Effects of dietary fiber sources on bacterial diversity in separate segments of the gastrointestinal tract of native and exotic pig breeds raised in Vietnam

Abstract Background and Aim: Dietary fiber has distinctive effects on the environment and microbiota of the pig’s intestinal tract. This study was conducted at the naturally ventilated facility of the experimental station, National Institute of Animal Sciences, Vietnam, to examine the effects of fiber sources in diets on the intestinal microbiota of two different pig breeds raised in Vietnam. Materials and Methods: A total of 18 native and 18 exotic pigs with average initial body weights of 9.5±0.4 and 16.5±0.4 kg, respectively, were each divided into three dietary treatments, including a low-fiber diet containing approximately 200 g NDF per kg dry matter (DM) and two high-fiber diets containing cassava by-products or brewer’s grains containing approximately 300 g NDF per kg DM. At the end of the experiment (28 days), the bacterial diversity of digesta samples collected from the stomach, ileum, and colon segments was analyzed through DGGE analysis of the V3 variable regions of 16S-rDNA and by cloning and sequencing. Results: Among the diets, significant differences were observed in the number of DNA bands in the stomach between the native and exotic pigs (p<0.05), but not in the ileum and colon. The dietary fiber affected the number of DNA bands in the ileum (p<0.05), but not in the stomach and colon. A significant interaction effect was found between diet and breed on the number of DNA bands in the ileum (p<0.05). Dietary fiber and breed had a greater effect on microbiota in the ileum and colon than that in the stomach. Conclusion: The fiber sources affected the number of DNA bands in the ileum, and breed affected the number of DNA bands in the stomach. The microbial compositions in the ileum and colon segments were significantly affected by the dietary fiber and breed.


Introduction
Fiber is a structural component of plants and constitutes an essential element in the diet of animals [1]. Dietary fiber consists of non-starch polysaccharides (NSP) and lignin that cannot be digested in the small intestine by endogenous digestive enzymes. The content of NSP and the soluble/insoluble fractions of NSP in the diets change the bacterial diversity in the large intestine and may be capable of preventing intestinal disorders in pigs [2]. Moreover, fructooligosaccharides are indigestible feed components that affect the gut microbial composition of animals [3] and tend to reduce the amount of Escherichia coli and increase the amount of Bifidobacterium in the distal colon of pigs [4].
Different fiber sources are known to affect the gut microbial composition of suckling piglets [5], rabbits [6], and pregnant sows [7]. In addition, the microbiota compositions of the gut and stomach are significantly affected by the breed and age of animals [8][9][10]. It has been reported that the breed has a significant effect on the rumen microbial properties [11].
Therefore, we hypothesized that the types of dietary fiber and breeds have important effects on microbiota in the gastrointestinal tract. This study aimed to explore the effects of dietary fiber sources on the bacterial diversity in different segments of the gastrointestinal tract of native and exotic post-weaning pigs.

Ethical approval
The experiment was approved by the Animal Science Committee of National Institute of Animal Sciences (NIAS), the Ministry of Agricultural and Rural Development, Vietnam. Available at www.veterinaryworld.org/Vol.14/October-2021/1.pdf

Experimental design
The study was carried out at the naturally ventilated facility of the experimental station, NIAS, located in Bac Tu Liem, Hanoi 29909, Vietnam (21°0'18" N latitude and 105°55'57" E longitude). A total of 36 animals, including 18 native Mong Cai and 18 exotic (Landrace × Yorkshire) pigs, aged approximately 60 days, were used in this experiment. The initial body weights (IBWs) of the native and exotic pigs were 9.5±0.4 and 16.5±0.4 kg, respectively. A completely randomized 2×3 factorial design was used with the two breeds (native and exotic) and three diets (a low-fiber diet [LFD] and two high-fiber diets consisting of either cassava by-products [HFC] or brewer's grain [HFB]). For each breed, the animals were evenly divided into three dietary treatments according to the IBW and sex by treatment. Each treatment had six replicate pens with one pig per pen. Animals were housed individually in concrete-floored pens (1.8 m×0.8 m), fully covered with wooden slats. All pigs were vaccinated against classical swine fever, pasteurellosis, pneumonia, and paratyphoid before starting the experiment.
The experimental diets were formulated on the basis of the primary raw materials, including maize, soybean meal, rice bran, fish meal, and cassava by-products (CB) or brewer's grain (BG) ( Table-1). The LFD diet was incorporated without CB and BG and contained approximately 200 g NDF/kg dry matter (DM), whereas CB and BG were incorporated into the HFC and HFB diets to obtain approximately 300 g NDF/kg DM (Table-2) [12]. The diets were incorporated to meet or exceed the nutrient requirements for post-weaning pigs [13]. The duration of the experiment was 28 days.

