Feeding dietary fermentable fiber improved fecal microbial composition and increased acetic acid production in a nursery pig model

Abstract

The objective of this study was to determine the fermentable fiber (FF) content of several common fibrous ingredients fed to nursery pigs, and then evaluate the effect of dietary FF level on growth performance and fecal microbial composition. In experiment 1, 54 nursery pigs were randomly allotted to be fed nine diets with six replicate pigs per diet. Dietary treatments included a corn–soybean meal basal diet and eight test diets based on a mixture of the corn–soybean meal diet and corn distillers dried grains with solubles, sunflower meal, oat bran, wheat bran, corn bran, sugar beet pulp (SBP), apple pomace (AP) or soybean hulls (SH). In experiment 2, 180 nursery pigs were housed in 30 pens (six pigs per pen) and randomly allotted to be fed five diets with different FF to total dietary fiber (TDF) ratios, which were 0.52, 0.55, 0.58, 0.61, and 0.64, respectively. Results showed that the FF content in SBP, AP, and SH was greater (P < 0.01) than that in other ingredients. Water binding capacity of fibrous ingredients was positively correlated (P < 0.05) to the digestibility of TDF, acid detergent fiber, and non-starch polysaccharides in test ingredients. Pigs fed the SBP, AP and SH diets had greater (P < 0.05) fecal acetic acid and total short-chain fatty acids (SCFAs) concentrations compared with pigs fed other diets. Fecal acetic acid and total SCFAs concentrations were positively correlated (P < 0.05) with FF content in experimental diets. Average daily weight gain and average daily feed intake of pigs quadratically increased (P < 0.01) as the ratios of FF to TDF increased. Pigs in FF64% group showed higher (P < 0.05) ACE index and fecal acetic acid concentration compared with pigs fed the dietary FF/TDF ratio of 0.52 to 0.61. Compared with the classification system of soluble dietary fiber and insoluble dietary fiber, FF could better describe the mechanism by which dietary fiber has beneficial effects on pig gut health.

This study determined the nutrient digestibility and fermentable fiber (FF) content of several common fibrous ingredients fed to nursery pigs, and then evaluate the effect of dietary FF level on growth performance and fecal microbial composition of nursery pigs. The results showed that classifying dietary fiber according to fermentability goes beyond the category of soluble or insoluble, and better relates its functional attributes. Moreover, this study quantified the FF of several common fibrous ingredients, which can help nutritionists make better use of these ingredients.

Introduction

There is a growing interest in including more crop-processing byproducts in pig diets due to their lower cost relative to corn as well as potential benefits in gut functions (Jaworski and Stein, 2017; Li et al., 2018). Dietary fiber has been considered an antinutritional factor due to its negative effects on nutrient digestibility and growth performance of pigs; however, it is still necessary to be supplied for pigs to maintain normal physiological functions and intestinal development (Lindberg, 2014). Dietary fiber cannot be digested by pig digestive enzymes but can be fermented by intestinal microbiota to produce short-chain fatty acids (SCFAs), which have an important role in regulating gut health, immune function, minerals and water absorption, and blood cholesterol concentration (Koh et al., 2016; Bai et al., 2020).

Dietary fiber is divided into soluble (SDF) and insoluble dietary fiber (IDF) according to the water solubility, of which SDF is considered the main substrate of microbial ­fermentation (Chen et al., 2019a; Jha et al., 2019). Insoluble dietary fiber is difficult to ferment by gut microbiota, but its presence in the gut can reduce endotoxin absorption and inhibit pathogen proliferation by decreasing retention time of intestinal digesta (Gerritsen et al., 2012; Li et al., 2019). However, a recent study found that the SCFA concentration in feces of growing pigs was positively correlated with IDF digestibility but not with SDF digestibility (Zhao et al., 2020). This suggested that dietary IDF fermentation may contribute more to producing SCFA in the intestine than SDF. Moreover, highly fermentable resistant starch and non-fermentable methylcellulose belong to typical examples of IDF and SDF, respectively, providing clear evidence that aqueous solubility is not directly related either to fermentability or to SCFA produced through fermentation (Gill et al., 2021). Therefore, simple classification of dietary fiber into soluble or insoluble, while helpful, is not sufficient to illustrate its fermentation characteristics and functional attributes.

Fermentable fiber (FF, the fiber digested by gut microbiota) is the main energy source for microbial fermentation, so it may be the link between intestinal mucosa and lumen symbiotic microbiota of piglets (Lindberg, 2014). Dietary inclusion of moderate FF reduced protein fermentation and pathogenic bacteria proliferation in the gut of weaned pigs (Bikker et al., 2006). It has also been reported that moderately FF sources with larger particle size had beneficial effects on the intestinal health of pigs after weaning (Molist et al., 2012). Moreover, inclusion of barley with different fermentability in piglet diets affected growth performance, nutrient digestibility, and diarrhea incidence (Wang et al., 2018). Our lab also reported that the proportion of FF in the substrate regulated microbial composition and production of SCFA (Tao et al., 2019). These findings suggested that the effect of dietary fiber on host metabolism and physiological function was mainly related to its fermentability. Thus, it is necessary to quantify the amount of FF in various feed ingredients to better understand the fiber properties of these feed ingredients and thus improve the growth performance of pigs. Many previous reports have shown the fiber digestibility and FF content of common fibrous ingredients fed to growing-finishing pigs and sows (Navarro et al., 2018b; Z. Wang et al., 2019c), but few studies have focused on nutritional values of these ingredients in nursery pigs.

In this study, piglets were used as model animals to evaluat the nutrient digestibility of eight fibrous ingredients with different physicochemical characteristics, and the relationship between the latter and nutrient digestibility. Based on the above data, diets with different ratios of FF and total dietary fiber (TDF) were fed to further evaluate the effects of FF levels on the growth performance and fecal microbiota composition of pigs.

Materials and Methods

Trial protocols implemented in this study were reviewed and approved by the Institutional Animal Care and Use Committee of China Agricultural University (Beijing, China; CAU No. AW52101201-1-1). The experiments were conducted at Fengning Swine Research Unit of China Agricultural University (Hebei, China). Eight high-fiber ingredients varying in contents of IDF and SDF were chosen. Corn distillers dried grains with solubles (DDGS), sunflower meal (SFM), oat bran (OB), wheat bran (WB), corn bran (CB), sugar beet pulp (SBP), and soybean hulls (SH) were provided by the Hefeng Company (Beijing, China). The apple pomace (AP) was purchased from Muguang Bio-technology Company (Henan, China). The analyzed chemical composition and physical characteristics of test ingredients are presented in Table 1.

Table 1.

Analyzed chemical composition and physical characteristics of high-fiber ingredients (as-fed basis)

Item, %1	Ingredient
	DDGS	SFM	OB	WB	CB	SBP	AP	SH
Gross energy, MJ/kg	18.47	17.60	18.55	17.15	17.59	16.00	18.36	16.12
Dry matter	88.62	89.55	91.63	89.22	92.37	90.02	93.63	90.99
Organic matter2	82.35	83.38	86.02	83.44	87.69	83.12	91.39	86.95
Crude protein2	23.34	31.20	21.58	14.87	17.29	10.33	6.63	8.68
Ether extract	6.91	0.84	7.69	3.72	2.90	0.58	3.03	1.54
NDF	32.68	37.45	39.17	43.78	46.99	42.81	46.18	64.01
ADF	19.35	25.40	8.74	12.90	11.82	24.02	33.04	45.54
TDF	33.65	44.26	45.87	46.01	52.80	61.02	63.23	73.67
SDF	2.60	4.10	18.06	4.01	1.85	14.44	17.73	6.10
IDF	31.05	40.16	27.81	42.00	50.95	46.58	45.50	67.57
Hemicellulose	13.32	12.04	30.43	30.89	35.17	18.79	13.15	18.47
Cellulose	13.37	17.69	5.53	10.44	11.47	20.54	29.23	44.08
ADL	5.15	6.48	2.89	2.08	0.11	1.75	0.57	1.10
NSP	28.49	37.78	42.98	43.93	52.69	59.27	62.66	72.57
Insoluble NSP	25.89	33.68	24.92	39.92	50.84	44.83	44.93	66.47
Non-cellulosic NSP	15.12	20.09	37.45	33.49	41.22	38.73	33.42	28.48
Swelling, mL/g	3.73	3.12	5.56	2.68	3.90	7.62	3.30	6.76
WBC, g/g	2.24	4.25	7.86	6.04	5.32	9.54	8.49	5.70

