested. This approach explains the average difference of 3 mo of age and almost 82 kg between the 2 groups at the beginning of the performance tests. After 2010, the 2 groups of females were evaluated simultaneously in different facilities. For males, only 1 group of animals was evaluated in individual pens per year between 2007 and 2011. In 2012, 31 animals were evaluated as done in the previous years and 87 animals were analyzed in collective pens using the GrowSafe system (GrowSafe Systems Ltd., Airdrie, Alberta, Canada).
Table 1. Groups of animals submitted to performance testing Year 2005 2005 2006 2006 2007 2008 2009 2009 2009 2010 2010 2010 2011 2011 2011 2012 2012 2012 2012 Overall
Sex1 F F F F M M M F F M F F M F F M M F F –
Duration, No. of Herd2 d animals NeC and NeS 56 32 NeC and NeS 56 32 NeC and NeS 70 32 NeC and NeS 70 32 NeC and NeS 112 61 NeC and NeS 112 62 NeT 70 60 NeT 85 32 NeT 84 32 NeT 73 60 NeT 84 32 NeT 84 24 NeT 72 62 NeT 84 31 NeT 84 25 NeT 91 31 NeC, NeS 91 87 and NeT NeT 86 26 NeT 86 32 – 83 ± 163 785
Initial Initial weight,3 age,3 d kg 263 ± 18 164 ± 28 341 ± 18 233 ± 34 267 ± 24 155 ± 30 350 ± 21 242 ± 48 268 ± 14 205 ± 31 249 ± 25 194 ± 36 300 ± 21 295 ± 32 272 ± 25 203 ± 28 391 ± 22 292 ± 26 262 ± 22 248 ± 40 294 ± 19 212 ± 28 281 ± 22 208 ± 25 261 ± 23 243 ± 40 294 ± 23 220 ± 33 286 ± 30 212 ± 28 277 ± 19 250 ± 31 270 ± 23 239 ± 41 225 ± 24 223 ± 25 287 ± 40
257 ± 29 256 ± 28 230 ± 49
1F
= female; M = male. = control herd; NeS = selection herd; NeT = traditional herd. 3Mean ± SD. 2NeC
Body weight was recorded at intervals of 14 d after a 12-h fast in 2005 and 2006 and at intervals of 28 d after fasting in 2007 and 2008 for males and from 2009 to 2011 for females. After 2009, males were weighed weekly without fasting, including 3 weight recordings per week on consecutive days in 2009 and 2010, 2 recordings per week on consecutive days in 2011, and 1 recording per week in 2012. The females tested in 2012 were weighed on 2 consecutive days at intervals of 14 d. Therefore, each animal was weighed at least 4 times with prior fasting and at least 11 times without prior fasting. The diets offered over the years differed in terms of the composition and source of feeds but were formulated to have isoenergetic and isoprotein levels (62% total digestible nutrients and 13% CP). In the individual pens, the diet was offered 2 times per day and adjusted daily and individually to permit an amount of leftovers of 5 to 10%, thus guaranteeing ad libitum intake. Weekly samplings of the diet offered and of leftover content were analyzed in composite samples at intervals of 28 d for the determination of DM. In individual pens, daily feed intake was calculated as the difference between the amount of food offered and leftovers. In collective pens, the troughs always contained food and were cleaned weekly. Individual daily intake (as-fed basis) recorded automatically by the GrowSafe system was multiplied by DM of the food offered. Daily feed intake records
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency
957
Table 2. Definition of the indicator traits of feed efficiency Formula1 ADG/DMI
Trait G:F, kg gain/kg DM
Definition Weight gain divided by the amount of feed consumed. Higher values are favorable. Partial efficiency of growth (PEG), kg ADG/(DMI – DMIm) Quantity of weight gain per kilogram feed available for this purpose, that is, subtracting total true gain/kg DM DMI from estimated DMI for maintenance. Higher values are favorable. Relative growth rate (RGR), kg BW/d 100 × [(logBWf – logBWi)/ Growth potential in relation to degree of maturity. days in test] Higher values are favorable. Kleiber’s ratio (KR), g gain/kg BW0.75 100 × (ADG/BW0.75) ADG, in grams, proportional to each kilogram of metabolic weight. Higher values are favorable. Residual feed intake (RFI), kg DM/d DMI – DMIe Difference between observed and estimated DMI based on ADG and BW0.75. Lower values are favorable. Residual weight gain (RWG), kg gain/d ADG – ADGe Difference between observed and estimated ADG based on DMI and BW0.75. Higher values are favorable. Residual intake and gain (RIG) RWG + (–1 × RFI) Simple index including RFI and RWG whose variance is adjusted at 1. Higher values are favorable. 1DMI
m
= DMI for maintenance; BWf = final BW; BWi = initial BW; BW0.75 = metabolic BW; DMIe = estimated DMI; ADGe = estimated ADG.
