Genetic changes in growth curve parameters in

Europ.Poult.Sci., 81. 2017, ISSN 1612-9199, © Verlag Eugen Ulmer, Stuttgart. DOI: 10.1399/eps.2017.205


Genetic changes in growth curve parameters in Japanese quail lines divergently selected for body weight Genetische Veränderung von Wachstumskurvenparametern bei in entgegengesetzten Richtungen auf Körpergewicht selektierten Linien Japanischer Wachteln 1

2

3

3

K. Karabağ* , S. Alkan , T. Karslı and M.S. Balcıoğlu

1

Faculty of Agriculture, Dept. of Agricultural Biotechnology, Akdeniz University, Antalya, Turkey

2

Faculty of Agriculture, Dept. of Animal Science, Ordu University, Ordu, Turkey

3

Faculty of Agriculture, Dept. of Animal Science, Akdeniz University, Antalya, Turkey

*Correspondence: [email protected] Manuscript received 23 May 2017, accepted 23 September 2017

Introduction The Japanese quail which has been used as a model animal for a long time has become an improtant poultry species in commercial production. Consumption of quail meat has increased significantly, especially in South America and Southern Europe. In previous studies, long-term selection was applied to increase body weight, but these studies were not sustainable since the flocks were small, and the focus was only on body weight (CARON et al., 1990; TOELLE et al., 1991; MARKS, 1996). Despite the fact that the results obtained from these studies assist the breeding of other poultry species, no superior quail genotype has been developed for a widespread use in commercial production of Japanese quail. However, there are some heavy quail meat lines, like French commercial lines, which may reach over 300 g body weight (MINVIELLE et al., 2000). It is believed that it is easy to obtain a hybrid genotype when the short generation interval in quail is considered (MINVIELLE, 1998; NARINÇ et al., 2016). Growth, feed efficiency, and carcass yield are the three main features in poultry species bred for meat production (MINVIELLE, 2004; NARINÇ et al., 2010a; NARINÇ et al., 2014a; KARAMAN et al., 2014). Weekly body weights, body weight gain and growth curves are suitable to describe the characteristics of growth. Gompertz, Logistic and Richards growth models were used in many studies in Japanese quail to model growth data (TZENG and BECKER, 1981; ANTHONY et al., 1991; AKBAŞ and OĞUZ, 1998; ALKAN et al., 2009; NARINÇ et al., 2010b; KARAMAN et al., 2013; NARINÇ et al., 2017). Significant changes in growth curves occurred in the selection of broilers. The point of inflection age of growth curve is now very close to the slaughter age (NARINÇ et al., 2017). Today, genetic improvement studies are carried out to enhance the meat efficiency level of quail in many countries such as in France, Brazil and Turkey. Currently, selection studies have been conducted to increase body weight at a fixed age or for a certain period. A fewer number of genetic studies considered carcass traits and feed efficiency (VARKOOHI et al., 2010; KHALDARI et al., 2011). Recently, studies aimed to develop dam and sire lines in terms of numerous traits by using the BLUP method. (NARINÇ and AKSOY, 2014; NARINÇ et al., 2016). Phenotypic mass selection was applied in many of the genetic breeding studies in Japanese quail, and phenotypic progressions were pointed out over generations (MARKS, 1991; HYANKOVA et al., 2001). It has also been argued

whether changes implemented by phenotypical progression are caused by other production traits (AKBAŞ and OĞUZ, 1998; TÜRKMUT et al., 1999; DURMUŞ et al., 2017). Many researchers emphasised that the selection implemented for body weight affects weight, feed efficiency and growth curve parameters (HYANKOVA et al., 2001, ALKAN et al., 2009; VARKOOHI et al., 2010; KHALDARI et al., 2010; NARINÇ et al., 2014b). Nevertheless, very few studies have discussed the effects of other traits during selection for body weight on the genetic variation (CARON et al., 1990; NARINÇ and AKSOY, 2014; HUSSAIN et al., 2014b; NARINÇ et al., 2016). The purpose of this study was to determine the effects of 11 generations of divergent selection on 5-week body weight on both the phenotypic and genetic variations of the

20.11.2017

1 / 10


Richards growth model parameters. The results will visualise the genetic and phenotypic changes in the different lines due to breeding.

