1. Introduction
Fats are usually used in poultry diets as they satisfy a large fraction of the energy requirements. Palm fatty acid distillate (PFAD) is a fat by-product from the production of refined palm oil (P) which is one of the most produced and consumed vegetable oils worldwide [
1]. Usually, P is obtained by a physical refining process that includes different steps, namely degumming, winterization (optional), bleaching, and deodorization [
2]. The latter step is a vacuum steam distillation process that removes the FFA that are accumulated in the fatty acid distillate [
3]. PFAD is characterized by having a high proportion of free fatty acids (FFA: 87–94%, being rich in saturated FA (SFA) and including other compounds such as tocopherols [
4]). Based on a circular economy and taking into account the rising cost of conventional fats, there is increased interest in upcycling by-products from the fat industry for animal feeding to reduce the cost of feed formulation and also the environmental impact [
5,
6].
Assessing the digestibility of a fat ingredient is one of the clearest ways of defining its nutritional value for an animal. Conventional fat and oil sources used in poultry feed mainly consist of triacylglycerol (TAG) molecules. During digestion, TAGs, and diacylglycerols (DAGs) are hydrolyzed into monoacylglycerol (MAG) and FFA, which are incorporated into dietary mixed micelles (DMM) to attain the enterocytes for their absorption. Therefore, studying the evolution of the lipid classes (TAG, DAG, MAG, and FFA) and FA digestibility along the different segments of the gastrointestinal tract (GIT) may be of great interest for understanding the dynamics of fat digestion [
7,
8], mainly in new alternative fat sources rich in FFA and also in fat blends.
The PFAD is rich in FFA and also in SFA. It is well known that the degree of FA saturation plays and important role in fat absorption. In broiler chickens, although SFAs are not digested as well as unsaturated FA (UFA), several authors have found a synergistic effect when saturated sources are blended with unsaturated ones [
9,
10,
11]. In fact, recent studies have found that the saturation degree of the dietary fat has more influence on fat digestibility than its FFA content [
5,
12]. Moreover, it has been suggested that there is a positive effect on FFA digestibility when there are increasing amounts of DAG or MAG, because their emulsifying effect enhances the inclusion of FFA in DMM [
13]. However, there are few studies on FA absorption using blends of crude (rich in TAG) and acid (rich in FFA) oils.
Furthermore, it is accepted that the absorption of FA is also affected by the age of the chickens. Better results in the hydrolysis–absorption process along with the GIT of conventional and alternative fats have been obtained in grower-finisher chickens compared to starter broiler chickens [
8].
Therefore, our hypothesis is that PFAD in combination with soybean oil (S) could be considered as an alternative energy source for broiler chicken diets, but the use of PFAD might be influenced by the age of the chicken. Thus, the aim of the present study was to research the effect of replacing S with graded levels of PFAD on lipid-class content and FA digestibility along the intestinal segments of the GIT (upper and lower jejunum, upper and lower ileum) and in the excreta in starter and grower-finisher broiler chickens.
2. Materials and Methods
2.1. Housing and Animals
The study was carried out at the animal experimental facilities of the Servei de Granges i Camps Experimentals (Universitat Autònoma de Barcelona; Bellaterra, Barcelona, Spain). All management practices and procedures were approved by the Animal Ethics Committee (CEEAH) of the same institution (number code: 3938), in accordance with the European Union guidelines for the care and use of animals in research (2010/63/EU).
A total of 480 newly hatched female broiler chickens (Ross 308) were obtained from a commercial hatchery (Pondex SAU; Lleida, Spain). On arrival, birds were wing-banded, individually weighed and randomly allocated to cages (16 birds per cage) and assigned to one of the five dietary treatments (six replicas per treatment). Birds were housed in metabolic cages, with a grid floor and excreta collection tray, located in an environmentally controlled room. Throughout the study, feed and water were offered ad libitum. The temperature, humidity, ventilation, and illumination were automatically controlled, as recommended by the specifications in the Ross 308 lineage management handbook [
14]. The animals and housing facilities were inspected, at least twice a day (d).
2.2. Experimental Design and Diets
All birds were raised with a starter feed until d 22 and a grower-finisher feed from d 23 to d 35, both in mash form. The wheat- and soybean meal-based diet was formulated to meet or exceed FEDNA’s (Fundación Española para el Desarrollo de la Nutrición Animal) requirements [
15] and to minimize basal fat levels, as shown in
Table 1. Titanium dioxide (TiO
2) was included (5 g/kg) as an inert marker for determining the digestibility of FA.