were excluded when there were no leftovers in the individual pens and when the GrowSafe system indicated irregularities in the collective pens, remaining a minimum of 49 feed intake records per animal. Dry matter intake was calculated based on the mean of all valid feed intake values obtained during the test period. Average daily gain of each animal was the coefficient of the linear regression of weights on days of test (DOT): yi = α + β × DOTi + εi, in which yi = weight of the animal in the ith observation, α = intercept of the regression equation corresponding to the initial weight, β = linear regression coefficient corresponding to ADG, DOTi = days of test for the ith observation, and εi = random error associated with each observation. Mean BW0.75 during the test was calculated as BW0.75 = [α + β × (duration of test)/2]0.75, with α and β assuming the values obtained by the equation described above. Twenty animals from the individual pens and 2 other animals from the collective pens of the 785 animals tested were excluded before having RFI, RWG, and RIG estimated by the regressions. The reasons for that were disease, a coefficient of determination less than 0.8 in the model used for the estimation of ADG, and ADG less than 330 g. In addition, animals with high food waste in individual pens and animals whose feed intake was below the reference of the group as indicated by the GrowSafe system were also excluded. Therefore, data from 411 males (56 C, 122 S, and 233 T) and 352 females (41 C, 86 S, and 225 T) were used for genetic evaluation, for a total of 763 animals. Another dataset was created that only contained the data
obtained in individual pens, that is, excluding the data of 85 males (10 C, 46 S, and 29 T) whose feed intake was measured in collective pens, corresponding to a total of 678 animals. Table 2 shows the definition and formula used for the calculation of feed efficiency measures. Dry matter intake for maintenance (DMIm) was estimated according to the NRC (1996) as the ratio between net energy for required maintenance of the animal and net energy required for maintenance of the diet. Considering a mean dietary content of 62.35% total digestible nutrients, corresponding to a net energy for maintenance of approximately 1,383 cal/kg DM, and that 69.3 cal/kg metabolic weight are necessary for maintenance of Zebu cattle, disregarding specific adjustments for sex, physiological condition, and compensatory gain, DMIm was estimated directly from BW0.75 as DMIm = 0.05011 × BW0.75. The models used for the estimation of RFI and RWG were derived from adjustments suggested by Koch et al. (1963) for DMI and ADG, respectively. Estimated DMI (DMIe) used for the calculation of RFI was obtained by the regression equation DMIe = β0 + βT × TG + βTG × TG × ADG + βTP × TG × BW0.75 + ε (i.e., RFI), in which β0 is the intercept, βT, βTG, and βTP are regression coefficients of the classificatory variable test group (TG) and of the interactions between TG and the covariates ADG and BW0.75, respectively, and ε is the residual of the equation (i.e., RFI). Estimated ADG (ADGe) used for the calculation of RWG was obtained by the regression equation
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
958
Grion et al.
