Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein supplemented with lysine, methionine, and threonine in tropical environment

- This experiment aimed to discover the effect of reducing dietary protein supplemented with lysine, methionine, and threonine on growth performance, volatile fatty acid profile, and intestinal villus height and crypt depth of broilers, as well as the microflora counts isolated from broiler chicken faeces. A total of 288-day-old broilers were allocated to eight treatments with six replicates consisting of 36 birds per treatment. The diets contained dietary protein from 21 to 18% in starter diet and 18 to 16% in finisher diet supplemented with L-lysine, DL-methionine, and L-threonine at the same ratio for all dietary treatments. Body weight and feed intake were determined, and feed conversion ratio was calculated. Blood, intestine, and digesta samples were collected at 21 and 42 days for further analysis. Dietary protein supplemented with amino acids improved growth performance, reduced pathogenic bacteria, and increased beneficial bacteria counts, small intestine villi height and crypt depth, and ileal-digesta volatile fatty acid concentrations of broiler chickens. However, reducing 2% of dietary protein supplemented with lysine, methionine, and threonine showed the best results, especially in growth performance, feed conversion ratio, microflora count, duodenal and jejunal villi height, and ileal-digesta volatile fatty acid concentrations, such as butyric and valeric acids. It is believed that by reducing the level of dietary protein in broiler diet while supplementing with synthetic amino acid may enhance the intestinal morphology, nutrient digestibility, and absorption in broiler chickens and will simultaneously result in better performance. were to compare mean at P<0.05 of treatments. Data as the mean ± standard error of the mean (SEM).


Introduction
Nowadays, people tend to be concerned with gastrointestinal tract health of broiler chickens due to its improvement for their health and performance (Mountzouris et al., 2007). About 60 to 70% of the overall poultry production cost is related to the price of feedstuffs and, therefore, the preferred poultry feed itself is economically important. The efficient use of feed is extremely important in broiler production (Attia et al., 2012). Dietary protein is a vital growth and reproductive performance regulator for broiler chickens. Nevertheless, it is also essential for the gastrointestinal tract features Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein... Shazali et al. 3 finisher: 15% CP). Diets were formulated based on their nutrient requirements by using least-cost feed formulation software (FeedLive, Thailand). All diets were formulated to be isocaloric (3,000 kcal/kg). All diets were supplemented with lysine, methionine, and threonine, except for the control, which was only supplemented with methionine. The control diet received amino acids from corn-soybean diets. Toxin binder 0.14 0.14 0.14 0.14 0.14 0.14 0.14 0.14 L-lysine 0.00 0.00 0. Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein... Shazali et al. 4 Methionine levels in corn-soybean diets were inadequate to support broiler requirements. The amino acids for starter and finisher diets in different treatments were adjusted to a similar ratio of lysine and methionine to maintain the level at or above, according to the Cobb Broiler Management Guide Toxin binder 0.14 0.14 0.14 0.14 0.14 0.14 0.14 0.14 L-lysine 0.00 0.09 0.14 0. Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein... Shazali et al. 5 (Cobb-Vantress, 2013). The L-lysine HCl, DL-methionine, and L-threonine used in the diets were of feed grade and were formulated based on total amino acids required for the chickens. The concentrations of dietary calcium and available phosphorus were equally maintained in all treatments. This experiment was conducted for six weeks, and feed intake was measured weekly by deducting the feed balance from the original quantity supplied to the chickens.
