Toxicity of canola-derived glucosinolates in pigs fed resistant starch-based diets

Abstract

A study was conducted to determine effects of reducing hindgut pH through dietary inclusion of high-amylose cornstarch (HA-starch) on growth performance, organ weights relative to live body weight (BW), blood thyroid hormone levels, and glucosinolate degradation products of nursery pigs fed cold-pressed canola cake (CPCC). A total of 240 pigs (initial BW: 7.1 kg), which had been weaned at 21 d of age, were housed in 40 pens (6 pigs per pen) and fed 4 diets (10 pens per diet) in a randomized complete block design for 28 d. Four diets were a basal diet with CPCC at 0 or 40%, and with HA-starch at 0 or 40% in a 2 × 2 factorial arrangement. The diets were fed in two phases: Phase 1 from day 0 to 14 and Phase 2 from day 14 to 28 and were formulated to have the same net energy, standardized ileal digestible AA, Ca, and standardized total tract digestible P contents. Dietary inclusion of CPCC and HA-starch was achieved by a partial or complete replacement of corn, soybean meal, and soy protein. At the end of the study, one pig from each pen was euthanized to determine organ weights, blood parameters, hindgut pH, and glucosinolate degradation products. Dietary CPCC reduced (P < 0.05) overall average daily gain (ADG) by 15%; increased (P < 0.05) relative weights of liver and thyroid gland by 27% and 64%, respectively; and reduced (P < 0.05) serum tetraiodothyronine (T4) level from 30.3 to 17.8 ng/mL. Heart, kidney, and gastrointestinal tract weights; serum triiodothyronine level; and hindgut pH of pigs were unaffected by dietary CPCC. Dietary HA-starch reduced (P < 0.05) overall ADG, relative weight of thyroid gland, cecal, and colonic pH; but increased (P < 0.05) relative weight of colon; tended to increase (P = 0.062) serum T4 level. Dietary CPCC and HA-starch interacted (P = 0.024) on relative weight of thyroid gland such that dietary CPCC increased (P < 0.05) weight of thyroid gland for HA-starch-free diet (120 vs. 197 mg/kg of BW) but not for HA-starch-containing diet (104 vs. 130 mg/kg of BW). Dietary CPCC and HA-starch interacted (P = 0.001) on cecal isothiocyanate content such that dietary CPCC increased (P < 0.05) level of isothiocyanates for HA-starch-containing diet but not for HA-starch-free diet. In conclusion, dietary CPCC reduced growth performance, increased liver, size and interfered with thyroid gland functions of pigs. However, the negative effects of dietary CPCC on thyroid gland functions of nursery pigs were alleviated by dietary HA-starch.

Introduction

Inclusion of canola co-products in diets for pigs is limited by glucosinolate degradation products, which interfere with thyroid gland, liver, and kidney functions of pigs (Bell, 1993). The two major glucosinolates present in canola co-products are progoitrin (aliphatic) and 4-hydroxyglucobrassicin (indolic; Slominski et al., 2012; Lee and Woyengo, 2018). Glucosinolates can be hydrolyzed by dietary myrosinases or myrosinases that are produced by gastrointestinal microorganisms (Fahey et al., 2001). Notably, most of the dietary myrosinases are inactivated by heat during cooking, pressing, and toasting of canola seeds during oil extraction (McCurdy, 1992), implying that microorganisms that reside in the hindgut of pigs are the major source of myrosinase that degrades glucosinolates into various metabolites.

During myrosinase-catalyzed hydrolysis of glucosinolates, progoitrin is degraded to isothiocyanates via a Lossen rearrangement, which spontaneously decompose to goitrin at neutral pH (pH 6 to 7), and to nitriles including crambene and epithionitrile in an acidic (pH ≤ 5.5) reaction medium that contained reducing iron (Galletti et al., 2001; Bernardi et al., 2003; Matusheski et al., 2006). Glucobrassicin is degraded to isothiocyanates and then to indole-3-carbinol by the release of thiocyanate at neutral pH (pH 6 to 7), and to indole-3-acetonitrile at acidic pH (4 to 5.6; Chevolleau et al., 1997; Agerbirk et al., 1998). Thus, the composition of glucosinolate degradation products is dependent on parent glucosinolate type and pH conditions. Nitriles are toxic when they are consumed in large amounts, and thus, their large consumption leads to the upregulation of metabolic activity in the liver and kidney as these organs are involved in their detoxification (Matusheski and Jeffery, 2001). An increase in metabolic activity in visceral organs results in their increased utilization of dietary nutrients at the expense of skeletal tissue deposition (Nyachoti et al., 2000). Nitriles are beneficial when they are consumed in small quantities because they increase the activity of hepatic antioxidant enzymes, which reduce oxidative stress. For instance, Tanii et al. (2008) reported increased activity of glutathione peroxidase (primary antioxidant enzyme) in the small intestine of mice due to administration of low dose of allyl nitrile at 50 μmol/kg/d by gastric intubation for 5 d. Also, Tanii et al. (2010) reported reduced neurotoxicity and skin inflammations through the upregulation of antioxidant and phase II enzymes in mice due to administration of subneurotoxic dose of allyl nitrile at 400 μmol/kg/d using gastric intubation for 5 d. Goitrin and thiocyanate interfere with thyroid functions of pigs and have to be detoxified by the liver and kidneys (Felker et al., 2016). Thus, it is apparent that the negative effects of goitrin and thiocyanate on performance and health of pigs are greater than those of nitriles and that the toxicity of glucosinolates in pigs can be reduced by a reduction of hindgut pH.

Resistant starch, which is starch that escapes enzymatic digestion in the small intestine, is highly fermented in the hindgut of pigs (Birt et al., 2013); thus, dietary resistant starch reduces hindgut pH. Amylose is a more resistant starch than amylopectin (Fouhse et al., 2015). In the study of Lee (2019), the inclusion of 40% high-amylose cornstarch (HA-starch) in corn-soybean meal (SBM)-based diet for nursery pigs reduced cecal digesta pH from 6.07 to 5.37. Thus, we hypothesized that the inclusion of HA-starch in canola co-products-based diets for pigs can reduce hindgut pH, leading to increased degradation of intact glucosinolates present in the hindgut of pigs into non-goitrogenic degradation products. However, there is limited information on the effects of including HA-starch in canola co-products-based diets for pigs on glucosinolates-induced toxicity. The objective of this study was to determine the effects of reducing cecal and colonic pH through dietary inclusion of HA-starch on growth performance, liver and thyroid gland weights, blood thyroid hormone levels, and glucosinolate degradation products in the hindgut of nursery pigs fed cold-pressed canola cake (CPCC)-based diet.