ADGe = β0 + βT × TG + βTG × TG × DMI + βTP × TG × BW0.75 + ε (i.e., RWG), in which β0 is the intercept, βT, βTG, and βTP are regression coefficients of the classificatory variable TG and of the interactions between TG and the covariates DMI and BW0.75, respectively, and ε is the residual of the equation (i.e., RWG). As suggested by MacNeil et al. (2011), W210 was included in the analyses to improve the genetic parameter estimates due to the large number of animals in the herd with this type of information. Regarding to the genetic correlations between traits, the use of a trait with a large number of records adds information to traits with a small number of animals analyzed, as is the case of performance test measures, feed intake, and feed efficiency. Weaning weight adjusted to 210 d of age is calculated based on weight gain from birth to weaning using the following equation: W210 = [(WW – BW)/AW] × 210 + FBW, in which WW is the weaning weight, FBW is the first BW or the birth weight, and AW is the age of the animal at weaning. The genetic parameter estimates were obtained by the restricted maximum likelihood method using multitrait analysis under an animal model that included 4 (W210, DMI, ADG, and BW0.75) or 5 traits (the previous traits plus an indicator trait of feed efficiency: G:F, PEG, RGR, KR, RFI, RWG, or RIG). The WOMBAT software was used for the analyses (Meyer, 2007). The model can be represented in matrix form as y1 X 1b1 Z1a1 V1m1 W1c1 e1 y X b Z a 0 0 e , 2 2 2 2 2 2 y3 = X 3b3 + Z 3a 3 + 0 + 0 + e3 y4 X 4 b4 Z 4a 4 0 0 e4 y5 X 5b5 Z 5a 5 0 0 e5
in which yi is the vector of observations for index i ranging from 1 to 5, respectively, for W210, DMI, ADG, BW0.75, and an indicator trait of feed efficiency, βi is the vector of fixed effects associated with the incidence matrix Xi, ai is the vector of random direct additive genetic effects associated with the incidence matrix Zi, mi is the vector of maternal additive genetic effects associated with the incidence matrix Vi, ci is the vector of maternal permanent environmental effects associated with the incidence matrix Wi, and ei is the vector of residual effects. The usual assumptions are
y1 X 1b1 yi X i bi a1 0 E a1 = 0 ; m 0 1 c1 0 e 0 1 ei 0 As2a1 Asai a1 a1 0 Á s A s2a1 a1 ai a1 0 0 0 m VAR 1 = c1 0 0 0 e1 0 0 0 0 0 0 ei 0 0 0
0
0
0
0 0
0 0
0 0
As2m1
0
0
0
Is
0
0
Ise21
0 0
0 0
Isei e1
2 c1
0
0 0 0 0 0 0 0 0 0 Isei e1 Ise21 0
in which A is the relationship matrix and I is an identity matrix. For W210, the models included the fixed effects of contemporary group (year of birth, sex, and herd) and month of birth and included age of dam (linear and quadratic effects) and age of animal at weaning (linear effect) as covariates. For the other traits, the fixed effects of contemporary group (test group) and linear effects of the covariates age of dam and age of animal in the middle of the test were included in the models. All 763 animals tested for performance and efficiency were born to 67 sires (12 with their own performance records) and 458 dams (63 with their own performance records), which provided a relationship matrix with 1,946 animals. The 678 animals tested in individual pens were born to 58 sires (9 with their own performance records) and 408 dams (52 with their own performance records), which provided a relationship matrix with 1,802 animals. For both datasets, the inclusion of 8,004 animals with W210 records, born to 305 sires and 1,969 dams, increased the relationship matrix to 8,357 animals. The expected response to direct selection was estimated for the traits of interest (W210, DMI, ADG, and BW0.75) using the following equation (Falconer and Mackay, 1996): ΔGY = rtiY × iY × σaY, in which ΔGY is the genetic gain in trait Y (W210, DMI, ADG, and BW0.75) per generation, rtiY is the accuracy of
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
959
Selection for feed efficiency
genetic prediction of Y (square root of heritability in the case of selection based on own performance), iY is the intensity of selection for trait Y, and σaY is the genetic variation (SD of the additive genetic effect) in trait Y. The correlated responses in W210, DMI, ADG, and BW0.75, considering selection for feed efficiency traits, were estimated by the following equation (Falconer and Mackay, 1996): ΔGY|X = rgXY × rtiX × iX × σaY, in which ΔGY|X is the genetic gain (ΔG) in trait Y (W210, DMI, ADG, and BW0.75) per generation given selection for X (feed efficiency trait), rgXY is the genetic correlation between X and Y, rtiX is the accuracy of genetic prediction of X, iX is the intensity of selection for trait X, and σaY is the SD of the additive genetic effect on trait Y. Estimates of ΔG were calculated considering a selection intensity of 1.2, corresponding to selection of 