¹DDGS, distillers dried grains with solubles; SFM, sunflower meal; OB, oat bran; WB, wheat bran; CB, corn bran; SBP, sugar beet pulp; AP, apple pomace; SH, soybean hulls; NDF, neutral detergent fiber; ADF, acid detergent fiber; TDF, total dietary fiber; SDF, soluble dietary fiber; IDF, insoluble dietary fiber; ADL, acid detergent lignin; NSP, non-starch polysaccharides, TDF − ADL; insoluble NSP, NSP − SDF; noncellulosic NSP, NSP − cellulose; WBC, water binding capacity.

²Organic matter was calculated as the difference between dry matter and ash.

Animals, diets, and experimental design

In experiment 1, 54 crossbred nursery barrows (Duron × Landrace × Yorkshire; initial body weight 14.76 ± 1.66 kg) were randomly allotted to be fed nine treatment diets in a completely randomized design with six individually-housed replicate pigs per dietary treatment. Dietary treatments fed included a corn–soybean meal basal diet and eight test diets which included the eight test fibrous ingredients replacing 19.36% corn grain and soybean meal in the basal diet (Table 2).

Table 2.

Ingredient composition and analyzed energy value and nutrient composition of experimental diets in experiment 1 (as-fed basis)

Item1	Diet
	Basal	DDGS	SFM	OB	WB	CB	SBP	AP	SH
Ingredients, %
Corn	67.76	54.21	54.21	54.21	54.21	54.21	54.21	54.21	54.21
Soybean meal	29.04	23.23	23.23	23.23	23.23	23.23	23.23	23.23	23.23
DDGS		19.36
SFM			19.36
OB				19.36
WB					19.36
CB						19.36
SBP							19.36
AP								19.36
SH									19.36
Dicalcium phosphate	1.50	1.50	1.50	1.50	1.50	1.50	1.50	1.50	1.50
Limestone	0.90	0.90	0.90	0.90	0.90	0.90	0.90	0.90	0.90
Premix2	0.50	0.50	0.50	0.50	0.50	0.50	0.50	0.50	0.50
NaCl	0.30	0.30	0.30	0.30	0.30	0.30	0.30	0.30	0.30
Analyzed composition, %
Gross energy, MJ/kg	16.16	16.38	16.63	16.48	16.28	16.42	16.55	16.57	15.94
Dry matter	87.71	87.68	89.50	88.32	87.71	88.23	89.52	88.93	87.93
Organic matter3	83.34	82.18	83.90	83.08	82.59	82.93	83.84	84.95	82.82
Crude protein3	17.78	17.73	21.03	18.10	16.75	17.32	17.27	15.55	16.26
Ether extract	2.60	3.38	1.84	3.92	2.72	2.93	1.87	2.65	1.89
NDF	8.90	13.76	13.92	16.41	16.52	17.46	14.85	16.36	19.68
ADF	3.79	6.72	6.96	4.90	5.74	5.53	6.80	9.66	11.87
TDF	12.15	15.69	18.10	18.90	18.46	20.03	21.63	22.37	24.35
SDF	1.03	1.26	1.48	4.44	1.55	1.17	3.18	4.37	1.96
IDF	11.13	14.43	14.62	14.46	16.90	18.85	18.45	18.00	22.39
Hemicellulose	5.11	7.04	6.95	11.52	10.79	11.93	8.05	6.70	7.80
Cellulose	3.65	5.49	5.56	4.05	5.12	5.14	6.25	9.08	11.54
ADL	0.01	0.74	1.18	0.31	0.23	0.06	0.31	0.06	0.27
NSP	12.14	14.94	16.92	18.59	18.23	19.97	21.32	22.31	24.07
Insoluble NSP	11.11	13.69	15.44	14.15	16.68	18.80	18.13	17.94	22.11
Non-cellulosic NSP	8.49	9.45	11.36	14.53	13.11	14.83	15.07	13.23	12.53

¹DDGS, distillers dried grains with solubles; SFM, sunflower meal; OB, oat bran; WB, wheat bran; CB, corn bran; SBP, sugar beet pulp; AP, apple pomace; SH, soybean hulls; NDF, neutral detergent fiber; ADF, acid detergent fiber; TDF, total dietary fiber; SDF, soluble dietary fiber; IDF, insoluble dietary fiber; ADL, acid detergent lignin; NSP, non-starch polysaccharides, TDF − ADL; insoluble NSP, NSP − SDF; noncellulosic NSP, NSP − cellulose.

²Premix provided the following per kilogram of feed: vitamin A as retinyl acetate, 12,500 IU; vitamin D₃ as cholecalciferol, 3,000 IU; vitamin E as dl-alpha-tocopheryl acetate, 30 IU; vitamin K₃ as menadione nicotinamide bisulfite, 3.0 mg; vitamin B12, 12.0 μg; riboflavin, 6 mg; pantothenic acid as dl-calcium pantothenate, 15 mg; niacin, 30 mg; choline chloride, 400 mg; folacin, 1.5 mg; thiamin as thiamine mononitrate, 3.0 mg; pyridoxine as pyridoxine hydrochloride, 3 mg; biotin, 50 μg; Mn as MnO, 40 mg; Fe as FeSO₄, 90 mg; Zn as ZnO, 75 mg; Cu as CuSO₄, 100 mg; I as KI, 0.3 mg; Se as Na₂SeO₃, 0.3 mg.

³Organic matter was calculated as the difference between dry matter and ash.

In experiment 2, 180 nursery pigs with initial body weight of 7.12 ± 0.94 kg (weaning at 28 d of age) housed in six pens of six pigs each were fed five diets with different FF to TDF ratios, which were 0.52, 0.55, 0.58, 0.61, and 0.64, respectively (Table 3). The FF content of eight test ingredients was calculated by TDF content multiplied by its digestibility. The TDF contents of corn, soybean meal, soy protein concentrates, and extruded full-fat soybean were 9.88%, 20.21%, 21.60%, and 17.64%, and the FF contents were 4.37%, 14.33%, 17.48%, and 13.97%, respectively (unpublished data in our laboratory). Vitamins and minerals were provided in all diets to meet the nutrient requirements of 7 to 25 kg pigs according to recommendation of NRC (2012).

Table 3.