Collection of digesta samples
All experimental pigs were slaughtered 4 h after morning feeding and thiopental injection. Individual digesta samples from the stomach, ileum, and colon sections were collected immediately into sterile tubes, placed on ice in an insulation box, and transferred to the laboratory. The digesta samples were frozen and stored at −20°C until further DNA extraction.
Determination of general microbial diversity DNA isolation DNA isolation of the stomach, ileum, and colon samples was conducted using the AccuPrep ® Stool DNA Extraction Kit (Bioneer, Korea) according to the manufacturer's instructions. Total RNA was treated with the enzyme ribonuclease. The concentration and purity of DNA were checked using the NanoDrop ® ND-1000 spectrophotometer (Wilmington, USA). After isolation, the DNA samples were stored at −20°C until polymerase chain reaction (PCR) amplification.

PCR amplification
PCR amplification of the V3 hypervariable region of the bacterial 16S rDNA gene was conducted using the forward primer 341F with GC clamp at the 5ʹ end (5ʹ-CGC CCGCCGCGCGCGGCGGGCGGGGCGGGG GCACGGGGGGCCTACGGGA GGCAGCAG-3ʹ) and the reverse primer 518R (5ʹ-ATTACCGCG-GCTGCTGG-3ʹ). The primers 341F and 518R were designed on the basis of the position of the nucleotides 341-518 of 16S rDNA in the E. coli gene [14,15]. Approximately 35 ng of DNA template was used for PCR. DNA was amplified as described by Hong et al. [16]. PCR products were tested by 2% agarose gel electrophoresis. Gels were stained with 0.5 µg/mL ethidium bromide, checked under ultraviolet (UV) light, and photographed under the Gel DOC TM 2000 Gel Documentation System (Bio-Rad, USA).  Available at www.veterinaryworld.org/Vol.14/October-2021/1.pdf

DGGE
The amplified PCR products were separated using the Bio-Rad Dcode TM Universal Mutation Detection System (Bio-Rad, USA), with the size of the gel being 16 cm× 20 cm. The gels consisted of 10% polyacrylamide and 30-60% linear gradients of the denaturant. The concentration of ammonium persulfate solution was 0.1% (w/v), and the concentration of tetramethylethylenediamine was 0.1% (v/v). Furthermore, 2% (v/v) glycerol was used to increase the durability of gels. A total of 20 µL of the PCR product was mixed with the loading dye and added to a well. Electrophoresis was conducted at 60°C for 17 h at 85 V, and the DGGE gels were then stained with 0.5 µg/mL ethidium bromide for 30 min, checked under UV light, and photographed under the Gel DOC TM 2000 Gel Documentation System (Bio-Rad, USA). Selected DGGE bands were cut out from a gel using a sterile scalpel, placed into tubes containing 30 mL distilled water, and stored overnight at 4°C. Next, pure DNA from these tubes was collected, transferred to new tubes, and stored at −20°C for further analysis.
Gel analysis DGGE images were analyzed using Quantity One 4.5.0 (Bio-Rad, USA). Ladder 1 kb marker was used as the reference system. The number of bands in each DGGE profile was determined as an indicator of richness, and comparisons between the dietary treatments were conducted. The Dice coefficient was used for comparing the fingerprint profiles of DGGE images using NTSYSpc version 2.1 (NTSYS-PC 2.10, Applied Biostatistics, Setauket, NY, USA).

Amplification of DNA bands from DGGE gels
A total of 1 µL of the eluate were used as a template in a PCR with the primers 341F without GC clamp (5ʹ-CCTACGGGAGGCAGCAG-3ʹ) and 518R. The PCR was conducted in a Bio-Rad thermos cycler (MJ Mini Personal Thermal Cycler, USA) using the conditions as described by Hong et al. [16]. The PCR products were stored at 4°C for cloning.

