The effect of amylase supplementation on individual variation, growth performance, and starch digestibility in broiler chickens

Abstract

The objective of this study was to evaluate the variance of starch digestibility in broilers individually fed diets without or with supplemental exogenous amylase. A total of 120 d-of-hatch male chicks were individually reared from 5 to 42 d in metallic cages and fed maize-based basal diets or diets containing 80 kilo-novo-α-amylase units/kg (60 birds or replicates per treatment). Beginning on d 7, feed intake, body weight gain, and feed conversion ratio were recorded; partial excreta collection was conducted every Monday, Wednesday, and Friday until 42 d, when all birds were sacrificed for individual collection of duodenal and ileal digesta. Lower feed intake (4,675 vs. 4,815 g) and feed conversion ratio (1.470 vs. 1.508) were observed in amylase-fed broilers during the overall period (7–43 d; P < 0.01), whereas body weight gain was not affected. Amylase supplementation improved total tract starch (TTS) digestibility (P < 0.05) on each day of excreta collection (except for d 28, where no difference was found), averaging 0.982 vs. 0.973 compared to basal-fed broilers from d 7 to 42. Both apparent ileal starch (AIS) digestibility and apparent metabolizable energy (AME_N) were increased (P <0.05) from 0.968 to 0.976 and from 3,119 to 3,198 kcal/kg, respectively, with enzyme supplementation. Activity of amylase in the duodenum was higher (18.6 vs. 50.1 IU/g of digesta) in supplemented birds. Amylase supplementation led to a reduced coefficient of variation for both TTS (averaged 2.41 vs. 0.92% from 7 to 42 d) and AIS digestibilities (1.96 vs. 1.03%), as well as AME_N (0.49 vs. 0.35%), when compared to the nonsupplemented group, indicating lower individual heterogenity. An age effect was detected for TTS digestibility, as both groups saw an increase during the first weeks (slightly more pronounced in the supplemented group); older birds (d 30 onwards) presented a lower TTS digestibility compared to ages between 7 and 25 d. In conclusion, amylase supplementation in maize diets for broilers can attenuate individual bird variation for starch and energy utilization by increasing amylase activity and enhancing starch digestibility.

INTRODUCTION

Starch is the primary source of energy in diets for broiler chickens and its rate of degradation in the gastrointestinal tract has a major impact on energy availability for the bird. Broilers are notorious for their efficiency in digesting starch, as ileal digestibility coefficients observed throughout literature are commonly above 0.95 (Svihus and Hetland, 2001; Gracia et al., 2003; Kaczmarek et al., 2014; Svihus, 2014; Herwig et al., 2020), even without dietary enzyme supplementation. However, lower digestibility values can be found in starch sources with higher proportions of amylose to amylopectin and higher amounts of nonstarch polysaccharides, such as wheat and peas, which may increase viscosity and impair overall diet solubility (Svihus and Hetland, 2001; Weurding et al., 2001). The starch digestion rate of maize-based diets, though, is noticeably high because of its low viscosity, as shown by Weurding et al. (2001) and Kaczmarek et al. (2014), but there has been evidence that the exogenous supplementation of amylase may improve starch and energy utilization of maize and maize-soybean meal diets (Gracia et al., 2003; Stefanello et al., 2015; Aderibigbe et al., 2020; Schramm et al., 2021). According to Schramm et al. (2021), the availability of maize starch, as well as of other nutrients, for broilers, is influenced by intrinsic grain characteristics, that is, variety and composition, or by processing parameters applied upon the grain. The use of amylase, in turn, can reduce these negative impacts and improve nutrient digestibility. Additionally, the endogenous production and secretion of pancreatic amylase and other digestive tract enzymes such as proteases and lipases can be suboptimal in posthatch chicks due to a nonfully developed gastrointestinal tract (Noy and Sklan, 1995; Uni et al., 1995), thus limiting nutrient digestion capacity.

