Influence of infection with Brachyspira hyodysenteriae on clinical expression, growth performance, and digestibility in growing pigs fed diets varying in type and level of fiber

Abstract

Research on the effects of different fiber types and levels on infection with Brachyspira hyodysenteriae on growth performance and nutrients digestibility in pigs is scarce. The objective of the current study was to investigate the effects of infection with B. hyodysenteriae when feeding diets varying in soluble and insoluble dietary fiber (DF) on the expression of swine dysentery, growth performance, and digestibility of organic matter (OM) nutrients. A total of 96 growing pigs (26.9 ± 2.5 kg) were used for the experiment and divided into six blocks. The growing pigs were fed one of four diets for 12 wk: low fiber (LF), high fiber (HF), high soluble fiber (HS), and high insoluble fiber (HI). After 2 wk, half of the pigs were inoculated with B. hyodysenteriae. Half of the pigs in each group were euthanized at week 6 for the measurement of the apparent digestibility at the ileum, cecum, colon, and total tract. The remaining pigs were maintained to observe and analyze the clinical expression of fecal score and excretion of B. hyodysenteriae, growth performance, and total tract digestibility up to 12 wk. In the current study, the experimental diets did not influence the expression of infection in the pigs. The body weight and average daily gain (ADG) were in line with the results of clinical expression from week 4 to 6. However, the ADG of the infected pigs started to recover from week 6 (P < 0.05) and then recovered from week 8 to 12 (P < 0.05). The infection with B. hyodysenteriae did not impair apparent ileal digestibility (AID; P > 0.05), whereas the apparent digestibility of OM, total non-starch polysaccharide, non-cellulosic polysaccharide, and cellulose in the cecum of the infected pigs was higher than non-infected pigs (P < 0.05). The apparent colonic digestibility of ash and nitrogen was higher in non-infected pigs than in infected pigs (P < 0.05). The pigs fed the LF diet had a higher digestibility in all segments of the intestinal tract, whereas the HS diet had the lowest AID but higher or similar to the LF diet in the cecum, colon, and the total tract (P < 0.05). The pigs fed the HF and HI diets, with a high proportion of insoluble fiber, had a lower digestibility in the hindgut than the other two diets (P < 0.05). In conclusion, the infection with B. hyodysenteriae negatively influenced clinical signs of swine dysentery and growth performance but did not impair AID, and neither soluble nor insoluble DF influenced the expression of the infection.

Lay Summary

Swine dysentery is a severe disease that can cause increased mortality and poor feed efficiency with bloody diarrhea. This disease can be treated with antibiotics, but there is a limitation of using antibiotics due to governmental policy, thereby the incidence of swine dysentery has been increased. We, therefore, try to find alternatives with diverse fiber sources and understand the mechanism of swine dysentery in growing pigs. In this study, infection of Brachyspira hyodysenteriae showed a negative influence on growth performance, but compensatory growth and recovery were observed in pigs after 6 wk of the infection. The apparent ileal digestibility was not affected by the infection, and the digestibility of non-starch polysaccharides in the cecum was rather increased than decreased probably because of interaction between B. hyodysenteriae and specific bacteria, which can stimulate fiber degradation in the cecum. However, fiber type and level did not influence the prevention and alleviation of the infection.

Influence of swine dysentery on growth performance, and digestibility of nutrients and fiber in growing pigs.

Introduction

Enteric bacterial infections are among the most common and economically significant diseases affecting swine production worldwide. Swine dysentery is caused by the anaerobic spirochete Brachyspira hyodysenteriae. It is an infection of the large intestine and affects mainly growing pigs. Affected pigs show muco-hemorrhagic diarrhea often with fibrin clots and/or specks of blood.

Swine dysentery is known to be influenced by dietary composition, and several studies conducted during the past two decades have focused on the effect of dietary fiber (DF) on the incidence of swine dysentery. DF can be classified as soluble and insoluble fiber; soluble DF is readily fermented in the cecum and proximal colon stimulating bacteria growth, which may prevent colitis and pathogen attachment on epithelial cells in the colon (Nakajima et al., 2020), whereas insoluble DF is less degraded and fermented compared with soluble DF in the large intestine due to structural constraints (Bach Knudsen et al., 2016). A number of studies have shown that feeding a diet low in fermentable carbohydrates protects against swine dysentery in an experimental challenge model (Pluske et al., 1996, 1998; Durmic et al., 2002; Thomsen et al., 2007), whereas diets containing insoluble DF increase the incidence of swine dysentery (Wilberts et al., 2014). However, other researchers have not been able to reproduce these results (Lindecrona et al., 2003), and recently, it was shown that a diet supplemented with highly fermentable carbohydrates offered complete protection against swine dysentery (Helm et al., 2021). The consequences of swine dysentery are increased mortality, decreased rate of growth, and poor feed efficiency during the outbreak of the disease. The effect of intestinal infection on the apparent ileal and/or total tract digestibility of nutrients including fermentation of DF has to our knowledge not been investigated so far. Therefore, in the current study, we will investigate the effect of infection on the digestibility of nutrients at ileal to fecal levels. It remains to be elucidated whether physiological or microbial factors are affected and cause changes in digestibility.

In the present study, the effect of four experimental diets differing in amount and composition of DF on the experimental B. hyodysenteriae challenge was investigated. It was hypothesized that a diet high in soluble DF would protect against swine dysentery. Furthermore, the digestibility of nutrients at the terminal ileum and the total tract was evaluated 4 wk after the experimental challenge. We did not expect the digestibility at the terminal ileum to be affected by the infection, whereas we anticipated that the total tract digestibility would be affected due to the negative influence of the infection with B. hyodysenteriae.

Materials and Methods

The protocol used in this experiment complied with the guidelines of the Danish Ministry of Justice concerning animal experimentation and care of experimental animals.

