Hindgut fermentation of starch is greater for pulse grains than cereal grains in growing pigs

Abstract

The nutritive value of starch, the major source of dietary energy in pigs, varies depending on its susceptibility for digestion. The botanical origin of starch determines starch structure, and therefore, digestibility. To compare digestibility of starch, fiber, gross energy (GE), crude protein, and amino acid (AA), and to characterize undigested starch of grains in growing pigs, seven ileal-cannulated barrows (initial body weight, 30 kg) were fed six diets containing 96% of one of six test ingredients (three pulse grains: zero-tannin faba bean, green field pea, or mixed-cultivar chickpea; three cereal grains: hulled barley, hard red spring wheat, or hybrid yellow, dent corn), or a N-free diet in a 7 × 7 Latin square at 2.8 × maintenance digestible energy. Grain samples were ground with a hammer mill through a 2.78-mm screen. Amylose content ranged from 29% to 34% for pulse grains and from 22% to 25% for cereal grains. The apparent ileal digestibility (AID) of starch was greater (P < 0.05) in cereal (94% to 97%) than pulse grains (85% to 90%) and was lowest (P < 0.05) in faba bean (85.3%) followed by field pea (87.2%) and chickpea (90.1%). However, apparent total tract digestibility (ATTD) of starch of all tested grains was close to 100%. Apparent hindgut fermentability (AHF, as ATTD − AID) of starch was greater (P < 0.05) in pulse grains (9.9% to 15%) than cereal grains (3.3% to 4.8%). The AHF of total dietary fiber tended to be the greatest (P < 0.10) for corn (43.5%) and lowest for wheat (25.3%). The AHF of GE was greater (P < 0.05) in pulse grains (17% to 20%) than in cereal grains (9% to 11%) and resulted in greater (P < 0.05) digestible energy (DE) contribution from hindgut fermentation for pulse grains than cereal grains (0.9 vs. 0.5 Mcal/kg). Wheat had the greatest standardized ileal digestibility of total AA (90.2%; P < 0.05). Confocal laser scanning microscopy images revealed that 20% to 30% of starch granules of pulse grains were entrapped in protein matrixes. In scanning electron microscopy images, starch granules were larger in faba bean and field pea than cereal grains. Digesta samples revealed pin holes and surface cracks in starch granules of corn and wheat, respectively. In conclusion, hindgut fermentation of starch and fiber was greater in pulse grains than cereal grains resulting in a greater DE value despite lower ileal DE for pulse grain than cereal grains. Defining the digestible and fermentable fractions of starch may enhance the accuracy of equations to predict the net energy value of these feedstuffs.

Introduction

Feed cost represents the greatest part of variable costs of swine production (Woyengo et al., 2014). Pulse grains, such as field pea, faba bean, and chickpea, are widely grown on the Great Plains for human consumption and provide both starch and protein and are alternative feed ingredients to cereal grains, such as corn, wheat, and barley for swine. However, the largest nutritive component, starch, in pulse and cereal grains differs in structure, amylose to amylopectin ratio (Tiwari and Singh, 2012), and characteristics of the protein matrix associated with starch granules (Rooney and Pflugfelder, 1986; Baldwin, 2001). Starch structure may directly affect the rate and extent of starch digestion in the small intestine (Stevnebø et al., 2006). Starch content is included in the equation to predict net energy (NE) value of feedstuffs, but NE value may vary greatly depending on the extent of ileal digestion vs. hindgut fermentation of starch (Blok, 2006). Reduced nitrogen (N) retention in pigs fed slowly digestible starch (SDS) indicates that SDS that is fermented may not support maximum protein deposition, hence limiting growth performance of pigs (Drew et al., 2012). However, comparative starch digestibility in pulse and cereal grains is rarely reported.

Confocal laser scanning microscopy (CLSM) is a novel microscopic technique to investigate microstructural changes in food and grains (Han and Hamaker, 2002). However, most researches focused on intact starch granules and limited qualitative details are available on their undigested residues (Naguleswaran et al., 2011). Thus, investigation of microscopic morphology of starch granules in pulse and cereal grains, and their undigested residues may help elucidate their nutritive relevance to pigs.

The null hypotheses of the present study were 1) that apparent ileal digestibility (AID), apparent total tract digestibility (ATTD), and apparent hindgut fermentation (AHF) of starch in pulse grains would not differ from cereal grains and 2) morphological structure of starch and its undigested residues would not differ among pulse and cereal grains. The objectives were 1) to compare the AID and ATTD of starch, fiber, gross energy (GE), crude protein (CP), and amino acid (AA) in three cereal grains (wheat, barley, and corn) and three pulses (faba bean, field pea, and chickpea) and 2) to investigate the morphology of starch granules in three ground pulse and three ground cereal grains and their ileal digesta.

Materials and Methods

Animal use was approved, and experimental procedures were reviewed by the University of Alberta Animal Care and Use Committee for Livestock and followed the guidelines of the Canadian Council on Animal Care (CCAC, 2009). The animal study was conducted at the Swine Research and Technology Centre at the University of Alberta (Edmonton, AB, Canada).

Experimental design, animals, and housing

The study was conducted as a 7 × 7 Latin square with seven dietary treatments fed over seven 9-d periods. Seven crossbred barrows (initial body weight [BW] 30 kg; Duroc × Large White/Landrace; Hypor Inc., Regina, SK, Canada) were housed in individual metabolism pens (1.2 m wide, 1.2 m long, and 0.9 m high). Each pen was equipped with a self-feeder and a single adjustable-height cup drinker attached to the front of the pen. Room temperature was maintained at 22 ± 2.5 °C. Pigs were surgically fitted with a simple T-cannula at the distal ileum.

Diets and feeding

At day 18 postsurgery, the seven pigs were switched to feed the seven experimental diets as mash. Daily feed allowance was 2.8 × maintenance of energy (110 kcal digestible energy [DE] per kg BW^0.75) provided as two equal weight meals at 0830 and 1500. Six test diets were formulated by including one of six test ingredients as a sole source of carbohydrates and protein in each diet (Table 1). The three pulse grains were zero-tannin faba bean (cultivar Snowbird) and green field pea sourced from W. A. Grain and Pulse Solution (Innisfail, AB, Canada), and chickpea (Kabuli type; mix of cultivars CDC Frontier, CDC Orion, CDC Luna, and CDC Leader) sourced from Midwest Investments (Moose Jaw, SK, Canada). The three cereal grains were hulled barley (CDC Austenson, two row feed barley), hard red spring wheat (CDC Utmost) sourced from the Crop Development Centre (Saskatoon, SK, Canada), and hybrid yellow, dent corn. Whole grain samples were ground with a hammer mill through a 2.78-mm screen. A N-free diet was prepared to measure basal endogenous losses of CP and AA. Chromic oxide was included in diets at 0.5% as an indigestible marker. Vitamins and minerals were formulated to meet or exceed the estimated requirements for growing pigs (NRC, 2012).