Nutrient utilization is also affected by individual animal heterogeneity, as traits like enzyme production and AME may have between-birds variation. Choct et al. (1999) and Choct et al. (2006) have shown that there is individual variation on the production of fiber-degrading enzymes by gut microbiota of broilers, that is, xylanase and β-glucanase, resulting in AME and starch degradation variability, which could then be reduced through enzyme supplementation. Svihus et al. (2010) demonstrated that birds can develop a habit of overconsuming feed, leading to a reduced feed retention time in the GIT and consequently to a flawed efficacy of enzymatic digestion, thus marked as one of the causes of high individual animal heterogeneity on nutrient digestibility and prone to be improved by, among other factors, exogenous enzymes. Impact of individuality on amino acid digestibility has also been noted by Cowieson et al. (2020) and likewise has been reduced by protease supplementation.

Therefore, the outset of such situations can make room for improvement of starch digestibility in amylase-supplemented broiler diets. The purpose of the study was to investigate the variance of starch digestibility coefficients in broilers fed maize-based diets with or without exogenous amylase on an individual bird basis. The hypothesis was that amylase improves starch digestion, and that the effect is at least partly governed by bird- and time-specific variations in baseline starch digestibility.

MATERIALS AND METHODS

All experimental procedures complied with the guidelines of the Local Ethical Committee for Experiments on Animals in Poznan regarding animal experimentation and animal care under study (European Union [EU] Directive 2010/63/EU for animal experiments).

Animal Husbandry and Experimental Design

A total of 560 d of hatch male Ross 308 chicks were obtained from Aviagen hatchery (Dan Hatch Poland S.A., Wolsztyn, Poland). On arrival, all birds were marked with a neck tag, randomly housed in 120 cages (4 birds per cage) and were provided the experimental diets. After 5 d, 1 bird per cage was randomly selected using a computer algorithm and the selected bird was placed in an individual cage (40 cm width × 50 cm depth × 50 cm height) until 28 d. Afterwards, birds were transferred to cages with similar dimensions but equipped with bigger feeders for the remaining period of the study (43 d), while keeping track of individual identifications.

The experiment comprised 2 dietary treatments: a basal diet and basal + amylase supplementation, each allocated to 60 individual caged birds (each bird considered a replicate), giving 120 birds in total. Birds were exposed to light for 24 h/d in the first 7 d, followed by 18 h light:6 h darkness throughout the experiment. The temperature was maintained at 32°C during the first week and was gradually decreased to ∼23°C by the end of the third week. Access to feed and water in the cages was provided ad libitum.

Experimental Diets

The maize- and soybean meal-based experimental feeds were manufactured at the Experimental Station of the Department of Animal Nutrition and Feed Management Gorzyń/Miedzychód—Poland, and the trial was divided into 3 feeding periods—starter (1–10 d), grower (11–24 d) and finisher (25–end). The composition of experimental diets and the calculated and analyzed nutrient values are presented in Table 1. The basal diet was prepared in a 100 kg horizontal mixer (model: Zuptor 100) with 4 min mixing time at a 27.4 rev/min speed. All ingredients, except minerals, vitamins, amino acids, and fat were ground in a Skiold Disc mill (SK2500) with disc distance set at 1.8 mm. Minerals, amino acids, vitamins, and fat were directly added to the mixer. The enzyme was previously mixed with 1 kg of maize, which was then added to the mixer. All diets were pelleted in a Scorpion pellet press (BMG Pelleting Experts, Gdańsk, Poland) equipped with a 22 kW engine and a 4 mm thick ring die with 3 mm diameter holes; pelleting conditions were monitored and maintained at a constant ampere draw of the load meter for the mill motor to the consistency of pelleting conditions, with a production rate around 100 kg/h.

Table 1.

Ingredients and nutritional composition of basal diets (% of dry matter).