Diets and feeding

Four diets were formulated: a low-fiber (LF) control diet based on wheat, a high-fiber (HF) control diet based on barley and barley hulls, and two experimental high-fiber diets: high fiber soluble (HS) and high fiber insoluble (HI) with the same amount of DF as the HF diet but different proportion between soluble and insoluble DF (Tables 1 and 2). The diets HS and HI were based on wheat and with increased amount of DF from different coproducts (residues from the industrial production of potato pulp, KMC, Kartoffelmelcentralen Amba, Brande, Denmark; sugar beet pulp, Danisco Sugar A/S, Assens, Denmark; pectin residue, CPKelco ApS, Lille Skensved, Denmark; brewer’s spent grain, Carlsberg A/S, Fredericia delivered by Agro-Korn A/S, Videbæk, Denmark; pea hull, Prodana Seeds A/S, Odense, Denmark; and seed residues, DLF Trifolium A/S, Roskilde, Denmark). Wet coproducts—sugar beet pulp, potato pulp, pectin residue, and brewer’s spent grain—were all dried to dry matter content above 87.5%. Chromic oxide (2 g/kg diet) was included in the diets as a marker for the determination of digestibility. The diets were cold pelleted at the feed mill at Aarhus University, Foulum, and did not contain any antimicrobials. The pigs were fed restrictively according to body weight (BW). The amount of feed was adjusted according to the BW of the non-challenged pigs every week. Feed residues were recorded daily. Feed residues of the non-challenged pigs were weighed. Due to extensive spillage of feed and mixing of the spilled feed with feces, the feed intake of the challenged pigs could not be calculated with any certainty. Water was available ad libitum. The pigs were adapted to the experimental diets for 2 wk prior to B. hyodysenteriae challenge.

Table 1.

Composition of the experimental diets

	Diet1
Item	LF	HF	HS	HI
Ingredients, g/kg (as-fed basis)
Wheat	744		493	618
Barley		673
Barley hulls		100
Sugar beet pulp			60
Potato pulp			60
Pectin residue			60
Seed residue				70
Pea hulls				70
Brewer’s spent grain				70
Soybean meal	199	178	276	120
Animal fat	20	20	20	20
l-Lysine	9	5	3	10
Methionine	3	1	2	2
Threonine	3	1	3	4
Ca(H2PO4)2	8	8	8	8
CaCO3	9	9	9	3
NaCl	4	4	4	4
Mineral–vitamin premix2	2	2	2	2
Chromic oxide	2	2	2	2

LF, low fiber; HF, high fiber; HS, high fiber soluble; HI, high fiber insoluble.

Provided the following quantities of vitamins and minerals per kilogram of complete diet: 4,400 IU vitamin A, 1,000 IU vitamin D₃, 60 mg alpha-tocopherol, 2.2 mg menadione, 2.2 mg thiamin, 4 mg riboflavin, 3.3 mg pyridoxine, 11 mg d-pantothenic acid, 22 mg niacin, 0.055 mg biotin, 0.022 mg vitamin B₁₂, 50 mg Fe as FeSO₄∙7H₂O, 80 mg Zn as ZnO, 27.7 mg Mn as MnO, 20 mg Cu as CuSO₄∙5H₂O, 0.2 mg I as KI, and 0.3 mg Se as Na₂SeO_3.

Table 2.

Chemical composition of the experimental diets

	Diet1
Item	LF	HF	HS	HI
Chemical composition, g/kg of DM
Dry matter	902	904	902	907
Protein (N × 6.25)	193	177	210	175
Fat	50	52	51	55
Ash	62	60	69	61
Total carbohydrates	635	635	593	644
Sugars	27	24	31	24
Starch	461	403	326	392
Total NSP2	134	201	225	216
Cellulose	31	58	77	83
Total NCP3	104 (35)	143 (52)	148 (65)	133 (25)
Arabinose	28 (9)	29 (6)	36 (13)	33 (7)
Xylose	37 (8)	37 (8)	31 (5)	57 (6)
Mannose	3 (1)	4 (1)	6 (0)	3 (1)
Galactose	13 (7)	12 (6)	29 (17)	11 (5)
Glucose	12 (5)	35 (27)	11 (4)	14 (2)
Uronic acid	9 (5)	10 (5)	32 (24)	12 (4)
Fructans	13	7	11	12
Total non-digestible CHO4	147	208	236	228
Klason lignin	27	51	34	45
Dietary fiber	174	259	270	273
Physicochemical properties
Viscosity, mPa·s	1.35	1.29	1.43	1.19
WBC5, kg/kg DM	1.80	2.09	3.18	2.10

LF, low fiber; HF, high fiber; HS, high fiber soluble; HI, high fiber insoluble.

Non-starch polysaccharides.

Non-cellulosic polysaccharides, values in brackets are soluble NCP.

Carbohydrates.

Water-binding capacity.

Animals and housing

A total of 96 pigs (Landrace/Yorkshire/Duroc, 26.9 ± 2.5 kg), half females and half castrates, were included in the experiment. The pigs were obtained from the specific pathogen-free swineherd at Aarhus University, Foulum, and the absence of B. hyodysenteriae was confirmed by analyses of fecal samples when the pigs were included in the experiment. The pigs were divided into six blocks with 16 pigs in each block. Within each block, the pigs were originating from four litters (4 pigs/litter). Pigs from the same litter were allocated to the same dietary treatment group. Half of the pigs (one female and one castrate from each litter), which were planned to be challenged with B. hyodysenteriae, were placed in a separate barn in the intensive care unit at the Department of Animal Science, Aarhus University, Foulum, in order to prevent the spreading of disease. The other half of the pigs were housed in the pig barn unit at the Department of Animal Science, Aarhus University, Foulum. The pigs were housed pair-wise.