Table 1.

Ingredient composition of experimental diets (as-fed basis)¹

Ingredient, % as fed	Test diets	N-free diet
Test ingredient1	95.6	—
Corn starch	—	84.6
Sugar	—	5.0
Solka Floc2	—	3.0
Canola oil	—	2.0
Limestone	1.3	1.3
Mono/dicalcium phosphate	1.1	1.1
Salt	0.5	0.5
Vitamin premix3	0.5	0.5
Mineral premix4	0.5	0.5
Cr2O3	0.5	0.5
K2CO3, 55% K	—	0.5
MgO, 58% Mg	—	0.1

¹Test ingredients were one of the following three pulse grains or three cereal grains, respectively: faba bean, field pea, chickpea, barley, wheat, and corn.

²Solka-Floc, International Fiber Corp., North Tonawanda, NY.

³Provided per kg diet: retinol, 2.5 mg; cholecalciferol, 20.6 µg; dl-α-tocopherol, 2.7 µg; niacin, 35 mg; d-pantothenic acid, 15 mg; riboflavin, 5 mg; menadione, 4 mg; folic acid, 2 mg; thiamin, 1 mg; d-biotin, 0.2 mg; and vitamin B12, 0.025 mg.

⁴Provided per kg diet: Zn, 100 mg as ZnSO₄; Fe, 80 mg as FeSO₄; Cu, 50 mg as CuSO₄; Mn, 25 mg as MnSO₄; I, 0.5 mg as Ca(IO₃)₂; and Se, 0.1 mg as Na₂SeO₃.

Experimental procedure

Each 9-d experimental period consisted of 5-d adaptation prior to specimen collections, followed sequentially by 2 d of feces collection and 2 d of ileal digesta collection. Feces were collected between 0830 and 1600 using plastic bags held between two snapped leather rings, the bottom one attached to a Velcro ring, which was spray-glued around the anus (van Kleef et al., 1994). Wet feces from individual pig were weighed, pooled for each period, and were frozen at below −20 °C. For digesta collection, plastic bags containing 15 mL of 5% formic acid were attached to the open T-cannula. Digesta were collected continuously from 0830 to 1800. Digesta bags were replaced when they were 50% to 70% full of digesta, or at least once every hour. Digesta samples from each pig observation were pooled and frozen at −20 °C. At the end of trial, frozen feces and digesta samples were thawed, homogenized, subsampled, and freeze-dried.

Chemical analyses

Ingredients, diets, and lyophilized feces and digesta were ground through a 1-mm screen using a centrifugal mill (Retsch ZM-1, Brinkmann Instruments Canada Ltd., ON, Canada) and analyzed for dry matter (DM; method 930.15), CP content (method 990.03; N × 6.25) as per AOAC (2006), and GE value using an adiabatic bomb calorimeter (model 5003; Ika-Werke GMBH & Co. KG, Staufen, Germany). Ingredients were analyzed for ash (method 942.05), ether extract (method 920.39A), acid detergent fiber (method 973.18), neutral detergent fiber (NDF; Holst, 1973), Ca (method 968.08), and P content (method 946.06) as per AOAC (2006). Ingredients, diets, and digesta were analyzed for AA content (method 982.30; AOAC, 2006). Diets, digesta, and feces were analyzed for total starch and soluble starch (method 996.11; AOAC, 2006) and total dietary fiber (TDF), soluble dietary fiber (SDF), and insoluble dietary fiber (IDF) content (method 985.29; AOAC, 2006) using assay kits (K-TSTA-100A and K-TDFR-200A, respectively; Megazyme, Wicklow, Ireland). Ingredients were analyzed for amylose using a method described previously (Hoover and Ratnayake, 2001) and resistant starch using an assay kit (K-RSTAR; Megazyme, Wicklow, Ireland) based on enzymatic analysis (method 2002.02; AOAC, 2006) and method 32-40.01; AACC International, 1976). Diets, digesta, and feces were analyzed for Cr₂O₃ content (Fenton and Fenton, 1979).

Microscopic analyses

The CLSM was conducted using a double-staining technique as previously described (Naguleswaran et al., 2011; Li et al., 2014). Using fluorescent dyes, aminofluorophore 8-amino-1,3,6-pyrenetrisulfonic acid (APTS) and Pro-Q Diamond phosphoprotein stain (Molecular Probes, Eugene, OR) were used to subjectively identify starch molecules and P-associated molecules (starch phosphate monoester, phospholipid, and inorganic phosphate) in ingredients and digesta samples (Naguleswaran et al., 2011). These samples (20 to 30 mg) were stained in 25 µL of APTS solution (20 mM APTS in 15% acetic acid), 25 µL of 1 M sodium cyanoborohydride, and 50 µL of pure ethanol at 30 °C for approximately 15 h. The APTS-stained samples were then washed with 0.5 mL 50% ethanol and centrifuge at 7,000 rpm for 5 min. Supernatant was removed and washing process was repeated for five times. After the final wash, 0.5 mL of Pro-Q Diamond solution was added and incubated for 1 h at room temperature. The washing process was repeated five times before finally suspension in 0.5 mL of 50% glycerol. Stained ingredient and digesta samples suspended in 50% glycerol (10 µL) were placed onto glass-bottom culture dish (MatTek Corporation, Ashland, MA), mixed with 0.1 mL of distilled water and then visualized under a CLSM (Zeiss LSM 710, Carl Zeiss MicroImaging, Jena, Germany) with a 40 × 1.3 oil objective lens. The excitation wavelengths were 488 and 561 nm with emission light interval of 490 to 563 nm. Images of optical sections of starch granules were captured with Zen 2009 software (CarlZeiss MicroImaging).