Ingredient composition and nutrient content of the experimental diets in experiment 2 (as-fed basis)

Item	FF to TDF ratios, FF/TDF
	0.52	0.55	0.58	0.61	0.64
Ingredients, %
Corn	53.72	53.72	53.72	53.72	53.72
Soybean meal	10.00	10.00	10.00	10.00	10.00
Extruded full-fat soybean	8.00	8.00	8.00	8.00	8.00
Whey powder	6.00	6.00	6.00	6.00	6.00
Soy protein concentrate	4.00	4.00	4.00	4.00	4.00
Fish meal	4.00	4.00	4.00	4.00	4.00
Corn bran	5.40	4.09	2.84	1.42	0.00
Sugar beet pulp	0.25	1.40	2.50	3.73	4.96
Soybean oil	3.25	3.00	2.80	2.50	2.25
Sucrose	2.00	2.00	2.00	2.00	2.00
Starch	0.00	0.29	0.50	0.95	1.11
Soy protein isolate	0.00	0.15	0.30	0.35	0.64
Dicalcium phosphate	1.00	1.00	1.00	1.00	1.00
Limestone	0.80	0.80	0.80	0.80	0.80
l-Lysine-HCl	0.57	0.56	0.55	0.54	0.53
Premix2	0.50	0.50	0.50	0.50	0.50
l-Threonine	0.21	0.20	0.20	0.20	0.20
NaCl	0.15	0.15	0.15	0.15	0.15
dl-Methionine	0.10	0.10	0.10	0.10	0.10
l-Tryptophan	0.05	0.04	0.04	0.04	0.04
Calculated nutrient levels
DE, Kcal/kg	3,542	3,542	3,542	3,542	3,542
CP, %3	19.07	19.15	19.13	19.16	19.09
TDF, %	12.61	12.62	12.63	12.63	12.63
FF, %	6.54	6.91	7.28	7.69	8.06
FF/TDF	0.52	0.55	0.58	0.61	0.64
SID Lys, %	1.35	1.35	1.35	1.35	1.35
SID Met, %	0.40	0.40	0.39	0.40	0.40

¹DE, digestible energy; CP, crude protein; FF, fermentable fiber; TDF, total dietary fiber; SID, standardized ileal digestibility. The TDF of corn, soybean meal, soy protein concentrate, and extruded full-fat soybean was 9.88%, 20.21%, 21.60%, and 17.64%, respectively, while the FF of these ingredients was 4.37%, 14.33%, 17.48%, and 13.97%, respectively. Digestible energy, TDF, and FF of corn bran and sugar beet pulp were measured values from experiment 1.

²Premix compositions were the same as those in experiment 1.

³Analyzed values.

Housing, feeding, and sample collection

In experiment 1, nursery pigs were individually housed in stainless-steel metabolism crates (1.4 × 0.7 × 0.6 m) equipped with a nipple drinker and a stainless steel feeder, and placed in six environmentally controlled rooms (nine metabolism crates per room) with temperature maintained at 24 ± 2 °C. All rooms were cleaned and rinsed every day to keep the environment sanitary. Before the trial, nursery pigs were fed a complete feed to adapt to metabolism crates for 3 d. Experiment 1 lasted 15 d, including 10 d of adaptation to the experimental diets and 5 d of total feces and urine collection. Pigs were provided ad libitum access to water. Daily feed allowance was equivalent to 4% body weight measured at the beginning of the diet adaptation period and divided into two equal-sized meals fed at 0830 and 1530 hours each day. Samples of test diets were collected and stored at −20 °C for further analysis of chemical composition and physical characteristics. During the 5-d collection period, feces were collected into bags (one pig per bag) immediately when they appeared in the metabolism crates and stored at −20 °C. Total urine was collected into buckets containing 50 mL of 6 N HCl to reduce nitrogen losses. The urine volume was measured daily, and 20% aliquot of the urine excreted from each pig was filtered through double-layer of gauze. Fifty milliliters of the mixed urine sample were transferred into a screw-capped tube and immediately stored at −20 °C. At the end of experiment 1, the 5-d collection of feces and urine were thawed and mixed separately per pig. Fecal sub-samples around 300 g were dried for 72 h in a 65 °C drying oven and then ground through a 1-mm screen for further analysis. In addition, fresh fecal samples from each pig were collected into plastic tubes by rectal palpation and immediately stored at −80 °C for SCFA analysis.

In experiment 2, pigs were housed in pens (1.2 × 2.1 m²) with natural lighting and at least 2 h of artificial lighting per day. All pigs had ad libitum access to feed and water throughout the 28-d experiment. Pigs were kept on slatted plastic floors in an environmentally controlled room. The initial temperature of the nursery barn was set at 28 °C and then gradually decreased by 1 °C/wk, and the relative humidity ranged between 60% and 70%.

Body weight of pigs and feed disappearance in each pen were recorded on days 14 and 28 to calculate average daily feed intake (ADFI), average daily weight gain (ADG), and gain-to-feed ratio (G:F). The clinical signs of diarrhea were assessed every morning and afternoon by technicians blinded to dietary treatments, and the visual observations were carried out for each piglet, with touching the butt of each piglet to assist a better assessment. The below equation was used to calculate diarrhea rate of weaned piglets:

$${\displaystyle \begin{array}{l}{\displaystyle Diarrhearate(\mathrm{\%})=\frac{\begin{array}{c}totalnumberofdiarrheapigs\end{array}}{\begin{array}{c}numberofpigs\times totalobservationaldays\end{array}}\times 100\mathrm{\%}.}\end{array}}$$

On day 28, representative fresh fecal samples of six pigs per treatment (one pig per pen) were collected by rectal palpation and stored at −80 °C to analyze SCFAs and microbial composition.

Chemical analysis

Samples of all diets and feces, and test ingredients were analyzed for dry matter (DM; method 930.15), crude protein (CP; method 984.13), ether extract (EE; method 920.39), ash (method 942.05), SDF (Method 991.43), and IDF (Method 991.43) according to the procedures of the Association of Official Analytical Chemists (2005). Organic matter (OM) was calculated as the difference between DM and ash. TDF was calculated as the sum of IDF and SDF. These samples were also analyzed for neutral detergent fiber (NDF) and acid detergent fiber (ADF) using fiber bags and fiber analyzer equipment (Fiber Analyzer, Ankom Technology, Macedon, NY). The content of NDF was analyzed using heat-stable amylase and sodium sulfite without correction for insoluble ash. The ADF fraction was expressed inclusive of residual ash. After ADF analysis, samples were submerged in 72% sulfuric acid for 4 h, rinsed with distilled water, and dried to determine cellulose content followed by ashing at 550 °C to determine acid detergent lignin (ADL; expressed exclusive of acid insoluble ash) content. Gross energy (GE) in feces, diets, urine, and test ingredients samples was analyzed using an Automatic Isoperibol Oxygen Bomb Calorimeter (Parr 6300 Calorimeter, Moline, IL). Physical characteristics of the feed ingredients were analyzed, including swelling and water binding capacity (WBC). Swelling was measured by weighing 0.3 g of sample into a 15-mL centrifuge tube. The sample was then dissolved in 10 mL of 0.9% NaCl solution containing 0.02% NaN₃ and incubated in a shaking water bath at 39 °C for 20 h. Samples were left standing for 1 h before the swelling capacity was determined by reading the volume occupied by fiber. Water binding capacity was analyzed using a modified method (Navarro et al., 2018a). In brief, 2 g of sample was dissolved in 50 mL ultrapure water for 18 h. After centrifugation at 2,500 rpm for 20 min, the supernatant was carefully removed and weight of the pellet was measured. All chemical analyses were conducted in duplicate.

An E.Z.N.A Soil DNA Kit (Omega Bio-tek, Norcross, USA) was applied to extract microbial DNA. For each sample, genes of bacterial 16S ribosomal RNA in the V4 to V5 variable region were amplified using polymerase chain reaction (PCR) with primers. Integrity of PCR amplicons was analyzed by electrophoresis using a Tapestation Instruction (Agilent Technologies, USA). The PCR amplicons were extracted and purified by DNA Gel Extraction Kit using 2% agarose gels (Axygen Biosciences, Union City, CA, USA). The output was quantified using QuantiFluor-ST and sequenced on an Illumina MiSeq system. The QIIME software (version 2.0: https://qiime2.org) was used to demultiplex and quality-filtered raw Illumina fastq files. RDP database (http://rdp.cme.msu.edu/) was used to implement the taxonomy-based analysis for operational taxonomic units using RDP classifier at a 90% confidence level.

