1. Introduction
The microbial population that inhabits specific niches in the body, developing symbiotic relationships with the host, is called the microbiota. This complex ecosystem of microbial communities is involved in important physiological processes in the host, such as nutrient digestion and absorption [
1], immune system development [
2], intestinal permeability [
3], and both direct and indirect protection against the development of dangerous or pathogenic bacteria [
4]. The gut microbiota plays an important role in healthy animals but can also be a cause of pathology [
5]. In recent decades, the development of high-throughput sequencing technologies, such as 16S ribosomal RNA (rRNA) amplicon sequencing, has reduced execution costs and facilitated knowledge of the microbiota composition of the different organs of the gastrointestinal tract (stomach, small, and large intestine) in various animal species such as pigs [
6], poultry [
7], and horses [
8]. Also, 16S rRNA sequencing can be used to study the changes induced by managerial, environmental, and individual factors in microbial populations [
9,
10,
11]. Recently, this technology has also been used in domestic rabbits (
Oryctolagus cuniculus) that are monogastric, hindgut-fermenting herbivores and rely on cecotrophy. The rabbit is a very interesting species because it is considered, at the same time, a farm animal, a pet, and also a laboratory animal [
12]. The composition of the microbiota of different intestinal tracts has been evaluated in livestock [
13], laboratory [
14], pet, and shelter rabbits [
15], finding a different microbiota composition in the diverse gastrointestinal tracts. In particular, the large intestine shows the highest richness and diversity in bacterial species, while the small gut has the highest variability in the gastrointestinal tract of rabbits [
13]. Several studies have been performed to evaluate the effects of the quality and levels of the diet [
16,
17], dietary supplementation with nutraceuticals [
18,
19,
20], the temperature of the drinking water [
21], age [
22], and weaning period [
23]. The fecal microbiota and its functional capacity associated with weaning weight in meat rabbits [
24], hygiene of the environment [
25], season [
15], and drug treatments [
26,
27] have been investigated. It was established that these factors influencing the microbiota composition can also affect the quality of the carcass and meat of the rabbit [
28,
29,
30]. Actually, only a few studies evaluated the effect of the breeds and the growth rate on the microbiota composition [
31]. Significant differences in the gut microbiota observed in two rabbit breeds and some families, Ruminococcus and Lachnospiraceae, could be considered biomarkers for improving the health and production performance of meat rabbits [
31]. The present study aimed at comparing, using a 16S rRNA-based analysis, the microbial populations in the jejunum and cecum of two breeds of rabbits: the Native Middle Egypt Rabbit—NMER (NM) as a local breed with a light body weight and the Giant Flander (GF) as an exotic breed with a heavy body weight. In addition, histology, scanning electron microscope (SEM) examinations, and biochemical parameters were used to compare NM with GF rabbits.
2. Materials and Methods
2.1. Experimental Design and Sampling
The experimental protocols were approved by the Animal Production Research Institute (APRI)’s animal care and use committee (ethical approval number: 2021920153429). The current research was conducted by APRI, the Agricultural Research Center (ARC), at the Rabbitry Research Farm in Sakha, Kafr el-Sheikh Governorate, which is located in the north of Egypt, during the period from June to September 2020.
A total of 200 Native Middle-Egypt Rabbit-NMER (NM) and Giant Flanders (GF) breeds, NM (n = 107) and GF (n = 93), rabbits in growing stages (5–12 weeks) were used in the present study. At 5 weeks of age, the young rabbits were weaned and housed in individual cages with 35 × 35 × 25 cm dimensions and equipped with feeding hoppers made of galvanized steel and nipples for automatic drinking. The animals were bred under the same environmental conditions and were watered and fed ad libitum from a commercial pelleted diet without probiotics or antibiotics. The ingredients and chemical composition of the diet used in the experiment are presented in
Table 1.
At 12 weeks, 20 male rabbits of the NM and GF breeds with similar weight at weaning (from 490.5 g to 542.5 g, average = 516.5 g) were selected for the study to avoid the effect of sex on studied traits. The rabbits were euthanized at 12 weeks. At euthanasia, the gastrointestinal tracts were removed shortly after death, and the contents of the jejunum and cecum were collected and immediately stored at −80 °C until genomic DNA extraction.
2.2. Blood Biochemical Parameters
Before euthanasia, blood samples were collected from all the rabbits (n = 10 for each breed) and centrifuged at 1500×
g for 20 min; the serum samples were then kept at −80 °C for further biochemical parameter analyses: glucose, total protein, albumin, globulin, triglycerides (TG), and urea were analyzed according to the manufacturing instructions of the Biodiagnostic company kits (Dokki, Giza, Egypt;
www.bio-diagnostic.com, accessed on 22 March 2021). The total protein and albumin differences in the collected samples were used to determine the globulin levels. The biochemical parameters were measured using a UV-VIS spectrometer (model T60UV, PG Instruments Limited, Lutterworth, UK).
2.3. Histological Characteristics
Immediately after the excision, the samples from the jejunum and cecum (n = 10 for each breed, NM and GF) were preserved in neutral buffered formalin. Then, the tissues were dehydrated in an ascending grade of ethanol and embedded in paraffin wax. Serial sections were cut at 5 μm with a microtome (Galileo SEMI, Diapath, Martinengo, Italy). The cross-sections of both the jejunum and cecum—three sections for each organ per rabbit—were stained with Mayer’s hematoxylin and eosin (H&E). Histological morphometric characteristics in the jejunum and cecum included villus height (VH), villus width (VW), and tunica muscularis (TM), according to [
32]. Ten individual villus and ten loci in the tunica muscularis were measured for each rabbit. Histological characteristics were studied using a Leica light microscope and imaging software from Leica Microsystems (Application Suite 3.1.0 software, Leica, Wetzlar, Germany).