For scanning electron microscopy (SCEM), the ground samples were mounted onto aluminum pin stubs, coated with carbon, and examined in a field emission scanning electron microscope (Zeiss Sigma 300 VP, Carl Zeiss Microscopy GmbH, Jena, Germany).

Calculations and statistical analyses

ATTD and AID of individual nutrients were calculated using the following equation (Adeola, 2001):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{T}\mathrm{T}\mathrm{D}\mathrm{o}\mathrm{r}\mathrm{A}\mathrm{I}\mathrm{D}=\{1-[(\mathrm{m}\mathrm{a}\mathrm{r}{\ker }_{\mathrm{d}}/\mathrm{m}\mathrm{a}\mathrm{r}{\ker }_{\mathrm{f}})/({\mathrm{N}}_{\mathrm{d}}/{\mathrm{N}}_{\mathrm{f}})]\}\times 100}\end{array}}$$

where marker_d and marker_f are Cr₂O₃ content in diet and feces or digesta, respectively, and N_d and N_f are nutrient content in diet and feces or digesta on DM basis. AHF was calculated as ATTD − AID. The NE values for the six ingredients were calculated using equations 1 to 8 (NRC, 2012) that were adapted from Noblet et al. (1994).

The AID and ATTD of fiber were corrected for nondietary interfering material present in ileal digesta and feces using values from feeding a fiber-free diet from a previous study, with the assumption that the nondietary interfering material was of endogenous or bacterial origin (Montoya et al., 2015), and presented as standardized ileal digestibility (SID) and standardized total tract digestibility of fiber. The N-free diet was used to correct for basal endogenous losses of AA, using the following equation (Stein et al., 2007):

$${\displaystyle \begin{array}{l}{\displaystyle I_{\mathrm{e}\mathrm{n}\mathrm{d}}=A{A}_{\mathrm{d}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{a}}\times \left(\frac{\mathrm{m}\mathrm{a}\mathrm{r}{\ker }_{\mathrm{f}}}{\mathrm{m}\mathrm{a}\mathrm{r}{\ker }_{\mathrm{d}}}\right)}\end{array}}$$

where I_end is the basal endogenous loss of an AA. The SID of AA in the diets were then calculated with the following equation (Stein et al., 2005):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{S}\mathrm{I}\mathrm{D}=\mathrm{A}\mathrm{I}\mathrm{D}+\{(I_{\mathrm{e}\mathrm{n}\mathrm{d}}/\mathrm{A}{\mathrm{A}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}})\times 100\}}\end{array}}$$

Pig was the sampling and experimental unit. Data were analyzed using the MIXED procedure of SAS (version 9.4; SAS Inst. Inc., Cary, NC) with diet as fixed effect, and pig and experimental period as random factors. Data were checked for normality using the UNIVARIATE procedure of SAS. Means were separated by a multiple comparison procedure using the Tukey–Kramer method if diet effect was significant (P < 0.05). Means are reported as least-squares means, plus pooled standard error of means (SEM). A contrast statement was used to compare pulse grains vs. cereal grains. A P-value of < 0.05 was considered significant and 0.05 ≤ P < 0.10 was considered a trend. A principal component (PC) analysis was performed using JMP software of SAS (version 16.1.0; SAS Inst. Inc.). The loading plot of PC 1 and 2 was used to determine correlations among grain characteristics and their digestiblity variables and energy values.

Results

All pigs remained healthy throughout the experiment. Sufficient ileal digesta and feces were collected from each pig for all period to properly quantify variables.

Chemical composition and digestibility of starch and fiber

Total starch content was greater in cereal (56% to 61%) than pulse grains (36% to 39%; Table 2). Amylose content of starch was greater in pulse grains (29% to 34%) than cereal grains (23% to 25%). Similarly, resistant starch content was greater in pulse grains (20% to 50%) than cereal grains (23% to 31%). The NDF content was greater for cereal grains (12% to 17%) than pulse grains (7.2% to 12%).

Table 2.

Analyzed chemical composition of the three pulse grains and three cereal grains ingredients fed to growing pigs (dry matter basis)¹

	Pulse grains			Cereal grains
Item2	Faba bean	Field pea	Chickpea	Barley	Wheat	Corn
Moisture, %	9.86	11.30	9.83	11.00	11.40	11.50
GE, Mcal/kg	4.28	4.28	4.49	4.23	4.28	4.23
Total starch, %	36.3	38.6	36.0	56.0	59.2	60.5
Resistant starch, % total starch	50.2	37.5	19.7	30.7	30.5	22.5
Amylose, % total starch	33.6	29.3	32.9	25.0	23.1	22.8
NDF, %	12.0	11.0	7.2	17.2	11.5	12.9
ADF, %	9.28	6.72	4.63	6.02	3.04	4.36
CP (N × 6.25), %	30.2	24.8	24.4	12.0	18.7	10.6
Ether extract, %	2.00	4.68	4.33	2.29	1.99	2.44
Ash, %	3.29	2.85	3.31	2.56	2.00	1.72
P, %	0.38	0.35	0.40	0.36	0.42	0.22
Ca, %	0.11	0.08	0.10	0.07	0.05	0.10
Indispensable AA, %
Arg	2.48	1.89	2.17	0.50	0.80	0.37
His	0.84	0.65	0.66	0.26	0.46	0.28
Ile	1.32	1.08	1.06	0.41	0.64	0.37
Leu	2.11	1.12	1.69	0.55	0.60	0.72
Lys	1.97	1.85	1.69	0.47	0.54	0.35
Met	0.24	0.22	0.35	0.20	0.29	0.18
Phe	1.32	1.22	1.39	0.59	0.86	0.50
Thr	1.00	0.87	0.80	0.37	0.48	0.35
Trp	0.25	0.23	0.24	0.11	0.20	0.06
Val	1.43	1.18	1.09	0.57	0.75	0.48
Dispensable AA, %
Ala	1.22	1.03	0.99	0.46	0.59	0.71
Asp	3.25	2.73	2.62	0.65	0.87	0.75
Cys	0.35	0.31	0.34	0.23	0.34	0.18
Glu	4.83	3.99	3.77	2.52	5.36	1.81
Pro	1.22	0.98	0.95	1.13	1.71	0.80
Ser	1.19	0.96	1.03	0.41	0.68	0.40
Tyr	0.88	0.68	0.52	0.23	0.46	0.21
Particle size, µm	473	506	511	481	509	576

¹One composite ingredient sample was collected by taking multiple samples across the single batch, and ground. Samples were analyzed in duplicate.

