1. Introduction
Rabbit production is improving in productivity, which has led to the design of more efficient diets, since rabbits’ feeds are formulated mainly based on by-products high in fiber, animal and vegetable fats, and other ingredients that contain nutrients sufficient for maintaining an efficient productivity [
1]. However, rabbit is a species that produces excellent meat, including nutritional characteristics and potential health properties, with this meat and its derivatives considered as functional foods due to their functional compounds [
2]. One of the benefits of this species is that it can be fed with different fibrous material, parts of herbs and spices as alternatives to additives or ingredients [
3]. In addition, feed costs associated with producing rabbit meat are high which is why alternatives are being sought in order to decrease them. One ingredient which could be an alternative is the fruit from the tree called
Pithecellobium dulce.
P. dulce is a fruit which originates from the Americas in countries including Brazil, Argentina, Colombia, and Mexico, but it is also distributed in several countries around the world, such as India or the tropical regions of Africa. This tree belongs to the Fabaceae family and is one of the 18 species of the genus
Pithecellobium [
4]. Murugesan et al. [
5] reviewed therapeutic and biological properties of
P. dulce, indicating that it has insecticide, anti-diabetic, anti-hyperlipidemic, antioxidant, antiulcer, antidiarrheal, antibacterial, and other properties. Dhanisha et al. [
6] demonstrated that an extract of
P. dulce fruit induced apoptosis in vivo and in vitro. Furthermore, Vargas-Madriz et al. [
7] reviewed the antioxidant capacity and phenol profile of
P. dulce indicating the main phenolics reported are caffeic acid, chlorogenic acid, ferulic acid, gallic acid, p-coumaric acid, protocatechuic acid, apigenin, catechin, daidzein, kaempferol, luteolin, quercetin, myricetin, naringin, and rutin. However, it was also mentioned that antioxidant capacity varies according to all the studies reviewed.
P. dulce are used in combination with other plants to feed goats, using leaves [
8,
9] or fruits [
10] The above-mentioned findings indicate that the fruit of
P. dulce is an ingredient that could be used to elaborate animal feed. The objective of this study was to evaluate the effect of
P. dulce fruit on productive performance, carcass traits, meat characteristics, biochemical and hematology analysis; as well as meatballs prepared with rabbit meat; as a potential alternative ingredient to feed fattening rabbits.
2. Materials and Methods
2.1. Raw Material and Proximate Analysis
The fruit of
P. dulce was collected in San Miguel de las Palmas, Guerrero, Mexico. The fruits were dried at 28 °C under shadow, and were then grounded in an Antarix grinder model THCF2800M13 (Antarix de México, Mexico City, Mexico). Afterwards, a proximate analysis was performed according to AOAC methodology [
11] to determine moisture (930.15), crude protein (945.01), and crude fat (954.02). Regarding fiber fractions (NDF and ADF), the technique described by Van Soest et al. [
12] was used.
2.2. Animals and Experimental Design
This study and animal management were carried out according to the institutional committee guidelines on animal care (protocol number CICUA/ICAP 001/2020). The experiments were conducted in a rabbitry located in Tulancingo, Hidalgo, Mexico. Ambiental conditions in the rabbit production house had an average temperature of 17 °C and 70% relative humidity. Seventy-two rabbits were used, which were 35 d of age, unsexed, California × English pot crossbreed, and weighed 650 g on average. The animals were selected and distributed randomly in two treatments, a control group (C, n = 36) and a group (G5, n = 36) fed with 5% of fruit of P. dulce, with nine repetitions (n = 4 rabbits). The fattening period was 29 d. In addition, the animals were housed in cages measuring 45 × 40 × 60 cm which were adapted with automatic drinkers and manual feeders. The rabbits were fed ad libitum.
2.3. Diets
Diets were prepared following the nutritional requirements of the National Research Council [
13], while the ingredient composition was based on the guidelines provided by Fundación Española para el Desarrollo de la Nutricion Animal [
14]. Diet formulations had to be isoproteic (16%), isoenergetic (2.5 Mcal·kg
−1 of digestible energy), and isofibrous (16% Neutro Detergent Fiber) as shown in
Table 1. The ingredients were mixed in an ASF model MZ50 double helicoidal mixer (Molinos y Mezcladoras Industriales S. A. de C. V. Mexico), and then pelletized in a SKJ-120 feed pellet machine (Yuezhen Machinery Co., Jinan, China) and finally stored in a hermetic container until use.
