The effect of diet composition, a probiotic and a symbiotic treatment on the ileal microbiota composition of one-week-old broiler chickens

A healthy microbiota present in the small intestine contributes significantly to small intestinal function, including digestion, nutrient absorption and health. The current study investigated the effects of a prebiotic, a probiotic and a symbiotic supplementation on ileal microbiota composition of broilers at 7 days of age. In a total of 574 male Ross 308 day-old chickens were divided into four groups using six replicate pens and 24 chickens per pen. A maise-soybean based control diet (C), a control diet supplemented with probiotics (Broilact; Br), a control diet supplemented with symbiotic (inulin, yeast, Bacillus subtilis; Sy) and a wheat based diet supplemented with wheat bran (W) were formulated. On day 7 of life, two chickens per pen were slaughtered and ileal chymus samples were collected. For microbiota analysis 16S rRNA (V3-V4 region) gene targeted Illumina MiSeq sequencing was used. Feeding all diets supplemented increased the diversity to varying degrees compared to the control (C) diet (p=0.006). As a conclusion, all supplementation substantially influenced ileal microbiota of broiler chickens at an early age. All these results could offer some information for the future study on the relationship between early intestinal microbiota and the compounds of the feed.


INTRODUCTION
The intestinal flora of broilers plays an important role for growth performance and health. However, knowledge about the composition of the gut microflora and microbial ecology of the gastrointestinal tract is restricted (Bjerrum et al., 2006). New molecular studies using 16S ribosomal DNA for phylogenetic analysis have resulted more detailed insight into the composition of this ecosystem (Bjerrum et al., 2006;Amit-Romach et al., 2004). Several studies demonstrated that different intestinal regions have different communities of bacteria and the diversity of the microbiota increases with the age of chickens (Gong, 2008). A healthy microbiota what present in the small intestine contributes significantly to small intestinal function, including digestion and nutrient absorption, which is the most limiting factors of growth rate (Moran, 1982). However, in commercial poultry production, the development of intestinal microbiota in the chickens is often impeded by modern practices. This includes facility hygiene, routine medication, and artificial egg incubation, hatching and chicken rearing. Because of this chickens are susceptible enough to colonization by bacterial pathogens (Barrow, 1992). The use of antimicrobial growth promoters is under debate in many regions mondial. Several countries, for example in the EU have already banned these compounds in animal diets. The use of therapeutic is still in the broiler industry worldwide and although they are fundamental to maintain animal welfare and to treat chickens from diseases, these compounds are also often used in a prophylactic manner (Van Immerseel, 2017). In order to improve the immunity and feed utilisation efficiency of livestock and poultry, research into environmentally friendly feed additives has become one of the main challenges of animal production in the 21 st century (Wang, 2017). It is well known that the composition and structure of the feed as well as the housing conditions influence the composition of the intestinal microbial community of chickens (Apajalathi et al., 2004;Engberg et al., 2004). Different studies trying to find alternatives to growth-promoting antibiotics suggest that the antibiotics can be replaced by probiotics, prebiotics, organic acids, or herbs. Of these animal and poultry feed additives, probiotics and prebiotics have drawn high attention (Zarei et al., 2017). It is generally accepted that shifts in the intestinal microbiota composition may be the result of dietary changes, such as the addition of cereal fibers which exert prebiotic-like effects (Torok, 2011). Nonstarch polysaccharides (NSPs) of wheat are present in the bran and endosperm of the seed and are composed of arabinoxylans (AX), β-glucans, cellulose and arabinogalactan-peptides (Parsaie et al., 2007). Soluble arabinoxylans increase viscosity in the digestive tract and impairs the digestibility of nutrients and increase the water content of excreta. Supplementation of broiler diets with exogenous enzymes, like xylanase decrease the negative effects of soluble AX. The degradation of AX also results in xylan oligosaccharides (XOS) that can exert prebiotic effect. Thereby feeding NSP-ases can influences the microbiome of both the ileum and caecum (Bedford, 2018). The most commonly used probiotics in the poultry industry are Lactobacilli, Bifidobacteria, Bacillus and Saccharomycetes. The host can benefit from each ones based on different mechanisms (Liu et al., 2012). Accordingly, the use of multistrain probiotics is better than using monostrain probiotics, because those different strains of the genera show symbiosis, additive relationships towards each other which enhances growth and metabolic activity of host animal (Kazemi et al., 2019). Nurmi and Rantala (1973) demonstrated that oral administration of caecal bacterial flora from adult birds into newly hatched chickens increased the chickens resistance to salmonella infection. Their results were later confirmed in different laboratories. Great interest was paid to the phenomenon, now often referred to as competitive exclusion or the Nurmi concept (Pivnick and Nurmi, 1982). Synbiotics consist of a combination of synergistically interacting probiotics and prebiotics. These can be aimed at improving the resistance and stability of health-promoting organisms in the gut of birds by providing a substrate for fermentation (Sobolevska et al., 2017).
To grant the greatest protection of the immune system for newly hatched chickens, external supplementation with bioactive substances should occur as early as possible. Therefore, the aim of this study was to investigate the effects feeding maize or wheat based dies and the supplementation of maize based diets with a multistrain probiotic and a symbiotic on the composition of the ileal microbial community of broiler chickens at 7 days of age using next generation sequencing technology.