10% of males and 60% of females in a single stage or as cumulative genetic gain in 2 stages. The first stage was simulated with selection of 50% of males and 80% of females (average selection intensity of 0.57) for W210 and the second stage with selection of 20% of males and 75% of females (average selection intensity of 0.91) for feed efficiency traits. In the case of 2-stage selection, all variances and covariances in the calculation of genetic gain based on selection for efficiency measures (second stage) were adjusted by previous selection for W210 using the following equation (Cochran, 1951; Cunningham, 1975): σxy* = σxy – [(σxz × σyz)/σzz] × iz × (iz – tz), in which σxy* is the covariance between traits X and Y adjusted by previous selection for Z, σxy is the covariance estimated before selection for Z, σxz is the covariance between X and Z, σyz is the covariance between Y and Z, σzz is the variance in Z, and iz and tz are the intensity of selection and truncation point of selection for Z, respectively. RESULTS AND DISCUSSION The phenotypic mean and SD of W210 was 185 ± 30.1 kg. The mean age of the animals at weaning and at the beginning of the efficiency test, after the adaptation period, was 199 ± 20 and 287 ± 41 d, respectively. The mean duration of the efficiency tests was 83 ± 16 d. Therefore, the mean age of the animals in the middle of the test was 329 ± 39 d. All animals were 7 to 15 mo old at the time of the test. This age range permits an adequate evaluation of the postweaning growth phase of Nellore beef cattle raised under tropical conditions.
Table 3. Phenotypic means and SD of growth, intake, and feed efficiency traits in Nellore cattle Trait1 DMI, kg DM ADG, kg BW0.75, kg G:F, kg gain/kg DM PEG, kg gain/kg DM RGR, kg BW/d KR, g gain/kg BW0.75 RFI, kg DM RWG, kg gain RIG, index
Individual pens (n = 678) Collective pens (n = 85) 6.63 ± 1.08 6.83 ± 1.21 0.930 ± 0.216 1.205 ± 0.201 65.5 ± 9.66 0.141 ± 0.0289 0.284 ± 0.0644 0.362 ± 0.0936 14.3 ± 3.30 0.00205 ± 0.354 0.000228 ± 0.0921 –0.00214 ± 1.76
70.8 ± 8.31 0.178 ± 0.0256 0.391 ± 0.106 0.417 ± 0.0586 17.0 ± 2.17 –4.47 × 10–11 ± 0.817 –9.41 × 10–12 ± 0.138 –1.65 × 10–11 ± 1.63
1BW0.75 = metabolic BW; PEG = partial efficiency of growth; RGR = relative growth rate; KR = Kleiber’s ratio; RFI = residual feed intake; RWG = residual weight gain; RIG = residual intake and gain.
The phenotypic means and SD of the feed efficiency traits (Table 3) agree with those reported for Nellore cattle raised in Brazil (Santana et al., 2012) but were generally lower than those described in studies on taurine breeds in which the diets are more energetic (Berry and Crowley, 2012). The means and SD obtained for individual pens (n = 678) differed from those obtained for collective pens (n = 85) since the former contain males and females of different ages evaluated in different years whereas the latter is a dataset of only 1 performance test with males. Heritabilities (Table 4) differed when estimated with or without information from collective pens. The greatest differences were observed for DMI and the feed efficiency traits that are phenotypically more correlated with DMI, such as PEG, RFI, and RIG. These differences suggest that feed intake in an individual pen is a distinct trait when compared to voluntary intake in the group. The evaluation of these 2 traits as a single trait may explain the residual variance inflation observed in DMI, PEG, and mainly RFI. This change in RFI may have also altered the variance component estimates for RIG. The evaluation of animals housed in individual pens is useful since it permits the measurement of voluntary intake in the absence of external interferences; however, it does not represent the natural pattern of feed intake of pasture-fed cattle. According to Crews et al. (2010), feed intake data obtained in individual pens are inadequate for genetic evaluation. In general, the heritability estimates for DMI, ADG, and BW0.75 were of high magnitude (>0.42) whereas the heritabilities for feed efficiency traits were moderate to low (0.42). ADG was poorly correlated with animal weight (0.8% in the growth traits of Nellore cattle during the postweaning period. A trait comprising RFI and RWG, such as RIG, may yield results similar to those reported by Rolfe et al. (2011) who used economic and empirical indices and observed
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency
963
Figure 1. Genetic change in weaning weight adjusted to 210 d of age (W210), DMI, ADG, and metabolic BW (BW0.75) when selecting for W210 followed by selection for W210, DMI, ADG, BW0.75, G:F, partial efficiency of growth (PEG), relative growth rate (RGR), Kleiber’s ratio (KR), residual feed intake (RFI), residual weight gain (RWG), or residual intake and gain (RIG).