SCFAs concentrations were analyzed according to a modified method described by Porter and Murray (2001). Briefly, fresh fecal samples (about 1 g) were put into centrifuge tube (10 mL) containing 0.10% hydrochloric acid (2.0 mL). Tubes were placed in an ice bath for 25 min, then mixed and centrifuged at 15,000 rpm to harvest supernatant. Supernatant was filtered through a 0.45-μm Nylon Membrane Filter (Millipore, Bedford, OH) and then analyzed using Gas Chromatograph System (Agilent HP 6890 Series, Santa Clara, CA, USA).

Calculation

Digestible energy (DE) and metabolizable energy (ME) value of each diet were calculated using the direct method as described by Adeola (2001). The DE and ME value of each test ingredient were subsequently calculated using the difference method

$${\displaystyle \begin{array}{l}{\displaystyle D{E}_d=\frac{G{E}_i-G{E}_f}{D_i}}\end{array}}$$ (1)

$${\displaystyle \begin{array}{l}{\displaystyle D{E}_{dc}=\frac{D{E}_d}{0.968}}\end{array}}$$ (2)

$${\displaystyle \begin{array}{l}{\displaystyle D{E}_t=\frac{(D{E}_d(100\mathrm{\%}-X\mathrm{\%})\times D{E}_{dc})}{X\mathrm{\%}}}\end{array}}$$ (3)

$${\displaystyle \begin{array}{l}{\displaystyle M{E}_d=\frac{(G{E}_i-G{E}_f-G{E}_u)}{D_i}}\end{array}}$$ (4)

$${\displaystyle \begin{array}{l}{\displaystyle M{E}_{dc}=\frac{M{E}_d}{0.968}}\end{array}}$$ (5)

$${\displaystyle \begin{array}{l}{\displaystyle M{E}_{\mathrm{t}}=\frac{(M{E}_d-(96.8\mathrm{\%}-X\mathrm{\%})\times M{E}_{dc})}{X\mathrm{\%}}}\end{array}}$$ (6)

where DE_d and ME_d are the DE and ME values from each diet, respectively; GE_i is the total GE intake of each pig calculated as the result of the GE value of the diet multiplied by D_i, which is the actual total diet intake during the collection period; GE_f and GE_u are the GE value of feces and urine of each pig over the collection period, respectively; DE_dc and ME_dc are the corrected DE and ME in the basal diet, respectively; 0.968 is the percentage of the ingredients that supplied energy in the diet; DE_t and ME_t are the DE and ME value in each test ingredient respectively; X% is the percentage of test ingredients on diets (19.36%).

The apparent total tract digestibility (ATTD) of nutrients and GE in diets (ATTD_d) and ingredients (ATTD_t) were calculated according to the following equation:

$${\displaystyle \begin{array}{l}{\displaystyle ATT{D}_d=\left(\frac{(N_i-N_f)}{N_i}\right)\times 100\mathrm{\%}}\end{array}}$$ (7)

$${\displaystyle \begin{array}{l}{\displaystyle ATT{D}_t=\frac{(A-B)}{P\mathrm{\%}}\times 100\mathrm{\%}+B}\end{array}}$$ (8)

where N_i represents the total intake of a certain nutrient in feed, and N_f represents the total fecal output of the homologous nutrient; A represents the ATTD of a certain nutrient in the test diet, and B represents the ATTD of the corresponding nutrient in the basal diet; P% is the percentage of nutrients supplied by ingredients in the test diets.

Statistical analyses

All data were checked for normality and outliers using the UNIVARIATE procedure of SAS 9.4 (SAS Institute Inc., Cary, NC, USA). Digestibility data were analyzed using PROC GLIMMIX procedure. Diet or ingredient as fixed effect, and replicate as random effect. Growth performance and SCFAs data were analyzed using the GLM procedure. The LSMEANS statement was used to calculate least squares means and statistical differences among the treatments were separated by Tukey’s multiple range test. Diarrhea incidence data were analyzed by a χ² test. The pen was the experimental unit for growth performance and diarrhea rate, whereas individual pig was the experimental unit for other indexes. Correlation coefficients between different variables were determined using the CORR procedure. Polynomial contrast was conducted to determine linear and quadratic effects of FF to TDF ratio. Microbial diversity and composition were analyzed using standardized operational taxonomic units reads according to the procedure of R software. The differential bacteria were classified using linear discriminant analysis (LDA) when the logarithmic LDA values of gut microbiota exceeded 2.0. Differences were considered significant when P < 0.05.

Results

Chemical composition and physical characteristics of high-fiber ingredients in experiment 1

The content of TDF ranged from 33.65% in DDGS to 73.67% in SH, and the content of IDF ranged from 27.81% in OB to 67.57% in SH (Table 1). OB, AP, and SBP had numerically greater content of SDF (18.06%, 17.73%, and 14.44%, respectively) than other ingredients. Swelling capacity of high-fiber ingredients ranged from 2.68 to 7.62 mL/g, and OB, SBP, and SH had greater swelling capacity than other five high-fiber ingredients. In addition, OB, WB, and CB showed greater hemicellulose and lower cellulose content compared with other five high-fiber ingredients.

Energy value and nutrient digestibility of diets in experiment 1

As shown in Table 4, diets including different test ingredients showed decreased (P < 0.01) DE compared with the basal diet. Diet including WB showed the lowest (P < 0.01) ME value. The content of FF ranged from 8.20% in the basal diet to 16.99% in the SBP diet. Diets including different test ingredients decreased (P < 0.01) the ATTD of GE, DM, and OM compared with the basal diet. Among diets, the diet containing CB showed the lowest (P < 0.01) ATTD of TDF, IDF, hemicellulose, non-starch polysaccharide (NSP), insoluble NSP, and non-cellulosic NSP. In addition, the diets containing OB, SBP or AP had greater (P < 0.01) ATTD of SDF than other diets.

Table 4.

Energy value and nutrient apparent total tract digestibility of experimental diets in experiment 1 (dry matter basis)

Item1	Diet									SEM	P-Value
	Basal	DDGS	SFM	OB	WB	CB	SBP	AP	SH
Energy content, MJ/kg
Digestible energy	16.01a	15.41b	15.00bcd	15.43b	14.93cd	14.62de	15.26bc	15.07bcd	14.38e	0.10	<0.01
Metabolizable energy	15.29a	14.50bc	14.45bc	14.80ab	13.64d	13.77cd	14.87ab	14.24bcd	13.76cd	0.17	<0.01
FF, %	8.20f	9.28ef	10.50cd	12.87b	11.18c	9.98de	16.99a	16.57a	16.62a	0.25	<0.01
ATTD, %
Gross energy	86.91a	82.49b	80.75bc	82.66b	80.43bc	78.54c	82.51bc	80.87bc	79.36c	0.53	<0.01
Dry matter	87.55a	82.69b	81.03bc	82.49b	80.97bc	78.07d	83.12b	82.29b	79.61cd	0.51	<0.01
Organic matter	88.91a	84.56b	82.72bc	84.10b	82.79bc	79.68d	85.02b	83.60b	80.77cd	0.50	<0.01
Crude protein	85.01a	82.08ab	81.50ab	81.07abc	80.54bc	81.38ab	77.23c	78.64bc	81.08abc	0.88	<0.01
Ether extract	53.99bc	62.39a	42.32d	58.46ab	41.97d	58.79ab	29.92e	46.67cd	47.50cd	1.72	<0.01
NDF	54.96bcd	50.84cde	48.69de	58.02bc	53.79bcd	43.42e	67.51a	60.32ab	55.88bcd	1.61	<0.01
ADF	58.55bc	44.78de	42.57e	52.46cd	49.00cde	45.91de	71.14a	65.09ab	49.99cde	2.11	<0.01
TDF	59.25cd	51.83e	51.91e	60.15bc	53.14de	43.94f	70.35a	65.87ab	60.02bc	1.35	<0.01
SDF	53.90c	56.66bc	65.20b	81.03a	59.24bc	56.66bc	75.68a	81.58a	63.54b	2.01	<0.01
IDF	59.74bc	51.41d	50.72d	53.73cd	53.41cd	43.13e	69.43a	62.05b	59.71bc	1.40	<0.01
Hemicellulose	52.30c	56.63bc	54.82bc	60.38ab	56.33bc	42.27d	64.45a	53.44c	64.84a	1.37	<0.01
Cellulose	62.62bc	50.12de	46.84de	56.51cd	53.91cde	45.11e	74.70a	66.88ab	51.12de	2.13	<0.01
NSP	59.99c	54.46de	53.68e	62.06bc	54.74de	44.11f	70.76a	66.63ab	60.36cd	1.35	<0.01
Insoluble NSP	60.55bc	54.26de	52.58e	56.11cde	54.32cde	43.33f	69.90a	62.99b	60.08bcd	1.36	<0.01
Non-cellulosic NSP	58.86bc	56.98c	57.03c	63.61ab	55.07c	43.77d	69.13a	66.45a	68.88a	1.31	<0.01