Experimental challenge and monitoring of infection

The B. hyodysenteriae strain used for inoculation was a Danish field isolate from a herd with clinical swine dysentery. The spirochaete was grown in a pre-reduced anaerobic medium consisting of trypticase soy broth (BBL, Cockeysville, USA) supplemented with 0% to 1% cysteine, 0.001% resazurin, 0.2% glucose, and 5% fetal bovine serum. The broth culture was incubated on a rocking platform at 37 °C for 24 h in an anaerobic chamber. On three consecutive days at week 2, pigs were inoculated with 50 mL of broth culture containing approximately 10⁹ colony-forming units per milliliter of B. hyodysenteriae according to Lindecrona et al. (2003). The inoculum was given intragastrically by a stomach tube. The pigs were fasted for 12 h before and 2 h after inoculation.

After inoculation, the pigs were observed daily with an evaluation of general health and appetite and for signs of swine dysentery. Consistency scoring of fecal samples was done daily (normal, loose, watery/mucoid diarrhea, or bloody diarrhea). Watery/mucoid diarrhea or bloody diarrhea and fecal excretion of B. hyodysenteriae were considered as clinical signs of swine dysentery. Fecal samples were collected on weeks 4, 5, 6, and 12 of the experiment and cultured for the presence of B. hyodysenteriae as described by Møller et al. (1998). The pigs were weighed every second week throughout the experimental period to determine the growth rate of challenged and control animals in the feeding groups. An overview of the experimental design and analyses is shown in Figure 1.

Figure 1.

The experimental design and analysis are described with the passage of experimental time with pig.

Sample collection

Half of the pigs were euthanized 4 wk after the experimental challenge (week 6 from the start of the experiment) to investigate the effect of the infection and the experimental diets on the digestibility of nutrients. Three days prior to the euthanization at week 6, fecal samples were collected from the pigs that were planned to be euthanized. The samples were pooled within pig. The pigs were killed by bolt pistol followed by exsanguination 3 h after the morning meal. The small intestine (SI), cecum, and colon were dissected free immediately after euthanasia. Digesta from the last third of SI and the cecum were collected. The colon was divided into four segments of equal length, and samples from the second segment were collected to estimate digestibility. Wet samples for the determination of chromic oxide in intestinal contents and feces were stored at −20 °C, while samples for the estimation of digestibility of organic matter (OM), nutrients, and non-starch polysaccharides (NSPs) components were freeze-dried. Fecal samples were collected for 3 d in week 12 from the remaining pigs. The samples were pooled within the pig and treated as described above.

Chemical analysis

All freeze-dried samples and the feed samples were milled through a 0.5-mm mesh screen (Cyclotec 1093 Sample mill, Foss Tecator, Hoeganaes, Sweden) before the analysis. The dry matter (DM) was determined by freeze-drying followed by drying at 105 °C for 20 h, and ash was determined by combustion at 525 °C for 6 h (Association of Official Analytical Chemists, 1990). Protein (N × 6.25) was determined as elementary N (Hansen, 1989). Fat was extracted with diethyl ether after acid hydrolysis (Stoldt, 1952). Starch was determined by enzymatic-colorimetric assays (Bach Knudsen, 1997; McCleary and Monaghan, 2002). Fructan was determined by an enzymatic-colorimetric method modified as previously described (Larsson and Bengtsson, 1983; Hindrichsen et al., 2004). Neutral NSP and constituent sugars were analyzed as alditol acetates by gas chromatography and uronic acids by colorimetry (Bach Knudsen, 1997). Klason lignin was measured gravimetrically as the residue resistant to 2 M H₂SO₄ (Theander and Aman, 1979). Chromic oxide was measured using the method of Schurch et al. (1950). The water-binding capacity (WBC) and viscosity in diets and digesta were determined following the methods described by Serena et al. (2008).

Calculation and statistical analysis

Apparent ileal, cecal, colonic, and fecal digestibility of OM and nutrients were calculated by the marker index method relative to dietary, ileal, and fecal concentrations of chromic oxide and calculated as:

$${\displaystyle \begin{array}{ll}{\displaystyle \mathrm{D}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\mathrm{o}\mathrm{f}\mathrm{X}\left(\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\right)=}& [1(\mathrm{C}{\mathrm{r}}_2{{\mathrm{O}}_3}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}\times {\mathrm{X}}_{\mathrm{d}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{a}/\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}})/\\ & (\mathrm{C}{\mathrm{r}}_2{\mathrm{O}}_{3\mathrm{d}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{a}/\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}}\times {\mathrm{X}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}})]\times 100\end{array}}$$

where X = the concentration of each nutrient or energy in intestinal digesta or feces and the diet.

Before performing the statistical analyses, we analyzed the data with full models including the random effects of block and sex of pigs. For none of the models, the random effects of block and sex were significant, and these effects are, therefore, omitted and included in the residual error effect.

Fecal excretion of B. hyodysenteriae and fecal score data were subjected to an analysis of variance with diets as the main effect.

$${\displaystyle \begin{array}{l}{\displaystyle Y_{ij}=\mu +{\alpha }_i+{\epsilon }_{ij},}\end{array}}$$

where Y_ij is the measured dependent variable, $\mu$is the overall mean, $\alpha$_i is the effect of the diets, and $\epsilon$_ij is the residual component (block and sex of pigs).

Growth performance and digestibility in the ileum, cecum, and colon data were subjected to an analysis of variance (ANOVA) with main effects of B. hyodysenteriae infection, the diets, and interaction between them using two-way ANOVA:

$${\displaystyle \begin{array}{l}{\displaystyle Y_{ij}=\mu +{\alpha }_i+{\beta }_j+\alpha {\beta }_{ij}+{\epsilon }_{ij},}\end{array}}$$

where Y_ij is the measured dependent variable, $\mu$is the overall mean, $\alpha$_i is the effect of the diets, and $\beta$_j is the effect of B. hyodysenteriae (inoculation), $\alpha \beta$_ij is the interaction between the diets and B. hyodysenteriae, and $\epsilon$_ij represents residual component (block and sex of pigs).