Ingredients (%)	Phases
	1–10 d		11–24 d		25–end
Maize	55.64		60.56		63.24
Soybean meal	36.63		31.27		27.69
Soybean oil	1.65		2.46		3.52
Fish meal	2.0		1.5		1.5
Monocalcium phosphate	1.22		1.27		1.21
Premix 1	1		1		1
Limestone	0.48		0.46		0.38
DL-Methionine	0.35		0.33		0.32
HCl-lysine	0.23		0.27		0.26
Sodium bicarbonate	0.36		0.43		0.43
Sodium chloride	0.05		0.03		0.04
L-threonine	0.07		0.1		0.09
Titanium dioxide	0.3		0.3		0.3
Calculated chemical composition
ME (kcal/kg)	2,850		2,950		3,050
Crude protein (%)	22.67		20.3		19.00
Calcium (%)	0.9		0.87		0.81
Available phosphorus (%)	0.45		0.44		0.41
Digestible AA:
Lysine (%)	1.27		1.15		1.06
Met + Cys (%)	0.94		0.87		0.83
Tryptophan (%)	0.22		0.2		0.18
Threonine (%)	0.83		0.77		0.71
Arginine (%)	1.4		1.23		1.13
Sodium (%)	0.16		0.16		0.16
Chloride (%)	0.16		0.16		0.16
Potassium (%)	0.98		0.88		0.82
Starch (%)	35.42		37.83		39.95
Analyzed chemical composition
	Basal	Basal + Amylase2	Basal	Basal + Amylase2	Basal	Basal + Amylase2
Starch	36.2	37.9	37.9	40.0	41.8	42.1
Crude protein	22.7	22.7	20.5	20.4	19.5	19.5
Crude fat	4.57	4.55	5.46	5.47	6.61	6.56
Amylase3 (KNU/kg)	17	70	13	70	10	88

Abbreviation: KNU, kilo-Novo-α-unit.

¹Provided per kg diet: IU: vit. A 11250, cholecalciferol 2500; mg: vit. E 80, menadione 2.50, vit. B12 0.02, folic acid 1.17, choline 379, D-pantothenic acid 12.5, riboflavin 7.0, niacin 41.67, thiamin 2.17, D-biotin 0.18, pyridoxine 4.0, ethoxyquin 0.09, Mn 73, Zn 55, Fe 45, Cu 20, I 0.62, Se 0.3.

²Basal diet + 80 amylase units (KNU)/kg (Ronozyme HiStarch CT, DSM Nutritional Products Ltd., Basel, Switzerland).

³Expected values in basal pelleted diet = 0 KNU/kg; Expected values in Basal + Amylase pelleted diet = 80 KNU/kg.

The enzyme used was Ronozyme HiStarch (CT; DSM Nutritional Products Ltd., Basel, Switzerland), a monocomponent amylase produced by submerged fermentation of Bacillus amyloliquefaciens, containing alpha-amylase as an active ingredient at a minimum of 600 kilo-Novo-α-amylase units (KNU) per g product was used. The product was added to the supplemented diets at 133 mg/kg, corresponding to a min. activity of 80 KNU/kg diet.

Sample Collection

The collection of excreta samples and growth performance measurements started at d 7. Excreta were collected every Monday, Wednesday, and Friday). Trays were cleaned at 8 am on the day of sample collection, and samples from individual birds were collected at 1 pm in the same order as the tray cleaning, avoiding contamination with droppings and feathers.

On d 43, an 8-h dark period was applied, followed by 3 h of light in the early morning to stimulate feed consumption. Subsequently, all 120 birds were euthanized by intracardial injection of Na-pentobarbitone to collect ileal content. The broilers were eviscerated, and whole duodenum and an ileal fraction from the lower half of the ileum were separated for content removal, defined as 4 cm below Meckel's diverticulum and 4 cm above the ileum-cecum-colon junction. Ileal and duodenal content was obtained by gentle flushing with distilled water, placed in identified plastic containers and immediate freezing with liquid nitrogen, followed by storage in a freezer at -30°C until analysis.

Chemical Analyses

Representative samples of feed were collected after pelleting. Prior to analysis, ileal digesta and excreta samples were homogenized using a stomacher homogenizer (Interscience, France), and then freeze-dried (Christ Epsilon-10D LSC plus, Medizinischer Apparatebau, Osterode/Harz, Germany). Feed, ileal, and excreta samples were then ground and passed through a sieve with a mesh size of 0.5 mm (Retsch, Ultra Centrifugal Mill ZM 200, Haan, Germany). The feeds were analyzed in duplicate for dry matter (method 943.01), crude protein (method 976.05), ether extract (method 920.39), acid detergent fiber (method 942.05, expressed inclusive of residual ash), and neutral detergent fiber (method 973.18, assayed with heat-stable amylase and expressed inclusive of residual ash) according to the Association of Official Agricultural Chemists (AOAC, 2005). Starch content was determined utilizing thermostable α-amylase and amyloglucosidase according to AOAC (2005; method 996.11). Dietary nitrogen content was analyzed using a Kjel Foss Automatic 16,210 analyzer (A/S N. Foss Electric, Denmark), and ether extract was determined using a Soxtec System HT 1043 Extraction Unit (Foss Tecator, Denmark). Gross energy (GE) was determined using an adiabatic bomb calorimeter (KL 12Mn, Precyzja-Bit PPHU, Poland) standardized with benzoic acid. Total starch and starch fractions (rapidly digestible starch, slowly digestible starch, available starch, resistant starch) in maize were determined using the method of Englyst et al. (1999). Content of salt-soluble protein in maize was determined using an official method (NF-V03–741; AFNOR, 2008) as described by Gehring et al. (2012).