¹DDGS, distillers dried grains with solubles; SFM, sunflower meal; OB, oat bran; WB, wheat bran; CB, corn bran; SBP, sugar beet pulp; AP, apple pomace; SH, soybean hulls; DE, digestible energy; ME, metabolizable energy; FF, fermentable fiber; ATTD, apparent total tract digestibility; NDF, neutral detergent fiber; ADF, acid detergent fiber; TDF, total dietary fiber; SDF, soluble dietary fiber; IDF, insoluble dietary fiber; ADL, acid detergent lignin; NSP, non-starch polysaccharides, TDF − ADL; insoluble NSP, NSP − SDF; noncellulosic NSP, NSP − cellulose.

^{a,b,c,d,e,f,g}Means within a row without a common superscript letter differ (P < 0.05), LSMEANS based on six replicates of six pigs per dietary treatment.

Energy value and nutrient digestibility of ingredients in experiment 1

As shown in Table 5, there were large variations on DE and ME among different high-fiber ingredients (P < 0.01), which ranged from 8.12 MJ/kg DM in SH to 13.52 MJ/kg DM in OB and 7.29 MJ/kg DDGS in WB to 13.26 MJ/kg DM in OB, respectively. The DE and ME in SH and CB were lower (P < 0.01) than those values in DDGS, OB, and SBP. Additionally, the FF content in SBP, AP, and SH was greater (P < 0.01) than other high-fiber ingredients.

Table 5.

Energy value and nutrient apparent total tract digestibility of test ingredients in experiment 1 (dry matter basis)

Item1	Ingredients								SEM	P-Value
	DDGS	SFM	OB	WB	CB	SBP	AP	SH
Energy content, MJ/kg
Digestible energy	13.40a	11.25b	13.52a	10.95bc	9.34cd	12.55ab	11.66ab	8.12d	0.41	<0.01
Metabolizable energy	11.34a	10.15ab	13.26a	7.29b	7.96b	12.30a	10.41ab	7.91b	0.73	<0.01
FF, %	15.72e	22.70d	30.62c	23.54d	16.72e	55.28a	48.19b	49.04b	1.08	<0.01
ATTD, %
Gross energy	66.66a	56.40bc	67.41a	55.15bc	46.55c	62.47ab	58.75ab	48.37c	2.18	<0.01
Dry matter	62.72ab	55.89bc	62.35ab	54.12bc	40.78d	66.75a	61.71ab	47.91cd	2.18	<0.01
Organic matter	66.46ab	58.75b	64.91ab	57.58bc	43.81d	70.65a	63.38ab	48.87cd	2.10	<0.01
Crude protein	73.50a	73.29a	67.92ab	59.00ab	66.24ab	19.59c	7.86c	47.01b	5.74	<0.01
Ether extract	75.24ab	-4.21c	65.77abc	8.62c	79.03a	−208.70d	20.91abc	12.98bc	13.79	<0.01
NDF	46.00d	43.49d	61.58bc	52.68cd	32.82e	78.65a	64.77b	56.42bc	2.28	<0.01
ADF	33.84c	38.90bc	40.93bc	36.61bc	28.01c	80.40a	68.44a	47.03c	2.86	<0.01
TDF	41.39d	45.92d	61.16c	45.65d	29.26e	81.55a	71.35b	60.57c	1.98	<0.01
SDF	60.78c	94.66a	88.35ab	53.77c	62.93c	83.27ab	89.12ab	69.91bc	4.29	<0.01
IDF	39.75c	42.13c	43.60c	46.74c	27.98d	81.33a	64.47b	59.69b	2.10	<0.01
Hemicellulose	64.13cd	52.38e	68.10bc	59.58cde	34.73f	75.85ab	55.30de	79.66a	2.34	<0.01
Cellulose	36.12cd	41.59c	39.51c	40.55c	22.11d	85.48a	69.46b	47.05c	3.11	<0.01
NSP	45.02e	48.46e	64.63bc	51.41ed	28.91f	82.35a	72.20b	60.62cd	2.15	<0.01
Insoluble NSP	46.64c	44.25c	47.54c	49.29c	27.67d	82.37a	65.58b	59.74b	2.11	<0.01
Non-cellulosic NSP	52.81c	53.93c	68.39b	51.41c	30.81d	80.30ab	74.39ab	81.63a	2.72	<0.01

¹DDGS, distillers dried grains with solubles; SFM, sunflower meal; OB, oat bran; WB, wheat bran; CB, corn bran; SBP, sugar beet pulp; AP, apple pomace; SH, soybean hulls; DE, digestible energy; ME, metabolizable energy; FF, fermentable fiber; ATTD, apparent total tract digestibility; NDF, neutral detergent fiber; ADF, acid detergent fiber; TDF, total dietary fiber; SDF, soluble dietary fiber; IDF, insoluble dietary fiber; ADL, acid detergent lignin; NSP, non-starch polysaccharides, TDF − ADL; insoluble NSP, NSP − SDF; noncellulosic NSP, NSP − cellulose.

^a,b,c,d,e,fMeans within a row without a common superscript letter differ (P < 0.05), LSMEANS based on six replicates of six pigs per dietary treatment.

The ATTD of GE, DM, and OM in CB and SH were lower (P < 0.01) than those in DDGS, OB, SBP, and AP. The ATTD of CP was lower (P < 0.01) in SBP and AP than other six high-fiber ingredients, whereas SBP exhibited the lowest (P < 0.01) ATTD of EE among ingredients. Among high-fiber ingredients, the ATTD of fiber fractions except for ADF, SDF, and cellulose was the lowest (P < 0.01) in CB, whereas the ATTD of fiber fractions except for ADF, SDF, hemicellulose, and non-cellulosic NSP was the greatest (P < 0.01) in SBP. In addition, the ATTD of SDF was greater (P < 0.01) in SFM, OB, SBP, and AP than those in DDGS, WB, and CB.

Fecal SCFAs concentrations in experiment 1

Fecal lactic acid concentration in the SH treatment group was greater (P < 0.05) than that in other treatment groups (Table 6). The SBP treatment group showed the highest (P < 0.05) fecal butyrate concentration, and the DDGS treatment group showed the lowest (P < 0.05) fecal propionate ­concentration. Pigs fed SBP, AP, and SH diets had greater (P < 0.05) fecal acetic acid and total SCFA concentrations compared with pigs fed other diets.

Table 6.