The total tract digestibility was subjected to an analysis of variance with main effects of the diets, B. hyodysenteriae infection, period (week), and the interactions among them using three-way ANOVA:

$${\displaystyle \begin{array}{l}{\displaystyle Y_{ijk}=\mu +{\alpha }_i+{\beta }_j+{\gamma }_k+\alpha {\beta }_{ij}+\alpha {\gamma }_{ik}+\beta {\gamma }_{jk}+\alpha \beta {\gamma }_{ijk}+{\epsilon }_{ijk}}\end{array}}$$

where Y_ijk is the dependent variable, $\mu$is the overall mean, and $\epsilon$_ijk is the residual component (block and sex of pigs).

The data were analyzed as standard least squares on the Fit Model platform of JMP pro version 15. 0. 0 (SAS Inst. Inc., Cary, NC). Statistical significance was determined at P < 0.05, and 0.05 < P < 0.10 was considered a trend. Student Turkey’s HSD (honestly significant difference) test was used for the determination of means separation.

Results

Incidence of swine dysentery and fecal score

All pigs included in the experiment were examined for the presence of B. hyodysenteriae, and no excretion of the spirochete and B. hyodysenteriae was found before the inoculation. Of the 48 pigs experimentally infected with B. hyodysenteriae, 10 pigs developed muco-hemorrhagic diarrhea. The pigs were randomly distributed among all dietary treatment groups. Two pigs, both receiving HS, died 2 wk after the inoculation. The fecal excretion of B. hyodysenteriae was the highest for pigs fed the LF diet and lowest for pigs fed the HI diet (P < 0.05; Table 3). As presented in Table 4, there was no significant difference in fecal score (P > 0.05).

Table 3.

Clinical signs and fecal excretion of Brachyspira hyodysenteriae fed the experimental diets

	Diet1
Period	LF	HF	HS	HI	SEM	P-value
Clinical signs of infected pigs2
4 wk	5/12	3/12	7/12	7/12	—	—
5 wk	12/12	6/12	10/12	5/12	—	—
6 wk	5/12	5/12	7/12	7/12	—	—
12 wk	0/12	0/12	1/6	0/4	—	—
Fecal excretion of B. hyodysenteriae3
4 wk	0.83	0.50	1.17	1.17	0.288	0.313
5 wk	2.00a	1.16b	1.67ab	1.00b	0.237	0.023
6 wk	1.00	0.83	1.17	1.40	0.295	0.610
12 wk	0.00	0.00	0.33	0.00	0.175	0.474
Overall (4 to 12 wk)	0.87	0.59	0.93	0.78	0.143	0.366

LF, low fiber with infected pigs; HF, high fiber with infected pigs; HS, high fiber soluble with infected pigs; HI, high fiber insoluble with infected pig.

Clinical signs indicate fecal excretion of B. hyodysenteriae and/or higher diarrhea score (greater than or equal to 1.5).

Excretion of B. hyodysenteriae in feces: 0, no detection; 2, detection; average score is presented in the table.

Within a row, values with different superscripts differ (P < 0.05).

Table 4.

Fecal score of infected pigs fed the experimental diets

	Diet1
Period	LF	HF	HS	HI	SEM	P-value
Fecal score2
3 to 4 wk	1.52	0.45	1.19	0.94	0.328	0.165
4 to 5 wk	1.85	0.96	1.88	1.67	0.340	0.227
5 to 6 wk	1.90	1.75	1.77	1.26	0.320	0.525
6 to 7 wk	1.71	1.63	1.36	1.26	0.253	0.555
7 to 8 wk	1.29	1.38	1.32	1.70	0.289	0.818
8 to 9 wk	0.99	1.14	0.94	1.11	0.267	0.946
9 to 10 wk	0.56	0.75	0.82	0.29	0.273	0.619
10 to 11 wk	0.39	0.25	0.45	0.23	0.236	0.903
11 to 12 wk	0.60	0.28	0.36	0.11	0.173	0.343
Overall (3 to 12 wk)	1.18	0.93	1.10	1.13	0.128	0.563

LF, low fiber with infected pigs; HF, high fiber with infected pigs; HS, high fiber soluble with infected pigs; HI, high fiber insoluble with infected pig.

Fecal score ranged from 1 to 3: a score of 0 is normal feces; 1 is soft feces; 2 is watery and mucoid; and 3 is bloody diarrhea.

Growth performance

The infection with B. hyodysenteriae had a strong influence on BW and average daily gain (ADG) during the 10 wk following infection (Table 5). Prior to infection (week 0 to 2), no effect of the experimental groups was observed (P > 0.05), whereas there was a strong effect of B. hyodysenteriae on BW (P < 0.05) during week 3 to 12. The ADG of inoculated pigs was lower compared with non-inoculated pigs until week 8, whereas opposite results were found from week 8 to 12 (P < 0.05). The ADG for the HS was lower than the ADG for the other diets in week 8 to 10 (P < 0.05), and there was a tendency of interaction between diets and B. hyodysenteriae in week 6 to 8 (P = 0.084).

Table 5.