Cloning
The amplified molecules were cloned into the vector pTZ57R/T using the InsTAclone TM PCR cloning Kit (Fermentas, Canada). Thereafter, the plasmids were transformed into E. coli DH5α. X-Gal and IPTG were used for blue/white colony selection on LB plates containing 50 µg/mL ampicillin.

Analyzing positive colonies
White colonies for each cloned band were randomly selected from the transformation plates. Cells from each colony were transferred using a sterile toothpick into a PCR tube. The PCR was conducted with the specific primers M13F (5ʹ-GTAAAACGACGGCCAG-3ʹ) and M13R (5ʹ-CAGGAAACAGCTATGAC-3ʹ). The program for amplification was exactly similar to the amplification with the primers 341F/518R without GC clamp as described earlier, but the final extension at 72°C was conducted for 7 min. The PCR products were checked by electrophoresis in 2% agarose gel and purified using the AccuPrep PCR Purification Kit (Bioneer, Korea).

Sequencing
Sequencing was conducted by Molecular Cloning Laboratories using the purified PCR product as the template and the primer M13F or M13R as described by Hong et al. [16].

DNA sequence analysis
Results obtained from the sequencing of DNA samples were compared with sequences available in the NCBI using the BLASTn tool [17].

Statistical analysis
Experimental data were analyzed using the MIXED procedure of the SAS software (Version 9.0, Institute Inc., Cary, NC, USA). An individual pig was considered as an experimental unit. Repeated measures conducted on the same experimental unit as a similar model were used but included the effect of a compound symmetry structure of covariance. Group means that showed statistical differences at the probability level of p<0.05 were compared using Tukey's multiple comparison procedures. In addition, the Dice similarity coefficient was calculated from the DGGE profiles of bacterial 16S rDNA PCR amplification products. The level of similarity between each pair of groups was named in their corresponding row and column. The Dice coefficient did not relate to the p-value, which is associated with statistically significant differences. Instead, the DGGE profile of each group comprised DNA bands that appeared on bands belonging to that group.

Bacterial diversity at separate segments in the gastrointestinal tract
The PCR products obtained from genomic bacterial DNA isolated from the segments of the stomach, ileum, and colon as the template were clear and without by-products. The final PCR product length was approximately 200 bp, as determined by 2% agarose gel electrophoresis. DGGE profiles were produced from samples obtained from the stomach, ileum, and colon of pigs fed with different experimental diets. One remarkable feature of the pig microbiota was relatively different among the individuals. As shown in Table-3, there was a significant difference in the number of DNA bands in the stomach between native and exotic pigs (p<0.05), with a higher value for exotic pigs, but not in the ileum and colon. The number of DNA bands in the ileum was significantly different among diets (p<0.05), with the highest value with the HFB diet, medium value with the HFC diet, and the lowest value with the LFD diet. An interaction effect was observed between diet and breed on the number of DNA bands in the ileum segment (p<0.05), such that diet had an effect on the number of DNA bands in exotic pigs, but not in native pigs. The number of DNA bands in the stomach samples was twice that in the colon and ileum samples within breed.

Gene diversity at different segments in the gastrointestinal tract
As shown in Table-4, the intestinal microbial diversity in native and exotic pigs fed with different diets was extremely different. However, the coefficient of gene diversity within breed was higher in the stomach site and lower in the ileum and colon sites. Fiber source and level had a greater impact on microbiota in the ileum and colon samples than that in the stomach samples. Within breed, the coefficient of gene diversity in the gastrointestinal tract was higher in each breed than in combination with two breeds.