²GE, gross energy; NDF, neutral detergent fiber; ADF, acid detergent fiber; CP, crude protein.

Ileal starch content was mostly insoluble and was greatest (P < 0.05; Table 3) in pigs fed field pea. The AID of starch was greater (P < 0.001) in cereal grains than pulse grains. The ATTD of starch was above 99% for all 6 grains. Hindgut fermentability of starch was greater (P < 0.001) for pulse grains than cereal grains. Greatest (P < 0.05) hindgut fermentability was for faba bean and field pea, followed by chickpea.

Table 3.

Starch content in diet, digesta, and feces, and apparent ileal digestibility (AID), apparent total tract digestibility (ATTD), and apparent hindgut fermentability (AHF) of starch in diets fed to pigs including three pulse grains and three cereal grains (dry matter basis)

	Pulse grains			Cereal grains				P-value
Item	Faba bean	Field pea	Chickpea	Barley	Wheat	Corn	SEM1	Diet	Pulses vs. cereals
Total starch, g/kg
Diet	428	449	417	570	639	680
Ileal digesta	187ab	214a	143b	94c	163b	91c	13.3	<0.001	<0.001
Feces	3.0	7.8	4.9	3.6	2.8	3.0	1.5	0.120	0.109
Soluble starch, g/kg
Diet	1.0	1.3	1.5	1.4	1.6	1.7
Ileal digesta	1.9b	2.4ab	1.4b	2.9ab	4.5a	2.5ab	0.62	<0.001	0.013
Insoluble starch2, g/kg
Diet	427	448	415	569	638	678
Ileal digesta	185ab	212a	142b	91c	158b	88c	13.2	<0.001	<0.001
Starch digestibility, %
AID	85.3c	87.2bc	90.1b	95.1a	93.9a	96.6a	0.97	<0.001	<0.001
ATTD	99.9	99.9	99.9	99.9	99.9	99.9	0.03	0.090	0.162
AHF3	14.6a	12.7ab	9.83b	4.77c	6.01c	3.31c	0.95	<0.001	<0.001

¹Least squares means based on seven pig observations per diet.

²Calculated as total starch minus soluble starch.

³ATTD − AID, apparent hindgut fermentability.

^a–cMeans within a row without a common superscript differ (P < 0.05).

Total dietary fiber content was greater in pulse grains than cereal grains (Table 4). Ileal TDF content was greater (P < 0.001) in pigs fed cereal grains than pigs fed pulse grains. The AID and SID of TDF was greatest (P < 0.05) for field pea and chick pea, followed by wheat, faba bean, barley, and corn. Hindgut fermentability of TDF did not differ between cereal grains and pulse grains. Hindgut fermentability of SDF and IDF ranged from 44% to 73% and 16% to 43%, respectively. Pigs fed barley had the greatest hindgut fermentability of SDF (P < 0.05) but had the lowest (P < 0.05) hindgut fermentability of IDF.

Table 4.

Fiber content in diets and digesta of pigs fed test diets including three pulse grains and three cereal grains (dry matter basis)

	Pulse grains			Cereal grains				P-value
Item	Faba bean	Field pea	Chickpea	Barley	Wheat	Corn	SEM1	Diet	Pulses vs. cereals
Total dietary fiber (TDF)2, g/kg
Diet	230	252	269	213	189	142
Ileal digesta	441bc	404c	402c	523a	486ab	541a	20.0	<0.001	<0.001
Soluble dietary fiber (SDF), g/kg
Diet	13	11	22	27	29	12
Ileal digesta	51c	64c	67bc	111a	90ab	53c	6.2	<0.001	<0.001
Insoluble dietary fiber (IDF), g/kg
Diet	218	240	248	186	161	130
Ileal digesta	390bc	340c	335c	412b	395bc	488a	17.6	<0.001	<0.001
AID, %
TDF	29.7b	52.5a	52.5a	22.0b	36.6ab	−0.8c	5.99	<0.001	<0.001
SDF	−37.2c	−46.4c	11.7ab	−26.9bc	28.2a	−7.0abc	10.22	<0.001	0.012
IDF	34.2b	57.8a	56.8a	20.0b	39.3ab	0.6c	6.02	<0.001	<0.001
ATTD, %
TDF	64.0b	81.6a	82.8a	42.7d	54.9c	32.9e	1.99	<0.001	<0.001
SDF	56.0d	59.5cd	79.9ab	82.5a	86.2a	69.8bc	3.07	<0.001	<0.001
IDF	64.1b	82.1a	82.7a	36.8d	49.2c	29.1e	2.20	<0.001	<0.001
SID3, %
TDF	34.9b	57.3a	57.0a	27.7bc	43.0ab	7.7c	5.99	<0.001	<0.001
SDF	9.5abc	5.2bc	38.9ab	−4.7c	48.8a	41.4ab	10.22	0.018	0.211
IDF	37.0b	60.4a	59.3a	33.3b	43.1ab	5.39c	6.02	<0.001	<0.001
STTD3, %
TDF	76.3b	92.8a	93.3a	55.9c	69.8b	52.7e	1.99	<0.001	<0.001
SDF	64.2c	68.5bc	84.6a	86.4a	89.8a	78.3ab	3.07	<0.001	<0.001
IDF	76.6b	93.5a	93.7a	51.4d	66.2c	50.1d	2.20	<0.001	<0.001
STTD− SID4, %
TDF	39.8	34.0	34.7	26.7	25.3	43.5	5.15	0.064	0.262
SDF	53.6ab	62.3ab	44.7ab	73.4a	40.0ab	35.9b	8.15	0.026	0.839
IDF	37.4ab	31.1abc	33.2abc	16.0c	20.9bc	42.5a	4.72	0.003	0.074

¹Least squares means based on seven pig observations per diet.

²Calculated as soluble dietary fiber plus insoluble dietary fiber.

³AID and ATTD values of fiber were corrected for interfering materials present in the ileal digesta and feces using values from a fiber-free diet (Montoya et al., 2015), with the assumption that these interfering materials were of endogenous or bacterial origin.

⁴STTD − SID, hindgut fermentability.

^a–dMeans within a row without a common superscript differ (P < 0.05).

