Effects of dietary energy and protein levels on growth curve parameters of Khazak native chickens


 This study was conducted to evaluate the effect of dietary energy (ME) and protein (CP) on growth parameters and absolute growth rates in the different ages of the Khazak chicks. A total of 360 one-day-old Khazak chicks were obtained from a local hatchery and in a 3×3 factorial experiment with completely randomized design, chicks were randomly allocated to experimental diets including 2,600, 2,800, and 3,000 kcal of ME/kg, and each containing 17, 19, and 21% CP from 7 to 98 days of age. Four growth model (Gompertz, Logistic, Lopez, and Richards) were fitted on weekly body weight data and the best model were selected by the goodness of fit criteria. Growth curve parameters were predicted for all chicks using the best model and other parameters including age (Ti) and weight (Wi) at the inflection point and absolute growth rate (AGR) in different ages were calculated from growth curve parameters. All parameters were analyzed using the general linear model procedure of SAS. Based on goodness of fit criteria, the Richards model had the lowest Akaike’s Information Criteria (AIC), root mean square error (RMSE), and highest adjusted determination coefficient (R2Adj) than other models and was selected as the best model. The effect of ME was significant on the mature index (k), Wi, Ti, and all AGR parameters (P<0.05) while CP levels were significant on final weight (Wf), Wi, and AGR parameters (P<0.05). The chicks fed with a diet containing 2,600 kcal of ME/kg and 17 % CP had the higher k parameter, and lower Wi, Ti, and AGR than those fed with other diets (P<0.05). Considering that the level of 2,800 kcal of ME/kg and 19 % CP had no significant difference with the level of 3,000 kcal of ME/kg and 21 % CP, therefore diet with 2,800 kcal of ME/kg and 19 % CP was suggested as optimum levels for change the growth curve parameter and having best performance for Khazak chickens during 7 to 98 days of old.


38
In rural areas of the tropical, rearing native poultry plays a pivotal role in the production of high-quality animal 39 protein resources with organic properties and income generation (Norris et al 2007;Padhi, 2016). Recently, public 40 concern was increased about the use of modern broiler genotypes for the production of chicken meat, and the tendency 41 toward the consumption of meat that produced from slow-growing broilers instead of fast-growing is increasing 42 (Dyubele et al., 2010). Furthermore, meat produced by modern broiler genotypes is less palatable than native breeds 43 (Wattanachant, 2008). Rural poultry has significant prospects for future development due to the easy and abundant 44 availability of all necessities input including land, labor, and feed resources in rural areas. This section can help in 45 increasing household income and improving family health through best nutrition (Shehbaz Anjum and Hassan Khan, characteristics of native chickens (Miah et al., 2014). In some studies, the effect of different levels of energy and 55 protein on growth performance in Korat (Maliwan et al., 2019), Venaraja (Perween et al., 2016), Desi (Miah et al., 56 2014), Arabi (Al-Khalifa and Al-Nasser, 2012), Assel (Haunshi et al., 2012) Betong (Nguyen et al., 2010), chickens 57 were reported. However, in these studies, the effect of energy and protein was investigated on body weight gain, feed 58 intake, and feed conversion ratio.

59
The growth of an animal can be defined as any change in body size per time (Narinc et al., 2017). These changes 60 can be measured as body weight in regular intervals and summarized by mathematical models fitted to growth curves 61 (Aggrey, 2002). Growth curves in animals are generally S-shaped Growth curves and can be divided into two phases 62 including the accelerated phase (the growth from hatching to the inflection point) and decelerating phase (the growth 63 rate decreased until to a mature weight) (Selvaggi et al., 2015). The mathematical models that are used to describe 64 growth curves have biological parameters such as body weight at a specific time, body weight at maturity, age and weight at the inflection point, growth rate (Masoudi and Azarfar, 2017). The growth curve can be useful in describing 66 the production of animals, especially when they can be estimated using the number of daily feed requirements (Abbas 67 et al., 2014). It has been shown that the shape of the growth curve was affected by the composition of the diet 68 (Mohammad, 2015). Prediction of production, as well as the nutritional requirements of birds of different ages, can 69 result in a restriction on the level of ad libitum access to feed (Lopez et al., 2000).