Faecal lactic acid bacteria (LAB) and Enterobacteriaceae (ENT) population were determined by using a method described by Foo et al. (2003), Loh et al. (2009), andShazali et al. (2014). The faecal sample was homogeneously mixed at a ratio of 1 g sample to 9 mL of peptone water in a universal bottle and incubated for 1 h at room temperature. The samples were then subjected to a series of 10-fold dilution by using 0.1% (v/v) peptone water, and 0.1 mL of each appropriately diluted sample was then plated onto Lactobacillus agar DE Man, ROGOSA, and SHARPE (MRS-agar, Merck, KgaA, Darmstadt, Germany) and incubated under anaerobic condition at 30 °C for 48 h. However, ENT was spread plated and enumerated on Eosin-methylene-blue lactose sucrose agar plates (EMBagar, Merck, KgaA, Darmstadt, Germany) and aerobically incubated for 24 h at 37 °C. The colonyforming unit (cfu) per gram of sample was expressed as a logarithm at base of 10 (Log 10 cfu/g). The enumeration of LAB/ENT was conducted in triplicates. Intestinal morphology changes procedure was applied as described by Choe et al. (2012) and Tenesa et al. (2016). Segments of 5 cm long were removed from the small intestines. The intestinal segments were flushed with 10% neutral buffered formalin solution and then used for the morphometric analysis. Then, the segments were fixed in 10% neutral buffered formalin solution and kept overnight. Intestinal samples were then excised and dehydrated with a series of alcohol in the tissue-processing machine (Leica Microsystems K. K., Tokyo, Japan) and embedded in paraffin wax. Sections of 4 µm were then stained with haematoxylin and eosin, and later examined under light microscope. The distance from the villus tip to villus-crypt junction represented the villus height, while crypt depth was defined as the depth of invagination between adjacent villi. The villi height and crypt depth were measured by using an image analyser.
Volatile fatty acid (VFA) concentrations in the ileal-digesta sample were determined according to the method outlined in Loh et al. (2014). One gram of ileal-digesta sample was mixed with 1 mL of 24% methaphosphoric acid, diluted in 1.5 M sulphuric acid, and kept overnight at room temperature. The internal standard 20 mM 4-methyl-valeric acid was added to the supernatant to achieve 10 mM in combination and stored at −20 °C until gas liquid chromatography (GLC) analysis. Volatile fatty acid was separated by using a silica capillary column in gas-liquid chromatography occupied with a flame ionisation detector.
Blood from two broilers in each cage was collected in ethylenediaminetetraacetic acid (EDTA) tubes. Plasma was separated following a centrifuge at 3000 rpm for 15 min and stored at −20 °C until further analysis. The plasma IgG and IgM concentrations were determined by using chicken IgM and IgG ELISA Kits (Alpha Diagnostic International). Four hundred and fifty nanometre wavelength was used to measure wavelength by using a microplate reader (Model 3550-UV, Bio-Rad). The concentrations of plasma IgM and IgG were calculated by using standard curves.
All data were analysed based on a completely randomised design by using the General Linear Model procedure of the SAS (Statistical Analysis System, version 9.2) software. Duncan's multiple range tests were used to compare mean at P<0.05 of treatments. Data were presented as the mean ± standard error of the mean (SEM).

Results
All diets supplemented with essential amino acids had significantly quadratically increased (quadratic, P<0.05) body weight (BW), average daily gain (ADG), and total weight gain (TWG) as compared with the control diet (T1) ( Table 3). As for the FCR, control diet presented the highest value among all treatments. However, reduction of CP up to 2% showed a lower (P<0.05) FCR among all dietary treatments. The overall results showed by reducing CP level up to 2% had a better growth performance as compared with other treatments.
Faecal LAB and ENT population were not affected (P>0.05) by reduction of CP level in the broiler diet (Table 4). The highest (P<0.05) faecal LAB population was observed in chickens fed 0% reduction of CP with equal amino acids supplementation (T2) as compared with other treatments. As for faecal ENT populations, by reducing 2% of CP in the diets and supplemented with amino acids (T6) had lowest (P<0.05) faecal ENT populations than the other treatments. As for the LAB to ENT ratio, reduction of 2% CP shows the highest (P<0.05) ratio compared with other treatments.