Digestibility of energy

The AID of GE was greater (P < 0.01; Table 5) in cereal grains than pulse grains with the greatest value (P < 0.05) for wheat. In contrast, the ATTD of GE was greater (P < 0.001) in pulse grains than cereal grains. The ATTD of GE was greatest (P < 0.05) in field pea and chickpea and lowest in barley. The hindgut fermentability of GE was greater (P < 0.001) for pulse grains than cereal grains. The total tract DE value was greatest (P < 0.05) for chickpea and lowest for barley. The DE value from hindgut fermentation was greater (P < 0.001) for pulse grains than cereal grains with the greatest value (P < 0.05) for faba bean and chickpea, followed by field pea. The calculated NE value was lower (P < 0.001) for pulse grains than cereal grains.

Table 5.

Apparent ileal digestibility (AID), apparent total tract digestibility (ATTD), and apparent hindgut fermentability (ATTD − AID) of nutrients and gross energy (GE), and digestible energy (DE) value (dry matter [DM] basis) in diets including three pulse grains and three cereal grains fed to pigs and calculated net energy (NE) value (DM basis) of these six ingredients

	Pulse grains			Cereal grains				P-value
Item	Faba bean	Field pea	Chickpea	Barley	Wheat	Corn	SEM1	Diet	Pulses vs. cereals
Dry matter
AID	65.2d	72.5bc	71.0bc	69.3c	76.2a	73.7ab	1.69	<0.001	0.009
ATTD	86.7b	91.5a	91.0a	81.9c	86.6b	85.5b	0.44	<0.001	<0.001
ATTD − AID	21.6a	19.0a	20.1a	12.4b	10.4b	11.8b	1.62	<0.001	<0.001
GE
AID	68.2d	74.4b	71.7c	71.4c	78.1a	74.4b	1.65	<0.001	0.016
ATTD	88.1b	91.8a	90.4a	82.5d	87.2bc	85.6c	0.44	<0.001	<0.001
ATTD − AID	19.9a	17.3a	18.7a	11.1b	9.2b	11.2b	1.54	<0.001	<0.001
Crude protein
AID	78.9a	81.2a	70.8b	66.8b	81.6a	59.8c	1.50	<0.001	<0.001
ATTD	91.5a	90.3ab	87.6c	76.9e	88.3bc	79.9e	0.66	<0.001	<0.001
ATTD − AID	12.6bc	9.0cd	16.8ab	10.1cd	6.7d	19.9a	1.50	<0.001	0.575
DE, Mcal/kg
Ileal	2.83d	3.13b	3.20ab	2.99c	3.30a	3.16b	0.072	<0.001	0.085
Total tract	3.76c	3.94b	4.12a	3.53e	3.76c	3.68d	0.017	<0.001	<0.001
Hindgut	0.93a	0.81b	0.92ab	0.54c	0.46c	0.51c	0.070	<0.001	<0.001
Calculated NE,2 Mcal/kg	2.09d	2.26c	2.43a	2.23c	2.34b	2.35b	0.010	<0.001	<0.001

¹Least squares means based on seven pig observations per treatment.

²Ingredient NE value calculated from diet DE value (×0.956) and analyzed macronutrient of ingredients using equation 5 developed by Noblet et al. (1994) and adapted as equation 1 to 8 by NRC (2012): NE = 0.700 × DE + 1.61 × ether extract + 0.48 × starch − 0.91 × crude protein − 0.87 × acid detergent fiber (energy values expressed as kcal/kg DM and nutrient content as g/kg DM).

^a-dMeans within a row without a common superscript differ (P < 0.05).

Digestibility of AA and protein

The AID and ATTD of CP was greater (P < 0.001; Table 5) in pulse grains than cereal grains. The AID of CP was greatest (P < 0.05; Table 6) for wheat, field pea and faba bean, intermediate for chickpea and barley, and lowest for corn. The ATTD of CP was greatest (P < 0.05) for faba bean and field pea and lowest (P < 0.05) for barley. The hindgut fermentability of CP was greatest (P < 0.05) for corn and lowest for wheat. The AID of Arg, His, Lys, Thr, Ala, Asp, Pro, and total AA was greater (P < 0.01) for pulse grains than cereal grains. The AID of indispensable AA was greatest (P < 0.05) in faba bean except for Met and Phe and in wheat except for Leu, Lys, and Thr. The AID of Lys and Thr were greatest (P < 0.05) for faba bean and field pea diets and lowest for corn. The AID of Arg and Trp was lowest (P < 0.05) for corn, and the AID of Leu, Met, Phe, and Val was lowest for field pea.

Table 6.

Apparent ileal digestibility of crude protein (CP) and amino acids (AA) in three pulse grain and three cereal grain diets fed to pigs (dry matter basis)

	Pulse grains			Cereal grains				P-value
Item	Faba bean	Field pea	Chickpea	Barley	Wheat	Corn	SEM1	Diet	Pulses vs. cereals
CP, %	78.9a	81.2a	70.8b	66.8b	81.6a	59.8c	1.50	<0.001	<0.001
Indispensable AA, %
Arg	88.5a	83.9ab	77.9bc	69.1c	81.1ab	53.7d	2.27	<0.001	<0.001
His	84.7a	80.9a	73.0b	74.0b	84.7a	72.2b	1.04	<0.001	0.005
Ile	78.7a	73.5b	59.8c	68.5c	80.9a	65.2c	0.86	<0.001	0.200
Leu	81.3a	71.2bc	51.8d	68.0c	73.9b	74.0b	0.90	<0.001	<0.001
Lys	82.6a	80.2a	65.7b	55.1c	66.4b	40.5d	1.63	<0.001	<0.001
Met	69.6c	74.0bc	60.4d	71.7bc	82.0a	75.5b	1.31	<0.001	<0.001
Phe	80.2b	75.9c	65.2d	74.8c	84.8a	75.2c	0.87	<0.001	<0.001
Thr	74.0a	70.0ab	58.2c	57.4c	69.1b	44.4d	1.08	<0.001	<0.001
Trp	79.0ab	71.5c	74.5bc	76.7bc	84.9a	59.9d	1.82	<0.001	0.431
Val	76.8a	71.3b	59.4d	65.7c	76.9a	61.8d	0.84	<0.001	0.138
Dispensable AA, %
Ala	74.7a	71.3ab	59.0c	55.1c	69.5ab	68.8b	1.39	<0.001	<0.001
Asp	81.8a	76.3b	63.5d	56.7e	71.3c	62.0d	1.07	<0.001	<0.001
Cys	58.5cd	61.6c	34.5e	67.6b	79.7a	55.5d	1.11	<0.001	<0.001
Glu	85.2b	80.1c	68.8d	81.1c	91.3a	79.2c	0.86	<0.001	<0.001
Pro	74.4a	64.9a	56.0a	36.1a	84.3a	-52.7b	13.8	<0.001	<0.001
Ser	79.0a	72.4b	59.0d	64.2c	80.7a	59.7d	0.88	<0.001	0.007
Tyr	78.2ab	75.3b	63.2d	68.8c	81.7a	70.3c	0.97	<0.001	0.082
Total AA, %	80.2a	75.4b	63.9c	64.7c	82.5ab	54.7d	1.65	<0.001	<0.001

¹Least squares means based on seven pig observations per diet.