that an increase in gain is even more important than a reduction in intake. The authors recommended an index consisting of RFI and ADG since it provides the best economic results. Archer et al. (2004) evaluated different cost scenarios of feed efficiency testing. Despite the relatively high cost of measuring feed intake, the authors concluded that the inclusion of indicator traits of feed efficiency is profitable, especially if preselection is performed to reduce costs by decreasing the number of animals evaluated in postweaning feed efficiency tests. Furthermore, in view of the need to select 10% of males as replacement sires and 60% of females as replacement cows, selection can be performed in 2 stages: in the first stage, 50% of males and 80% of females are selected at weaning for W210; in the second stage, 20% of males and 75% of females are selected after evaluation of feed efficiency. Figure 1 shows the cumulative genetic gains as a result of selection for weight at weaning in the first stage and for the other traits in the second stage. Positive genetic gains in W210 are expected in all cases, even if selection to reduce DMI is performed in the second stage. However, only selection for PEG and RFI provides favorable genetic gains in both indicator growth traits and DMI. Selection for PEG in the second stage would reduce DMI
by 0.77% of the mean in the next generation and would provide gains of 1.32% in W210 and of 0.26% in ADG. In contrast, selection for RFI in the second stage would provide gains of 2.11% in W210 and of 0.46% in ADG, in addition to reducing DMI by 0.49%. Taken together, the results highlight the importance of PEG and RFI because they permit the best responses to selection for reduced DMI proportionally to high genetic gains in growth traits in 2-stage selection schemes, including preselection during the postweaning period. Conclusions The additive genetic variability observed in feed efficiency traits of Nellore cattle can provide moderate genetic gains through selection. Among the indicator traits of feed efficiency, the combination of residuals of the prediction equations of intake and gain, represented by RIG, should result in the best responses to selection designed to reduce the use of feed and to increase growth traits. This finding suggests that the use of selection indices combining more than 1 efficiency trait or even measures of direct interest (DMI, ADG, and weights) is the most effective solution for the improvement of Nellore cattle when the
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
964
Grion et al.