70
Studies on long-term growth curves of animals can be helpful in the dynamically understanding of growth patterns 71 and responses to dietary nutrient density as well (Russo, 2009, Yun et al., 2015. In a study by Nahashon et al. (2010) 72 reported that dietary protein and energy affect the growth parameters of the French guinea fowl by using of Gompertz-73 Laired model. Growth curve parameters of commercial broiler and native chickens that fed by different energy levels 74 by four growth models were studied and a significant effect of energy level was reported on some growth parameters 75 (Moharrery and Mirzaei, 2014.).

76
The Khazak breed is one of the small native breeds in the Sistan (Sistan region, IRAN), with short legs relatively 77 high potential for egg production. A tendency toward the native chicken consumption in this region was more than 78 industrial chickens, so an improvement of the performance for this breed in terms of growth by demining appropriate 79 levels of energy and protein could be useful (Faraji-Arough et al., 2019). Therefore, this study aimed to evaluate the 80 growth curve parameters of Khazak chickens fed diets containing different levels of energy and protein. were raised together until 7 days of age in floor pens containing litter composed of wood shaving. At one week of age, 88 chicks were weighed and randomly distributed into nine groups. Each group had 40 chicks that were allocated into 4 89 replicate with 10 birds in each. The chicks were fed with a maize-soybean meal-based diet supplying three levels of 90 metabolizable energy (2,600, 2,800, and 3,000 kcal/kg) and three levels of crude protein (17%, 19%, and 21%) in a 91 3×3 factorial experiment with a completely randomized design ( Four non-linear mathematical models including Gompertz, Logistic, Lopez, and Richards were fitted to the body 98 weight data to recognize the best model. The age and weight at the inflection point and Absolute growth weight in 99 different ages for each model were calculated based on the model parameters. The equations of fitted models and 100 biological parameters are shown in Table 2. In all models, W is the body weight of a bird at age t, W0, Wf, and k are 101 initial and final weights, and coefficient of relative growth or maturing index, respectively. The parameter b indicates 102 the age at approximately half the maximum body weight, and m represents the shape parameter. The fitting of models 103 on body weight data was performed by nlme package of R software (Pinheiro et al. 2014

105
After fitting the models, four goodness of fit criteria were used to compare the models and selection of the best

116
The parameters of the model for each bird were obtained by the nlsList package of R software using the best model.

126
The mean and standard error of growth parameters for studied models for all populations is shown in 136 Table 3.

137
The mean growth curve parameters and absolute growth rate in different ages for ME and CP levels are presented 138 in Table 4. The initial body weight (W0) was high in birds fed a diet containing 3000 kcal of ME /kg than other levels 139 of ME, but the difference between ME levels was not significant (P˃0.05). A similar trend was observed for final 140 body weight (Wf) so that the final body weight was high in higher levels of ME (P˃0.05). However, the predicted 141 maturing index (k) was significantly lower (P˂0.05) in birds fed with a diet containing 3000 kcal of ME/kg compared 142 with those fed the 2600 and 2800 kcal of ME/ kg diets.

144
The difference between shape parameter (m) among dietary ME levels was not significant (P˃0.05), but the 145 difference of age (Ti) and weight (Wi) at the inflection point among ME levels were significant (P˂0.05) so that birds 146 fed with a diet containing 3000 kcal of ME /kg arrived at age at the inflection point in higher ages than other ME 147 levels. Forasmuch as the birds fed with higher levels (3000 and 2800 kcal/kg) of ME had a higher Ti so these birds 148 showed a higher weight at the inflection point (Wi) than those fed diet containing 2600 kcal of ME/kg (P˂0.05). The 149 effect of diet CP levels was significant on Wf and Wi parameters were significant (P˂0.05,