Materials and Methods

The experimental animal procedures were reviewed and approved by the Institutional Animal Care and Use Committee at South Dakota State University (#18-076E).

Animals and housing

A total of 240 pigs [initial body weight (BW) of 7.1 ± 1.2 kg; Large White-Landrace female (line 42) × Duroc male (line 280); Pig Improvement Company], which had been weaned at 21 d of age were obtained from Swine Education and Research Facility, South Dakota State University (Brookings, SD). Pigs were fed commercial starter diets during the first 7-d postweaning. Pigs were then individually weighed and group-housed in 40 pens (6 pigs per pen). Pens (1.8 × 2.4 m) had fully slated-concrete floors, metal spindle walls (1.0-m high), and solid polyvinyl chloride gates. Each pen was equipped with a cup drinker, a double-space dry feeder, and a heat lamp. Room temperature was maintained at 24 ± 1°C during the first week. Thereafter, the room temperature was maintained at 23 ± 2°C throughout the experiment.

Experimental diets

Four experimental diets were a basal diet with CPCC at 0% or 40% and with HA-starch at 0% or 40% in a 2 × 2 factorial arrangement (Table 1). Dietary CPCC fed in the current study was derived from Brassica napus seed and was sourced from Dakota Lakes Field Station (Pierre, SD) in one lot and had been produced by pressing the canola seed at less than 50°C (barrel temperature) for 2 min using a screw press expeller (Model KEK-P0020, Remscheid, Germany). The CPCC was delivered to South Dakota State University within 7 d after its production. The HA-starch fed in the current study was Hylon VII, which contains 70% amylose, and was obtained from Ingredion Incorporated (Westchester, IL). The experimental diets were formulated to (1) have similar net energy (NE), crude protein (CP), Ca, standardized total tract digestible P, and standardized ileal digestible Lys, Met, Thr, and Trp contents; and (2) meet or exceed NRC (2012) nutrient recommendations for nursery pigs. The experimental diets were fed as mash and in two phases; Phase 1 for the first 14 d, and Phase 2 for the last 14 d.

Table 1.

Ingredient composition and analyzed nutrient content of experimental diets (as-fed basis)¹

Item	Phase 1				Phase 2
	− HA-starch2		+ HA-starch		− HA-starch		+ HA-starch
	−CPCC	+CPCC	−CPCC	+CPCC	−CPCC	+CPCC	−CPCC	+CPCC
Corn	59.80	46.05	11.05	–	69.12	56.14	20.84	8.36
Cold-pressed canola  cake	–	40.00	–	40.00	–	40.00	–	40.00
Whey permeate	10.00	10.00	10.00	10.00	–	–	–	–
Soybean meal	16.00	–	23.50	–	26.00	–	26.50	–
HA-starch	–	–	40.00	40.00	–	–	40.00	40.00
Soy protein3, 56% CP	10.00	–	10.00	5.33	–	–	7.00	7.00
Soybean oil	0.28	–	1.65	0.78	0.90	–	1.88	0.92
Limestone	1.22	0.80	1.13	0.74	1.17	0.77	1.12	0.70
Monocalcium phosphate	1.00	0.80	1.08	0.88	0.94	0.70	1.00	0.80
Salt	0.65	0.72	0.67	0.73	0.63	0.68	0.64	0.69
l-Lysine HCl	0.55	0.93	0.42	0.83	0.53	0.86	0.36	0.71
l-Threonine	0.15	0.29	0.14	0.28	0.23	0.33	0.19	0.30
dl-Methionine	0.12	0.10	0.15	0.14	0.19	0.16	0.21	0.18
l-Tryptophan	0.03	0.11	0.01	0.09	0.09	0.16	0.06	0.14
Vitamin premix4	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.05
Mineral premix5	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15
Dry matter	87.07	89.27	88.55	90.28	85.55	87.42	88.19	89.61
Crude protein	18.16	22.66	16.87	18.40	16.55	18.56	19.68	19.98
Ash	6.19	6.39	5.51	5.94	5.92	5.55	5.75	5.55
Ether extract	2.87	7.14	3.40	6.43	2.04	7.67	1.97	8.24
Neutral detergent fiber	8.29	13.69	5.19	10.54	8.94	21.00	16.52	11.33
Acid detergent fiber	4.79	8.79	3.75	7.67	4.92	8.86	4.03	7.53

¹The experimental diets were fed in two phases: Phase 1 from day 1 to 14 and Phase 2 from day 14 to 28.

²HA-starch (Hylon VII; 70% amylose) was obtained from Ingredion Incorporated.

³Soy protein was a hydrolyzed soy protein product (HP 300) from Hamlet Protein (Horsens, Denmark).

⁴Provided the following per kilogram of diet: 11,011 IU vitamin A, 1,652 IU vitamin D₃, 55 IU vitamin E, 0.04-mg vitamin B₁₂, 4.4-mg menadione, 9.9-mg riboflavin, 61-mg pantothenic acid, 55-mg niacin, 1.1-mg folic acid, 3.3-mg pyridoxine, 3.3-mg thiamine, and 0.2-mg biotin

⁵Provided the following per kilogram of diet: 165-mg Zn as ZnSO₄, 23-mg Fe as FeSO₄; 17-mg Cu as CuSO₄, and 44-mg Mn as MnSO₄.