The experimental diets consisted in a basal diet supplemented with 6% of the different fat sources (
Table 2). The S was included at 6% (S6) and increasing amounts of PFAD were added in replacement of S: S4-PA2 (4% of S and 2% of PFAD), S2-PA4 (2% of S and 4% of PFAD) and PA6 (PFAD at 6%). The P was included at 6% (P6) as a control treatment for PFAD. Thus, 5 different diets were compared that were replicated 6 times. The composition of the experimental diets is shown in
Table 3. The basal diet was manufactured at Pinsos Molinet S.A., (Prats de Lluçanès, Barcelona, Spain) and the addition of the experimental fat sources or fat blends to manufacture the experimental diets was performed at Lindo Pet Global S.A. (Castellar del Vallès, Barcelona, Spain).
2.3. Controls and Sampling
Individual body weight (BW) and feed consumption by cage were measured at 11, 22 and 35 d of age to calculate the average daily gain (ADG), average daily feed intake (ADFI) and the feed conversion ratio (FCR) throughout the experiment. Mortality was recorded and weighed to correct these parameters.
Two digestibility balances were performed in young animals from 9 to 11 d and in older animals from 33 to 35 d. At 11 d of age, 12 birds per cage were killed by cervical dislocation, and the jejunum (from the distal-most point of insertion of the duodenal mesentery to the junction with Meckel’s diverticulum), and ileum (from the junction with Meckel’s diverticulum to a point 1 cm proximal to the ileocecal junction) were carefully excised. Then, both segments (jejunum and ileum) were divided into 2 equal portions, named as upper and lower. Thus, for each cage, samples were taken of the digestive content from the upper and lower jejunum and the upper and lower ileum. The samples from the 12 birds from each cage were then homogenized and pooled (n = 6 per type of sample and dietary treatment). A representative sample of excreta (free of contaminants, such as feed or feathers) was also taken from each cage. Samples were frozen at −20 °C, and lyophilized. Thus, 5 different digesta samples were taken: 4 intestinal segments and excreta. Samples of diets, digesta and excreta were ground to pass through a 0.5-mm sieve and kept at 4 °C until further analyses. At 35 d of age, 2 birds per cage were euthanized, and the same procedure described above was carried out for sampling. In addition, at 35 d the abdominal fat pad (from the proventriculus surrounding the gizzard down to the cloaca) of each bird was removed and weighed. Abdominal fat pad weights were expressed in absolute values and as a percentage of BW.
2.4. Chemical Analysis
Oil samples were analyzed in triplicate for moisture and volatile matter according to the AOCS official method Ca 2d-25 [
16], insoluble impurities [
17], unsaponifiable matter according to the AOCS official method Ca 6b-53 [
18], lipid-class composition according to IUPAC (2508 method) [
19], and total FA composition [
20], that were adapted to acid oils by Varona et al. [
4]. The chemical analyses of the experimental fats are shown in
Table 2.
Analytical determinations of the diets were performed according to the methods of AOAC International [
21]: dry matter (Method 934.01), ash (Method 942.05), crude protein (Method 968.06), crude fiber (Method 962.09), and ether extract (EE) by Soxhlet analysis (Method 920.39). Gross energy was determined by an adiabatic bomb calorimeter (IKA C-4000, Janke-Kunkel; Staufen, Germany).
TiO
2 in feed, digestive content and excreta was analyzed following the procedures of Short et al. [
22] and determined by spectrophotometry ICP-OES (Optima 3200 RL, Perkin Elmer, Waltham, MA, USA).
The FA content of the feed, digestive content, and excreta was determined according to the method of Sukhija and Palmquist [
23]. A direct extraction-transesterification procedure using nonadecanoic acid (C19:0; Sigma Aldrich Chemical Co.; St. Louis, MO, USA) as internal standard was performed. Then, the lipid extract was injected in a gas chromatograph (HP6890, Agilent Technologies; Waldbronn, Germany) under the conditions of the method previously described by Cortinas et al. [
24]. FAs were identified based on the retention times of commercial standards of major FA (Supelco 37 component FAME Mix; Sigma-Aldrich Co.). Quantification was carried out by internal normalization. The macronutrient and FA compositions of the experimental diets are presented in
Table 3.