Identification of cloned 16S rDNA sequence in DGGE profiles
To obtain more detailed information about microbiota at different segments in the gastrointestinal tract of the two pig breeds fed with diets with different fiber sources and levels, the DNA bands were cut from DGGE gels, cloned, and sequenced. A total of 120 predominant DNA bands from six gels (20 dominant DNA bands from each gel) were examined and reamplified with the primers 341F (without GC clamp) and 518R. Then, the PCR products were cloned into the vector pTZ57R/T and transformed into E. coli DH5α. The result showed that all DNA bands cut from DGGE gels were cloned. The results of the sequence of DNA bands from different gastrointestinal sites of experimental post-weaning pigs revealed that most of the sequences of DNA bands matched with the V3 variable region of 16S rDNA of bacteria in the GenBank database.
In the stomach of native pigs, among 20 sequences matching the dominant bands (Table-5A), seven demonstrated high similarity (99-100%) with a known sequence of Lactobacillus in GenBank; six demonstrated 100% similarity with a known sequence of Megasphaera elsdenii; four showed 97% and 96% similarity with the known sequences of Prevotella ruminicola and Prevotella spp., respectively; two demonstrated 97% similarity with a known sequence of Selenomonas spp.; and one clone did not match any known sequence in GenBank. For exotic pigs, among the 20 sequences matching the dominant bands (Table-6A), one sequence demonstrated high similarity with a known sequence of Bifidobacterium in GenBank; six sequences showed 99-100% similarity with a sequence of Lactobacillus; one sequence displayed 97% similarity with a sequence of Helcococcus ovis; two sequences showed 98% similarity with a sequence of M. elsdenii; three sequences demonstrated 98% similarity with a sequence of Selenomonas spp.; three sequences showed 96%, 98%, and 99% similarity with the sequences of P. ruminicola, Mitsuokella jalaludinii, and Syntrophococcus spp., respectively; and four other clones did not match any known sequence in GenBank.  In the ileum of native pigs, most sequences matched with Clostridium (six sequences), followed by Lactobacillus (four sequences) and Bifidobacterium (three sequences) in GenBank, with a similarity of 100% (Table-5B). For exotic pigs, 10 of 20 sequences matched with known sequences in GenBank, whereas the remaining ten sequences belonged to uncultured bacteria. Six sequences matched with a full similarity of 100%, and four sequences demonstrated 98-99% similarity (Table-6B). The ileum samples showed a group of beneficial bacteria such as Veillonella magna, Bifidobacterium, and Lactococcus lactis.
In the colon samples of native pigs, 20 sequences matched with the V3 variable region of 16S rDNA of bacteria in GenBank, in which 12 sequences matched with known bacteria were divided into ten different bacterial groups, and the remaining eight sequences matched with uncultured and unknown bacteria (Table-5C). The sequences that matched known bacteria belonged to Clostridium (two sequences), Lactobacillus (two sequences), and Dialister (two sequences). This result was similar to that observed in the ileum; the dominant bands in the DGGE gel matched with the sequence of Lactobacillus, and other   (Table-6C).
Three sequences were recognized as belonging to some species (L. johnsonii, Eubacterium coprostanoligenes, and M. elsdenii), and six sequences could not be distinguished between species (Streptococcus spp., Lactobacillus spp., and Shigella spp.).

Discussion
This study showed that Lactobacillus, Megasphaera, and Prevotella were present in the stomach of native and exotic pigs fed with the three experimental diets. Supporting this finding, the previous studies have reported that these bacterial species  [20] reported that Lactobacillus and M. elsdenii were dominant in the stomach of pigs fed with a coarse-non pelleted diet, consistent with our study. Interestingly, Lactobacillus and Bifidobacterium were detected in the stomach samples of exotic pigs, whereas only Lactobacillus was detected in native pigs. This result can be attributed to the environmental factors in the stomach between the native pig breed and exotic pig breed, causing a difference in microbiota composition. Lactobacillus and Bifidobacterium are known to have important roles in the digestive tract of pigs as they can prevent contamination with harmful bacteria, reduce diarrhea, and regulate the immunological system [21]. The development of these bacteria creates a better environment for the growth of beneficial bacteria and the inhibition of harmful bacteria [22]. In the present study, beneficial bacteria were detected in the stomach samples of both native and exotic pigs, whereas most harmful bacteria were not detected. This could be due to the lower pH value in the stomach of pigs [23]. Both beneficial bacteria (Bifidobacteria and Lactobacillus) and harmful bacteria (Clostridium spp.) were detected in the ileum of native and exotic pigs. This result suggested that microbiota compositions in the ileum have changed compared with those in the stomach. A recent study by Konstantinov et al. [24] reported that the neutral pH in the ileum was a good condition for the development of harmful bacteria. Among the native pigs fed with the HFC diet, three pigs showed