Experimental design and procedure

The four experimental diets were allotted to 40 pens with 6 pigs per pen in a randomized complete block design (10 pens per diet). Diets and fresh water were offered to pigs ad libitum during the entire period. Pig weights and feed intake were determined by phase to calculate average daily gain (ADG), average daily feed intake (ADFI), and gain-to-feed ratio (G:F). At the conclusion of the study, one pig from each pen (eight pigs per diet) with BW that was close to the pen average BW was selected and then euthanized by captive bolt penetration, followed by exsanguination. During exsanguination, a 10-mL blood sample was collected from each pig into two sets of vacutainer serum tubes (BD Vacutainer, Plymouth, UK; 6 mL per tube) and immediately stored on ice. The collected blood samples were centrifuged at 2,000 × g for 10 min at 4°C (~30 min after their collection) to recover serum, which was stored frozen at −20 °C for latter determination of serum triiodothyronine (T3) and tetraiodothyronine (T4) concentrations. After euthanasia, visceral organs including heart, kidneys, liver, and thyroid gland were isolated from each pig, blot dried, and weighed. The small intestine, cecum, and colon of the euthanized pigs were also isolated. The colon was divided into three equal sections: proximal, middle, and distal sections. The digesta contents of the cecum; and proximal, middle, and distal colon of each euthanized pig were collected and immediately divided into two equal portions; one portion for determining glucosinolate degradation metabolites, and another portion for determining hindgut digesta pH. Digesta pH was determined immediately using a pH meter (AE 150; Fisher Scientific, Pittsburgh, PA). Digesta samples for determining glucosinolate degradation metabolites were snap-frozen using liquid nitrogen and stored −80°C. After the collection of the digesta samples, the small intestine, cecum, and colon of pigs were emptied, blot-dried, and weighed to determine their empty weights.

Sample preparation and analyses

The CPCC and experimental diets were ground to pass through a 1.00-mm screen using a Wiley mill (Thomas-Wiley Laboratory Mill Model 4, Thomas Scientific, Swedesboro, NJ). The ground CPCC and diet samples were analyzed for dry matter (DM), CP, ether extract (EE), ash, acid detergent fiber (ADF), and neutral detergent fiber (NDF). The samples were analyzed for DM by oven drying at 135°C for 2 h (method 930.15), CP by a combustion procedure (method 990.03), EE (method 2003.06), dry ash (method 942.05) as per AOAC (2007); and for ADF and NDF on a Ankom 200 Fiber Analyzer (Ankom Technology, Fairport, NY) according to Van Soest et al. (1991). Samples were analyzed for AA (method 982.30 E [a, b, and c]; AOAC, 2006) at the University of Missouri Experiment Station laboratories (Columbia, MO). Glucosinolate content was quantified by gas chromatography (POS Pilot Plant Corp., Saskatoon, SK, Canada) according to the method of Daun and McGregor (1981). Blood serum concentration of T3 was determined using an immunoassay analyzer (Immulite 1000, DPC, Los Angeles, CA), whereas blood serum concentration of T4 was determined using Clinical Chemistry Auto-Analyzer System (Vet Axcel Chemistry Analyzer, Alfa Wassermann Diagnostic Technologies, West Caldwell, NJ).

Determination of isothiocyanates using high-performance liquid chromatography

Isothiocyanate concentrations in cecal digesta samples were determined by cyclocondensation reactions as described by Zhang et al. (1996) with some modifications. Briefly, 5 mL of cecal digesta samples were sonicated for 5 min. About 15 mL acetonitrile was added to the sonicated samples, vortexed for 2 min for protein precipitation, and then centrifuged at 4,000 rpm for 20 min at 4°C. After centrifugation, 10 mL of the supernatant was mixed with 15-mL 100-mM potassium phosphate buffer (pH 8.5), and 20-mL acetonitrile with 1,2-benzenedithiol (4.0 mM final concentration) in 50-mL conical centrifuge tubes and vortexed for 2 min. The mixtures were incubated for 2 h at 65°C to convert all isothiocyanates in the samples into 1,3-benzodithiole-2-thione by cyclocondensation reactions. After incubation, the reaction mixtures were lyophilized for 48 h. One hundred milligrams of the lyophilized reaction mixtures were extracted with 2 mL dichloromethane three times and evaporated under air. The dried residues were dissolved in 200 μL of acetonitrile for HPLC/UV analysis. The analysis was performed using mobile phases (solvent A: 30% distilled deionized water and solvent B: 70% acetonitrile) at a rate of 1 mL/min using an Apollo C18 column (250 mm × 4.6 mm i.d., 5 m; Alltech, Deerfield, IL). The total amount of 1,3-benzodithiole-2-thione was determined at the wavelength of 365 nm. The cyclic condensation product (1,3-benzodithiole-2-thione) was purchased from ALFA chemistry (Ronkonkoma, NY). Indole-3-acetonitrile was determined as described by Śmiechowska et al. (2010) with some modifications.

Isothiocyanate concentrations in serum samples were determined by cyclocondensation reactions as described by Zhang et al. (1996) and Choi et al. (2014) with some modifications. Briefly, 200 µL of serum samples, 200 µL of 100-mM phosphate buffer (pH 8.5), and 400 µL of acetonitrile with 1,2-benzenedithiol (50-mM final concentration) were mixed in 2-mL vial and vortexed for 2 min. The mixtures were incubated for 12 h at 65°C to convert all isothiocyanates in the samples into 1,3-benzodithiole-2-thione by cyclocondensation reactions. After incubation, the reaction mixtures were centrifuged at 10,000 × g for 2 min at 4°C. Twenty microliters of the supernatant of the reaction mixtures were analyzed for HPLC/UV analysis. The analysis was performed using mobile phases (solvent A: 30% distilled deionized water and solvent B: 70% acetonitrile) at a rate of 1 mL/min using an Apollo C18 column (250 mm × 4.6 mm i.d., 5 m; Alltech). The total amount of 1,3-benzodithiole-2-thione was determined at the wavelength of 365 nm.

Determination of thiocyanate using gas chromatography mass-spectrometry

Thiocyanate concentrations in the cecal digesta and serum were quantitated by simultaneous determination as described by Bhandari et al. (2012) with some modifications. Briefly, 150 μL of cecal digesta and serum samples were transferred to 2-mL micro-centrifuge vials. A 100 μL of KS¹³C¹⁵N (50 μM) internal standard, 800 μL of 10 mM TBAS (sodium tetraborate decahydrate at pH 9.5) and 500 μL of pentafluorobenzyl bromide (20 mM solution in ethyl acetate) as a derivatizing reagent were added to the sample vials. The solution was vortexed for 2 min and incubated at 70°C in a heating block for 1 hr. The incubated solution was centrifuged at 13,000 rpm for 4 min. A 200 μL of the supernatant (organic layer) was added to autosampler vials containing 200-μL glass inserts for gas chromatography–mass spectrometry injection.