Body weight and average daily gain of pigs fed the experimental diets¹

	Diet2					B. hyo 3			P-value
Period	LF	HF	HS	HI	SEM	−	+	SEM	Diet	B. hyo	D × B
Body weight, kg
Initial	26.8	27.1	26.5	27.2	0.52	26.8	27.0	0.37	0.742	0.809	0.876
2 wk	33.8	34.5	34.4	33.0	0.85	34.0	33.8	0.60	0.609	0.815	0.619
3 wk	38.9	40.6	39.6	38.9	1.09	40.7x	38.3y	0.77	0.642	0.029	0.635
4 wk	41.4	44.9	42.3	42.4	1.49	45.4x	40.1y	1.05	0.361	0.001	0.634
6 wk	50.4	52.9	51.0	52.8	1.57	58.3x	45.5y	1.10	0.422	0.001	0.893
8 wk	66.7	66.8	65.5	63.7	2.15	74.0x	57.7y	1.57	0.222	0.001	0.495
10 wk	84.5	84.4	78.9	80.7	2.98	88.5x	75.3y	2.10	0.451	0.001	0.918
12 wk	95.8	96.6	91.7	95.0	2.28	100.0x	90.0y	1.65	0.428	0.001	0.765
Average daily gain, g/d
0 to 2 wk (pre-challenge)	510	531	554	436	38.7	514	503	27.3	0.132	0.788	0.482
2 to 4 wk	546	695	467	580	69.2	767x	435y	47.6	0.170	0.001	0.555
4 to 6 wk	629	440	474	475	116.3	884x	231y	80.1	0.721	0.001	0.900
6 to 8 wk	1,037	889	908	707	122.8	1,034x	791y	91.5	0.165	0.014	0.084
8 to 10 wk	1,148a	1,205a	905b	1,230a	81.9	1,033y	1,184x	58.3	0.031	0.037	0.318
10 to 12 wk	1,126	1,135	1,099	1,177	108.8	975y	1,253x	78.4	0.940	0.014	0.966
0 to 6 wk	558	573	536	550	38.6	731x	420y	25.1	0.795	0.001	0.958
6 to 12 wk	1,109	1,065	972	1,014	51.5	1,015	1,063	37.3	0.321	0.436	0.360
0 to 12 wk	814	824	768	803	23.6	884x	746y	16.6	0.271	0.001	0.522

n = 11 per mean by each diet; n = 22 per mean by infection (− and +), unless otherwise noted.

LF, low fiber; HF, high fiber; HS, high fiber soluble; HI, high fiber insoluble.

B. hyo, Brachyspira hyodysenteriae.

Within a row, values with different superscripts differ (P < 0.05).

Physicochemical properties in the stomach and SI

The viscosity in the SI and WBC in the stomach of infected pigs was higher compared with non-infected pigs (P < 0.05), whereas there was no effect of diet on the viscosity either in the stomach or SI (P > 0.05; Table 6). The WBC in the stomach of the pigs fed the LF diet was lower than of the pigs fed the high-fiber diets, and the WBC in the SI of pigs fed the HS diet was higher compared with pigs fed the other diets (P < 0.05).

Table 6.

Physicochemical properties of digesta in intestinal segments of pigs fed the experimental diets in week 6¹

		Diet4					B. hyo 5			P-value
Item2	Segment3	LF	HF	HS	HI	SEM	−	+	SEM	Diet	B. hyo	D × B
Viscosity	Stomach	1.41	1.04	1.29	1.39	0.109	1.34	1.24	0.078	0.097	0.341	0.791
	SI	1.80	1.88	1.85	1.98	0.144	1.54y	2.21x	0.102	0.836	0.001	0.531
WBC	Stomach	1.06b	1.60a	1.85a	1.62a	0.148	1.32y	1.74x	0.105	0.004	0.007	0.101
	SI	3.04b	2.92b	3.60a	3.03b	0.143	3.13	3.17	0.101	0.007	0.794	0.301

n = 12 per mean by each diet; n = 24 per mean by infection (− and +), unless otherwise noted.

Water-binding capacity.

SI, Small intestine.

LF, low fiber; HF, high fiber; HS, high fiber soluble; HI, high fiber insoluble.

B. hyo, Brachyspira hyodysenteriae.

Values within a row with different superscripts differ significantly (P < 0.05).

OM and nutrient digestibility in the ileum, cecum, and colon

The apparent ileal digestibility (AID) of OM, ash, sugars, and total non-cellulosic polysaccharide (NCP) was influenced by diets (P < 0.05; Table 7). The AID of sugars in HF was higher than the other diets, whereas the AID of OM, ash, and total NCP in the LF was higher compared with the HS and HI diets, and the AID in the HF diet was similar to the LF or positioned in between the LF and HI diets. The AID of sugars and OM in HS was lower than the other diets and with a tendency of lower AID of protein (P = 0.078). The apparent cecal and colonic digestibility had a different pattern compared with AID. The apparent cecal and colonic digestibility of OM, total NSP, total NCP, and cellulose in the HS was higher than the other diets or in line with the LF except for the OM digestibility in the cecum (P < 0.05). No effect of B. hyodysenteriae infection was found for the AID of OM, ash, nitrogen, starch, sugars, fructan, total NSP, and total NCP except for cellulose (P > 0.1). However, the apparent cecal digestibility of OM, total NSP, total NCP, and cellulose in the inoculated pigs was greater compared with the non-inoculated pigs (P < 0.05). The apparent colonic digestibility of ash and nitrogen in the challenged pigs was lower than the non-challenged pigs (P < 0.05), and there was a similar tendency in the apparent colonic digestibility of OM (P = 0.092).

Table 7.

Digestibility of organic matter, nutrients, and non-starch polysaccharides components in the different intestinal segments of pigs fed the experimental diets in 6 weeks¹