2.4. Productive Performance
Feed consumption was registered daily (offered and rejected) while live weight was measured every week using a Mettria MTNUV-40 digital scale (Mettria México, Ciudad de México, Mexico). From the data obtained, the average daily feed intake (DFI), daily weight gain (DWG), and feed conversion rate (FCR) were calculated between ages 35 and 63 days. In addition, total weight gain and total feed intake were also determined, including initial and final weight of the experiment.
2.5. Apparent Digestibility
Dry matter, organic matter, neutral detergent fiber, and acid detergent fiber were determined according to Perez et al. [
15]. Briefly, 8 rabbits by group were selected to perform apparent digestibility, then feces were collected from each cage every morning during the last 4 days of the fattening period. Afterwards, feces were dried in a Riossa model HCF82D oven (RSU Labsupply, Monterrey, NL, Mexico). Content for moisture, ash, neutral detergent fiber (NDF), and acid detergent fiber (ADF) was determined in both feces and feed as indicated above. Subsequently, the digestibility coefficient was calculated.
2.6. Carcass Traits
After the fattening period, animals (63 days of age,
n = 32 rabbits by group) were weighed and transported to the meat laboratory belonging to the Instituto de Ciencias Agropecuarias and then slaughtered without previous fasting and mechanical concussion stunning according to national legislation [
16]. Before slaughtering, the dorsal length and the lumbar circumference of the animals was measured (from the atlas to the last ischia vertebra) using a measuring tape. After evisceration, the weights of the skin, feet, hot carcass, viscera (including esophagus, trachea, digestive apparatus, heart, lungs, kidneys, and liver), carcass length, and lumbar circumference were obtained. Carcasses were stored in refrigeration at a temperature of 4 °C for 24 h. Afterwards, cold carcass and main cuts (head, forequarter, thoracic cage, foreleg, and legs) were obtained according to the indications provided by Blasco et al. [
17]. Then, legs were dissected into meat, fat, and bone using a Scout Pro model SP402 scale (Ohaus Corporation, Pine Brook, NJ, USA).
2.7. Hematological and Biochemical Analysis
During the exsanguination procedure, 3 mL of blood were collected in vacutainer tubes (n = 9 by treatment) to determine blood biochemical analysis using BA400 Biosystem equipment. Another tube was used to collect blood and perform blood biometry using a hematology analyzer Procyte Dx (Idexx laboratories Inc., Westbrook, ME, USA).
2.8. Meat Characteristics
The carcasses were kept for 24 h under refrigerated conditions, then color was measured using a Minolta colorimeter model CM-580d (Konica-Minolta, Tokyo, Japan) with a CIEL*a*b* color space using an illuminant D65, and 0.8 cm aperture size. The observer was set to 10° according to the American Meat Science Association meat color measurement guidelines [
18]. For measuring pH, a Hanna HI99163 meat pHmeter (Hanna Instruments, Cluj-Napoca, Romania) was used. Furthermore, in order to determine water holding capacity (WHC), a technique described by Honikel [
19] was employed. Cooking loss was measured by cutting half of a loin which was then weighed and cooked in a hot water bath at 80 °C for 20 min. Subsequently, 1 cm
3 meat cubes were analyzed for a texture profile analysis in a Brookfield CT3 texture analyzer (Brookfield, Middleboro, MA, USA). The equipment was adapted with a TA3/1000 probe and set up to compress the sample at 50% using a crosshead speed of 1 mm·s
−1. The sample was compressed twice. Force–time graph parameters of hardness, resilience, cohesiveness, springiness, and chewiness were obtained using Texture Pro CT software (Brookfield, Middleboro, MA, USA).