Short-chain fatty acid concentrations (mg/g) in feces of pigs fed various fiber-rich diets in experiment 1

Item1	Diets								SEM	P value
	DDGS	SFM	OB	WB	CB	SBP	AP	SH
Lactic acid	0.40b	0.38b	0.52b	0.47b	0.24b	0.30b	0.45b	1.01a	0.12	<0.01
Acetic acid	2.94cd	3.02cd	3.52c	2.75d	3.14cd	4.46ab	4.19b	4.79a	0.16	<0.01
Propionic acid	1.03e	1.82cd	1.88cd	1.73d	1.49d	2.19bc	2.37b	2.86a	0.12	<0.01
Butyric acid	0.54b	0.70b	0.90b	0.70b	0.60b	1.37a	0.98b	0.74b	0.13	<0.01
Isobutyric acid	0.13	0.27	0.15	0.15	0.24	0.24	0.28	0.13	0.06	0.39
Valeric acid	0.12c	0.29bc	0.24bc	0.25bc	0.14c	0.61a	0.43ab	0.12c	0.07	<0.01
Isovaleric acid	0.12bc	0.27abc	0.14bc	0.14bc	0.14bc	0.30ab	0.37a	0.08c	0.05	<0.01
Total SCFA	5.31d	6.78bc	7.38b	6.22cd	6.00cd	9.51a	9.09a	9.75a	0.29	<0.01

¹DDGS, distillers dried grains with solubles; SFM, sunflower meal; OB, oat bran; WB, wheat bran; CB, corn bran; SBP, sugar beet pulp; AP, apple pomace; SH, soybean hulls; SCFA, short-chain fatty acid.

^a,b,c,d,eDifferent superscript within a row means significant different (P < 0.05), LSMEANS based on six replicates of six pigs per dietary treatment.

Correlation coefficients in experiment 1

There were no significant correlations (P > 0.05) between DE or ME with physical characteristics of feed ingredients, but the ME value was positively (r = 0.71, P < 0.05) correlated with the ATTD of SDF (Table 7). In addition, swelling capacity was positively correlated (r = 0.71, P < 0.05) with the ATTD of hemicellulose in high-fiber ingredients. WBC of ingredients was negatively correlated (r = −0.76, P < 0.05) with the ATTD of CP, but was positively correlated (P < 0.05) with the ATTD of TDF, NSP, and ADF (r = 0.80, 0.78, and 0.78, respectively) in high-fiber ingredients.

Table 7.

Correlation coefficients between physicochemical characteristics, energy value, and nutrient digestibility of test ingredients in experiment 1 (dry matter basis)

Item1	Swelling	WBC	DE	ME	ATTD, %
					CP	ADF	TDF	SDF	IDF	Hemicellulose	NSP
Swelling, mL/g	1
WBC, g/g	0.49	1
DE	−0.04	0.23	1
ME	0.35	0.40	0.90**	1
ATTD, %
CP	−0.34	−0.76*	−0.09	−0.20	1
ADF	0.53	0.78*	0.30	0.47	−0.91**	1
TDF	0.63	0.80*	0.35	0.56	−0.80*	0.93**	1
SDF	0.20	0.45	0.43	0.71*	−0.31	0.50	0.56	1
IDF	0.61	0.70	0.16	0.33	−0.85**	0.95**	0.94**	0.36	1
Hemicellulose	0.71*	0.28	0.12	0.33	−0.29	0.48	0.69	0.11	0.67	1
NSP	0.58	0.78*	0.41	0.57	−0.76*	0.91**	0.99**	0.53	0.92**	0.71*	1

¹Chemical compositions were expressed as %. WBC, Water binding capacity; DE, digestible energy; ME, metabolizable energy; ATTD, apparent total tract digestibility; CP, crude protein; ADF, acid detergent fiber; TDF, total dietary fiber; SDF, soluble dietary fiber; IDF, insoluble dietary fiber; NSP, non-starch polysaccharides.

*, ** represent P < 0.05 and P < 0.01, respectively, LSMEANS based on six replicates of six pigs per dietary treatment.

As presented in Table 8, fecal acetic acid and propionic acid concentrations were both positively correlated (P < 0.05) with the contents of FF, TDF, IDF, ADF, and NSP in experimental diets. The fecal butyrate concentration was positively correlated (P < 0.05) with the contents of FF and SDF (r = 0.79 and 0.81, respectively) in experimental diets. Additionally, the fecal SCFA concentration was positively correlated (P < 0.05) with the contents of FF, TDF, and NSP (r = 0.98, 0.92, and 0.87, respectively) in experimental diets.

Table 8.

Correlation between fiber fractions in diets and fecal short-chain fatty acid concentrations in experiment 1

Item1	Acetic acid	Propionic acid	Butyric acid	Total SCFAs	FF	TDF	SDF	IDF	NDF	ADF	NSP
Acetic acid	1.00
Propionic acid	0.87**	1.00
Butyric acid	0.64	0.51	1.00
Total SCFA	0.96**	0.93**	0.73*	1.00
FF	0.94**	0.88**	0.79*	0.98**	1.00
TDF	0.91**	0.91**	0.57	0.92**	0.89**	1.00
SDF	0.56	0.46	0.81*	0.64	0.71*	0.43	1.00
IDF	0.72*	0.72*	0.19	0.65	0.61	0.87**	-0.01	1.00
NDF	0.48	0.62	-0.09	0.42	0.38	0.67	0.01	0.80*	1.00
ADF	0.73*	0.75*	0.07	0.69	0.63	0.68	0.06	0.69	0.50	1.00
NSP	0.87**	0.89**	0.48	0.87**	0.85**	0.99**	0.41	0.89**	0.76*	0.68	1.00

¹Short-chain fatty acid concentrations were expressed as mg/g and fiber fractions concentrations were expressed as %. SCFA, short-chain fatty acid; FF, fermentable fiber; TDF, total dietary fiber; SDF, soluble dietary fiber; IDF, insoluble dietary fiber; NDF, neutral detergent fiber; ADF, acid detergent fiber; NSP, non-starch polysaccharides.

*, ** Represent P < 0.05 and P < 0.01, respectively, LSMEANS based on 6 replicates of 6 pigs per dietary treatment.

Growth performance and diarrhea rate in experiment 2

For days 14 to 28 and the entire trial, the ADG and ADFI of weaned pigs responded quadratically (P < 0.01) as the ratios of FF to TDF increased (Table 9). Compared with pigs fed the dietary FF/TDF ratio of 0.58, ratio of 0.52 and 0.64 groups showed greater (P < 0.05) ADFI for days 14 to 28, whereas ratio of 0.64 group showed greater (P < 0.05) ADFI and ADG for the whole period. However, no differences (P > 0.05) in diarrhea rate and gain-to-feed ratio were observed when pigs were fed diets containing different proportions of FF.

Table 9.

Effects of feeding diets with different fermentable fiber to total dietary fiber ratio on growth performance and diarrhea rate of nursery pigs in experiment 2

	FF to TDF ratio, FF/TDF						P-value
Item1	0.52	0.55	0.58	0.61	0.64	SEM	Treatment	Linear	Quadratic
Days 0 to 14
ADG, g/d	335	323	314	334	343	12.67	0.52	0.46	0.14
ADFI, g/d	506	521	481	506	514	16.01	0.50	0.99	0.38
G:F	0.66	0.62	0.65	0.66	0.67	0.02	0.10	0.18	0.14
Diarrhea rate, %	5.56	7.46	6.82	6.03	6.35		0.69
Days 14 to 28
ADG, g/d	484	443	429	455	496	17.64	0.08	0.51	<0.01
ADFI, g/d	799a	755ab	724b	762ab	835a	22.31	0.03	0.27	<0.01
G:F	0.61	0.59	0.59	0.60	0.60	0.02	0.03	0.85	0.54
Diarrhea rate, %	4.76	5.87	4.29	3.81	4.92		0.51
Days 0 to 28
ADG, g/d	409ab	383ab	371b	395ab	420a	10.25	0.03	0.31	<0.01
ADFI, g/d	652ab	638ab	603b	634ab	674a	13.30	0.02	0.34	<0.01
G:F	0.63	0.60	0.62	0.62	0.63	0.01	0.57	0.73	0.32
Diarrhea rate, %	5.16	6.67	5.56	4.92	5.63		0.37

¹FF, fermentable fiber; TDF, total dietary fiber; ADG, average daily gain; ADFI, average daily feed intake; G:F, gain-to-feed ratio.

