Effect of feeding less shell, extruded and enzymatically treated palm kernel cake on expression of growth-related genes in broiler chickens

Abstract Palm kernel cake (PKC) is a residue of palm kernel after oil extraction and can be used as a by-product feedstuff for livestock. In the current study, various post-treated PKC fed to broiler chickens and their growth performance and expression of genes related to growth trait studied. A total of 2500-day-old broiler chicks were randomly assigned to following five isocaloric and isonitrogenous diets: 0% PKC (control: corn-soybean meal), 25% PKC in the forms of either untreated or less-shell or extrudedor enzymatically-treated. Each treatment group consisted of five replicates with 100 chicks each. Growth performance and expression of 27 selected genes related to growth and metabolism pathways were investigated. Chickens were fed with treated PKC diets had lower (p < .05) body weight gain, inferior feed conversion ratio (FCR) and higher feed intake. The expression of genes involved in mTOR (GRB2, GRB10), FoxO (FOXO3) and insulin (PRKCZ) signalling pathways, glycolysis/gluconeogenesis (ENO1), fructose and mannose metabolism and apoptosis (RHOBTB2, LOC101750363) were significantly up-regulated in broilers fed with less-shell PKC. It can be concluded that the shell reduction, enzymatic and extrusion treatment practiced in this study did not change PKC feeding potentials in broiler chickens. Highlights Significant up-regulation of eight genes which involved in energy metabolism and growth related functions were observed in broilers fed with less-shell palm kernel cake (PCK). The DEGs (differentially expressed genes), which were upregulated in less shell palm kernel cake fed broilers constituted the pathways related to energy and metabolism such as mTOR, glycolysis, fructose and mannose metabolism, FOXO, apoptosis, and insulin signaling pathway. Higher bulk density and lower water holding capacity of palm kernel cake may lead to higher feed consumption in palm kernel cake incorporated diets.


Introduction
The poultry industry makes up 75% of the livestock industry in Malaysia. The poultry industry plays an important role in Malaysia's economy by providing a cheap source of meat protein. Malaysia produces 1.5 million chickens a day, which is worth of RM10 billion a year as a net exporter (NST 2017). However, the poultry industry is heavily dependent on imported feed ingredients such as corn and soybean meal as the source of energy and protein, respectively. As a result of this dependency, the cost of poultry production is dictated by the fluctuations in the cost of these materials on the global market. The cost of feed represents over 70% of the poultry production cost (Ravindran 2013). Nevertheless, the import demand for corn and soybean meal in Malaysia continues to increase in tandem with the demand for poultry products. In lieu of this dependency on foreign imports of feed, poultry producers in Malaysia have to find alternative local feed ingredients to sustain the poultry industry.
Malaysia is ranked as the second-largest producer of palm oil after Indonesia (Mekhilef et al. 2011). One of the by-products of this industry is palm kernel cake (PKC), with a yearly production of more than 2 million tons (Mohamed et al. 2012). Palm kernel cake contains high fibre (18-21%) and a moderate amount of protein (15-17%) which makes it highly suitable as a ruminant feed (Alimon and Wan Zahari 2012). However, the use of PKC as poultry feed is limited to about 15%. The broilers' performance was significantly reduced possibly due to high fibre content which reduces feed digestibility (Abidah and Nooraida 2017). Another aspect that limits the usage of PKC in monogastric animal diets is the presence of broken kernel shells (about 12%), which contributes to the fibre and lignin content of PKC (Roslan et al. 2015).
Numerous studies on the gene expression on concerning regulation of adipocyte proliferation in chickens with the cDNA microarray have been reported (Bohannon-Stewart et al. 2014;Hausman et al. 2014). However, little is known about the expression of genes that regulate growth traits in broiler chickens using the PCR array approach. Since growth performance is the most important trait in production, knowledge of the effects of nutrients on the expression of growth regulatory genes would be beneficial in understanding the effect of dietary treatments on growth performance.
Hence, in the present study, a feeding trial with broiler chickens was conducted to evaluate the effects of feeding untreated PKC, extruded PKC (EXPKC), lessshells PKC (LSPKC) and enzyme-treated PKC (ETPKC) on their growth performance and the expression of genes related to the growth trait.

Experimental birds and diets
The experimental protocol was approved by the Institutional Animal Care and Use Committee of Universiti Putra Malaysia. A total of 2500 day-old (Cobb 500) broiler chicks were obtained from a local commercial hatchery. The birds received a starter feed from day 1-18 and finisher feed from day 19-39. The chicks were weighed and randomly assigned in five groups of 500 each and raised in slated floor pens in a naturally-ventilated open house. From day 1-18, all the chicks were provided a standard commercial broiler starter diet. Commencing from day 19, five replicates of 100 chicks each were assigned to one of the following isocaloric, isonitrogenous diets: (1) 0% PKC (control; basal diet); (2) 25% untreated PKC; (3) 25% EXPKC; (4) 25% LSPKC and (5) 25% ETPKC. All diets were free of antibiotics or growth promoters. The experimental diets were formulated based on the ME values of the respective PKC to meet or exceed the requirement by Cobb for broilers of this age (Table 1). Feed and water were provided ad libitum. Feed intake, body weight and feed conversion ratio (FCR) were determined at day 21 and 39. All performance parameters were recorded and calculated as pen-basis. All performance data were subjected to one-way ANOVA using the GLM procedure of SAS (V 9.1, SAS Institute Inc., NC).