The most important VFA for broiler chickens are acetic, propionic, and butyric acids. Total VFA was significantly quadratically increased (quadratic; P<0.05) with decreasing levels of CP as compared with  Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein... Shazali et al. 7 the negative control (Table 5). However, individual VFA was not affected (P>0.05) by the different levels of CP in the diets. Reduction of 0.5% CP (T3) led to higher (P<0.05) acetic acid concentration compared with other dietary treatments. Moreover, the diet with decreased 2.5% CP with equal amino acid supplementation (T7) had a significantly higher (P<0.05) propionic concentration as compared with other treatments. Furthermore, the highest (P<0.05) butyric acid concentration was observed in diet containing less than 2% CP as compared with other treatments. In terms of total VFA concentration, the reduction of 2.5% CP in the diet showed the highest (P<0.05) total VFA concentration as compared with all treatments.
The duodenal, jejunal, and ileal of chickens fed lower CP levels quadratically increased (quadratic; P<0.05) villi height with reduced CP levels in the diets (Table 6); meanwhile, the highest (P<0.05) duodenal and jejunal villi heights were observed in chickens fed diet containing less than 2% CP with equal amino acid supplementation as compared with the other treatments. In the case of ileal villi height, 0% reduction CP level with equal amino acid supplementation was significantly higher (P<0.05) as compared with other treatments. Concerning crypt depth, the duodenal crypt was linearly shallow (P<0.05) with reduction of CP in the diets. Howbeit, jejunal and ileal crypt depth was not affected by the different levels of CP. In terms of jejunal crypt depth, chickens fed the control diet had significantly deeper (P<0.05) jejunal crypt depth than chickens fed other diets. There was significant difference (P<0.05) in 0% reduction CP level with equal amino acid supplementation in the ileal crypt depth as compared with all treatment chickens. Furthermore, a significant quadratic increase (quadratic; P<0.05) was observed in the duodenal and jejunal villi height to crypt depth ratio with the reduction of CP in the diets. In contrast, ileal villi height to crypt depth ratio was linearly increased with the different levels of CP. The lowest villus height to crypt depth ratio was observed in chickens fed the control as compared with other treatments.
Plasma IgG showed a significant quadratic increase (quadratic; P<0.05) with the reduction of CP levels ( Figure 1). Blood plasma of chickens fed the T6 showed significantly higher (P<0.05) IgG concentration than other treatments. On the other hand, plasma IgM was not affected by the reduction of CP in the diets. Reduction of 2% CP in the dietary treatment had significantly higher (P<0.05) plasma IgM than other dietary treatments.   T1  T2  T3  T4  T5  T6 T7 T8 Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein... Shazali et al. 9

Discussion
In this experiment, reduction of CP up to 2% from the normal broiler CP requirement and supplementation with available amino acid improved body weight, average daily gain, feed intake, and feed conversion ratio in broiler chickens (Table 3). However, there were no significant differences in growth performance between the 0% CP level reduction with equal amino acid supplementation and the 2% reduced CP in the dietary treatment. These results agree with Han and Lee (2000), who reported that a reduction of 2% CP supplemented with 0.1% extra lysine did not affect broiler chicken performance. Better performance of animals fed the low dietary CP supplemented with amino acid could be due to the fulfilment of basic amino acid and CP requirement in the dietary treatments (Banerjee et al., 2013). In previous studies, many researchers reported a reduced CP level with crystalline amino acid supplementation in broiler diets without affecting the chicken performance.
Lysine is the first limiting amino acids in common broiler diets, and its requirements for broilers are higher in low-protein diets for maximum weight gain and feed efficiency (Labadan et al., 2001;Corzo et al., 2002;Saima et al., 2010). Threonine is considered as the third limiting essential amino acid in low-CP diets (Kidd et al., 1999;Baylan et al., 2006;Ayasan et al., 2009). To balance the broiler amino acid requirement, L-threonine was added to the low-protein diets (Ayasan et al., 2009). The studies included 22.7% CP , 17.6% CP (Corzo et al., 2005), 19% CP (Namroud et al., 2008), and 16% CP (Aletor et al., 2000), respectively, in the starter diets and 18.2% CP (Dari et al., 2005), and 16% CP (Gomide et al., 2012) in finisher diets. In this study, the growth performance of broilers in T6 (19% CP in starter and 16% CP in finisher diet) supplemented with essential amino acid had higher body weight gain, average daily gain, and feed intake and lower feed conversion ratio as compared with the 21% CP in starter and 18% CP in finisher diet supplemented with amino acids. Furthermore, Saima et al. (2010) reported that CP levels did not significantly affect the weight gain of broilers until four weeks of age in hot climate condition.