^a,bMeans within a row without a common superscript letter differ (P < 0.05), LSMEANS based on six replicates of 36 pigs per dietary treatment.

Fecal microbial community and SCFAs concentrations in experiment 2

The fecal acetic acid and total SCFA increased linearly (P < 0.05) with the increasing ratio of FF to TDF (Table 10). Pigs fed the dietary FF/TDF ratio of 0.64 showed greater (P < 0.05) fecal acetic acid concentration than those of pigs fed other diets. Pigs in the FF64% group had higher ACE index and relative abundance of Bacteroidota, and lower relative abundance of Firmicutes compared with pigs fed the dietary FF/TDF ratio of 0.52 to 0.61 (Figure 1). In addition, compared with pigs fed the dietary FF/TDF ratio of 0.64, ratio of 0.55 and 0.58 groups showed lower (P < 0.05) Chao index of fecal microbiota.

Table 10.

Effects of feeding diets with different fermentable fiber to total dietary fiber ratio on fecal short-chain fatty acid concentrations (mg/g) in experiment 2

	FF to TDF ratio, FF/TDF						P-value
Item1	0.52	0.55	0.58	0.61	0.64	SEM	Treatment	Linear	Quadratic
Acetic acid	4.78b	4.93b	5.01b	5.14b	5.77a	0.18	<0.01	<0.01	0.16
Propionic acid	2.29	2.03	2.10	2.33	2.57	0.20	0.33	0.17	0.12
Butyric acid	1.50	1.40	1.25	1.35	1.74	0.22	0.62	0.40	0.20
Isobutyrate	0.15	0.09	0.13	0.19	0.18	0.03	0.15	0.09	0.27
Valerate	0.31	0.22	0.19	0.41	0.45	0.07	0.16	0.44	0.41
Isovalerate	0.18	0.19	0.11	0.21	0.21	0.03	0.15	0.18	0.11
Total SCFA	9.21	8.86	8.79	9.62	10.92	0.52	0.06	0.02	0.10

¹FF, fermentable fiber; TDF, total dietary fiber; SCFA, short chain fatty acid.

^a,bMeans within a row without a common superscript letter differ (P < 0.05), LSMEANS based on six replicates of six pigs per dietary treatment.

Figure 1.

Effects of the different ratios of fermentable fiber to total dietary fiber on fecal microbial community in experiment 2. (A) Chao index of different treatments; (B) ACE index of different treatments; (C) the abundance of Firmicutes; (D) the abundance of Bacteroidota. The FF52%, FF55%, FF58%, FF61%, and FF64% represented the ratios of fermentable fiber to total dietary fiber as 0.52, 0.55, 0.58, 0.61 and 0.64, respectively.

Taxa with LDA values > 2.0 are depicted in Figure 2. The linear discriminant analysis effect size (LEfSe) cladogram showed that a total of 22 differential bacterial taxa, such as Christensenellaceae, Christensenellaceae_R-7_group, Prevotellaceae_UCG-001, and Parabacteroides, had a higher abundance in the FF64% group. The relative abundance of 16 differential bacterial taxa, such as Prevotella, Butyrivibrio, and Eubacterium_nodatum_group, was found in the FF55% group, while seven differential bacterial taxa were enriched in the FF61% group. The family Butyricicoccaceae and three genera Dorea, Butyricicoccus, and Senegalimassilia were enriched in the FF52% group. Only the two genera Shuttleworthia and Erysipelotrichaceae_UCG_003 were more abundant in the FF58% group.

Figure 2.

Cladogram of LEfSe demonstrates taxonomic profiling for the distinct bacteria among different dietary groups (LDA score > 2.0). The FF52%, FF55%, FF58%, FF61%, and FF64% represented the ratios of fermentable fiber to total dietary fiber with 0.52, 0.55, 0.58, 0.61, and 0.64, respectively.

Discussion

Fiber degradation among different high-fiber feed ingredients

Effects of fiber sources on the digestibility of dietary fiber fractions varied in the present study, which was specifically reflected in the fact that addition of SBP or AP to the diet increased the ATTD of TDF, whereas addition of DDGS, SFM, or CB decreased the ATTD of TDF. Previous report showed a positive correlation between fiber digestibility and SDF intake (Renteria-Flores et al., 2008). Greater intake of SDF enhanced the diversity and richness of intestinal microbiota, thereby improving the fermentation of dietary fiber fractions, and overall, increasing their digestibility (Chen et al., 2019b). Since the greater SDF content in SBP and AP, the fermentation amount of fiber was larger. On the contrary, due to the greater IDF, DDGS, SFM, and CB shortened the retention time of intestinal digesta, and thus the level of intestinal fermentation, resulting in reduced fiber digestibility.

Energy contents and its correlation with fiber composition

There were large variations in DE and ME among eight different high-fiber feed ingredients in nursery pigs, which were not only associated with different fiber composition, but also varying concentrations of CP and EE. Contents of fiber ­fractions in this study were roughly comparable to the reported values for DDGS (Navarro et al., 2018b), SFM (Rodriguez et al., 2013), SH (Lyu et al., 2018), CB (Anderson et al., 2012), and WB (Zhao et al., 2018a), whereas DE and ME values were lower than these reported in growing pigs, which supports that the available energy measured in growing pigs may overestimate utilization capacity of feed ingredients compared with piglets (Xie et al., 2017). The negative value of EE digestibility in the SBP group indicated a greater endogenous loss of EE, which is caused by higher microbial lipid synthesis in the hindgut due to higher viscosity and fermentability of SBP (Kil et al., 2010). The relatively high TDF concentration with low EE concentration in SBP was responsible for negative EE digestibility. Similarly, a previous report also determined that the ATTD of acid-hydrolyzed EE in SBP was negative when the inclusion level of SBP was 15.0% to 30.0% in diets (Navarro et al., 2019).

The ATTD of TDF of different high-fiber ingredients ranged from 29.26% to 81.55%, which was consistent with the results reported in previous studies (Navarro et al., 2018b; Zhao et al., 2020). In general, SDF is more easily fermented by intestinal microbiota than IDF (Jaworski and Stein, 2017), and thus compared with DDGS, SFM, WB, and CB, the ATTD of TDF and NSP, as well as the FF content was greater in OB, SBP, and AP. Higher fiber digestibility was conducive to release of more nutrients enwrapped in the cell wall, and then improved the digestibility of GE and OM. Furthermore, except for SDF, ADF, and cellulose, CB had the lowest ATTD of other fiber fractions among feed ingredients, which was related to a low SDF/IDF ratio, indicating that the specific composition of dietary fiber also affected the fermentation of intestinal microbiota.

Chemical characteristics and its correlation with fiber composition

The swelling capacity of SBP with a higher SDF concentration in this study was considerably greater than that given by Navarro et al. (2018a). Additionally, the swelling capacity and WBC values of feed ingredients were generally within the range reported before (Slama et al., 2019). Swelling capacity occurs when the fiber structure of plant cell wall dissolves and is dispersed by incoming water (Bach Knudsen et al., 2012). The expansion and dispersion of fiber components may increase the surface area and allow more rapid access for microbial enzymes, thus improving fermentation efficiency. The WBC is an ability of fiber components to combine with water, which has positive responses on activity of intestinal microbiota, resulting in greater fiber fermentation (Bachmann et al., 2021). Therefore, there is a strong positive correlation between the digestibility of fiber fractions and swelling capacity or WBC. Similarly, a previous report also found that swelling capacity and WBC of ingredients were positively correlated with the ATTD of IDF, TDF, and NSP (Navarro et al., 2018b). The WBC was negatively correlated with the ATTD of CP, which might be due to the improvement of fiber fermentation by the increased WBC, resulting in increased microbial protein synthesis in the gut (Jha and Leterme, 2012). The above results indicated that in addition to fiber composition, the physicochemical properties of feed ingredients also affect fiber fermentation and nutrient digestibility. Therefore, these physicochemical characteristics are expected to help nutritionists better control the fermentation process that occurs in the pig’s intestine and predict the energy contribution of feed ingredients.