Statistical analysis

Data were subjected to ANOVA using the MIXED procedure of SAS (ver. 9.3, SAS Inst. Inc., Cary, NC) in a randomized complete block design with pen as the experimental unit. The model included dietary CPCC and HA-starch as the fixed factors and block as a random factor:

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

where $Y$ is the dependent variable to be determined, $\mu$ is the overall mean, ${\alpha }_i$ is the main effect of dietary CPCC (i = 1, 2), ${\beta }_j$ is the main effect of dietary HA-starch (j = 1, 2), ${(\alpha \beta )}_{ij}$ is the interaction effect between dietary CPCC and HA-starch, ${\gamma }_k$ is the random block effect (k = 1, 2, 3, 4, 5), and ${\epsilon }_{ijk}$ is the experimental error. Least squares means were determined for each independent variable. Treatment means were separated by the probability of difference when interactions between dietary CPCC and HA-starch were significant. To test the hypotheses, P < 0.05 was considered significant. If pertinent, trends (0.05 ≤ P < 0.10) are also reported.

Results

Progoitrin (2-OH-3-butenyl) was the most abundant aliphatic glucosinolate, whereas 4-hydroxyglucobrassicin (4-OH-3-CH₃-indolyl) was the most abundant indolic glucosinolate present in dietary CPCC fed in the current study (Table 2). Data on growth performance of pigs are presented in Table 3. Dietary CPCC and HA-starch did not interact on pig BW. Dietary inclusion of CPCC tended to reduce (P = 0.070) BW of pigs at d 14 of the study and reduced (P < 0.05) BW of pigs at d 28 of the study. Dietary inclusion of HA-starch did not affect BW at d 14 of the study but tended to reduce (P = 0.058) BW of pigs at d 28 of the study. Inclusion of CPCC or HA-starch in diets reduced (P < 0.05) ADG of pigs for day 0 to 14, day 14 to 28, and for the entire study period (day 0 to 28). There was an interaction (P = 0.007) between dietary CPCC and HA-starch on the ADG of pigs for d 0 to 14 such that dietary inclusion of HA-starch resulted in a greater reduction (P < 0.05) in the ADG of pigs fed CPCC-containing diet than that of pigs fed CPCC-free diet. Dietary inclusion of CPCC reduced (P < 0.05) ADFI of pigs for day 0 to 14, day 14 to 28, and for the entire study period. Dietary inclusion of HA-starch did not affect ADFI of pigs for day 0 to 14, but reduced (P < 0.05) ADFI of pigs for day 14 to 28 and for the entire study period. There was an interaction (P = 0.031) between dietary CPCC and HA-starch on ADFI of pigs for day 14 to 28 such that dietary HA-starch reduced the ADFI for pigs fed CPCC-containing diet, but not for pigs fed CPCC-free diet. There was an interaction (P < 0.05) between dietary CPCC and HA-starch on G:F for day 0 to 14 such that dietary HA-starch reduced (P < 0.05) G:F for pigs fed CPCC-containing diet, but not for pigs fed CPCC-free diet. The G:F for day 14 to 28 and for the entire study was unaffected by dietary CPCC or HA-starch. There were no effects of dietary CPCC or HA-starch on heart and kidney weights relative to live BW of pigs (Table 4). However, the liver weight relative to live BW of pigs was increased (P < 0.05) by dietary inclusion of CPCC, but not of HA-starch. Dietary CPCC increased (P < 0.05) thyroid gland weight relative to live BW of pigs. However, dietary inclusion of HA-starch reduced (P < 0.05) thyroid gland weight relative to live BW of pigs. Dietary CPCC and HA-starch interacted (P = 0.024) on the thyroid gland weight relative to live BW of pigs such that dietary CPCC increased (P < 0.05) thyroid gland weight of pigs fed HA-starch-free diet, but not of pigs fed HA-starch-containing diet. The small intestine and cecum weights relative to live BW of pigs were not affected by dietary CPCC or HA-starch. Dietary CPCC did not also affect colon weight relative to live BW of pigs. However, inclusion of HA-starch in diets resulted in an increase (P < 0.05) in the colon weight relative to live BW of pigs. Dietary CPCC and HA-starch tended to interact (P = 0.091) on the colon weight relative to live BW of pigs such that the magnitude by which dietary HA-starch increased colon weight relative to live BW of pigs fed CPCC-free diet was greater than the magnitude by which dietary HA-starch increased colon weight relative to live BW of pigs fed CPCC-containing diet. The serum T3 level of pigs was not affected by dietary CPCC or HA-starch. However, the serum T4 level of pigs fed CPCC-containing diet was lower (P = 0.016) than that of pigs fed CPCC-free diet. Dietary CPCC reduced (P < 0.05) serum T4 concentration of pigs fed HA-starch-free diet but not of pigs fed HA-starch-containing diet. The pH of cecal and colonic digesta of pigs was not affected by dietary CPCC (Table 5). However, dietary HA-starch reduced (P < 0.05) pH of cecal and colonic digesta of pigs regardless of dietary levels of CPCC. Dietary CPCC and HA-starch interacted (P = 0.001) on the concentration of isothiocyanates in the cecum of pigs such that dietary CPCC increased (P < 0.05) concentration of isothiocyanates in cecal digesta of pigs fed HA-starch-containing diet but not of pigs fed HA-starch-free diet. Dietary CPCC did not affect isothiocyanate concentration in serum of pigs. However, dietary HA-starch resulted in an increase (P < 0.05) in serum concentration of isothiocyanate in pigs. Thiocyanate concentration in cecum of pigs was numerically increased by 128% due to inclusion of CPCC in HA-starch-free diet, and was numerically decreased by 47% due to inclusion of HA-starch in CPCC-containing diet. Dietary CPCC increased (P < 0.05) concentration of thiocyanate in serum of pigs, whereas inclusion of HA-starch in CPCC-containing diet numerically reduced thiocyanate concentration in serum by 39%. Nitriles (indole-3-acetonitriles) were not detectable in the cecal digesta of pigs.

Table 2.