^a–cMeans within a row without a common superscript differ (P < 0.05).

The SID for CP was greatest (P < 0.05; Table 7) for wheat and faba bean, then field pea, intermediate for barley and chickpea, and lowest for corn. The SID of Arg, Lys, Thr, and Asp was greater (P < 0.001) for pulse grains than cereal grains. The SID of indispensable AA was greatest (P < 0.05) in wheat except for Lys, and in faba bean except for Ile, Met, and Phe. The SID of Lys was greatest (P < 0.05) in faba bean and field pea. The SID of Arg, Lys, and Thr in field pea and Trp in barley was also among the highest. The SID of indispensable AA, except for Ile, Met, Phe, and Val, was the lowest (P < 0.05) in corn.

Table 7.

Standardized ileal digestibility of crude protein (CP) and amino acids (AA) in three pulse grains and three cereal grains

	Pulse grains			Cereal grains				P-value
Item	Faba bean	Field pea	Chickpea	Barley	Wheat	Corn	SEM1	Diet	Pulses vs. Cereals
CP, %	86.5ab	86.3b	75.3c	76.4c	90.5a	69.7d	1.46	<0.001	0.310
Indispensable AA, %
Arg	91.0a	87.2ab	80.6b	80.7b	88.9ab	69.6c	2.30	<0.001	<0.001
His	87.1ab	83.9bc	75.9e	80.7cd	88.9a	78.9de	1.08	<0.001	0.557
Ile	80.8b	76.1c	62.4d	74.7c	85.1a	73.0c	0.88	<0.001	<0.001
Lys	84.7a	82.4a	68.2bc	63.7c	74.3b	52.9d	1.70	<0.001	<0.001
Met	73.1c	77.2bc	62.3d	75.6bc	84.7a	79.7b	1.32	<0.001	<0.001
Phe	82.3b	78.1c	67.1d	79.2bc	87.9a	80.5bc	0.91	<0.001	<0.001
Thr	79.2a	75.9a	64.6c	70.9b	80.1a	60.0c	1.37	<0.001	<0.001
Trp	82.7ab	75.6cd	77.9bc	83.1ab	88.9a	70.5d	1.92	<0.001	0.164
Val	80.0a	75.0b	63.4d	73.4bc	82.8a	71.0c	0.93	<0.001	<0.001
Dispensable AA, %
Ala	79.1a	76.5a	64.2b	66.3b	78.5a	76.5a	1.42	<0.001	0.597
Asp	84.0a	78.8b	66.1d	66.9d	79.3b	71.4c	1.06	<0.001	<0.001
Cys	64.9c	68.2c	40.6d	76.3b	85.6a	66.9c	1.34	<0.001	<0.001
Glu	86.9b	82.1c	70.9d	84.2bc	92.7a	83.6bc	0.88	<0.001	<0.001
Pro	120.1a	122.4a	112.9a	83.9a	115.7a	15.5b	14.93	<0.001	<0.001
Ser	82.8b	77.2c	63.4e	75.2c	87.5a	71.5d	1.02	<0.001	<0.001
Tyr	80.9b	78.6bc	67.1d	76.0c	86.6a	77.2bc	1.01	<0.001	<0.001
Total AA, %	85.1ab	81.3bc	69.6d	76.9c	90.2a	68.7d	1.79	<0.001	0.877

¹Least squares means based on seven pig observations per ingredient.

^a–dMeans within a row without a common superscript differ (P < 0.05).

CLSM and SCEM of ingredients and ileal digesta

In pulse grains (faba bean, field pea, and chickpea), 20% to 30% of the starch granules were entrapped in the protein matrix (red color; Figure 1A–C). Around 40% to 50% of the large granules of faba bean, field pea, and chickpea contained prominent central cavities or cracks. The overlay CLSM images of wheat, barley, and corn grain showed that most starch granules (green color) existed as individual granules (Figure 1D–F). The starch granules of corn were irregular and polyhedral in shape compared with other grains. In ileal digesta, some large pieces of fiber, mainly cell wall and proteins, were stained green, orange, and red depending on the concentration of P and reducing sugars in the matrix (Figure 2). Approximately 20% to 40% fewer individual starch granules were observed in all digesta samples compared with ingredients. In digesta from pigs fed faba bean, small starch granules embedded in the protein matrix were observed (Figure 2A). The SCEM images revealed that most starch granules in ground faba bean and field pea were larger in size and clearly separated from nonstarch components (Figure 3A and B). In contrast, starch granules in ground chickpea, wheat, and barley were smaller and more likely to be surrounded by nonstarch components (Figure 3C–E). The scans revealed that starch granules in ground corn were mostly in the form of aggregates instead of individual granules (Figure 3F). Pitted surfaces or pin holes in the starch granules of corn (Figure 4B) and cracks on the surface of wheat starch granules (Figure 4C) were observed in ileal digesta.

Figure 1.

Confocal laser scanning micrographs of three pulse grains and three cereal grains stained with aminofluorophore 8-amino-1,3,6-pyrenetrisulfonic acid and Pro-Q Diamond phosphoprotein stain. Panels include three pulse grains: (A) faba bean, (B) field pea, and (C) chickpea, and three cereal grains; (D) barley, (E) wheat, and (F) corn. Starch granules are stained fluorescent green (E and F; arrow). Nonstarch components (protein, lipid, or fiber) are stained red. Starch granules embedded in the protein matrix (red color) could be easily distinguished (B; arrow).

Figure 2.