SCFA production and its correlation with fiber composition

In the present study, the concentration of fecal total SCFA in corn DDGS, CB, and WB groups was markedly lower than that in OB, SBP, AP, and SH groups, which might be because recalcitrant fiber fractions such as cellulose and arabinoxylans in CB, corn DDGS, and WB are resistant to microbial fermentation (Urriola and Stein, 2010; Chen et al., 2017), and the contents of SDF easily fermented in these ingredients are low. Although the SDF content in SH was not very high, it showed high fermentability due to the large amount of highly fermentable oligosaccharides (Molist et al., 2014). Similarly, Zhao et al. (2019) found that the fermentability of WB and CB was lower than that of SBP and SH containing highly fermentable substances. Chen et al. (2013) found that the colonic concentrations of acetic acid, propionic acid, and total SCFA in the soybean fiber diet were numerically higher than CB and WB. The SDF in OB is mainly β-glucan, which is almost completely fermented in the hindgut of pigs, thus increasing the SCFA concentration in the OB group. For the SFM group, fiber fermentation and SCFA production were limited due to the high cross-linking of polysaccharide components and lignin (Lannuzel et al., 2022). Thus, diets supplemented with different fibrous ingredients provide different types of dietary fiber fractions thereby resulting in varying SCFA profiles. In particular, FF was found to correlate more strongly with acetic acid and total SCFA concentrations than with other fiber fractions in this study, suggesting that FF may be more useful for predicting the concentration of SCFAs in the gut and the physiological effects mediated by them. Therefore, quantifying FF in feed ingredients and incorporating this parameter into pig feed formulation will be needed to help better exploit the potential probiotic functions of high-fiber ingredients.

Effects of different FF ratios on piglet performance and microbial composition

Fermentation in the gut is important for animal health and the fermentability of dietary fiber can be used for optimizing diet formulations to stimulate beneficial microbial with positive effects on gut health (Heyer et al., 2022). In the experiment 2, to further explore the effects of FF on gut microbiota and its response to SCFAs, we selected CB with lower FF content and SBP with higher FF content as dietary fiber sources, and formulated diets with different proportions of FF. The total amount of CB and SBP added to the diets was about 5%, as the diet containing 5% fibrous ingredients has been shown to improve growth performance and gut health of weaned piglets (Zhao et al., 2018a, 2018b).

Consistent with a previous study (Chen et al., 2019b), the diet with more FF increased fecal α-diversity in nursery pigs, which leads to more production of SCFAs, especially acetic acid. The SCFAs help to maintain a stable intestinal environment by regulating intestinal pH, and increased acetic acid production driven by intestinal microbes has been shown to have a significant protective effect on intestinal inflammation (Koh et al., 2016; Sun et al., 2016). Our results also showed that the diet with the FF/TDF of 0.64 increased Bacteroidetes abundance but decreased Firmicutes abundance, which agrees with previous findings that more FF reduced a ratio of Firmicutes to Bacteriodetes (Helm et al., 2021). The phylum Firmicutes and Bacteroides predominate in the gut and are closely related to carbohydrate metabolism (Zhang et al., 2018). Bacteroidetes can produce a variety of fiber-degrading enzymes and participate in the metabolic pathway of multiple soluble polysaccharides (White et al., 2014). Parabacteroides, Christensenellaceae_R-7_group and Prevotellaceae_UCG-001 were previously shown to be increased due to the consumption of highly fermentable SDF (Whisner et al., 2016; Shang et al., 2021), which is similar to our results. Study has shown the Parabacteroides genus are saccharolytic and produce the major end products, such as acetic acid and succinic acid (Tan et al., 2012). Members of the Parabacteroides, such as Parabacteroides distasonis and Parabacteroides goldsteinii are known probiotics that alleviate obesity and metabolic dysfunctions, and enhance intestinal immunological and barrier function (Wu et al., 2018; Wang et al., 2019a). The Christensenellaceae, within the Firmicutes phylum, has been reported to be increased by dietary interventions involving prebiotic such as polydextrose (Hibberd et al., 2019). It has been shown that an increase in Christensenellaceae was positively correlated with improvements in acetic acid and butyric acid, which might be beneficial in protecting the intestinal barrier of pigs (Waters and Ley, 2019). Members in family Prevotellaceae, such as Prevotellaceae_UCG-001, are known to be important for polysaccharide degradation and SCFAs formation (T. Wang et al., 2019b). As a result, the increased abundance of these taxa may partly explain the greater concentrations of acetate and total SCFA in the FF64% group. Microbial fermentation of FF produces large amounts of SCFAs, leading to a positive shift in the microbiota, thus improving the growth performance and health of pigs.

The ADG and ADFI of piglets were decreased quadratically with the increasing ratios of FF to TDF for days 15 to 28 and the whole experimental period. This result may be ascribed to different mechanisms of action between inert fiber and FF. Inclusion of more inert fiber in the nursery pig diets promoted intestinal maturation and improved digestion and absorption (Gerritsen et al., 2012; Chen et al., 2019a), so the ADFI in the FF52% group was markedly higher than that in the FF58% group for days 15 to 28. Previous studies also have reported that dietary supplementation with ingredients rich in inert fiber improved growth performance and intestinal barrier function (Zhao et al., 2018b; Chen et al., 2019a). Inert fiber could decrease the intestinal transit time and reduce the proliferation of pathogens in the intestine, reducing the risk of diarrhea (Chen et al., 2019a). The lower abundance of Erysipelotrichaceae_UCG_003 in the FF52% group than in the FF58% group also supports this view, as Erysipelotrichaceae_UCG_003 was found to be associated with inflammation-related disorders in the gastrointestinal tract (Kaakoush, 2015).

Conclusions

In summary, this study showed that compared with SDF and IDF systems, fermentable fiber can better describe the mechanism by which dietary fiber has beneficial effects on piglet gut health. Swelling capacity and WBC are the physicochemical characteristics highly related to fiber degradation, and could help nutritionists better control the fermentation process that occurs in the pig’s intestine and predict the energy contribution of feed ingredients in the future. Moreover, the FF of several common fibrous ingredients was quantified using nursery pig as a model, and this will undoubtedly make a contribution to our understanding of how fibrous ingredients affect overall health in piglets.

Glossary

Abbreviations

ADF

acid detergent fiber

ADG

average daily gain

ADFI

average daily feed intake

ADL

acid detergent lignin

AP

apple pomace

ATTD

apparent total tract digestibility

CB

corn bran

CP

crude protein

DDGS

distillers dried grains with solubles

DE

digestible energy

DM

dry matter

EE

ether extract

FF

fermentable fiber

GE

gross energy

G:F

gain-to-feed ratio

IDF

insoluble dietary fiber

LDA

linear discriminant analysis

ME

metabolizable energy

NDF

neutral detergent fiber

NSP

non-starch polysaccharides

OB

oat bran

OM

organic matter

SBP

sugar beet pulp

SCFA

short-chain fatty acid

SDF

soluble dietary fiber

SFM

sunflower meal

SH

soybean hulls

TDF

total dietary fiber

WB

wheat bran

WBC

water binding capacity

Ethics Statement

All animal procedures such as ethical and animal welfare issues were approved by the ethics committee of China Agricultural University.