2.9. Analysis of Meatball
Meat obtained from legs was ground in a Torrey grinder (Torrey, Monterrey, NL, Mexico); the meat was separated into two batches. The meatballs were prepared by adding 10 g of salt and 200 mL of water to 1 kg of rabbit meat, which were then mixed. Afterwards, 50 g portions were made and stored on plastic trays, covered with film, and then stored at 4 °C until analysis.
To determine the effect of
P. dulce, microbiological and physicochemical analysis was performed on days 0, 7, and 14, with dilutions and bacterial counts analyzed using indications according to national legislation [
20]. Total viable counts of bacteria
Enterobacteriaceae and
Staphylococcus were tested. Antioxidant activity was determined according to Brand-Williams et al. [
21] with 2,2-Diphenyil-1-picrylhydrazyl (DPPH) as radical, while antioxidant activity was expressed in mg·mL
−1. The pH was measured using a Hanna meat pHmeter model HI99163 (Hanna Instruments, Cluj-Napoca, Romania). Finally, water activity (Aw) was determined with a HP23-aw HygroPalm (Rotronic Measurement Solutions, Bassersdorf, Switzerland).
Meatballs were subjected to a sensory analysis using an affective hedonic test to determine the acceptability of the meat’s taste. A total of 120 consumers with an average of 21.5 years participated, of which 41.7% were female and 57.5% were male. A hedonic 7-point scale affective test (7 like very much and 1 dislike very much) was undertaken to determine acceptability. The test was developed according to the indications provided by Drake [
22].
2.10. Statistical Analysis
In this work, a completely randomized design was used to analyze productive performance parameters, including total weight gain, total feed consumption, and feed conversion ratio; for these variables, treatment was the fixed effect and cage was random term. On apparent digestibility of dry matter, organic matter, neutral detergent fiber, and acid detergent fiber, all carcass traits, all meat characteristics, all biochemical and hematology analysis, an analysis of variance following the general linear model procedure was carried out, continuing with a lsmeans option, using treatment as fixed effect and cage as random term. Statistical model was following:
where Y
ij = dependent variable, μ = mean of the variable, β
i = the fixed effect of i-th rabbit of the group, δ
k = the random effect of k-th cage, and ε
ij = experimental error associated with the observation Y
ij.
A repeated time one-way design was used to analyze feed consumption and daily weight gain through the fattening period; treatment and time was used as fixed effect, with the nested time in treatment, then cage was utilized as random term. Also, all variables measured to determine meatballs’ shelf-life were analyzed through a repeated time one-way design, where treatment and time was used as fixed effect, with the nested time in treatment, then batch was utilized as random term.
where Y
ij = dependent variable, μ = mean of the variable, β
i = the fixed effect of i-th rabbit of the group, τ
j = time and β
i(τ
j) time inside treatment, δ
l = the random effect of l-th cage or batch, and ε
ijk = experimental error associated with the observation Y
ijk.
For sensory analysis, firstly a statistical descriptive analysis was carried out, subsequently a Student’s t-test was conducted to determine differences between samples of meatballs elaborated with meat from rabbits given feed with or without P. dulce, one session in approximately two hours was conducted, then data collected as indicates above.
The statistical models were the following:
where
t is
t-test, the numerator is the difference between the two consumers groups to taste meat, denominator is an estimate of the standard error. All analyses were performed using SAS software ver. 9.0.
4. Discussion
There is little information about
P. dulce fruits’ proximate composition. However, there are some studies that revealed chemical composition is diverse; when the fruit is divided into its main parts (seed, aril, and husk), the seed has 39% protein, while dry aril is between 12 and 15%, and fresh aril is 3%; which indicates that composition is influenced by the part of the fruit analyzed [
23,
24]. Yet, whole
P. dulce fruits have a high quantity of protein (21.5%) and NDF (78%); according to these amounts, those fruits can be used to feed animals as an additive or ingredient.