153
The mean absolute growth rate in different ages for various levels of ME and CP are presented in Table 4. The 154 effect of ME and CP levels was significant on absolute growth rate in different ages (P>0.05). The highest absolute 155 growth rate was observed for birds fed a diet containing 3000 kcal of ME/kg and 21 % CP, while the lowest absolute 156 growth rate was related to the diet containing 2600 kcal of ME/kg and 17 % CP. However, the mean absolute growth 157 rate for birds that received a diet with 2800 kcal of ME/kg and 19 % CP was not significant with a high level of energy 158 (3000 kcal/kg) and protein (21 %).
159 Table 5 shows the interaction effect of ME and CP on growth curve parameters and Absolute growth rate in 160 different ages. The interaction effect of ME and CP was significant on the k and Wi parameters (P<0.05) and the 7 differences of W0, Wf, m, and Ti were not significant (P>0.05). The mean of the mature index (k) for birds fed with a 162 diet containing 2600 kcal of ME/kg and 16 % CP was higher than other treatments, but only its difference was 163 significant than diets with 3000 kcal of ME/ kg and 19 and 21 % CP. The highest value of weight at the inflection 164 point was birds fed with a diet containing 3000 kcal of ME/kg and 21 % CP that was significantly higher than the 165 2600 kcal of ME/kg and 17 % CP diet. Interaction of ME and CP had a significant effect on absolute growth rate in 166 different ages except for absolute growth rate at 77 and 98 days of age (P<0.05). In all ages, the mean absolute growth 167 rate for birds fed with a diet containing 2600 kcal of ME/kg and 17 % CP was lower than other diets.

174
Cumulative feed consumption until slaughter weight in poultry depends on growth rate and the shape of the growth 175 curve, therefore the use of a mathematical model in combination with feed intake data can be useful in bioeconomic 176 studies (Pasternak and Shalev, 1983).

177
In this study, the Richards model was selected as the best model to describe the growth curve of Khazak chickens.

178
In agreement with this finding, the superiority of the Richards models to describe the growth curve of broiler and 179 native chickens that fed with different levels of ME was reported (Tompic et al., 2011;Moharrery and Mirzaei, 2014). chickens that in contrast with present results. However, the data of body weight in this study was related to chicks that 189 fed with diets with different levels of ME and CP, but the used data in a study by Faraji-Arough et al. (2019) was 190 collected from chicks that used the same diet (2,800 kcal of ME/ kg and 16% CP) that could a reason for this difference.

191
Our findings indicates that the maturity index (k) decreased linearly with increasing of ME and CP levels (Table   192 4) whereas the final weight (Wf) increased with the increment of ME and CP levels. Inverse association between Wf

203
Effect of ME and CP on growth curve parameter of French guinea fowl by Gompertz model was studied by 204 Nahashon et al. (2010) and no significant effect of ME on growth curve parameter except for W0 was reported that 205 was contrary to the results of the present study, but similar to our result, the effect of CP was significant on Wi and 206 growth rate. It has also been reported that the final body weight of guinea fowl broiler fed diet containing 3,100 or 207 3,150 kcal of ME /kg was significantly higher than those fed the 3,050 kcal of ME /kg diet (Nahashon et al., 2005) 208 that was opposite with our finding.

209
Based on the Richards model, the decreased dietary ME concentration cause a linear decrease in Wf in broiler 210 chickens (P<0.05) (Moharrery and Mirzaei, 2014) but no effect of ME on Wf and other growth parameters in native 211 chickens was reported that was similar with present results. Adaptation of native chickens in consuming feed of lower 212 ME concentration can be a reason for the lack of significant effect of dietary ME on Wf. On the other hand, the native 213 chickens do not need high energy concentration in their feed due to having a slower growth rate and small size at maturity. Therefore, their feed intake was reduced with increasing ME in diet, which can be another reason for the 215 non-significant effect of ME on Wf (Moharrery and Mirzaei, 2014).

216
After hatch, the growth of birds is accelerated till a certain age which this time showed by age at the inflection 217 point (Ti) and growth rate of bird in this maximum in this phase. The growth of birds decreased gradually after this 218 age to achieve their mature weight (Sakomura and Rostagno, 2016). Our finding shows that this age for chicks fed 219 with a diet containing a high level of ME significantly increased than a low level of ME. The increase of Ti provides 220 an opportunity for the chicks to gain more body weight (Wi W in all models is the body weight of bird at age t, W0, Wf and k are initial and final weights, and coefficient of relative growth or maturing index, respectively. The parameter b is the age at approximately half maximum body weight, and m represents the shape parameter.  , and m are initial weight, final body weight, coefficient of relative growth or maturing index, and the shape parameter, respectively. Wi; weight at the inflection point (g); Ti; Age at the inflection point (d); AGR: Absolute growth rate in different age (g/d); SEM: standard error of mean.
a-b: Different superscripts within a row shows significant different between energy and protein levels (P˂0.05).