Study designed
A total of 576 ROSS 308 one-day-old male broiler chickens were randomly devided into four groups of 24 chickens each in 4 replicate floor pens with chopped wheat straw bedding. Feed and water were available ad libitum. In the experiment, four dietary treatments were used as followes. A maize soybean based diet was used as a control (C). Beside the control diet, three different diets were fed: control diet with Broilact supplementation (Br), control diet supplemented with symbiotic feed additive (Sy) and a wheat-based diet containing 3% of wheat bran (W). The product Broilact® (Broilact, Europharmavet Kft. 1077 Budapest, Rózsa str. 10-12. Hungary) is an undefined but limited mixture of normal intestinal bacteria derived from a healthy adult hen and screened to ensure the absence of specific pathogens (Nurmi et al., 1987). This product was given the first two days to chickens via crop inoculation, two times at 1.25x10 7 CFU/0.5 ml dosage. The symbiotic treatment was contained three product: Bacillus subtilis, DSM17299 bacterial strain (0.4 g kg −1 , 1.6x10 6 CFU/g; Gallipro, Biochem GmbH, Küstermeyerstrasse 16. 49393 Lohne, Germany), inulin (5 g kg −1 , Oratfi HSI, Beneo GmbH, Aandorenstraat 1, B. 3300 Tienen Belgium) and yeast (Saccharomyces cerevisiae boulardii, 1x109 CFU/kg Levucell SB 20, Lallemand GmbH. Ottakringer Straße 89, A-1160 Wien, Austria). Diets were formulated to be isoenergetic and isonitrogenous, fitting to the requirements of Ross 308 broiler chickens (Aviagen, 2019). Each diet contained the NSP-degrading enzyme supplementation, Econase XT (AB Vista Ltd., Marlborough, Wiltshire, SN8 4AN). The composition and nutrient content of the experimental diets are shown in Table 1-2. Computer controlled housing and climatic conditions were maintained according to the breeder suggestions (Aviagen, 2019). On day 7 of life, 2 chickens per pen were slaughtered and digesta samples were taken from ileum (about 3 cm proximal to Meckel's diverticulum). All samples were immediately snap-freezed and stored at -80 o C until further processing.

16s rRNA analysis
Bacteria were identified by the analysis of the V3-V4 region of the 16S rRNA gene using Illumina Miseq platform. Sequences were analyzed using Quantitative Insights Into Microbial Ecology 2 (QIIME2) version 2020.2 software package (Bolyen et al., 2019). Sequences were filtered based on quality scores and the presence of ambiguous base calls using the qualityfilter q-score options. Representative sequences were found using a 16S reference as positive filter as implemented in the deblur denoise-16S method. Finally, sequences were classified by taxon in Operational Taxonomic Units (OTUs) using a fitted classifier base on SILVA 132 database (Quast et al., 2013).
Alpha diversity (e.g., Observed, Shannon and Simpson indices) for individual samples was estimated using MicrobiomeAnalyst (Chong et al., 2020). Samples analyzed with MicrobiomAnalyst were filtered for low abundance based on prevalence of OTUs, and for low variability using the inter-quantile range assessment. Alpha diversity analysis is used to study diversity within a specific environment or within a single sample. Bacterial alpha diversity in the ileal microbiota of broiler chickens was estimated by calculating the Identified OTU numbers, Chao's, Shannon's, and Simpson's indices of richness and diversity. The Chao1 was used to estimate species richness; Shannon's and Simpson's index were used to indicate species diversity. The Shannon index is more sensitive to rare species and the Simpson index to dominant species.
Beta diversity analysis is used to compare the differences of samples groups in terms of species diversity. It revealed that the microbial composition of the ileum in the control and treated group was different.