The lipid-class composition (TAG, DAG, MAG, and FFA) of the feed, digestive content, and excreta was determined according to the IUPAC, 2508 method [
19] by size-exclusion chromatography on an Agilent 1100 series HPLC chromatograph equipped with an isocratic pump, with the oven and a Refractive Index Detector (RID) both set at 35 °C (Agilent Technologies, Santa Clara, USA). Lipid extraction was previously performed following the methodology described by Rodriguez-Sanchez et al. [
8] with slight modifications. Briefly, 0.1 g of lyophilized sample was weighed to extract the fat content with diethyl ether after acidification with HCl 1N. After lipid extraction, lipids were dissolved in 2 mL of tetrahydrofuran and filtered through a Nylon filter (13 mm, 0.45 μm), then 100 μL were injected (20 μL loop) into the HPLC. Separation was conducted using 2 Styragel columns (Styragel HR 1 and Styragel HR 0.5) of 30 cm × 0.78 cm i.d., filled with a spherical styrenedivinylbenzene copolymer of 5-μm particle size and pore sizes of 100 Å and 50 Å, respectively (Water Associates; Milford, MA, USA), connected in series. The mobile phase consisted of tetrahydrofuran (HPLC quality grade) at 1 mL/min. Lipid classes were identified by using standards for each lipid-class (trioleoylglycerol for TAG, dioleoylglycerol for DAG, oleoylglycerol for MAG and oleic acid for FFA; Sigma-Aldrich GmbH; Steinheim, Germany) and they were quantified according to their calibration curves.
2.5. Calculations
The apparent digestibility coefficients (ADC) of FA in each intestinal segment and the excreta were calculated according to the following formula using the TiO
2 marker ratio in the diet and digesta or excreta.
where (TiO
2/FA)
d is the concentration of the inert marker and the FA in the diet (g/g DM), and (TiO
2/FA)
e is the concentration of the inert marker and the FA in the digestive content or excreta (g/g DM).
The apparent metabolizable energy (AME) was calculated with the following equation:
To determine the lipid-class content in the different GIT segments and excreta, the content of each lipid class present in the digestive tract of the chickens was estimated according to the following formula [
12]:
where [LC] is the concentration of the lipid-class in the digesta of the GIT segment or excreta (mg/g DM) and [TiO
2] is the concentration of TiO
2 in the digesta of the GIT segment or excreta (mg/g DM).
2.6. Statistical Analysis
The study design included 2 main factors: diet (5 treatments) × intestinal segment (5 types, being 4 intestinal segments and the excreta). The effect of the age was also compared as described below (11 vs. 35 d). The normality of the data and homogeneity of variance were verified. For each age, the effect of the diet on productive parameters (including abdominal fat depot at 35 d) and AME were statistically analyzed by one-way ANOVA using the GLM procedure of SAS (version 9.4, SAS Institute Inc., Cary, NC, USA) (n = 30; 5 diets × 6 replicas). For each age, and for each intestinal segment and excreta, the effect of the diet on the lipid-class content, FA digestibility, and its contribution on FA absorption was also evaluated by one-way ANOVA (n = 30; 5 diets × 6 replicas).
For each age, the effect of the intestinal tract on the lipid-class content was also analyzed by one-way ANOVA with the intestinal segments and the excreta as the main factor (n = 150; 30 samples × 5 types of digesta samples).
On each intestinal segment, the effect of the age (11 or 35 d) on FA absorption was statistically analyzed by one-way ANOVA using the age as the main factor (n = 60; 5 dietary treatments replicated 6 times × 2 ages). Additionally, for each dietary treatment, one-way ANOVA was used to test the effect of the age on feed AME, and at lower ileum level on lipid-class content and FA digestibility, (n = 12; 6 replicas of lower ileum × 2 ages).
The differences between treatments means were tested using Tukey’s correction for multiple comparisons. The cage served as the experimental unit, so there were six units per diet.
The results shown in tables are reported as least-square means, and in all statistical analyses, differences were considered significant at p < 0.05.