Productive performance results indicate the use of
P. dulce fruits to feed rabbits saw similar results between groups. However, there are several studies on the use of fruits, vegetables, extracts, or essential oils to feed fattening rabbits. Some of these investigations concluded that there are no effects on productive performance, such as the studies by Perna et al. [
25] which used cauliflower leaves, Mancini et al. [
26] who fed rabbits with quebracho and chestnuts tannins mixes, as well as Kovitvadhi et al. [
27] who focused on
Lythrum salicaria. Other studies indicated an increase in some productive performance parameters, such as Khalid et al. [
28] who observed that Moringa oleifera leaf powder increased daily gain and improved the feed rate conversion. In another study, Elwan et al. [
29] fed rabbits with 1 and 2% of Citrus limon powder and found that the parameters related to final weight, weight gain, and average daily gain all increased. Perhaps,
P. dulce fruits did not improve productive performance, but they can be used to feed rabbits.
Apparent digestibility coefficients were similar to other ingredients in rabbit feeds, including bilberry pomace [
30]. It is notable that cellular content has high digestion influenced by
P. dulce fruits. It is possible that this fruit has small quantities of tannins (according to Roselin and Parameshwari [
31]); their study also reviewed bioactive compounds in
P. dulce which indicated the presence of tannins. Also, according to Mancini et al. [
26], tannins did not influence the palatability of the feed. However, in contrast to this study, several studies indicate that dry or organic matter digestibility is not influenced by tannins. For example, Kovitvadhi et al. [
27] stated that low levels of tannins did not affect digestibility, while similar results were obtained by Dabbou et al. [
30]. Dalle-Zotte et al. [
32] did not find any differences in dry matter and organic matter digestibility when feeding rabbits with chestnut hydrolysable tannins.
In general, the reference values for these biochemical parameters were normal with regard to rabbits; however, control groups had low levels for alanine transferase, aspartate transferase, and phosphorus compared to G5 group. It is possible that a hepatic malfunction was a consequence of intensive production or use of
P. dulce. However, biochemical parameters were normal as indicated by Brandão et al. [
33]. Aljohani and Abduljawad [
34] reported similar concentrations of alanine and aspartate transferase in rabbits fed with
Moringa oleifera, although values reported in their study were higher than those obtained in this experiment. In contrast, Imbabi et al. [
35] found aspartate transferase decreased in rabbits fed with fennel oil, although alanine transferase was similar.
Most of the carcass traits were similar between groups, except feet and drip loss (the growth of animals is related to productive performance; if these last parameters are analogous between groups, it is possible that organs and body composition were similar.) There are other studies that report similar results, such as Wolf and Cappai [
36] who used rabbit feed incorporated with acorns (
Quercus pubescens Willd.) and there was no difference in carcass traits. However, the use of Moringa olefira leaves increased carcass traits in a study by Aljohani and Abduljawad [
34]. Moreover, drip loss is related to protein and the accumulation of lactic acid [
37], meaning that it is possible that
P. dulce slightly increased protein content while the pH also remained slightly high.
Meat characteristics of rabbits fed with
P. dulce fruits were similar between groups, except a* value and hardness. The maximum shear force is correlated with the connective tissue [
38]. It is possible that
P. dulce induced a lower proportion of collagen instead of structural proteins; Sembiring et al. [
39] demonstrated that
Muntingia calabura extract leaves increase collagen density. However, other studies using leaves or fruits did not find differences, such as Pałka et al. [
40] who supplemented rabbit feed with
Urtica dioica L. or
Trigonella foenum-graecum L., or Garcia-Vazquez et al. [
41] who used an infusion of
Chenopodium ambroisoides. Also, the change in redness color value is associated to Fe and anthocyanins content in the aryl of
P. dulce fruits, as indicated by Rao et al. [
23].
P. dulce fruits feed to rabbits decreases Staphylococcus counts. Koné et al. [
42] observed that plant extracts or essential oils supplemented into rabbits’ diets decreased bacterial content in rabbit meat under anaerobic conditions. Mancini et al. [
43] used 3.5% of turmeric powder to extend the shelf-life of rabbit burgers.
The acceptance of rabbit meat burgers was similar between treatments. Other studies have observed similar results using bilberry pomace [
44] and goji berries [
45], while Mancini et al. [
46] used
Zingiber officinale roscoe powder in rabbit meat burgers to improve sensorial characteristics.