Materials and Methods

Animal Husbandry and Selection Experiment In this study Japanese quail (Coturnix coturnix japonica) lines were used which have been selected divergently for either high (HL: High Line) or low (LL: Low Line) 5-week body weight for 11 generations, as well as a random bred control line (C). The selection lines were derived from the same base population, which was randomly mated for

several years and was not subjected to any breeding program at the Faculty of Agriculture of the Akdeniz University in Turkey. The trial was conducted in accordance with the guidelines of the Animal Experiment Local Ethics

Committee of Akdeniz University. The lines were established by applying individual selection with 10% and 40% selection intensity for males and females, respectively. A one-to-three mating rate was applied. Fertilised eggs were collected from lines HL, LL and C for a week and were stored at 15–20°C and 75–80% humidity. These eggs were

incubated at 37.5°C and 55% humidity for the first 14 days and at 37.2°C and 70% humidity for the last 4 days. Chicks were weighed individually using 0.01 g precision scales, and an aluminum ID number was attached to the left wings of chicks after incubation. Quail were reared as straight-run flocks under standard brooding and rearing temperatures. Birds were housed individually in 20x20x29 cm (length × width × height) cages in a quail house with windows at both sides, exposed to 16 h of light and 8 h of darkness. During the experiment, the birds were fed a diet containing 11.7 MJ/kg metabolisable energy and 21% g crude protein/kg ad libitum and had free access to drinking water.

Statistical Analyses To obtain estimates of individual growth curve parameters, all quail from the three lines were weighed weekly from th

hatching to 5 week of age. The Richards nonlinear regression model [1] was used to estimate the growth curve of each quail. [1] Where yt is the weight of bird at time t, day),

is the asymptotic (mature) weight,

is the age at maximum rate of growth (IPA, day), and

the relative weight at the inflection point (AGGREY et al., 2003).

is the maximum relative growth (per

is a shape parameter, with the property that

is

Parameters of Richards function were estimated by using PROC NLIN (Marquart algorithm) procedure of SAS 9.3 software (SAS Institute Inc., Cary, NC). Descriptive statistics and Shapiro-Wilks normality test of the traits were performed using the UNIVARIATE procedure of SAS software (version 9.3; SAS Institute Inc., Cary, NC). Firstly, the parametric Box-Cox transformation was used for non-normally distributed data. Then a non-parametric Rank transformation was performed for the traits that were not normally distributed according to the Box-Cox transformation (NARINÇ et al., 2014a). The restricted maximum likelihood (REML) estimator was used to estimate variance-covariance components for the following multi-trait model [2]; [2] where y, vector of observations for the trait; β, vector of fixed effects (gender and cage floor) for the trait; u, vector of random animal effects for the trait; e, vector of random residual effects for the trait; and X and Z are incidence matrices relating records of the trait to fixed and random animal effects, respectively. The sire, dam, residual variance components and additive genetic, environmental covariance matrices for multivariate analyses were estimated using a mixed-model equation by SAS PROC MIXED. Heritabilities ( ) and genetic correlations (rg(ii´)), were calculated by the means of variance and covariance parameters as follows: Where i and i´ represents the trait(s) of interest and respectively. Also,

20.11.2017

and

are the diagonal elements of G0 and R0 matrices,

stands for the additive genetic covariance between the traits i and i´. Estimates of genetic

2 / 10


correlation and heritability and their approximate standard errors for the traits were obtained by SAS interactive matrix language (IML) procedure (SAS Institute Inc., Cary, NC).

Results The results of the analysis of variance and multiple range test (Duncan) and least square means for body weight at 5

weeks of age (BW), the parameters of the Richards growth curve (β0, β1, β2) and the coordinates of the inflection point (IPA and IPW) traits by line, gender and interaction effects are presented Table 1. A significant difference between all

lines was found for all characteristics when parameter β2 was excluded. In contrast, when excluding parameters β1 and β2 a significant difference was found for all characters in terms of gender (for all signs P