Statistical analysis
The significance of differences in alpha diversity among sample groups was tested using analysis of variance (ANOVA) with a Benjamini-Hochberg post hoc test and correction of p-values for multiple comparisons. Both platforms were also used to generate beta diversity indices and to visualize community (dis)similarities using Jensen-Shannon divergence with tests of significance using permuted analysis of variance (PERMANOVA). The mean relative abundances of microbial communities, richness, and evenness were analyzed by ANOVA and Benjamin-Hochberg FDR correction (q) using q<0.05 value.

RESULTS AND DISCUSSION
A total 478.619 quality controlled sequences were generated from each of the 24 samples. The read number were with an average of 20.608 reads per control (C), 22.152 reads per Br, of 20.555 reads per Sy and 16.452 reads per W treatment samples. The sequences were clustered into 1.558 operational taxonomic units (OTU) using 97% similarity. Results were assigned into 5 taxonomic levels (phylum, class, order, family and genus). A total of 9 phyla, 15 classes, 30 orders, 47 families, and 87 genera were found in the examined sample groups.

Alpha-diversity
The Shannon's and Simpson's indices apparently indicated that the three treatment contained more diverse bacterial communities compared to control group. The Chao's and Observes indices were higher at tendency level in the Br group than in the C group (Table 3).

Beta diversity
The Principal coordinate analysis (PCoA) plots analyzed using Permanova for Jensen-Shannon divergence ß-diversity demonstrated significance difference (F-value: 3.4206, R-squared: 0.3391, p=0.001) in ileal samples among the four treatment. The diversity in the chickens of control group was lower than that in the the three other treatments (Figure 1).
Several previous studies examined the composition of ileal microflora of broiler chickens. Xiao et al. (2017) reported that the microbiota of ileum showed the lowest diversity among the different gastrointestinal segments. This finding was in accordance with the other studies showing that the microbiota in caecum were more diverse than in ileum (Bjerrum et al., 2006;Dumoneeaux et al., 2006;Mohd Shaufi et al., 2015). Rodrigues (2020) found that the alpha-diversity measurements decreased in the ileum with increasing age. The microbiome was less diverse on the tenth day compared to the third day of life. To their knowledge, the drop in ileal diversity was potentially accounted for by the relative overabundance of few dominating species as as the age of the birds increased. In face of declining biodiversity, the host-microbiota may lose redundant species (Foster et al., 2017). This means that members of the community which are phylogenetically unrelated, but have similar functional niches can be substituted for one another with little impact on the ecosystem (Lozupone et al., 2012;Tian et al., 2017).
In our study throughout the four dietary groups, 9 bacterial phyla were identified with Firmicutes, Proteobacteria and Cyanobacteria being the most abundant ones. The listed three phyla represented more than 98% of the examined bacterial population. Bacterial composition at the phylum level did not differed (q > 0.05) among dietary treatments (Figure 2). Relative abundance of the Planctomycetes (p=0.02; q=0.15) and Bacteroidetes (p=0.03; q=0.15) phylum tended to be higher in the Sy and W group compared to the C group.
The phylum Planctomycetes is a group of budding, peptidoglycan-less bacteria of increasing significance for microbial evolution, ecology, cell biology and genomics. This phylum which represents a deepbranching group within the Bacteria on the basis of 16S rRNA sequence phylogenetics (Fuerst, 2004). These bacteria have been identified in diverse freshwater, marine and soil habitats and even invertebrate animals (Fuerst, 1995;Wang et al., 2002). The experimental data, however, remain scarce due to the low number of characterized representatives of this phylum, but some Planctomycetes may be involved in degrading polymeric organic matter. (Ivanova et al., 2017). The rate of the Planctomycetes phylum was higher in the Sy and W groups than the control group. These groups received prebiotic components to the feed. This could correlate to the polymeric degrading attitude of Plactomycetes.