Confocal laser scanning micrographs of ileal digesta from pigs fed diets including three pulse grains and three cereal grains stained with aminofluorophore 8-amino-1,3,6-pyrenetrisulfonic acid and Pro-Q Diamond phosphoprotein stain. Panels include three pulse grains: (A) faba bean, (B) field pea, (C) chickpea, and three cereal grains; (D) barley, (E) wheat, and (F) corn. Starch granules are stained fluorescent green. Nonstarch components (protein, lipid, or fiber) are stained varying shades of orange, red, yellow, or brown.

Figure 3.

Scanning electron microscopic images (500×) of three pulse grains and three cereal grains. Panels include three pulse grains: (A) faba bean, (B) field pea, (C) chickpea, and three cereal grains; (D) barley, (E) wheat, and (F) corn. Starch granules were larger in size for faba bean and field pea.

Figure 4.

Scanning electron microscopic images (2,500×) of digesta of pigs fed diets containing selected pulse grains and cereal grains. Panels include (A) digested and partially digested faba bean, (B) pin holes on starch granules of corn, and (C) surface crack on starch granules of wheat.

PC analysis

In the loading plot, amylose and resistant starch were closely associated in a cluster that was positively correlated with AHF of starch and hindgut DE but negatively correlated with AID of starch, total starch, and calculated NE value (Figure 5). The TDF was closely related to ATTD of GE. The AID of starch is positively correlated with AID of GE, ileal DE, and calculated NE value.

Figure 5.

Loading plot of principal component (PC) analysis showing the correlations among chemical characteristics of pulse and cereal grains of interest (solid arrows; including total dietary fiber [TDF]), digestibility of nutrients (dashed arrows; including apparent ileal digestibility [AID]; apparent total tract digestibility [ATTD]; apparent hindgut fermentability [AHF]; and standardized ileal digestibility [SID]), and digestibility of energy (semidashed arrows; including energy values) of the first two eigenvalues (PC 1 and PC 2). Angles between lines describe associations: the length, direction, and angle between arrows indicates the relationship between variables or between variables and PC axes (e.g., α = 0° and r = 1; α = 90° and r = 0; and α = 180° and r = −1). Percentages on x and y axes indicate proportions of variability of data that are described with the corresponding PC in the model.

Discussion

In the present study, the AID of starch was lower in pulse grains than cereal grains, but the ATTD of starch was nearly 100% for all grains. The AHF of starch was greater in pulse grains than cereal grains. The CLSM images used to study morphological structure revealed that starch granules in pulse grains were larger and embedded in protein matrixes.

Composition and digestibility of starch and fiber

Starch constitutes the major fraction of carbohydrates in pulse and cereal grains. Although α-amylase and brush-border enzymes can digest most of the starch in the small intestine in pigs, a part of starch is resistant to digestion (Englyst et al., 1992). Greater AID of starch for cereal grains than pulse grains indicated that starch digestion in the small intestine was more effective for cereal grains than pulse grains. Previous studies reported AID of starch averaging 90% for cereal grains (Bach Knudsen et al., 2006) and ranging between 75% and 90% for pulse grains such as field pea (Sun et al., 2006). The ATTD of starch was close to 100% for both pulse grains and cereal grains indicating that undigested starch was almost completely fermented in the hindgut. Hindgut fermentation of starch was greatest in faba bean and field pea likely because greater amylose:amylopectin ratio in pulse grains than cereal grains (Woyengo et al., 2014, Hoover and Zhou, 2003). The crystalline packing of double helices differs among botanical sources of starch. Cereal grains exhibit A-type with crystallites more densely packed than B-type. In contrast, pulse grains exhibit C-type, which is a mix of A-type and B-type (Singh et al., 2008). Previous studies reported that C-type crystalline structure in starch granules of pulse grains have lower digestibility than A-type structures in cereal grains (Singh et al., 2008; Magallanes-Cruz et al., 2017; Martens et al., 2018). Consistent with previous studies, the amylose content of pulse grains ranged 30% to 40% in contrast to cereal grains that contain around 25% amylose (Singh et al., 2010). Specifically, the amylose content was previously reported to be 21% to 58% in field pea starch (Singh et al., 2010), 32% in faba bean starch (Hoover and Sosulski, 1986), and 28% to 34% in chick pea starch (Lineback and Ke, 1975; El-Faki et al., 1983; Singh et al., 2004). In pulse grains containing more amylose, the longer, straight chains of amylose compared with shorter, branched amylopectin is associated with increased starch resistance to hydrolysis by digestive enzymes (Magallanes-Cruz et al., 2017). Dietary amylose that resists enzymatic digestion in the small intestine may stimulate hindgut fermentation of starch that favors production of butyrate and thereby alters gut microbiota composition by promoting the proliferation of beneficial bacteria such as Bifidobacteria spp. (Martinez et al., 2010; Regmi et al., 2011; Fouhse et al., 2015).

The three tested pulse grains contained 23% to 27% TDF and had a lower AID of starch. In a previous study, pulse grains (black bean, red kidney bean, lentil, navy bean, black-eyed pea, split pea, and northern bean) with high TDF (mean 36.5%) had a lower in vitro ileal digestibility of starch (21%) than cereal grains (Bednar et al., 2001). Results from the present study indicated that the TDF content of the grains did explain part of the variation in AID of starch indicating that factors other than fiber played a role in starch digestion. The calculated values of AID of SDF were mostly negative likely because of the presence of nondietary material in the gut content, such as microbial cells and mucins that may have interfered in the dietary fiber determination (Montoya et al., 2015). The fiber digestibility values were corrected for these endogenous, nondietary interfering material; therefore, the reported fiber digestibility data are physiologically more relevant.

Digestibility of energy

Given that starch is the main macronutrient by mass in pulse and cereal grains, the AID of DM in diets were reflected in the AID of GE. In the present study, field pea and chickpea had the greatest ATTD of GE, followed by faba bean, wheat, barley, and corn. The greatest AID of GE in wheat (78%) matched its greatest AID of TDF and high AID of starch likely because the greater amount of amylopectin in wheat starch is more rapidly digested than amylose (Rooney and Pflugfelder, 1986).