objective is postweaning feed efficiency. Genetic gains in feed efficiency are expected even after preselection of animals at weaning and subsequent feed efficiency testing of the preselected portion. LITERATURE CITED Archer, J. A., P. F. Arthur, R. M. Herd, P. F. Parnell, and W. S. Pitchford. 1997. Optimum postweaning test for measurement of growth rate, feed intake, and feed efficiency in British breed cattle. J. Anim. Sci. 75:2024–2032. Archer, J. A., S. A. Barwick, and H. U. Graser. 2004. Economic evaluation of beef cattle breeding schemes incorporating performance testing of young bulls for feed intake. Aust. J. Exp. Agric. 44:393–404. Arthur, P. F., J. A. Archer, D. J. Johnston, R. M. Herd, E. C. Richardson, and P. F. Parnell. 2001a. Genetic and phenotypic variance and covariance components for feed intake, feed efficiency, and other postweaning traits in Angus cattle. J. Anim. Sci. 79:2805–2811. Arthur, P. F., G. Renand, and D. Krauss. 2001b. Genetic and phenotypic relationships among measures of growth and feed efficiency in young Charolais bulls. Livest. Prod. Sci. 68:131–139. Barwick, S. A., M. L. Wolcott, D. J. Johnston, H. M. Burrow, and M. T. Sullivan. 2009. Genetics of steer daily and residual feed intake in two tropical beef genotypes, and relationships among intake, body composition, growth and other post-weaning measures. Anim. Prod. Sci. 49:351–366. Bergh, L., M. M. Scholtz, and G. J. Erasmus. 1992. Identification and assessment of the best animal: The Kleiber ratio (growth/ metabolic mass) as a selection criterion for beef cattle. In: Proc. Aust. Assoc. Anim. Breed. Genet., vol. 10, Rockhampton, Australia. p. 338–340. Berry, D. P., and J. J. Crowley. 2012. Residual intake and body weight gain: A new measure of efficiency in growing cattle. J. Anim. Sci. 90:109–115. Bouquet, A., M. N. Fouilloux, G. Renand, and F. Phocas. 2010. Genetic parameters for growth, muscularity, feed efficiency and carcass traits of young bulls. Livest. Sci. 129:38–48. Bourdon, R. M. 2000. Understanding animal breeding. 2nd ed. Prentice Hall, Upper Saddle River, NJ. Cochran, W. G. 1951. Improvement by means of selection. In: Proc. 2nd Berkeley Symp. Math. Stat. and Prob., Berkeley, CA. p. 449–470. Crews, D. H., Jr., M. E. Dikeman, S. L. Northcutt, D. J. Garrick, T. T. Marston, M. D. MacNeil, L. W. Olson, J. C. Paschal, G. H. Rouse, R. Weaber, T. L. Wheeler, S. D. Shackelford, R. E. Williams, and D. E. Wilson. 2010. Animal evaluation. In: L. V. Cundiff, L. D. Van Vleck, and W. D. Hohenboken, editors, Guidelines for uniform beef improvement programs. Beef Improvement Federation, Raleigh, NC. p. 16–55. Crowley, J. J., R. D. Evans, N. Mc Hugh, D. A. Kenny, M. McGee, D. H. Crews Jr., and D. Berry. 2011. Genetic relationship between feed efficiency in growing males and beef cow performance. J. Anim. Sci. 89:3372–3381. Crowley, J. J., M. McGee, D. A. Kenny, D. H. Crews Jr., R. D. Evans, and D. P. Berry. 2010. Phenotypic and genetic parameters for different measures of feed efficiency in different breeds of Irish performance-tested beef bulls. J. Anim. Sci. 88:885–894. Cunningham, E. P. 1975. Multi-stage index selection. Theor. Appl. Genet. 46:55–61.
Del Claro, A. C., M. E. Z. Mercadante, and J. A. I. I. V. Silva. 2012. Meta-análise de parâmetros genéticos relacionados ao consumo alimentar residual e a suas características componentes em bovinos. (In Portuguese.) Pesqui. Agropecu. Bras. 47:302–310. Falconer, D. S., and T. F. C. Mackay. 1996. Introduction to quantitative genetics. 4th ed. Addison Wesley Longman, Harlow, UK. Fan, L. Q., D. R. C. Bailey, and N. H. Shannon. 1995. Genetic parameter estimation of postweaning gain, feed intake, and feed efficiency for Hereford and Angus bulls fed two different diets. J. Anim. Sci. 73:365–372. Fitzhugh, H. A., Jr., and C. S. Taylor. 1971. Genetic analysis of degree of maturity. J. Anim. Sci. 33:717–725. Giannotti, J. D. G., I. U. Packer, and M. E. Z. Mercadante. 