4. Discussion
Studying the lipid classes and FA digestibility along the intestinal tract leads to a better understanding of the dynamics of the hydrolysis-absorption process of PFAD alone and in blends in broiler chickens. Our results show that lipolysis, based on the disappearance of TAG and DAG in the digesta, is extended until the ileum. In addition, the results obtained support that hydrolysis is not the most limiting step for fat utilization when compared with the absorption process, which is in accordance with our previous studies in vitro [
26] and in vivo [
12]. Furthermore, our results suggest that hydrolysis efficiency is mainly affected by the age of the bird, whereas the lipid composition of the diet (saturation and FFA level) has less influence on this process. An improvement in the hydrolysis capacity with chicken age was demonstrated by the higher disappearance of TAG and DAG at 35 d compared to 11 d. In starter broiler chickens, some limitations in the hydrolysis process due to low bile and lipase secretion have been described [
27]. However, Noy and Sklan [
28] reported an increase of 80% in the duodenal bile acid secretion between 10 d and 21 d, and a 20-fold increase in lipase secretion between 4 d and 21 d.
The absorption process takes place as a dynamic process parallel to the hydrolysis of fat. The content of the end lipolysis products, mainly FFA, decreased from the upper jejunum to ileum and the maximum digestibility coefficients of FA were reached in the lower ileum. These results show that the lower ileum is the last segment where FA absorption occurs. Moreover, the evolution of the FFA content and the digestibility values throughout the gut confirmed that the jejunum was the main site of FA absorption, in line with previous studies on broiler chickens [
7,
12]. However, the absorption dynamics along the GIT are different according to the FA, and the stearic acid is the one that is absorbed later with no absorption observed until the lower jejunum at both ages. This is related to the lower solubilization into DMM for long-chain saturated FA compared to unsaturated long-chain FA [
29]. This is also reflected in the lower digestibility coefficients for SFA along the GIT compared to MUFA or PUFA, regardless of the age of the chicken or the diet.
The results on the effect of the diet provide evidence of the detriment to the dietary AME values, FA digestibility coefficients, especially SFA, and FFA absorption associated with both the higher SFA% and FFA% of the broiler chicken diet. The lower FA absorption together with the higher residual FFA content in the digesta at the lower ileum obtained for chickens fed the higher level of PFAD (6%; PA6), could be explained by two factors. First, the association of FFA, mainly SFA, with minerals to form insoluble soaps has been described, so that both the FFA and the mineral become unavailable for the absorption [
30]. This has a greater impact on young birds than on older ones [
31]. In our last in vitro study [
26], we found that fat content from PFAD compared to other fat sources (P, S, or soybean acid oil) was less available for micellar solubilization. Second, and related to the lipid-class content, the lowest MAG content in PA6 diets (
Table 3; 1.5% on average for both ages) may hinder the absorption of many FFAs [
9] since the emulsifying properties of MAG improve the rate of FA incorporation into DMM [
32]. This in turn could explain that birds fed PA6 tended to show the lowest abdominal fat weight (%).
The potential inclusion of PFAD in feed for broiler chickens is influenced by the age of the bird. In 11-d-old broiler chicks, the supplementation of PFAD at any level studied had a negative effect on fat utilization compared to S. Consistent with our results, several authors (Wiseman and Salvador, [
33]; Vilarrasa et al. [
5]; Rodriguez-Sanchez et al. [
12]) have found a negative impact of dietary saturation and FFA level on fat utilization in broiler chickens. At 35 d the PA6 showed the worst fat utilization, however, adding PFAD in substitution of S with a feed FFA content up to 30% and a UFA:SFA ratio higher than 2.6 made it possible to achieve a high level of fat digestibility, similar to that obtained using S. This could be partially related to the higher FA digestibility coefficients and higher dietary AME values obtained for the S4-PA2 diet compared to those calculated from the proportions of the components. This suggests a positive synergic effect by the presence of UFA together with the presence of different lipid-class structures provided by S, since UFA and MAG obtained from the lipolysis of TAG are natural emulsifiers, which might enhance the incorporation of SFA, mainly FFA of PFAD, in the DMM and increase its absorption [
34]. This synergistic effect is in agreement with the reported positive results of blending saturated and unsaturated conventional lipid sources [
11] and acid oils [
10]. However, the similarities obtained in feed AME values, lipid-class content in digesta, and apparent FA digestibility coefficients for S2-PA4 (UFA: SFA ratio: 1.7; FFA%: 53) and P6 (UFA: SFA ratio: 1.3; FFA%: 9) suggest that changes in the saturation degree might have a greater impact on FA utilization than the changes in the FFA level of the diet, as reported by Vilarrasa et al. [
5] and Rodriguez-Sanchez et al. [
12].