Analyzed nutrient and glucosinolate content of cold-pressed canola cake (% DM basis)¹

Item	Cold-pressed canola cake
Nutrient composition, %
Moisture	6.92
Crude protein	35.93
Ash	5.86
Ether extract	15.32
Neutral detergent fiber	28.50
Acid detergent fiber	19.66
Indispensable amino acids
Arg	2.32
His	0.96
Ile	1.57
Leu	2.60
Lys	2.21
Met	0.71
Phe	1.51
Thr	1.54
Trp	0.38
Val	1.96
Dispensable amino acids
Ala	1.61
Asp	2.79
Cys	0.91
Glu	6.33
Gly	1.87
Pro	2.15
Ser	1.36
Tyr	1.04
Glucosinolates2, µmol/g
Gluconapin (3-butenyl)	2.03
Glucobrassicanapin (4-pentenyl)	0.16
Progoitrin (2-OH-3-butenyl)	3.81
Gluconapoleiferin (2-OH-4-pentenyl)	0.04
Total aliphatics	6.04
Glucoerucin (CH3-thiobutenyl)	<0.02
Gluconasturtiin (Phenylethyl)	0.15
Glucoberteroin (CH3-thiopentenyl)	0.63
Glucobrassicin (3-CH3-indolyl)	0.19
4-Hydroxyglucobrassicin (4-OH-3-CH3-indolyl)	3.91
Total glucosinolates	11.1

¹The analysis was done in duplicate and, hence the presented data are means of two values.

²Glucosinolate contents were analyzed by gas chromatography (Daun and McGregor, 1981) at POS Pilot Plant Corp.

Table 3.

Growth performance of nursery pigs fed cold-pressed canola cake without or with resistant starch¹

Item2	− HA-starch3		+ HA-starch		SEM	P-value
	−CPCC	+CPCC	−CPCC	+CPCC		CPCC	Starch	C × S
Day 0	7.1	7.1	7.1	7.1	0.593	0.990	0.971	0.971
Day 14	11.4	10.8	11.3	9.6	0.836	0.070	0.321	0.290
Day 28	18.0	16.4	17.2	14.0	1.206	0.007	0.058	0.354
Day 0 to 14	0.307a	0.264b	0.303a	0.177c	0.020	<0.001	0.005	0.007
Day 14 to 28	0.446	0.398	0.415	0.311	0.025	<0.001	0.005	0.124
Day 0 to 28	0.388	0.331	0.359	0.246	0.023	<0.001	0.003	0.161
Day 0 to 14	0.468	0.361	0.451	0.304	0.031	<0.001	0.113	0.379
Day 14 to 28	0.714a	0.631b	0.677a	0.470c	0.041	<0.001	0.003	0.031
Day 0 to 28	0.600	0.496	0.564	0.381	0.039	<0.001	0.011	0.169
G:F
Day 0 to 14	0.677b	0.731a	0.673b	0.594c	0.015	0.419	<0.001	<0.001
Day 14 to 28	0.629	0.631	0.616	0.646	0.012	0.184	0.954	0.228
Day 0 to 28	0.648	0.668	0.638	0.675	0.019	0.120	0.909	0.672

¹Data are means of 10 pens of pigs with 6 pigs per pen.

²The pig BW and feed intake were determined by phase to calculate ADG, ADFI, and G:F by phase and for the entire experiment.

³HA-starch (Hylon VII; 70% amylose) was obtained from Ingredion Incorporated.

^a,b,cWithin a row, means without a common superscript differ (P < 0.05).

Table 4.

Organ weights and serum thyroid hormone concentrations of nursery pigs fed cold-pressed canola cake without or with resistant starch¹

Item	− HA-starch2		+ HA-starch		SEM	P-value
	−CPCC	+CPCC	−CPCC	+CPCC		CPCC	Starch	C × S
Organ weights
Heart, g/kg of BW	5.47	5.16	5.31	5.40	0.253	0.618	0.868	0.378
Kidneys, g/kg of BW	5.67	5.65	5.71	5.83	0.297	0.856	0.689	0.797
Liver, g/kg of BW	29.2	37.0	28.3	35.5	1.144	<0.001	0.254	0.768
Thyroid gland, mg/kg of BW	119.7b	196.5a	103.5b	130.1b	11.512	<0.001	<0.001	0.024
Gastrointestinal tract weights, g/kg of BW
Small intestine	46.3	43.0	45.3	44.1	1.274	0.100	0.999	0.442
Cecum	2.69	2.23	2.65	2.51	0.247	0.226	0.609	0.506
Colon	15.8b	16.9b	22.6a	20.4a	0.979	0.530	<0.001	0.091
Serum, ng/mL
T3	0.735	0.739	0.639	0.559	0.085	0.611	0.102	0.551
T4	30.3	17.8	32.1	28.3	3.320	0.016	0.062	0.176

¹Pigs were housed in 40 pens (6 pigs/pen) and 1 pig from each pen (8 pigs/diet) with BW that was close to the pen average BW were euthanized to collect heart, kidneys, liver and thyroid gland, followed by the collection of 10 mL of blood from each pig to determine serum T3 and T4 concentrations.

²HA-starch (Hylon VII™ ^; 70% amylose) was obtained from Ingredion Incorporated.

^a,bWithin a row, means without a common superscript differ (P < 0.05).

Table 5.

Hindgut pH and glucosinolate degradation metabolites in cecum and serum of nursery pigs fed cold-pressed canola cake without or with resistant starch¹

Item	− HA-starch2		+ HA-starch		SEM	P-value
	−CPCC	+CPCC	−CPCC	+CPCC		CPCC	Starch	C × S
Hindgut pH
Cecum	5.60	5.49	5.27	5.27	0.088	0.530	0.004	0.548
Proximal colon	5.62	5.79	5.29	5.25	0.109	0.550	<0.001	0.307
Mid colon	5.70	5.81	5.50	5.28	0.129	0.677	0.006	0.175
Distal colon	5.94	5.99	5.50	5.62	0.138	0.519	0.006	0.792
Isothiocyanates, µmol/mL
Cecum	0.043b	0.041b	0b	0.388a	0.053	0.001	0.009	0.001
Serum	0.124	0.130	0.162	0.208	0.023	0.287	0.026	0.417
Thiocyanate, µmol/mL
Cecum	18.20	41.41	25.54	21.87	9.816	0.289	0.507	0.147
Serum	9.75	45.51	4.87	27.80	11.004	0.041	0.389	0.620

¹Pigs were housed in 40 pens (6 pigs per pen) and 1 pig from each pen (8 pigs per diet) with BW that was close to the pen average BW were euthanized to collect cecal and colonic digesta to determine gastrointestinal tract pH and glucosinolate degradation metabolites.