		Diet2					B. hyo 6			P-value
Item	Segment3	LF	HF	HS	HI	SEM	−	+	SEM	Diet	B. hyo	D × B
Starch	SI3	96.5	97.4	92.7	96.3	1.82	95.5	96.1	1.30	0.363	0.747	0.705
Sugars	SI3	87.6b	95.2a	88.9b	88.7b	1.46	89.8	90.4	1.04	0.003	0.674	0.712
Fructan	SI3	40.5	49.4	31.0	44.4	8.41	47.8	35.1	6.01	0.488	0.136	0.793
Organic matter	SI3	67.8a	59.3b	47.0c	54.8bc	2.76	56.8	58.1	1.97	0.001	0.732	0.569
	Ce	74.8a	55.0c	66.4b	63.7b	2.27	62.6y	67.4x	1.62	0.001	0.043	0.645
	Co	79.4a	65.2c	75.4ab	70.8bc	2.14	74.5	70.7	1.52	0.001	0.092	0.627
Ash	SI3	19.7a	7.6ab	4.5b	−7.3b	7.20	1.2	7.0	5.15	0.051	0.443	0.817
	Ce	14.0	7.4	7.8	−9.6	8.51	5.7	3.9	6.10	0.248	0.881	0.392
	Co	13.8	13.7	23.5	18.9	7.83	27.2x	7.5y	5.54	0.804	0.017	0.494
Nitrogen	SI3	72.3	63.4	58.2	65.2	3.49	62.1	67.8	2.49	0.078	0.117	0.501
	Ce	70.6	66.8	71.5	71.0	2.27	70.7	69.1	1.62	0.454	0.537	0.559
	Co	68.3	64.1	68.3	69.0	4.36	71.6x	62.8y	3.05	0.812	0.049	0.846
Total NSP4	SI3	6.3	9.7	−4.7	−3.9	5.37	5.5	−1.7	3.85	0.171	0.191	0.368
	Ce	27.0ab	−16.4c	30.8a	8.4b	6.64	5.2y	19.2x	4.71	0.001	0.035	0.906
	Co	48.4a	23.0b	58.2a	34.3b	3.94	41.8	39.6	2.79	0.001	0.709	0.610
Total NCP5	SI3	9.7ab	18.6a	1.3b	−0.4b	4.81	8.5	6.3	3.44	0.033	0.648	0.588
	Ce	39.0ab	10.4c	47.6a	26.3b	4.86	25.8y	35.4x	3.48	0.001	0.043	0.652
	Co	57.0b	40.1c	67.1a	46.0c	3.13	53.4	51.2	2.21	0.001	0.587	0.599
Cellulose	SI3	−5.0	−12.3	−16.2	−9.5	7.75	−2.3x	−19.3y	5.55	0.774	0.037	0.202
	Ce	−13.0a	−82.5b	−1.5a	−20.2a	11.16	−41.8y	−17.9x	7.93	0.001	0.031	0.998
	Co	19.9b	−19.3c	41.1a	15.5b	5.94	15.0	12.3	4.23	0.001	0.849	0.607

n = 12 per mean by each diet; n = 24 per mean by infection (− and +), unless otherwise noted.

LF, low fiber; HF, high fiber; HS, high fiber soluble; HI, high fiber insoluble.

SI3, end of small intestine; Ce, Cecum; Co, colon.

Non-starch polysaccharides.

Non-cellulosic polysaccharides.

B. hyo, Brachyspira hyodysenteriae.

Within a row, values with different superscripts differ (P < 0.05).

Total tract digestibility of nutrients

The diets had a significant influence on the apparent total tract digestibility (ATTD; P < 0.05), whereas there was no influence of B. hyodysenteriae (P > 0.05) and no interactions between the two effects (P > 0.05; Table 8). However, there was an effect of week in the ATTD of nitrogen and fat (P < 0.05), and tendency of interaction between effects of diet and week in the ATTD of ash and cellulose (0.05 < P < 0.1). The ATTD of OM in weeks 6 and 12 in the LF was higher compared with the other diets, and the ATTD in the HF and HI was almost similar but lower than the HS (P < 0.05). The ATTD of nitrogen in week 12 in the pigs fed the LF was higher and the HF was lower than the other diets, with the other two diets positioned in between LF and HF (P < 0.05). As opposed to the ATTD of OM and protein, the ATTD of total NSP, total NCP, and cellulose in the HS was higher compared with the other diets regardless of week. The ATTD of the LF was in line with the HF and HI, or slightly lower than the two diets except for total NCP in week 12 (P < 0.05).

Table 8.

Digestibility of organic matter, nutrients, and non-starch polysaccharides components in feces of pigs fed the experimental diets in 6 and 12 wk¹

		Diet2					B. hyo 7			P-value
Item	Period	LF	HF	HS	HI	SEM	−	+	SEM	Diet	B. hyo	D × B
Organic matter	6 wk	81.5a	76.9bc	78.8ab	74.1c	1.23	77.4	78.5	0.75	0.001	0.265	0.643
	12 wk	85.5a	74.8c	81.1b	77.0bc	1.42	79.4	80.8	1.02	0.001	0.232	0.991
Ash4	6 wk	29.7	39.0	37.2	30.1	4.34	33.1	36.1	3.00	0.134	0.417	0.568
	12 wk	43.5	33.1	41.1	34.7	2.88	37.5	39.7	4.04	0.219	0.456	0.821
Nitrogen3	6 wk	73.1	70.4	73.3	69.5	2.36	70.5	74.0	1.63	0.294	0.114	0.446
	12 wk	82.8a	75.0b	78.2ab	77.1ab	2.05	78.2	78.8	1.45	0.047	0.673	0.613
Fat3	6 wk	64.2	65.7	64.3	66.4	3.01	63.9	68.1	2.08	0.716	0.095	0.116
	12 wk	71.6	67.8	69.9	70.1	1.66	68.5	71.3	1.21	0.422	0.101	0.987
Total NSP5	6 wk	53.7b	53.8b	67.6a	47.5c	2.27	55.1	57.4	1.55	0.001	0.417	0.516
	12 wk	59.1b	46.5c	70.2a	52.5bc	3.51	54.4	60.2	2.54	0.001	0.120	0.848
Total NCP6	6 wk	61.9b	62.9b	73.8a	56.3c	1.90	63.4	64.8	1.35	0.001	0.602	0.756
	12 wk	66.1	58.9	75.4	59.6	2.94	62.6	68.0	2.13	0.001	0.106	0.857
Cellulose4	6 wk	26.7b	31.3b	55.5a	33.4b	3.54	35.6	41.0	2.51	0.001	0.302	0.384
	12 wk	36.0b	16.0c	60.3a	41.1ab	5.89	34.1	40.8	4.27	0.001	0.225	0.787

n = 12 per mean by each diet; n = 24 per mean by infection (− and +), unless otherwise noted.