Figure 2. Relative proportions (%) of the most abundant phyla in the ileum of broiler chickens (day 7 of life) fed different diets
Abbreviations: Maize based diet; Control group supplemented with Broilact (Europharmavet Kft. 1077 Budapest, Rózsa str. 10-12. Hungary); Control group supplemented with Bacillus subtilis, inulin and yeast; Wheat-based diet supplemented with 3% wheat bran; At the class level, the Alphaproteobacteria was higher (q=0.007) in W compared the C group and at the order level. Relative abundance of the Rickettsiales was significant higher (q=0.008) in W compared than that in the C group.
At the family level the Lactobacillaceae was the dominant in all treatment, but the relative abundance tended to decrease in the chickens fed the by the Br diet. (p=0.239 q=0.492). The Mitochondria (q=0.01) was significant higher in the W group, and the Bacillaceae was higher in the Sy group than in the control group (p=0.001; q=0.04). Enterobacteriaceae had a trend to be lower (p=0.05; q=0.48) in control comparising to the Br group (Table 4). The Lactobacillaceae is a highly prevalent member of ileal microbiota during the poultry lifetime, producing short-chain fatty acids and/or lactic acid (Rivière et al., 2016). This contribute to the inhibition of many acid-sensitive bacteria, such as Enterobacteriaceae, by lowering the pH of the intestinal contents (Cisek and Binek, 2014). During the growth of the chicken takes about 2 week for Lactobacilli to become the predominant bacteria (Barnes et al., 1973). The predominance and endurance of most species of Enterobacteriaceae within the intestine is a standard marker of dysbiosis (Rivera-Chávez et al., 2017), even though it is well-known that this family is a pioneer colonizer in the gut (Donaldson et al., 2017). In addition, members of Enterobacteriaceae may also delay or block the growth of beneficial microbiota due to their utilization of important early ecological niches (Pedroso et al., 2016). In agreement of this study, Juricova et al. (2013) demonstrated that chicken infected with Salmonella Enteritidis at one and 4 days of age delayed the microbiota development. The reason is probably that the number of Enterobacteriaceae population increased and at the same time the Clostridiales and Lactobacillales relative abundances decreased. In previous studies the Broilact were able to inhibit Campylobacter spp. binding to chicken intestinal mucus in the exclusion assay (Ganan et al., 2012), and became established in the gut of the newly hatched chickens while the colonization of Salmonella was inhibited (Schneitz et al., 2016). In our experiment the members of Enterobacteriaceae may also delay the growth of beneficial microbiota in the Br group, because here was the highest rate of this family and the lowes rate of the family Lactobacillaceae.
The family Mitochondria is the member of the Rickettsiales order and the Alphaproteobacteria class. Rickettsiae are obligate intracellular endosymbionts. These are parasites of eukaryotic cells. (Emeyanov, 2001). The Mitochondria are important organelles in eukaroytic cells, not a bacterial family. The alphaproteobacterial origin of mitochondria is generally indisputable, but attempts to resolve the phylogenetic position of mitochondria in the alphaproteobacterial species tree have failed to reach consensus. Most studies support the idea that mitochondria evolved from an ancestor related to Rickettsiales (Fitzpatrick et al., 2006;Ferla et al., 2013). Alphaproteobacterial and mitochondrial sequences display a high degree of compositional heterogeneity, they are sensitive to the compositional bias artefact in which unrelated lineages with similar sequence compositions falsely group together (Martjin et al., 2018). The sequences of Mitochondria were analysed by the SILVA Sequence Database. According to the results, the sequences are same to plant sequences and the assigned Rickettsiales possibly represent mitochondria of wheat not bacterial sequences. This may skew the results.

Merdibacter
(p=0.04; q=0.5), Ruminococcaceae_UCG_013 (p=0.04; q=0.5), Romboutia (p=0.04; q=0.5), and Rothia (p=0.04; q=0.5) genera were present with a relatively higher frequency in the Br than in the C group. The abundance of Tyzzerella was (p=0.02; q=0.5) higher in the C than in the other three group. The Bacillus genus (p=0.001; q=0.1) was higher in the Sy than the control group.
In a study, Farkas et al. (2020) identified the composition of the Broilact product also with 16s rRNA analysis. According to that results, the dominant phyla were the Firmicutes, Proteobacteria and the Bacteroidetes (43.5%, 43.7% 12.6%, respectively). Though most genera showing changes belonged to the Clostridia phyla.
Some previous study reported that the B. subtilis modulates the intestinal microflora (La Ragione and Woodward, 2003) and selectively supports the growth of lactic acid producing bacteria (Knarreborg et al., 2008). The symbiotic treatment in this study contained Bacillus subtilis, DSM17299 bacterial strain. This can be explanation that the Bacillus genus abundance was relative higher in this treatment group than in the other three groups.

CONCLUSIONS
This study examined the impact of early intestinal colonisation by a prebiotic, a probiotic and symbiotic treatment on the development of ileal microbiota. An increase in the rate of Bacillus genus in the small intestine has been observed with Symbiotic treatment. Broilact tended to decrease lactobacilli and increased the proportion of the Enterobacteriaceae family. Enterobacteriaceae strains, as pioneer colonizers, delayed the microbial consortium establishment in the ileum of young broilers. The treatment in W and Sy groups resulted higher rate of the Planctomycetes phylum. Finally, an essential contribution of this study is that all of the three treatments increased the diversity compared to the control treatment, but the Broilact was resulted the highest difference. In future researches, it would be worth to study whether how this significant diversity changes with the age of broilers. All these results could offer some information for the future study on the relationship between early intestinal microbiota and the compounds of the feed.