In the ileal digesta and excreta samples, dry matter and starch contents were analyzed using the same methods previously described; the GE was also measured in excreta. Titanium dioxide (TiO₂) was used as a dietary marker, and its content in diets, ileal digesta, and excreta was determined according to the procedure of Short et al., (1996). The activity of amylase in duodenum digesta was determined using commercial kits as was presented by Pruszynska-Oszmalek et al. (2015). In-feed determination of the added alpha-amylase was conducted by Biopract GmbH, (Berlin, Germany), on behalf of DSM Nutritional Products Ltd. (Basel, Switzerland), based on the assay of alpha-amylase using red-starch (S-RSTAR 03/06, Megazyme, Bray, Ireland).

Calculations and Statistical Analysis

Growth performance variables were recorded by individually weighing all the broilers and feed leftovers at d 7, 11, 25, and 42 to calculate feed intake (FI), body weight gain (BWG), and feed conversion ratio (FCR) in the periods and overall period.

The following equation was used to calculate apparent total tract starch (TTS) and apparent ileal starch (AIS) digestibility:

$$Starchdigestibility=\left\{1-\left[\left(TiO2\%\frac{diet}{excretaordigesta}\right)x\left(Starch\%\frac{excretaordigesta}{diet}\right)\right]\right\}$$

The GE values in the samples were used to calculate apparent metabolizable energy (AME_N) through the following equation:

$$AM{E}_N\left[MJ/kg\right]=G{E}_{excreta}-\left[G{E}_{exreta}\times \left(Ti{O}_{2diet}/Ti{O}_{2exreta}\right)\right]-0.0344\times \left\{N_{diet}-\left[N_{excreta}\times \left(Ti{O}_{2diet}/Ti{O}_{2excreta}\right)\right]\right\},$$

where GE is the gross energy (MJ/kg), N is nitrogen, and TiO₂ is the dietary marker. AME_N was corrected to a zero nitrogen balance assuming that 34.4 MJ/kg N was retained (Hill and Anderson, 1958).

Cages/birds were considered the experimental unit and P < 0.05 was considered to be significant. One mortality was observed in the enzyme-supplemented group on d 38th, and the TTS data from this bird was considered up until this date, after which the supplemented group comprised 59 replicates. All the obtained results for experimental groups were compared using the Student t test on the basis of the following formula:

$$t=\frac{x_1-x_2}{\sqrt{\frac{S_1^2}{n_1}}+\frac{S_2^2}{n_2}}$$

Where t is the Student t test value; X₁-X₂ the difference in the means between the 2 groups being compared; $S_1^2$ and $S_2^2$are variance estimates from each independent group; n₁, n₂ are the respective sample sizes for each independent group. In this study, the standard error of the mean was established as a measure of error.

To verify the effect of age and diet effects considering repeated records of individual birds, a mixed model in ASReml (Gilmour et al., 2015) was used:

$$\text{Yijk}=\mu +\text{di}+\text{hij}+\text{wk}+\text{eijk}$$

where µ = overall mean, di = fixed effect of diet, hij = random effect of chicken within diet (treatment; j = 1 to 60, number of chickens per treatment), wk = fixed effect of day of sampling (k = 1−16), and eijk = random error variation (residual error).

To identify consistent differences in individual bird's digestibility patterns, medians were calculated for the individual bird data and tested for significance. Repeatability was calculated as the ratio of bird variance to phenotypic variance. Principal components analysis was performed in R to reduce dimentionality of the data.