Nevertheless, the levels of lysine formulated with corn and soybean meal diet had a significant effect on the live weight gain and the feed conversion ratio of broiler (Oliveira et al., 2013). Dietary methionine and lysine deficiency were shown to impair chicken growth (Zhai et al., 2016a). According to Zhai et al. (2016b), the highest supplemental levels of methionine and lysine in broiler diets were not affected by the feed intake, but body weight was increased, and this resulted in a lower feed conversion ratio of broiler chickens. Ayasan and Okan (2010) also mentioned that threonine and lysine supplementation to broiler diets improved feed intake and body weight gain. Nevertheless, supplemented threonine and lysine did not affect their feed conversion ratio (Rezaeipour and Gazani, 2014). This outcome was different from that reported by Bregendahl et al. (2002), who found that decreasing the dietary CP resulted in a decreased average daily gain, feed efficiency, and growth rate of broiler chickens due to slow growth, lower feed efficiency, and less nitrogen retention than chickens fed the commercial diet (Bregendahl et al., 2002). In commercial feeds, diets are always formulated to minimum CP levels, and consequently, amino acids will be supplemented at a higher level than required (de Beer, 2011). However, increase in threonine to lysine ratio in the diets improved broiler FCR and increased feed intake and weight gain compared with the lower level of threonine, and that showed a significant difference in weight gain of Clostridium-infected chickens (Valizade et al., 2016).
Previously, Bowmaker and Gous (1991) mentioned that amino acid composition in the diet was crucially important for broiler chicken performance, suggesting that CP requirement could be reduced once the amino acid requirements were met (Lopez and Leeson, 1994). Moran Jr. (2007) mentioned that deficiency of amino acids may cause marginal deficiencies in the offspring. On the other hand, Wilson (1997) stated that higher levels of a particular nutrient in the diets, particularly amino acids, could be deposited and may cause health problem due to toxicity. Defective body weights and greater FCR were found in the chickens offered the dietary CP lower than 22%. In this study, 2% lower dietary CP supplemented with synthetic amino acids had similar growth compared with the 0% reduction CP level with equal amino acid supplementation. This result indicated that amino acid supplementation could be used to replace and meet the amino acid loss that originated from dietary CP .
Previous studies conducted by Waldroup et al. (2005) claimed that the supplementation of synthetic amino acids could partially improve the loss in body weight gain.
In this study, faecal LAB count in chickens fed 0% reduction CP level with equal amino acid supplementation diet was significantly different as compared with the other treatments (Table 4). Control (T1) showed a higher ENT count as compared with other treatments. However, reduction of 2% CP in the diets had a slightly lower ENT count than other amino acid supplementation in the low CP diets. This result was in agreement with Laudadio et al. (2012), who claimed that reduction of CP up to 18.5% decreased faecal pH values and decreased the number of pathogenic bacteria, such as aerobic mesophilic and Escherichia coli in broiler excreta. However, Nyachoti et al. (2006) claimed that the reduction of 23 to 17% CP did not affect the microbial counts in both beneficial and pathogenic bacteria in ileal digesta. Additionally, diet with low CP did not influence the coliforms and Lactobacilli counts in jejunum and colon of the chickens (Bikker et al., 2006). Generally, the increase in Lactobacilli counts may help suppress the coliform counts, and this could be attributed to the microbial colonisation resistance in the small intestine (Bikker et al., 2006). Thus, the increased Lactobacilli population may reduce the challenge of the Lactobacilli population to fight for space and nutrient availability in the intestinal tracts (Bikker et al., 2006). This finding is similar to our current results by showing that lactic acid-producing bacteria suppress pathogenic bacteria activities, such as Enterobacteriaceae. The result showed that reducing the dietary CP up to 2% and supplementing with amino acids increased LAB and reduced ENT population. This may improve and provide a favourable environment in the intestinal tract. The favourable environment will contribute to higher nutrient digestibility and absorption that enhance the chicken to grow subsequently, recording a better growth performance. Thus, it also contributes to reduce foodborne pathogens from animal origin foods (Rasekh et al., 2005).