2005. Meta-análise para estimativas de herdabilidade para características de crescimento em bovinos de corte. (In Portuguese.) R. Bras. Zootec. 34:1173–1180. Gunsett, F. C. 1984. Linear index selection to improve traits defined as ratio. J. Anim. Sci. 59:1185–1193. Hegarty, R. S., J. P. Goopy, R. M. Herd, and B. McCorkell. 2007. Cattle selected for lower residual feed intake have reduced daily methane production. J. Anim. Sci. 85:1479–1486. Herd, R. M., and S. C. Bishop. 2000. Genetic variation in residual feed intake and its association with other production traits in British Hereford cattle. Livest. Prod. Sci. 63:111–119. Hoque, M. A., P. F. Arthur, K. Hiramoto, and T. Oikawa. 2006. Genetic relationship between different measures of feed efficiency and its component traits in Japanese Black (Wagyu) bulls. Livest. Sci. 99:111–118. Kleiber, M. 1936. Problems involved in breeding for efficiency of food utilization. J. Anim. Sci. 1936b:247-258. Koch, R. M., L. A. Swiger, D. Chambers, and K. E. Gregory. 1963. Efficiency of feed use in beef cattle. J. Anim. Sci. 22:486–494. Lancaster, P. A., G. E. Carstens, D. H. Crews Jr., T. H. Welsh Jr., T. D. A. Forbes, D. W. Forrest, L. O. Tedeschi, R. D. Randel, and F. M. Rouquette. 2009. Phenotypic and genetic relationships of residual feed intake with performance and ultrasound carcass traits in Brangus heifers. J. Anim. Sci. 87:3887–3896. MacNeil, M. D., N. Lopez-Villalobos, and S. L. Northcutt. 2011. A prototype national cattle evaluation for feed intake and efficiency of Angus cattle. J. Anim. Sci. 89:3917–3923. Mercadante, M. E. Z., and A. G. Razook. 2010. Programa de melhoramento genético do Instituto de Zootecnia. In: A. V. Pires, editor, Bovinocultura de Corte. Vol. 2, 1st ed. (In Portuguese.) Fundação de Estudos Agrários Luiz de Queiroz, Piracicaba, Brazil. p. 847–860. Meyer, K. 2007. WOMBAT – A tool for mixed model analyses in quantitative genetics by restricted maximum likelihood. J. Zhejiang Univ. Sci. B 8:815–821. Nkrumah, J. D., J. A. Basarab, Z. Wang, C. Li, M. A. Price, E. K. Okine, D. H. Crews Jr., and S. S. Moore. 2007. Genetic and phenotypic relationships of feed intake and measures of efficiency with growth and carcass merit of beef cattle. J. Anim. Sci. 85:2711–2720. NRC. 1996. Nutrient requirements of beef cattle. 7th ed. National Academy Press, Washington, DC. Razook, A. G., L. A. Figueiredo, J. N. S. G. Cyrillo, L. J. Pacola, L. M. Bonilha Neto, J. B. F. Trovo, A. C. Ruggieri, and M. E. Z. Mercadante. 1997. Prova de Ganho de Peso: Normas adotadas pela estação experimental de Zootecnia de Sertãozinho. Boletim Técnico, n. 40. (In Portuguese.) Instituto de Zootecnia, Sertãozinho, Brazil. Robinson, D. L., and V. H. Oddy. 2004. Genetic parameters for feed efficiency, fatness, muscle area and feeding behavior of feedlot finished beef cattle. Livest. Prod. Sci. 90:255–270.
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
Selection for feed efficiency Rolfe, K. M., W. M. Snelling, M. K. Nielsen, H. C. Freetly, C. L. Ferrel, and T. G. Jenkins. 2011. Genetic and phenotypic parameter estimates for feed intake and others traits in growing beef cattle, and opportunities for selection. J. Anim. Sci. 89:3452–3459.
965
Santana, M. H. A., P. Rossi Jr., R. Almeida, and D. C. Cucco. 2012. Feed efficiency and its correlations with carcass traits measured by ultrasound in Nellore bulls. Livest. Sci. 145:252–257. Schenkel, F. S., S. P. Miller, and J. W. Wilton. 2004. Genetic parameters and breed differences for feed efficiency, growth, and body composition traits of young beef bulls. Can. J. Anim. Sci. 84:177–185.
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
References
This article cites 27 articles, 14 of which you can access for free at: http://www.journalofanimalscience.org/content/92/3/955#BIBL
Downloaded from www.journalofanimalscience.org at Mansfield Library - Serials on December 14, 2014