A calculation of frequency of distribution within different categories of ileal starch digestibility was also carried out. Eight classes, based on AIS digestibility results, were set as follows: 0 to 93; 93 to 94; 94 to 95; 95 to 96; 96 to 97; 97 to 98; 98 to 99; and 99 to 100% digestibility. The percent of the total number of observations (n = 960) from each dietary treatment that fit into each category was then determined according to their values for AIS digestibility. No outlier removal method was applied to the dataset as a means to observe true variability between the groups.

RESULTS AND DISCUSSION

The chemical analysis of maize is presented in Table 2. Starch has been commonly classified into different categories according to its rate of digestibility and glucose release in the GIT, namely from most to least digestible: rapidly digestible starch, slowly digestible starch, and resistant starch (Englyst et al., 1996). Maize is known to have higher concentrations of rapidly and slowly digestible fractions compared to resistant starch, which increases its potential of digestibility (Weurding et al., 2001; Wiseman, 2006). In our analyses, the amount of resistant starch in maize samples was remarkably low, in addition to a higher fraction of rapidly digestible starch over slowly digestible, indicating a high starch digestion rate and fast glucose release. These results differ from previously reported values in other studies (Wiseman, 2006; Zhang et al., 2006), where resistant starch in maize was not as low, and the slowly digestible fraction was predominant, thus indicating a between-study variation that could be caused by several factors, such as intrinsic grain variability, environmental conditions, and others.

Table 2.

Proximate chemical analyses of maize.

Nutrient1	g/100 g
Crude protein	8.09
Salt-soluble protein (eq. mg albumin)	11
	g/100g monosaccharide equivalent
Rapidly digestible starch	39.4
Slowly digestible starch	27.0
Available starch	66.4
Resistant starch	1.5
Total starch	67.9

¹Starch fractions were determined according to Englyst et al. (1999).

Growth Performance

The growth performance data are presented in Table 3. The broilers fed amylase-supplemented diets were characterized by lower FI in the periods of 7 to 11 d (P < 0.001), 25 to 42 d (P < 0.05), and the whole experiment (7−42 d, P < 0.01), with a trend for lower FI from 11 to 25 d (P < 0.09). The FCR was lowered by amylase supplementation on periods from 7 to 11 d (P < 0.05), 25 to 42 d (P < 0.01) and during the total experimental period (P < 0.05), whereas from d 11 to 25 there was a tendency (P < 0.1) to lower FCR with amylase supplementation. There were no differences in BWG on any evaluated period.

Table 3.

Body weight gain (BWG), feed intake (FI), and feed conversion ratio (FCR) of broilers fed basal or amylase-supplemented diets.

Variable	Basal	Basal + Amylase1	P-value
FCR [g:g]
7–11 d	1.305	1.241	0.034
11–25 d	1.188	1.178	0.324
25–42 d	1.779	1.704	0.003
7–42 d	1.508	1.470	0.013
FI [g/bird]
7–11 d	228	216	<0.001
11–25 d	1526	1496	0.090
25–42 d	3060	2949	0.026
7–42 d	4815	4675	0.017
BWG [g/bird]
7–11 d	177	175	0.505
11–25 d	1288	1273	0.423
25–42 d	1732	1736	0.904
7–42 d	3197	3184	0.736

¹80 amylase units/kg (Ronozyme HiStarch CT, DSM Nutritional Products Ltd., Basel, Switzerland).

The current study demonstrated a reduction of FCR by an average of 2.52% over the period of 7 to 42 d with the supplementation of 80 KNU/kg, compared to the basal diet. Other studies have also reported improvements on FCR with the use of amylase in maize-based diets for broilers, mostly correlated to an increase in BWG (Gracia et al., 2003; Stefanello et al., 2015; Yuan et al., 2017). However, BWG was not affected by treatments in the present study. It is likely that the dietary nutrient density was sufficient to allow for maximal growth in both diets, and thus a reduced FI was the response to the improved nutrient availability caused by amylase. This was demonstrated by the reduction of FI (2.9% lower) in supplemented birds, being aligned with improvements on TTS and AIS digestibility, as well as on dietary AME_N. Energy is known to be a FI regulation factor (Richards and Proszkowiec-Weglarz, 2007; Massuquetto et al., 2020), thus indicating that broilers were able to self-regulate consumption on account of an increased net energy availability resulting from the higher amylase-induced starch degradation. Similar results were reported by Castro et al. (2020) regarding lower FI on broilers fed amylase at 500 g/ton of feed, due to an increased energy supply.