LF, low fiber; HF, high fiber; HS, high fiber soluble; HI, high fiber insoluble.

Effect of week (P < 0.05).

Tendency of interaction between diet and week (0.05 < P < 0.1).

Non-starch polysaccharides.

Non-cellulosic polysaccharides.

B. hyo, Brachyspira hyodysenteriae.

Within a row, values with different superscripts differ (P < 0.05), and there was no interaction among diet, week, and B. hyodysenteriae (P > 0.05).

Discussion

Recently, the occurrence of swine dysentery has been increasing in the world compared with the 1990s (Wilberts et al., 2014; Williamson, 2020). The increasing incidence of swine dysentery has been related to the increased use of feed ingredients high in insoluble fiber such as maize distiller’s dried grains with soluble (DDGS) and wheat bran, and a lower use of antibiotics compared with the past (Wilberts et al., 2014). The European Union (EU) legislation prohibits the routine use of antibiotics in animal farms from 2022, meaning that outbreaks of swine dysentery in the EU may be more common as seen in the United States and the United Kingdom. Thus, the prevention of swine dysentery and alleviation of the symptoms are necessary with proper alternatives instead of antibiotics. In many experiments, researchers have tried to find alternatives by modulating the composition of carbohydrates and DF instead of using antibiotics. However, contradictory results have been found. Fermentable carbohydrates such as inulin from chicory roots and sweet lupins completely inhibited the occurrence of swine dysentery after experimental inoculation in the pig (Thomsen et al., 2007; Hansen et al., 2010), whereas insoluble fiber from DDGS has been shown to worsen the severity of swine dysentery (Wilberts et al., 2014). Contrary to the findings of Thomsen et al. (2007) and Hansen et al. (2010), Kirkwood et al. (2000) reported that high fermentable fiber or highly digestible diets did not protect against infection with swine dysentery in pigs. Moreover, diets providing rapidly fermentable carbohydrates to the large intestine could increase the incidence of swine dysentery in infected pigs (Pluske et al., 1998). The most likely cause for the diverse findings is that different structures of carbohydrates and fibers influence microbial activity and composition to different degrees. Moreover, strains of B. hyodysenteriae are different depending on regions and nations, and the composition and abundance of the microbiota differ among hosts. Therefore, the differences between the host and strain can produce diverse results to dietary factors and B. hyodysenteriae infection in growing pigs.

The data from the current experiment did not confirm extensive effects of the diets on the clinical signs of swine dysentery, and the infection with B. hyodysenteriae had a strong influence on growth performance and the ATTD. Clinical signs of B. hyodysenteriae were deteriorated from week 4, and the most severe symptoms were shown in week 5 except for the HI. These results are in line with previous results, which showed that the symptoms were found at 3- to 4-wk intervals after infection with swine dysentery (Bilkei, 1996). In week 5, the pigs fed the diets with the highest content of insoluble fiber—HF and HI diets—had lower fecal excretion of B. hyodysenteriae compared with the other diets. One explanation could be probably that the higher insoluble fiber reduced the feed intake compared with the other two diets (Owusu-Asiedu et al., 2006). In support for this interpretation is that restriction feeding can reduce the clinical outcome of pathogen and virus in comparison to ad libitum feeding in pigs (Le Flocʹh et al., 2014). Thus, insoluble fiber in the HF and HI diets restricting the feed intake could influence the excretion of B. hyodysenteriae.

In the present study, infection with swine dysentery clearly reduced the growth performance of infected pigs, and already a week after inoculation (week 3), a difference in BW between inoculated and non-inoculated pigs was identified. The negative effect of the infection on BW lasted until week 12, and the largest difference, 16.3 kg, between inoculated and non-inoculated pigs was observed in week 8. Similarly, the ADG of infected pigs was much lower from week 2 to 6 compared with uninfected pigs, whereas the difference diminished from week 6 to 8 after which compensatory growth was observed. Compensatory growth is potential growth that can occur after malnutrition or restriction feeding, and this type of growth is strongly connected with feed intake (Skiba, 2005). Generally, swine dysentery infection can decrease feed intake with loss of appetite showing limitation of growth performance due to lack of ingestion of protein and energy. However, feed intake might be increased after the recovery period from week 8, and it was reflected in ADG at week 8 to 12 showing higher ADG of the infected pigs. Thus, despite the infection of the pigs, the pigs could recover showing compensatory growth after week 8.

Results of the present experiment revealed that infection with B. hyodysenteriae did not impair AID regardless of fiber type and level. Schweer et al. (2019) reported that B. hyodysenteriae has minimal impact on basal endogenous loss and thereby did not impact the AID of amino acids. In this study, the AID of total NSP, NCP, and cellulose was not influenced by B. hyodysenteriae, indicating that the colonization of B. hyodysenteriae occurs only in the large intestine. However, the infection with B. hyodysenteriae did increase the viscosity of ileal digesta probably because infection with B. hyodysenteriae increased the mucus production in the SI. However, this increase in ileal viscosity did not influence AID, and hence, higher viscosity caused by the mucus might not be related to AID, unlike viscosity from DF. The apparent cecal digestibility of OM, total NSP, total NCP, and cellulose was higher in the inoculated pigs, which might be a result of interaction between host-microbiota and pathogen stimulating microbial activity in the cecum. It has earlier been found by Varel (1987) that high microbial activity caused by a high-fiber diet increased fiber degradation in the large intestine. Leser et al. (2000) reported synergistic effects between B. hyodysenteriae and Bacteroides uniformis, Bacteroides. fragilis, and Prevotella melaninogenica. These species are positively correlated with fiber degradation in the intestine (Nakajima et al., 2020; Steimle et al., 2021), and especially Prevotella species can ferment fiber polysaccharides with high short chain fatty acids concentration as a result (Myhrstad et al., 2020). Thus, microbes may interact with B. hyodysenteriae, thereby increasing the NSP digestibility in the cecum. In contrast, the apparent digestibility of ash and nitrogen in the colon of infected pigs was lower than that of non-infected pigs. A possibility could be that the induced necrosis of the epithelium in the cecum and colon by B. hyodysenteriae could lower mineral absorption in the cecum and colon and thereby lower the apparent digestibility of ash in the colon. For the nitrogen, the most likely cause for the lower apparent digestibility of nitrogen in the colon is the endogenous losses of mucin and degeneration of epithelial cells in the cecum and colon (Schweer et al., 2019).