Volatile fatty acids (VFA) consist of carbon chain molecules and some called short chain fatty acids.
Common compounds in VFA are acetic, propionic, and butyrate acids. These compounds are typically absorbed by the inner lining of the intestine and pass into the bloodstream before passing through the liver. Acetic acid is often used to build energy, as well as create lipids. The liver will not use much of the acetic acid, but propionic acid is generally removed from the blood by the organ. Bikker et al. (2006) reported that butyric acid acts as a microbial energy source and maintains the structure and function of human epithelial cells. Generally, VFA concentrations indicate the microbial activity along the small intestine and, usually, VFA concentration is higher in ileal digesta than in the jejunum and duodenum. Franklin et al. (2002) observed similar findings and reported that the increased VFA concentration in the cecum coincided with the increased number of microbial populations in the cecum. Bacteria often produce VFA in the intestines by digesting cellulose. Cherian et al. (2013) reported that the growth of the pathogen in the gut could be inhibited by lowering the gut pH . Pluske et al. (2003) claimed that by lowering the dietary CP content, a reduction in fermentation and formation of the microbial metabolite in the ileum and cecum could be expected. The results of this study showed that by lowering the dietary CP there is a reduction in nitrogen and fermentable energy supply to the intestinal microbes. Another study showed that lowering the dietary CP reduces the ammonia concentrations (Bikker et al., 2006;Nyachoti et al., 2006) and individual VFA (Nyachoti et al., 2006) in ileal digesta of pigs weaned at 18 days old. In this study, the dietary CP content did not affect the total concentrations of VFA in ileal digesta but showed a higher (P<0.05) concentration in the dietary CP supplemented with amino acids as compared with the dietary CP without supplementation of amino acids (Table 5).
Reducing dietary CP with amino acid supplementation resulted in significant linear reductions in VFA in ileal digesta (Franklin et al., 2002). However, in this study, reducing CP in the diet and supplementing it with amino acids contributed to fluctuated results in the total VFA concentration. In addition, VFA concentration was not consistent among the dietary CP studies. Htoo et al. (2007) speculated that the absence of an effect of dietary CP content on the concentration of VFA in the ileum might be affected by the rapid passage rate of digesta through the small intestine, lack of bacterial fermentation occurring in the ileum part, and nutritive value and nutrient digestibility of certain selected feed ingredients. Furthermore, Bikker et al. (2006) reported that by reducing the 21.1% dietary CP to 15.3% does not influence the individual and total VFA in the ileal digesta of pigs weaned at 26 days old.
Gut microflora and intestinal morphology changes of broiler chickens fed reducing dietary protein... Shazali et al. 11 An indirect relation between lactic acid bacteria and butyrate was suggested by Tsukahara et al. (2002), whereas the higher counts of lactic acid bacteria produced more lactic acid and were converted into butyrate in the lumen. Hence, the current result showed that by reducing the dietary CP supplemented with amino acids had no significant difference in total VFA concentrations as compared with dietary CP without amino acid supplementation. This may create a favourable environment for beneficial bacteria and better nutrient digestibility of broiler chickens.