Fermentation products, mainly volatile fatty acids (VFA), can provide energy to the host (Stein et al., 2005). Indeed, carbohydrate fermentation can cover up to 15% of maintenance energy requirement in growing pigs (Varel and Yen, 1997). The present study indicated that the value of energy digested was lower from the hindgut than from the ileum (0.4 to 0.9 vs. 2.8 to 3.3 Mcal/kg) among grains, but the value of energy digested was greater in pulse grain than cereal grains. Thus, hindgut fermentation of pulse grain starch contributed to its DE value. However, hindgut fermentation of starch to VFA is at least 14% less efficient in yielding NE than enzymatic digestion of starch to disaccharides and glucose (Jørgensen et al., 1997; Gerrits et al., 2012). Equations to predict the NE value of feed ingredients based on DE value and macronutrient content (NRC, 2012) was adapted from Noblet et al. (1994). Thus, using these equations including total starch content for starch-rich ingredients containing fermentable starch may overestimate their NE value because 10% to 15% of total starch in pulse grains is fermented and not digested (Fouhse and Zijlstra, 2017). The quantity of energy digested in the hindgut can also be attributed to the increase TDF inclusion in the diets (Huang et al., 2015). Similarly, in the present study the greater TDF content and hindgut fermentation of TDF of pulse grains likely contributed to its DE value.

Apparent and SID of AA and CP

The protein content and AA composition of the grains varies with cultivar and agronomic conditions (Singh et al., 2007). Pulse protein mainly comprise of 70% globulins and 10% to 20% albumin (Boye et al., 2010). In vitro protein digestibility of raw faba bean, field pea, and chick pea ranged from 34% to 80% (Bessada et al., 2019). The combination of dietary soluble and insoluble fibrous components may partly explain the variation observed in the SID of AA in the grains (Chen et al., 2014). In the present study, wheat contained the most soluble fiber (2.9%) and had the greatest AID and ATTD of soluble fiber, and SID of total AA. Soluble fiber is more quickly fermented in the small intestine and can decrease digesta passage rate allowing for greater enzymatic digestion and thereby increase AID of AA (Hooda et al., 2011; Chen et al., 2014). Insoluble fiber is associated with increased rate of passage reducing time for aminopeptidase to digest substrate thereby reducing ileal AA digestibility (Souffrant, 2001; Hooda et al., 2011). In the present study, pigs fed corn had the greatest insoluble fiber content in digesta that may explain the lower SID of CP and AA than observed for other grains. In addition, the ATTD of CP was lower in pigs fed corn with high resistant starch than pigs fed regular corn, brown rice, and sticky rice (Li et al., 2007). In the present study, cereal grains contained more starch than pulse grains. The ATTD of CP however, increased in inverse order, with faba bean having the greatest ATTD of CP at 91.5%. Combined, these examples align with the studies, indicating that starch type (in terms of amylose content, protein-starch interface, and starch granules types) affect the ATTD of CP (Li et al., 2007; Gunawardena et al., 2010).

CLSM and SCEM of ingredients and ileal digesta

Morphology of starch granules (size, shape, and structure) varies depending on plant species (Singh et al., 2007). The CLSM and SCEM images in the present study showed similarity for wheat, barley, and corn with smaller starch granules than the pulse grains. Smaller starch granules have larger surface area, pores, and channels that enhance water uptake (Cornejo-Ramírez et al., 2018). Smaller starch granules of barley and wheat grain support faster enzymatic digestion than larger pulse starch granules with a smaller surface area (Lindeboom et al., 2004, Tester et al., 2006). Surface characteristics such as presence of equatorial grooves in starch granules of wheat, barley (Fannon et al., 1992), and corn (Li et al., 2001) or cracks in starch granules can enhance starch hydrolysis by facilitating amylase access (Li et al., 2007). The surface of starch granules in cereal grains was relatively smooth when viewed under SCEM similar to previous research (Vasanthan et al., 2012). In contrast, CLSM scans of pulse grains showed prominent cracks that could be due to mechanical grinding. Nevertheless, these surface cracks likely indicated the different stages of digestion in the collected digesta samples. Pitted surfaces (pin holes) observed for some of the pulse and cereal grains could be due to the enzymatic degradation. Digestion of starch granules starts at surface pores and interior channels, which allows for α-amylase to enter the interior and digest the granule gradually from the inside out (Zhang et al., 2006).

Starch granule size, shape, and surface features may influence its functionality and digestibility (Magallanes-Cruz et al., 2017). The lower AID of starch in pulse grains indicates that the rate of starch digestion is attributed not solely to morphology but also to a combination of other factors. Recently, cell wall permeability was identified as the rate-limiting step for digestion of starch in intact grains (Li et al., 2019). However, structural features of starch granules such as cracked surface may control the digestion kinetics in grains that was processed mechanically. Pulse grains contain more dietary protein and fiber than cereal grains; thus, CLSM images of faba bean, field pea, and chickpea showed more prominent staining of starch-associated components such as protein, fiber, and lipid. The increased presence of these components may reduce enzymatic access to starch granules and thereby reduce ileal digestibility of starch to varying extent (Singh et al., 2007).

In conclusion, cereal grains (barley, wheat, and corn) had greater ileal digestibility of starch than pulse grains (faba bean, field pea, and chickpea). The larger starch granules and presence of starch-associated components such as protein and fiber in pulse grains may reduce ileal digestibility of pulse grain starch. Although starch in pulse and cereal grains was almost completely digested at the total tract, more of pulse grain starch was fermented in the hindgut contributing to their DE value. Fermentation of starch to VFA is known to be less efficient at yielding energy compared with digestion of starch to glucose. Thus, starch with varying rate and site of digestion may influence its dietary energy value. Defining the digestible and fermentable fractions of starch would enhance the accuracy of equations to predict the NE value of feedstuffs.

Glossary

Abbreviations

AA

amino acid

AHF

apparent hindgut fermentability

AID

apparent ileal digestibility

APTS

aminofluorophore 8-amino-1,3,6-pyrenetrisulfonic acid

ATTD

apparent total tract digestibility

BW

body weight

CLSM

confocal laser scanning microscopy

CP

crude protein

DE

digestible energy

DM

dry matter

GE

gross energy

IDF

insoluble dietary fiber

NDF

neutral detergent fiber

NE

net energy

PC

principal component

SCEM

scanning electron microscopy

SDF

soluble dietary fiber

SID

standardized ileal digestibility

STTD

standardized total tract digestibility

TDF

total dietary fiber

Conflict of interest statement

The authors declare no real or perceived conflicts of interest.