Although the AID in the pigs fed the HS diet was lower compared with the AID of the other diets, the apparent digestibility in the cecum and colon was opposite. The lower AID is related to the relatively higher soluble NSP content influencing luminal viscosity and WBC that delay gastric emptying and hinder nutrients absorption in the ileum (Bach Knudsen et al., 2016), whereas, in the cecum and proximal colon, the soluble NSPs are almost completely degraded. In contrast, the high proportion of insoluble fiber with a rigid structure in HF and HI diets limits the degradation and fermentation by microbiota in the hindgut (Bach Knudsen et al., 2016). Although the viscosity of ileal digesta was not influenced by the diets unlike the result of the AID, WBC in the ileal digesta of the pigs fed HS diet was higher than other diets. Tan et al. (2017) reported that soluble DF increases viscosity and WBC in digesta, but there was no difference by the diet in the ileal viscosity. We speculate that the effect of the infection results in high production of mucus in the infected pigs and thereby reduced diet effect in the viscosity of ileal digesta. In addition, WBC might not be influenced by mucus and thus WBC can be a better indicator for effects of DF, especially in the infected pigs with intestinal disease.

The ATTD of NSP was not different between infected and non-infected pigs in weeks 6 and 12, unlike the results for the apparent digestibility in the cecum. The likely explanation is that the degradation and fermentation in the colon of non-infected pigs will level out the differences seen in the cecum, whereas for the infected pigs in week 6, the larger degradation and fermentation of NSP in the cecum result in less degradation and fermentation in the colon.

The ATTD of the experimental diets in week 6 was similar to the ATTD observed in the cecum and colon. The ATTD of nitrogen and fat increased significantly at week 12 regardless of the diets in contrast to ash, OM, and NSP and its components, where there was no week effect. Although the ATTD of fiber is positively related to age and BW, the ATTD of NSP, NCP, and cellulose for diets LF, HS, and HI was only numerically higher at week 12 than week 6. However, the pigs fed the HF diet had significantly higher OM, NSP, NCP, and cellulose digestibility in week 6 than in week 12. This has also been found in a previous study where younger pigs were found to have higher ATTD of DM compared with older pigs on barley-based diet (Beames and Ngwira, 1978). Angelis et al. (2015) reported that Bacteroides, Porphyromonas, and Prevotella spp., which are related to DF degradation, were decreased due to barley β-glucan. In the current study, HF diet consists of 67.3% barley and 10% barely hull, and it cannot be excluded that the β-glucan in barley could be related to lower OM and fiber ATTD in older pigs (week 12).

In this study, experimental diets (DF level and type) did not influence infection with B. hyodysenteriae. Thomsen et al. (2007) used chicory roots (138 g/kg) and sweet lupins (221 g/kg), which contain a high amount of galactan and fructan, and Hansen et al. (2010) included inulin (80 g/kg), and in both studies, the diets completely prevented swine dysentery. Moreover, Helm et al. (2021) reported that a diet containing sugar beet pulp (10%) and resistant starch (10%) showed a protective effect against infection. The preventive effects of the ingredients against the infection may be potentially related to their bifidogenic effect (Queiroz-Monici et al., 2005; Meyer and Stasse-Wolthuis, 2009). In the present study, the HS diet contained pectin residue (60 g/kg), sugar beet pulp (60 g/kg), and potato pulp (60 g/kg), which probably would only to a limiting extent stimulate bifidobacterium spp. and cause bifidogenic effect (Gebbink et al., 1999; Panasevich et al., 2015). Therefore, the preventive effect against infection may be related not only to fermentation but also to abundance of specific bacteria in the large intestine.

In conclusion, the current study clearly showed a negative influence of infection by B. hyodysenteriae on growth performance regardless of fiber type and level. However, the negative effect of infection on ADG was eliminated after 6 wk after which compensatory growth was observed. The ileal digestibility was not impaired by B. hyodysenteriae, whereas the digestibility of NSP, NCP, and cellulose in the cecum of infected pigs was higher probably because of an unknown interaction between B. hyodysenteriae and species of the microbiota at this site of the gastrointestinal tract. The difference seen in the cecum was not reflected in the total tract digestibility. Although the pigs fed the HS diet had the lowest ileal digestibility, the digestibility in the cecum, colon, and total tract was similar or higher compared with the pigs fed the LF diet because of higher degradation of NSP in the hindgut. However, the experimental diets did not cause prevention or alleviation of B. hyodysenteriae infection, suggesting that the infection of B. hyodysenteriae is not strongly related to the fermentation caused by neither fiber type nor level probably because the ingredients did not specifically stimulate bifidobacterium spp.

Glossary

Abbreviations

ADG

average daily gain

AID

apparent ileal digestibility

ATTD

apparent total tract digestibility

BW

body weight

DF

dietary fiber

HF

high fiber

HI

high insoluble fiber

HS

high soluble fiber

LF

low fiber

NCP

non-cellulosic polysaccharide

NSP

non-starch polysaccharide

OM

organic matter

SI

small intestine

WBC

water-binding capacity

Conflict of interest statement

The authors declare that there is no conflict of interest.