In contrast, a lack of effect of exogenous amylase on performance of broilers fed maize diets has also been reported. Schramm et al. (2021) found no improvements on performance of 15-to-25-d-old broilers supplemented with increasing amylase doses (up to 160 KNU/kg), presumably because of a short evaluation period. Kaczmarek et al. (2014) assessed enzymatic blends for broilers with amylase and amyloglucosidase or amylase, amyloglucosidase, and protease, but did not observe any effects on growth performance. In a second experiment, though, the authors reported higher BWG and lower FCR with amylase (10,000 U/kg) + amyloglucosidase (10 U/kg) supplementation when broilers were fed finely ground maize relatively to coarse maize, evidencing the impact of extrinsic characteristics such as particle size on starch degradation.

Total Tract and Ileal Starch Digestibility, and Duodenal Amylase Activity

The TTS and AIS digestibility as well as AME_N results are presented in Table 4. Amylase supplementation improved TTS digestibility (P < 0.05) at each day of excreta collection. A single exception was on d 28 where the difference in TTS digestibility between the basal and supplemented was insignificant. In average, TTS digestibility throughout the experimental period (from d 7 to 42) was 0.973 and 0.982 for the basal and amylase-supplemented fed broilers, which were statistically different (P < 0.001). The Principal Component Analysis showed differentiation in digestibility patterns between the basal and supplemented diet (Figure 1) with a few birds overlap based on component 1, which explained 30.3% of variation in the data.

Table 4.

Total tract starch digestibility, ileal starch digestibility, AME_N, and duodenal amylase activity of broiler chickens fed diets with or without amylase from 0 to 42 d.

Day:	Basal		Basal + Amylase1		Diet P-value
	Starch total tract digestibilty:
	Mean	CV2 (%)	Mean	CV (%)
7	0.968	2.41	0.976	0.92	0.0199
9	0.974	1.14	0.986	0.39	<0.0001
11	0.978	0.89	0.985	0.31	<0.0001
14	0.973	0.48	0.986	0.32	<0.0001
16	0.982	0.95	0.987	0.31	0.0008
18	0.970	1.07	0.987	0.32	<0.0001
21	0.979	0.62	0.985	0.77	<0.0001
23	0.976	0.93	0.986	0.8	<0.0001
25	0.978	1.21	0.984	0.52	0.0005
28	0.968	4.97	0.976	1.02	0.2006
30	0.962	1.4	0.975	1.23	<0.0001
32	0.976	1.16	0.980	0.76	0.0173
35	0.976	1.04	0.980	0.79	0.0079
37	0.969	1.13	0.978	1.08	<0.0001
39	0.963	1.62	0.976	0.84	<0.0001
42	0.972	1.29	0.979	0.78	0.0004
7–42	0.973	2.41	0.982	0.92	<0.0001
		Ileal starch digestibility
42	0.968	1.96	0.976	1.03	<0.0001
		AMEN (kcal/kg)
	3,119	0.49	3,197	0.35	<0.001
	Amylase activity in duodenum (IU/g of fresh digesta)
	18.7	5.51	51.6	6.01	<0.0001

Abbreviations: AMEN, apparent metabolizable energy; CV, coefficient of variation.

¹80 amylase units/kg (Ronozyme HiStarch CT, DSM Nutritional Products Ltd., Basel, Switzerland).

²Coefficient of variation.

Figure 1.

Principal component analysis of individual values of total tract starch digestibility of broilers fed diets with or without amylase supplementation.

Both AIS digestibility and AME_N measured on d 42 were increased (P < 0.05) by 0.031 and 78 kcal/kg, respectively, with amylase supplementation. These findings of greater starch utilization and AME are in agreement with other findings (Gracia et al., 2003; Stefanello et al., 2015, 2019; Yuan et al., 2017; Aderibigbe et al., 2020). Because starch is the primary source of energy in broiler diets, the increase in AME_N is related to the higher starch digestion provided by amylase. Using the same dose of 80 KNU/kg in maize-based diets for broilers, Stefanello et al. (2015) reported an increase of jejunal and ileal digestibility 2.2 and 0.6%, respectively, along with a 70 kcal increase in AME, compared to nonsupplemented diets—notably, starch and energy utilization were further increased with the use of a combined supplementation of amylase and xylanase, suggesting synergic effects between both enzymes.