²HA-starch (Hylon VII; 70% amylose) was obtained from Ingredion Incorporated.

^a,bWithin a row, means without a common superscript differ (P < 0.05).

Discussion

In the current study, dietary CPCC reduced the ADG of nursery pigs. The reasons for the reduction in the ADG of pigs due to dietary inclusion of canola co-products such as CPCC were discussed in our previous study (Lee and Woyengo, 2018); they included reduced ADFI, increased metabolic activities in the liver as evidenced by the increased liver size, and reduced production of thyroid hormones. The reduced feed intake of pigs due to dietary CPCC is attributed to the bitterness of dietary glucosinolate degradation products. The increased liver size of pigs due to dietary CPCC is attributed to increased activity of hepatic detoxification enzymes as a result of an increase in the absorption of toxic glucosinolate degradation products from the gastrointestinal tract of pigs. The presence of nitriles in the feed or in the stomach can lead to the increased liver size of pigs. This is because heat treatment resulted in increased thermal degradation of aliphatic glucosinolates into nitriles (Hanschen et al., 2012). Also, Frandsen et al. (2019) reported increased degradation of glucosinolates to nitriles at acidic pH found in the stomach of pigs regardless of the presence and absence of myrosinases. The increase in liver size of pigs could have been due to increased absorption of glucosinolate-derived goitrogenic compounds generated in the lower part of the small intestine and in the large intestine of pigs (where the neutral pH favors the degradation of glucosinolates to goitrin or thiocyanate), which need to be detoxified in the liver. The reduced thyroid hormone production due to dietary CPCC is attributed to impaired uptake of iodine by the thyroid gland by dietary glucosinolates. The magnitude of reduction in the ADG of pigs due to inclusion of CPCC in diets at 40% in the current study was lower than the magnitude that we (Lee and Woyengo, 2018) observed in our previous study when CPCC was included in the diet at 40% (57 vs. 152 g). This lower magnitude of reduction in the ADG of pigs due to dietary inclusion of CPCC at 40% observed in the current study can be explained by the lower level of total glucosinolates in CPCC fed in the current study than in the study of Lee and Woyengo (2018; 11.1 vs. 14.9 µmol/g).

Dietary inclusion of HA-starch resulted in a reduction in the ADG of nursery pigs. Dietary HA-starch fed in the current study had a greater content of amylose, and hence resistant starch content than conventional corn that it replaced in the HA-starch-free diet. Resistant starch is poorly digested in the small intestine but highly fermented in the hindgut. For instance, Fouhse et al. (2015) reported lower apparent ileal digestibility of starch for HA-starch than that for high amylopectin-starch (78.4 vs. 97.5%) in weanling pigs. Glucose is the end product of starch digestion by digestive enzymes in the small intestine, whereas VFA are the end products of starch fermentation in the gastrointestinal tract. Regmi et al. (2011) reported reduced appearance of glucose but increased appearance of total VFA in the portal vein of pigs due to replacement of conventional starch with 70% of HA-starch (63% amylose), confirming that HA-starch yields less glucose but more VFA in the gastrointestinal tract of pigs. Glucose is a more efficient source of energy than VFA (Jørgensen et al., 1997). Also, the inclusion of resistant starch in diets for pigs can result in reduced digestibility of nutrients other than starch. For instance, De Schrijver et al. (1999) reported reduced ileal digestibility of CP, fat, and energy in weaning pigs due to dietary inclusion of HA-starch. In addition to reduced glucose yield, increased VFA yield, and reduced digestibility of nutrients, dietary HA-starch results in a reduction in voluntary feed intake of pigs as evidenced by the reduced ADFI by dietary inclusion of HA-starch in the current study. Thus, the reduced ADG of pigs due to dietary HA-starch could be attributed to the reduction in energetic efficiency, ileal digestibility of nutrients, and voluntary feed intake of pigs. Fouhse et al. (2015) and Li et al. (2007) similarly reported a reduction in ADG of pigs due to dietary inclusion of HA-starch. There is a need to identify optimal dietary level of HA-starch that results in reduced hindgut pH without significant effects on ADG pigs fed canola co-product-based diets or to identify another strategy of reducing hindgut pH of pigs fed canola co-product-based diets without significant effects on ADG.

Inclusion of HA-starch in diets resulted in a reduction in the ADFI of nursery pigs. The reduced ADFI of pigs due to dietary HA-starch can be attributable to increased gastrointestinal secretions of feed intake-inhibiting hormones as a result of increased VFA production in the hindgut and the high content of amylose in the HA-starch product fed in the present study. Production of VFA in the hindgut results in increased secretion of glucagon-like peptide-1 (food intake-regulating incretin) from L cells in the distal small intestine and colon (Moran and Dailey, 2011). Regmi et al. (2011) reported increased production of glucagon-like peptide-1 due to the inclusion of HA-starch that contained 63% amylose at 70% in the diet, implying that the increased VFA production due to dietary HA-starch results in increased secretion of glucagon-like peptide-1, leading to reduced voluntary feed intake of pigs. In the current study, dietary HA-starch reduced the hindgut pH of pigs, implying that it increased VFA production. In addition to glucagon-like peptide-1 hormone, increased production of VFA in the hindgut resulted in increased secretion of leptin (Xiong et al., 2004), which is a hunger-inhibiting hormone that is produced by white adipose cells and cells of the small intestine (Klok et al., 2007). Barb et al. (1998) reported reduced feed intake in gilts due to the injection of 100 µg of porcine leptin.