Small intestine consists of the duodenum, jejunum, and ileum, which they help to absorb nutrients from feed. The longer the length of villi, the higher is the value of nutrient absorption. Lower villus height results in decreased intestinal absorption area. Gut morphology plays an important role in dietary nutrient absorption efficiency and can be determined by measuring the villi height and crypt depth of the intestine (Swatson et al., 2002). The height of villi and depth of crypt are mostly depended on the diet provided ). In the present study, reducing dietary CP up to 2% plus supplementation with amino acids led to better duodenum and jejunum villi height as compared with the other treatments (Table 6). Thus, this result was in agreement with Sritiawthai et al. (2013), who claimed that ducks that received 18% CP also had higher villus height in duodenum and jejunum as compared with higher CP levels. However, this result contradicted Abbasi et al. (2014), who found that a reduction in dietary CP led to significantly lowered jejunal villi height and shallowed crypt depth in broiler chickens. Furthermore, Abbasi et al. (2014) reported that decrease in CP diet had a similar effect on the villi height to crypt depth ratio, whereby this result also contradicted our findings. Gu and Li (2004) reported that increasing dietary CP caused a higher number of goblet cell in the distal jejunum of a piglet that was associated with the development of villus height, crypt depth, and epithelium cell size. It is believed that high dietary CP rewards the small intestinal mucosa with higher digestive and absorptive capacity compared with low dietary CP in piglet (Gu and Li, 2004). Wu (1998) also pointed out that small intestine may use about 30 to 50% of essential amino acids. In addition, some of the amino acids were not available for extra intestinal tissue (Wu, 1998). Schaart et al. (2005) found that dietary threonine was utilised by small intestine mucosa for protein synthesis, which was related to a decrease of threonine levels in faecal samples.
Among the essential amino acids, threonine is important for maintaining the gut barrier integrity and has an important role in the structure and function of gastrointestinal tract (Valizade et al., 2016). Threonine participates in protein synthesis, generates metabolic products such as glycine and serine, and serves as a component of body protein for broiler chickens (Ayasan et al., 2009). Production of glycine and serine derived from threonine catabolism is needed in gastrointestinal mucin production (Lemme, 2003). Law et al. (2000) and Ball (2001) reported that piglets that received deficient diets in threonine had lower intestinal weight and developed less intestinal structure. In the current study, the results are in agreement with Law et al. (2000) and Ball (2001), whereby different intestinal structures were found in different dietary CP supplemented with essential amino acids. Wu et al. (1996) mentioned that an increase in glutamic supplementation prevented jejunal atrophy as indicated by villus height. At the same time, it also increased the gain:feed ratio. However, in our study, amino acid supplementation in the low dietary CP showed significantly different villi height and crypt depth of duodenum, jejunum, and ileum as compared with dietary CP without amino acid supplementation. However, by reducing dietary CP up to 2% and supplementing with amino acids led to higher duodenal and jejunal villi height and crypt depth as compared with other dietary CP supplemented with amino acids.
The interaction between nutrition and immunity showed an implication on animal growth and productivity (Golian et al., 2010). In the current study, the plasma IgM and IgG were higher (P<0.05) for chickens fed low CP diets supplemented with amino acids (Figure 1). L-threonine was added to broiler diets to balance the unique nutritional requirements of the animals and served as a component to body protein that is involved in immune responses of the animals (Lemme, 2003;Ayasan et al., 2009). Golian et al. (2010) reported that low dietary CP was not significantly different in anti-SRBC, IgG, and IgM antibody titers as compared with high-protein diet. Takahashi et al. (1995) reported that plasma acute-phase of α1-acid glycoprotein concentration and interleukin-1-like activity after Escherichia coli lipopolysaccharide injection in chicks fed the low-protein diet was greater than those of fed the high-protein diet.

Conclusions
Broilers fed a low crude protein diet supplemented with amino acids have improved growth performance, reduced Enterobacteriaceae, increased lactic acid bacteria counts, and increased small intestine villi height and crypt depth, as well as ileal digesta volatile fatty acid concentrations. However, reducing dietary crude protein up to 2% shows the best results, especially in growth performance, feed conversion ratio, Enterobacteriaceae count, duodenal and jejunal villi heights, ileal digesta volatile fatty acid concentrations, and immunity status. It is believed that by reducing the level of crude protein in the broiler diet while supplementing it with commercial amino acids may enhance the nutrient digestibility and absorption in broiler chickens, and simultaneously will give better performance.