Gene expression and PCR array
On day 39, one average size bird was randomly selected from each replicate pen per treatment group and slaughtered humanely. Liver samples were collected immediately and snap-frozen in liquid nitrogen and stored at À80 C for RNA extraction. Total RNA extractions of liver samples were conducted using RNeasy Lipid Tissue Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer's instruction. The purity of extracted RNA was monitored by (Alimon and Wan Zahari 2012) measuring the optical density ratio 260/280 nm. The ratio between 1.8 and 2.0 was considered acceptable. One microgram of RNA was reverse transcribed using RT 2 First Strand Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. The reverse transcription process involved genomic DNA elimination step.
To determine the expression of the growth and metabolism-related genes, a custom-made chicken RT 2 Profiler PCR Array, was designed (SABiosciences, Qiagen). Each 96-well array contained 3 identical sets of genes/control which include 27 genes of interest (Table 2), 2 internal control genes, a positive PCR control, a reverse transcription control, and a well to test for genomic DNA contamination. All reactions were performed in triplicate. Real-time PCR was performed  using RT 2 SYBR Green Mastermix (Qiagen, Hilden, Germany) and was quantified by Bio-Rad CFX96 Touch Real-time PCR Systems (Bio-Rad, USA). The PCR was performed according to the following workflow: initial denaturation at 95 C for 3 min, 40 cycles of denaturation for 15 s at 95 C, primer annealing for 30 s at 60 C and extension at 72 C for 30 s. The data normalisation was performed using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and b-actin as the housekeeping genes. Alterations in mRNA levels with mean difference of equal to or greater than two-fold change with p value less than .05 were considered to be up-or down-regulated. The statistical calculation was based on the web-based programme of RT 2 Profiler PCR Array Data Analysis (https://www.qiagen. com/us/products/genes%20and%20pathways/data-analysis-center-overview-page/?UID¼44dedd6b-4002-4924-81bb-dfb2b330d20e). Table 3 shows the growth performance of broilers fed with basal diet (control), treated and untreated PKC diets. There was no significant difference in feed intake among broilers fed with treated and untreated PKC with the control group during the treatment period (day 21-39). The feed intake of broilers fed  with treated PKC diets was significantly (p < .05) higher than the control and untreated PKC groups throughout the experimental period (day 1-39). In terms of body weight gain, birds fed PKC containing diets had significantly (p < .05) lower values than the control birds. The FCR of the birds were significantly (p < .05) higher in all PKC fed chickens compared to the control, however broilers fed with untreated PKC showed better FCR than broilers fed with treated PKC.

PCR array analysis
All the genes that play important roles during chicken growth and development were expressed but not significantly up-regulated in untreated PKC and EXPKC groups ( Table 4). The genes IL6 and TNFRSF1A that were involved in FOXO and apoptosis signalling pathway, respectively, were not up-regulated in all the dietary treatments. Significant up-regulation in the expression of energy and metabolism gene markers such as GRB2, GRB10, ENO1, PFKL, FOXO3, LOC1017750363, PRKCZ and RHOBTB2 were observed in LSPKC-fed chicken based on the p values and fold-change cut-off as indicated in Table 4. However, only two genes; PRKCZ and RHOBTB2 (p < .05) were significantly up-regulated in broilers fed with ETPKC (Table 4).