The present results also demonstrated that replacing S with PFAD led to a delay in the FA absorption along the GIT, which was more evident in starter animals (11-d-old chicks), and for the absorption of linoleic acid in 35-d-old chickens. Thus, even though the jejunum is the main site of fat absorption, the differentiation between the upper and the lower segments should be considered for future studies, at least in starter broiler chickens.
The comparison of the results between starter (11 d) and grower (35 d) broiler chickens confirms that the age has a positive effect on the FFA lipid-class absorption, FA absorption, and AME values of all the diets, which is consistent with the findings of Batal and Parsons [
35], Tancharoenrat et al. [
11], Roll et al. [
13], Rodriguez-Sanchez et al. [
8], and Viñado et al. [
6]. However, it is important to highlight that the observed improvement with age in FA digestibility (especially for SFA), FFA absorption, and dietary AME values, was higher for those chickens fed the most saturated diets (especially with higher FFA%) than for those fed the most unsaturated diets. That there were no differences among diets in grower chickens in the contribution of intestinal segments to FA absorption suggests that the absorption of diets with higher SFA% and FFA% is advanced to the upper intestinal segments at 35 d. This was especially evident for the absorption of stearic acid, and the contribution of the lower jejunum increased due to the absorption of this FA acid in grower broiler chickens fed the PA6 diet. In addition, the higher contribution of the upper ileum in TFA and linoleic acid absorption at 35 d suggests that this segment plays a key role in improving FA absorption with age. In starter chickens, the limited capacity of fat absorption [
27] together with the shorter feed retention time in the GIT (3.15 h in 11-d-old chicks and 5.10 h in 42-d-old chickens) [
36] could explain the lower efficiency in the absorption process. This in turn could explain the higher implication of the lower ileum in young chicks as it is the last part of the GIT for the remaining FA to be absorbed. Therefore, it may be recommendable to separate the ileum into upper and lower segments for further studies. Determining the maximum fat utilization at the lower ileum level instead of from the pool of digesta of the whole ileum may give more accurate results.
The maximum digestibility coefficients of SFA reached at the lower ileum, show that at 11 d both the dietary FFA content (PA6: 0.18 vs. P6: 0.48) and the SFA level (P6: 0.48 vs. S6: 0.69) had a great impact. At 35 d the magnitude of the negative effect was lower than at 11 d, for the dietary FFA content (PA6: 0.64 vs. P6: 0.78) and for the SFA level (P6: 0.78 vs. S6: 0.90). This in turn suggests that as age increases, the digestibility of SFA improves and, most notably, the utilization of FFA improves.
5. Conclusions
The present study confirms that determining the lipid classes together with the FA digestibility along the GIT provides valuable information for better understanding the dynamics of FA utilization in diets with different FA profiles and FFA contents. The results demonstrate that the effect of dietary saturation degree (UFA:SFA) on dietary fat utilization is higher than the effect of dietary FFA level. A clear improvement in the efficiency of both the lipolysis and the absorption process was observed with age. Fat hydrolysis is more affected by the age of the chicken than by the saturation degree and/or FFA content of the diet. The absorption results demonstrated that most of the FA absorption occurs in the jejunum (from 73% to 92%), but the ileum also plays a key role (from 8% to 27%). The contribution of the upper and lower segments of the jejunum and ileum to FA absorption is influenced by the characteristics of the dietary FA (degree of saturation, chain length, and FFA%), and the age of the chicken. There is a notably higher implication of the upper ileum for grower broiler chickens.
Replacing soybean oil with palm fatty acid distillate affected the extent and the site of FA absorption. The results show that the increase in SFA% and FFA% in the diet reduced and delayed the absorption of the dietary FA, especially the SFA in starter broiler chicks. As age increased, the FA absorption increased, and advanced to the upper intestinal segments, especially in the most saturated and rich FFA diets. Age has a positive effect on the digestibility of SFA and, above all, on the FFA utilization. For 11-day-old starter broiler chickens, it is not recommended to use this by-product alone or in blends. For grower broiler chickens, it is possible to include palm fatty acid distillate blended with a conventional unsaturated oil, such as soybean oil, in feed formulation, when the blend has from 2.6 UFA:SFA and the FFA% does not exceed 30%, without impairing FA utilization or growth performance. This potential strategy for using palm fatty acid distillate without negatively impacting fat utilization by the animal implies a reduction in costs and a way to upcycle and valorize this by-product.