It has been shown by Osman (1982) that the jejunum is the main site of digestion of starch, due to a superior endogenous amylase activity. When reaching the ileum, up to 95% and above of the starch has already been digested (Svihus, 2014; Herwig et al., 2019). Even so, Aderibigbe et al. (2020) evaluated the digestion of starch at different sites of the GIT with increasing amylase doses from 80 to 160 KNU/kg in maize-based diets for broilers, and found greater starch digestibility in both jejunum and ileum with the use of both doses compared to a control diet. According to the authors, an improved breakdown of starch in the jejunum could also lead to a sparring effect of amino acids for energy production, further improving diet utilization with the use of amylase.

As seen in the current study, and throughout literature, the increment on ileal digestibility of maize starch with amylase usually ranges from 1.5 to 2.5% (Gracia et al. 2003; Stefanello et al., 2015; Woyengo et al., 2019), and whereas it seems marginal, it can have substantial benefits in the long term for growth performance and overall nutrient utilization, given that maize is included at 60 to 70% in the majority of diets for broilers, and it may contain up to 69% starch (Bach Knudsen, 1997). Even though the low viscosity of maize is not as detrimental to starch digestibility as other high-viscosity grains with higher levels of nonstarch polysaccharides such as faba bean, wheat, and barley (Aftab and Bedford, 2018; Zaworska-Zakrzewska et al., 2022), there is still room for improvement when targeting maize starch as substrate for exogenous amylases. Stefanello et al. (2019) and Schramm et al. (2021) made use of the partial substitution method to estimate the effects of amylase on maize digestibility; both studies observed linear increases in maize starch and energy digestibility when including up to 160 KNU/kg.

Endogenous secretion of pancreatic enzymes is restrained in young broilers. Studies by Nitsan et al. (1991), Noy and Sklan (1995), and Uni et al. (1995) have demonstrated that net duodenal secretion of amylase, lipase, and trypsin is low at posthatch, rapidly increasing up to 21 d. Concurrently, changes on enzyme secretion are complemented by changes to the morphology of the small intestine during the first days of life (Uni et al. 1995). This is considered as one of the bottlenecks that could limit starch digestibility in young chicks, combined with a seemingly low capacity of extracting, absorbing, and transporting glucose in the lumen (Aderibigbe et al., 2020), thereby driving forth the dietary inclusion of exogenous amylase intended at supporting the endogenous enzyme activity. Additionally, Zelenka and Čerešňáková (2005) and Cowieson et al. (2019) suggest that enzyme activity at an older age is equally or even more important, as the intestinal mass, and, ergo, enzyme secretion relative to body weight is many times lower in older birds.

Amylase supplementation substantially increased (P < 0.05) amylase activity in duodenal digesta (from 18.7 to 51.6 U/g) at 42 d. This agrees with observations by Jiang et al. (2008) with 21-d-old broilers, whose amylase, protease, and trypsin activities in the anterior intestine were increased with the inclusion of up to 750 mg/kg of an amylase preparation. As suggested, this is likely related to additive effects of both endogenous and exogenous enzymes, and as in the current study, it benefitted digestion. The authors reported, though, that increasing the dose to 2,250 mg/kg, reduced amylase and protease activity due to what seemed to be a negative feedback in amylase mRNA expression and inhibition of pancreatic amylase production. Mahagna et al. (1995) also observed lower pancreatic and intestinal amylase and trypsin activities in 7- and 14-d-old broilers fed diets with amylase. Both studies indicate a long-term unwanted effect of exogenous amylase on depressing pancreatic synthesis and secretion. In this study, broilers had a higher amylase activity even at a considerably advanced age (42 d), thus negative impacts on enzyme activity could be more closely associated to the utilized dose.

Individual Variation and Age Effect

The individual variation for TTS digestibility as expressed by the coefficient of variation (CV) in Table 4 was higher for broilers fed the basal diets on all days of collection, whereas amylase supplementation effectually lowered the average CV (d 7–42) by 61.83%; the same was observed with individual variation of AIS digestibility and AME_N, whose CV was reduced by approximately 28 and 48%, respectively, although the CV for duodenal amylase activity was higher on supplemented birds. The repeatability of digestibility after correction for diet and day was low, but different from zero (0.04 ± 0.01). Despite low repeatability, consistent patterns were observed and 9 groups differing in the mode of digestibility (P < 0.05) were identified. The digestibility pattern of birds with extreme medians (96.6 for basal vs. 98.8 for amylase-supplemented) is illustrated on Figure 2.