Discussion
Chickens' growth performance is a significant economic factor in broiler production. A high growth rate has been one of the main goals of broiler breeding for the past few decades. Early recommendations for using PKC in poultry diets were 15% for broiler chickens; a higher inclusion rate was thought to result in reduced poultry performance (Yeong 1981). Onwudike (1986) suggested that the inclusion of PKC in broiler starter and finisher diets could be as high as 28% and 35%, respectively without any adverse effect on the growth performance. A higher rate of inclusion of up to 50% PKC was suggested by Panigrahi and Powell (1991), but in contrast, the inclusion rate of 10-15% PKC resulted in lower body weight (Soltan 2009). In the present study, the body weight of all PKC fed chickens was lower when compared to the control group. There was no significant effect of any of the PKC treatments, and for that matter, broilers fed untreated PKC performed equally well compared to the other groups fed treated PKC.
As observed by Sundu et al. (2006) the feed intake of birds fed diets containing PKC is usually higher than those fed a corn-based diet. In the present study, treated PKC groups showed a higher feed intake compared to control and untreated PKC groups. Nevertheless, those that consumed more feed showed poorer FCR than those in the control and untreated PKC groups. The higher feed intake was probably due to a faster passage rate of the feed caused by the presence of crude fibre (3.55-5.23%) as compared to the control (2.92%) which increased the contraction rate of the gizzard. This may have sped up the peristaltic movement of digesta in the duodenum and throughout the small intestine, which, in turn, resulted in a higher feed intake (Sundu et al. 2006). The higher bulk density and lower moisture capacity of PKC compared to the bulk densities of many poultry feedstuffs may lead to higher feed consumption in PKC incorporated diets (Sundu et al. 2006).
In the present study, broilers fed with PKC diets showed a similar growth performance. The respective ME values for each PKC feed in this study were made to be isocaloric. Nevertheless, the effects of the various treatments on nutrients availability were not observed. Although extrusion reduced the fibre content of PKC from 18 to 15%, the high extrusion temperature and pressure during the extrusion process may have caused the loss of digestible protein and amino acids that subsequently affect nutrient digestibility Martinchik and Sharikov (2015).
A number of studies have indicated the positive effects of enzymes treated PKC on the performance of broilers (Zakaria et al. 2010;Makhdum et al. 2013;Hanafiah et al. 2017). However, the results of the present study were otherwise. A similar observation was reported by Saenphoom et al. (2013) and Chen et al. (2018), who found that although enzyme treatment increased the reducing sugar and reduced the fibre content of PKC, the growth performance of broilers compared to the control (without PKC) or untreated PKC was not significantly enhanced. The lack of improvement in the growth performance of the broilers fed enzyme treated PKC is not clearly understood. However, the possible explanation for the above results may be because, PKC contains a high level of non-starch polysaccharides (NSPs) (46.6%), in which the main component is mannan (78%) (Dusterhoft and Voragen 1991;Knudsen 1997). Mannans impair the digestibility and utilisation of nutrients either by directly capsuling the nutrients or by increasing the viscosity of the intestinal contents. This causes a reduction of the hydrolysis rate and the absorption of nutrients in the diet (Sundu and Dingle 2003). Although extrusion reduced the NSPs and increased the manno-oligosaccharides (MOS), nevertheless the reduction of about 6% (Roslan et al. 2017) was not substantial enough to provide a significant impact on the digestion of PKC.
In terms of number of DEGs (differentially expressed genes), significantly up-regulated gene expression was observed in broilers fed LSPKC and ETPKC (Table 4). None of the DEGs were significantly down-regulated in all dietary treatments. The DEGs, which were upregulated in LSPKC fed broilers constituted the pathways related to energy and metabolism such as mTOR, glycolysis, fructose and mannose metabolism, FOXO, apoptosis, and insulin signalling pathway. The mTOR pathway involved in the energy-sensing/signalling regulates the feed intake and energy balance (Cota et al. 2006). Chang et al. (2015) reported that the inclusion of amino acid leucine in chickens diet activates the mTOR signalling, leading to a decrease in feed intake and body weight. Elevation of the mTOR signalling pathway reduced the feed intake and body weight through increasing the energy expenditure (Chang et al. 2015). In the present study, LSPKC stimulated the mTOR signalling, but the feed intake was comparable to that of untreated PKC group. This indicated that the mTOR signalling could be modulated by LSPKC and its effect on food intake is mTOR independent. The mTOR pathway plays a key role in nutrient sensing and cellular response to growth factors (Sengupta et al. 2010). Activation of the mTOR pathway could lead to up-regulation of the glycolysis via several ways including, increasing the influx of glucose and transcription activation of key glycolytic genes expression (Ward and Thompson 2012). Hence, in our findings, this explains why ENO1 and PFKL in LSPKC that are involved in the regulation of glycolytic pathway were up-regulated (p < .05). FOXO pathways play an important role in promoting protein degradation and thereby contributing to muscle atrophy (Wen et al. 2017). Thus, the increased expression of FOXO3 (p < .05) may have increased the loss of muscle protein in LSPKC fedchickens (Zhou et al. 2015). The growth rate in meat producing animals is related to the muscle protein accretion, that is, the higher the protein accretion, the higher the growth rate of the animal (Therkildsen and Oksbjerg 2009). The up-regulation of LOC101750363 and RHOBTB2 (p < .05) which involved in apoptosis pathways (St. Paul et al. 2014;Monson et al. 2015;Zhang et al. 2015) in LSPKC and ETPKC fed-chickens could influence the growth responses.
The insulin signalling pathway mechanisms involve glucose storage and uptake, protein synthesis, and regulation of lipid synthesis (Kim et al. 2014). Another gene that was found to be up-regulated in LSPKC and ETPKC fed-chickens was PRKCZ (p < .05), which plays an important role in the modulating insulin signalling pathways (Hassan et al. 2016). However, the energy produced may not be efficiently used for growth and development as it was observed that these broilers did not show improvement in growth performance compared to the control group.

Conclusions
The growth performance of broilers fed with untreated and treated PKC groups was significantly (p < .05) lower than the broilers fed the control diet. The expression of genes involved in the energy metabolism and growthrelated functions evaluated were expressed accordingly in all dietary treatments. However, only 29.6% of the genes evaluated were significantly up-regulated in broilers fed LSPKC. Therefore, a more systematic analysis to the evaluation of nutritional strategies of feed via assessing the gene responses would allow a rapid understanding of the nutrigenomic aspect.