Amylose is poorly digested in the small intestine (Sajilata et al., 2006), leading to its increased availability for microbial fermentation in the hindgut and hence increased production of VFA in the hindgut of pigs as evidenced by a reduction in the hindgut pH in the current study due to dietary inclusion of HA-starch. Production of VFA in the hindgut of pigs results in reduced voluntary feed intake (Souza da Silva et al., 2014). For instance, Fouhse et al. (2015) reported a reduction in ADFI of pigs due to dietary inclusion of 67% of HA-starch that contained 80% amylose. Thus, the reduction in the ADFI of nursery pigs due to dietary HA-starch in the current study can be partly attributed to increased glucagon-like peptide-1 and leptin secretions in response to increased VFA production and high amylose content in the HA-starch product fed in the present study. Results of the current study are in agreement with the result from the study of Li et al. (2007), who reported reduced ADFI of weanling pigs by 18% due to dietary inclusion of HA-starch. However, the reduction in the ADFI of pigs observed in the current study due to dietary HA-starch is contrary to the result by Fouhse et al. (2015), who reported unaffected ADFI of pigs due to dietary inclusion of 67% of rice and field pea starch that contained 20% and 36% amylose, respectively. Notably, dietary HA-starch used in the current study contained 70% amylose and was included in the diets at 40%, implying that HA-starch-containing diets fed in the current study contained more amylose (28% vs. 13% and 24%, respectively) than those diets fed in the study of Fouhse et al. (2015). Thus, a reduction in the ADFI of pigs observed in the current study, but not in the study of Fouhse et al. (2015) could be explained by the differences in amylose content in diets and hence VFA production in the hindgut of pigs.

Dietary HA-starch increased colon weight relative to live BW of pigs, which could be ascribed to increased production of VFA (including butyric acid) as a result of dietary inclusion of HA-starch. Kripke et al. (1989) reported increased colonic mucosal weight and crypt depth in the proximal colon of rats due to infusion of butyric acid at 40 or 150 mM through colonic infusion catheter. Dietary raw potato starch (that has a high content of resistant starch) inhibited apoptosis and promoted proliferation of colonocytes of growing pigs (Claus et al., 2003), implying that butyric acid increases colon weight by preventing apoptosis and promoting proliferation of colonocytes. The increased colon weight of pigs observed in the current study is in agreement with the results from the studies of Martinez-Puig et al. (2003) and Bird et al. (2007), who reported an increase in colonic length and weight of growing pigs due to dietary inclusion of HA-starch. Dietary inclusion of HA-starch reduced hindgut digesta pH of nursery pigs, which could be attributed to increased production of organic acids due to starch fermentation in the hindgut of pigs. Bird et al. (2007) and Fouhse et al. (2015) also reported reduced cecal and colonic digesta pH of pigs due to dietary inclusion of HA-starch.

In the current study, it was hypothesized that dietary HA-starch results in reduced cecal and colonic pH of nursery pigs, leading to increased microbial myrosinase-induced degradation of glucosinolates into less toxic but more health-promoting metabolites. Thus, the objective of this study was to determine the effects of reduction of cecal and colonic digesta pH through dietary inclusion of HA-starch on various indicators of glucosinolate toxicity in nursery pigs fed CPCC-based diet. Dietary inclusion of CPCC resulted in a greater reduction in the ADG of pigs fed HA-starch-containing diet (0.303 vs. 0.177 kg) than that of pigs fed HA-starch-free diet (0.307 vs. 0.264 kg). The greater reduction in the ADG of pigs fed CPCC-based diet due to dietary HA-starch can be explained by a combination of negative effects of dietary CPCC and HA-starch on growth performance. As previously noted, dietary CPCC reduces voluntary feed intake of pigs because it can result in increased production of glucosinolate degradation products that are bitter, whereas the high content of amylose in dietary HA-starch fed in the current study can reduce voluntary feed intake of pigs. Also, dietary CPCC can increase metabolic activities in the liver and thyroid gland as evidenced by enlargement of these organs in pigs, leading to increased utilization of energy in these organs at the expense of growth, whereas dietary HA-starch can reduce nutrient digestibility.

Dietary CPCC increased thyroid gland weight relative to live BW of pigs fed HA-starch-free diet, but not of pigs fed HA-starch-containing diet. Also, dietary CPCC reduced serum T4 level for HA-starch-free diet, but not for HA-starch-containing diet. The reduced thyroid hormone synthesis increases the release of thyroid-stimulating hormone by the pituitary, which increases metabolic activity in the thyroid gland, thereby resulting in enlarged thyroid gland of pigs (Schöne et al., 1997b). Thus, the failure of dietary CPCC to increase the thyroid gland weight of pigs fed HA-starch-containing diet in the current study could be explained by the lack of effect of dietary CPCC on the serum T4 level for HA-starch-containing diet. The lack of the effect of dietary CPCC on the serum T4 concentration of pigs fed HA-starch-containing diet could be attributed to decreased degradation of glucosinolates to thiocyanate as evidenced by a numerical reduction in the concentration of thiocyanate in cecal digesta and serum of pigs due to dietary inclusion of HA-starch. As previously mentioned, progoitrin and glucobrassicin are the major glucosinolates found in canola co-products. Goitrin and thiocyanate are the major goitrogenic degradation products that are, respectively, derived from progoitrin and glucobrassicin (Felker et al., 2016). Goitrin and thiocyanate interfere with thyroid hormone synthesis, leading to reduced blood concentration of T4 and enlargement of thyroid glands (Fahey et al., 2001; Bones and Rossiter, 2006). For instance, Kelley and Bjeldanes (1995) reported reduced serum T4 level of rats due to dietary goitrin supplementation at 200 mg/kg of diet. Also, Schöne et al. (1997a) reported reduced serum T4 concentration of growing pigs due to supplemental thiocyanate at 1,000 mg/kg of diet. It should be noted that thiocyanate concentration in cecal digesta and serum for CPCC-containing and HA-starch-free diet was numerically lower than that for CPCC- and HA-starch-containing diet. Thus, the lack of the effect of dietary CPCC on serum T4 level of pigs fed HA-starch-containing diet could also have been due to increased degradation of glucosinolates to nitriles at the expense of goitrin and thiocyanate. Goitrin concentration in the cecal digesta and serum of pigs fed CPCC-based diet due to dietary inclusion of HA-starch was not determined in the current study. However, a trend for the production of goitrin in cecal digesta and serum would be expected to be similar to those, which were observed for the production of thiocyanate in cecal digesta and serum. This is because glucosinolates are degraded to goitrin and thiocyanate at neutral pH and changes in goitrin production in pigs due to change in diet composition can be inferred from the change in thiocyanate production in pigs due to the same change in diet composition. Thus, further research is needed to determine the effect of dietary HA-starch on the production of goitrin from glucosinolates in the gastrointestinal tract of pigs. As previously mentioned, nitriles do not interfere with thyroid hormone synthesis. A reduction in gastrointestinal pH results in increased production of nitriles at the expense of goitrogenic glucosinolate degradation products (Chevolleau et al., 1997; Agerbirk et al., 1998; Galletti et al., 2001; Bernardi et al., 2003); dietary HA-starch reduced the hindgut pH of pigs in the current study.