Figure 2.

Total tract starch digestibility medians of broilers fed diets with or without supplementation of amylase.

According to Cowieson et al. (2020), individual animal heterogeneity comes from natural differences in absorptive and digestive capacities, microbiome, feeding behavior, and others, and accounts as one of the sources of variance for nutrient digestibility, along with ingredient variation. Whereas the latter is well acknowledged, though, the former has yet to be better understood, as only a handful of studies have addressed the impact of individual broiler variation on dietary nutrient digestibility. Similarly to the current study, Choct et al. (1999) determined the effect of xylanase supplementation on individual bird variation, and observed a reduction of the variability in ileal starch digestion from 9 to 1.5% with enzyme inclusion. A reduced variation of AME with xylanase supplementation was also found in a following study (Choct et al., 2006). Svihus and Hetland (2001) explain that differences between-birds for starch digestion capacity can lead to a high number of broilers with low starch digestibility, which was then reduced due to an increased starch digestibility when utilizing whole wheat.

Therefore, it can be stated that amylase reduced individual variation in the current study by increasing starch digestibility and thus reducing the gap between broilers with low and high capacities of digesting starch. This is further highlighted in Figure 3, as a high number of broilers shifted from lower classes (from 93 to 98%) to higher classes (above 98%) of starch digestibility when fed enzyme-supplemented diets. As suggested by Choct et al. (1999), lower between-birds variation can be fruitful not only by increasing diet utilization, but also by resulting in more uniform flocks.

Figure 3.

Frequency of distribution of starch digestibility of birds fed diets with or without amylase supplementation, according to individual values of total tract starch digestibility sampled during 16 d (% of 960 samplings per treatment).

An age effect was observed for TTS digestibility on both basal and amylase-fed birds (Figure 4). Overall, TTS digestibility for both groups scaled up during the first week of the experiment (approximately 7–14 d of age). This observation concurs with the reports on how young broilers can present low nutrient digestibility due to a limited endogenous enzyme activity and immature GIT, which then increases after the first week posthatch (Nitsan et al., 1991; Noy and Sklan, 1995; Uni et al., 1995). This increase in starch digestibility after 2 wk of age was slightly more pronounced in the amylase-supplemented group; birds in this group presented a higher TTS digestibility coefficient since d 1 of the experiment, and remained superior during the entire period.

Figure 4.

Effect of age on individual total tract starch digestibility of broilers fed diets with or without amylase supplementation.

Between d 28 and 30, a reduction on TTS digestibility was noted, gradually increasing back up in the following days. This drop on starch digestion occurred in both groups, so rather than being related to the dietary treatments, it was most likely caused by high environmental stress, as it coincided with the day all birds were transferred to new cages (d 28). Noteworthy, the CV in the basal group on d 28 was exceedingly higher (4.97%) compared to the supplemented group or even to previous and posterior days of collection in the same group; a bird with much lower TTS digestibility (0.60) was identified as the reason behind this, and when removing it from calculations, the CV was reduced to 1.05%. Past d 30, digestibility rose again, but despite this increase, older birds from both groups were overall characterized by a slightly lower TTS digestibility than prior to 25 d. Zelenka and Čerešňáková (2005) demonstrate that starch digestibility linearly decreases with age in fast-growing broilers. This seemingly starts at around 30 d of age, possibly due to the birds reaching a maximum synthesis capacity of amylase, which becomes insufficient to keep up with an increasing feed consumption, further emphasized by Cowieson et al. (2019). Nonetheless, at 42 d amylase-fed birds had a higher duodenal amylase activity, hence why AIS and TTS digestibility was persistently higher than the basal group.

The results of the current experiment clearly showed that maize starch is well digested, even by very young birds. The low repeatability of starch digestibility (around 30%) indicates that more research is needed to verify the hypothesis about inherent low starch digestibility in some of the birds. Additionally, amylase supplementation improved consistency of starch digestibility.

DISCLOSURES

The authors declare no conflicts of interest.