Dietary CPCC increased cecal isothiocyanate concentration of pigs fed HA-starch-containing diet, which was surprising. It had been hypothesized that dietary HA-starch will reduce cecal and colonic pH of pigs fed CPCC-based diet, leading to increased conversion of progoitrin and glucobrassicin into nitriles. This is because the composition of degradation products derived from glucosinolates is partly dependent on parent glucosinolate type and gastrointestinal pH of pigs. For instance, myrosinase-catalyzed hydrolysis of progoitrin favored goitrin production at neutral pH (pH 6 to 7), but nitrile production at acidic pH ≤ 5.5 (Bernardi et al., 2003; Matusheski et al., 2006); whereas myrosinase-catalyzed hydrolysis of glucobrassicin favored thiocyanate production at neutral pH (pH 6 to 7), but indole-3-acetonitrile production at acidic pH (4 to 5.6; Chevolleau et al., 1997; Agerbirk et al., 1998). Thus, a reduction in the hindgut pH of pigs due to dietary HA-starch was expected to promote nitrile production from dietary glucosinolates at the expense of goitrin and thiocyanate productions. However, it should be noted that the production of isothiocyanates from glucosinolates is also partly dependent on the presence of iron. For instance, in vitro production of isothiocyanates from glucosinolates decreased with a reduction in the pH of incubation medium from 6.5 to 5.5 in the presence of iron, but increased with a reduction in the pH when the incubation medium in the absence of iron (Uda et al., 1986). The presence of resistant starch (such as HA-starch) results in increased absorption of iron in the upper part of the gastrointestinal tract (Morais et al., 1996), implying that resistant starch reduces the availability of iron in the hindgut. Thus, the increased concentration of isothiocyanates in the cecal digesta due to the inclusion of CPCC in HA-starch-containing diet could have been due to reduced iron availability in the cecal digesta of pigs. Although some studies have reported that isothiocyanates are unstable and that they are readily converted to other degradation products such as goitrin and indole-3-carbinol (Agerbirk et al., 1998; Bernardi et al., 2003), several other studies have reported their production in the gastrointestinal tract and their absorption into the body (Elfoul et al., 2001; Johnson, 2002). Isothiocyanates promote the health of animals and humans by detoxifying chemical carcinogens through induction of Phase II detoxification enzymes and by exhibiting antimicrobial activity against pathogenic bacteria (Tierens et al., 2001; Johnson, 2002). Thus, the dietary HA-starch-induced increase in the production of isothiocyanates from glucosinolates in the gastrointestinal tract of pigs implies that HA-starch can be included in glucosinolates-containing diets for pigs to improve their health. Thiocyanate, the major metabolite of cyanide, is present in all biological samples due to the endogenous production of cyanide (Logue et al., 2010). For instance, Bhandari et al. (2012) reported the presence of endogenous thiocyanate in plasma of pigs at concentration of 8.2 to 46.6 μM. Thus, thiocyanate levels detected in cecal digesta and serum for pigs fed CPCC-free diets could have been due to the presence of endogenous thiocyanate in pigs. The 4-hydroxyglucobrassicin, which is the major aromatic glucosinolate found in CPCC, is degraded into thiocyanate at neutral pH. Also, progoitrin is degraded into goitrin at neutral pH, implying that the effect of reducing gastrointestinal pH through dietary inclusion of HA-starch on the degradation of progoitrin into goitrin can be inferred from the effect of reducing gastrointestinal pH though dietary inclusion of HA-starch on the degradation of 4-hydroxyglucobrassicin into thiocyanate.

In the current study, nitriles (indole-3-acetonitriles) were not detectable in the cecal digesta of pigs, which could have been due to increased degradation and absorption of nitriles in the upper part of the gastrointestinal tract. Frandsen et al. (2019) reported increased degradation of glucosinolates to nitriles at acidic pH found in the stomach of pigs both in the presence and absence of myrosinases, implying that the degradation of glucosinolates to nitriles is partly independent of the presence of myrosinases. In the same study, it was reported that nitriles are absorbed in the stomach and the upper part of the small intestine. Also, 4-hydroxyindole 3-acetonitrile and indole-3-acetonitrile are both derived from aromatic glucosinolates (indol-3-ylmethylglucosinolate) at acidic pH. Thus, a change in the production of indole-3-acetonitrile due to change in the reaction medium pH implies change (in the same direction) in the production of 4-hydroxyindole 3-acetonitrile.

In conclusion, dietary inclusion of CPCC resulted in reduced growth performance, increased metabolic activities of liver and interfered with thyroid gland functions of nursery pigs. Inclusion of HA-starch in diets for nursery pigs reduced growth performance, extent of thyroid gland enlargement, cecal and colonic digesta pH, and numerically reduced cecal digesta and serum concentrations of thiocyanate but increased colon weight relative to live BW, serum T4 concentration, and cecal and serum concentration of isothiocyanates. It appears that the adverse effects of dietary CPCC on thyroid gland functions of nursery pigs can be ameliorated by dietary HA-starch fed in the current study, which resulted in increased isothiocyanate and reduced thiocyanate productions in pigs. Since dietary HA-starch at 40% reduced growth performance of pigs in the current study, there is a need to identify optimal dietary level of HA-starch that does not compromise growth performance of pigs fed canola co-product-based diet, or to identify alternative strategy that can be used to reduce pH in the hindgut of pigs fed canola co-product-containing diet without compromising growth performance.

Conflict of interest statement

The authors declare no real or perceived conflicts of interest.

Glossary

Abbreviations

ADF

acid detergent fiber

ADFI

average daily feed intake

ADG

average daily gain

BW

body weight

CP

crude protein

CPCC

cold-pressed canola cake

DM

dry matter

EE

ether extract

G:F

gain-to-feed ratio

HA

-starch high-amylose cornstarch

NDF

neutral detergent fiber

NE

net energy

SBM

soybean meal

T3

triiodothyronine

T4

tetraiodothyronine