2.4. Electron Microscopic Examination
The collected jejunum and cecum samples (n = 10 for each breed, NM and GF, and one sample for each organ) were fixed in 3% glutaraldehyde and 0.1 M sodium cacodylate buffer (pH 7.0) for 2 h at room temperature, then rinsed in the same buffer, and finally post-fixed in 1% osmium tetroxide for another 2 h at room temperature. The scanning electron microscope (SEM) samples were dehydrated in an ethanol series ranging from 10% to 90% for 15 min in each alcohol dilution, followed by 30 min in absolute ethanol. The SEM samples were critical-point dried by using liquid carbon dioxide. Specimens were mounted on aluminum stubs with silver paint and coated with gold/palladium in a SPI-Module Sputter Coater device (SPI Supplies, West Chester, PA, USA). The SEM was performed in the electron microscope JEM-2100 (JEOL, Ltd., Tokyo, Japan) at a 20 kV accelerating voltage for studying the villi of the cecum and jejunum.
2.5. DNA Extraction, Library Generation, and Sequencing
Total bacterial genomic DNA was isolated from the content of the jejunum and cecum (n = 10 for each breed, NM and GF) by using the Easy Pure Stool genomic kit protocol (TransGen Biotech, Haidian District, Beijing, China) following the manufacturer’s instructions. The 16S ribosomal RNA (rRNA) gene was amplified using primers targeting the V3-V4 hypervariable regions according to the 16S Metagenomic Sequencing Library Preparation (
https://support.illumina.com/documents/documentation/chemistrydocumentation/16s/16s-metagenomic-library-prep-guide-15044223-b.pdf, accessed on 3 July 2023, Illumina, San Diego, CA, USA). All of the PCR amplifications were performed in 25 µL. A total of 12.5 µL of KAPA HIFI Master Mix 2× (Kapa Biosystems, Inc., Wilmington, MA, USA) and 0.2 µL of each primer (100 µM) were added to 2 µL of genomic DNA (5 ng/µL). Blank controls (no DNA template added to the reaction) were also performed. The PCR conditions were as follows: 94 °C for 5 min, 98 °C for 30 s, 56 °C for 60 s, and 72 °C for 60 s for a total of 25 cycles, with a final extension step at 72 °C for 10 min.
Amplicons were cleaned with Agencourt AMPure XP (Beckman, Coulter Brea, CA, USA), and libraries were prepared following the 16S Metagenomic Sequencing Library Preparation Protocol (Illumina, San Diego, CA, USA). The libraries obtained were quantified by real-time PCR with KAPA Library Quantification Kits (Kapa Biosystems, Inc., Wilmington, MA, USA), pooled in equimolar proportion, and sequenced in one MiSeq (Illumina) run with 2 × 250-base paired-end reads.
2.5.1. Bioinformatics—Sequence Processing
Raw paired reads from each sample were merged into one single sequence per fragment by PandaSeq [
33]; then, low-quality bases (Phred quality score < 3) were trimmed from the 3′-end and fragments having a length < 75% of the initial fragment length were discarded. Filtered reads were clustered into zero-radius operational taxonomic units (zOTUs) by USEARCH (v. 11.0.667, Edgar), retaining only those supported by 5 or more reads. All downstream analyses were performed in the QIIME 1.9.0 suite [
34]. Taxonomic assignment of zOTUs was performed by the RDP classifier [
35] against the SILVA 132 database [
36], using 0.5 as a confidence threshold.
The dataset comprised a total of 40 samples, derived from 2 organs (jejunum and cecum) and 2 rabbit breeds (NM and GF). Each combination had 10 independent replicates. The comparison among the experimental categories comprised three levels of analysis: (a) Breed: comparison between microbiota from GF and NM rabbits (as a whole, considering both organs); (b) Organ: comparison between microbiota from caecum and jejunum (as a whole, considering both breeds); and (c) stratified analyses comparing organs within the same breed and breeds in samples from the same organ.
2.5.2. Statistical Analysis Diversity
In order to have a comparable representation of the bacterial communities, read counts per sample were normalized to the least-sequenced one, at 18,561 reads per sample. The analysis of the sample biodiversity (i.e., alpha-diversity) was based on different metrics (i.e., Shannon’s diversity, chao1 diversity index, observed species, and Faith’s phylogenetic diversity index (PD whole tree)). A non-parametric permutation-based t-test (equivalent to the Mann–Whitney U-test) with 999 random permutations was made to assess whether the samples belonging to one experimental class were statistically different from those of a different class.
Beta-diversity analysis was performed according to the unweighted and weighted UniFrac distances among samples and represented via a Principal Coordinate Analysis (PCoA). A statistical test (i.e., the “Adonis” test, Permutational Multivariate Analysis of Variance Using Distance Matrices, using pseudo-F ratios) was used in order to define whether there was a significant difference among the experimental groups using 999 random permutations. Statistical analyses and graphs were performed in Matlab (v. 2008a, Natick, MA, USA).
2.5.3. Co-Abundance Analysis
This analysis was aimed at identifying groups of bacterial genera whose abundance was correlated with each other. Spearman’s rank correlation was calculated for all the bacterial genera having abundance > 0.5% in at least 50% of the samples in each experimental category: cecum NM, cecum GF, jejunum NM, and jejunum GF. On the basis of the correlations, bacterial genera were clustered in co-abundance groups (CAGs), following the procedure originally developed by Claesson and co-workers [
37] and using Euclidean distance and average linkage; Cytoscape v. 3.0 [
38] was used to graphically represent CAGs, as well as the relative abundance of bacterial genera and strength of correlation.
2.5.4. Microbiota-Body Weight Correlations
Spearman’s rank correlation was used to analyze the correlation between bacterial genera, the body weight, and the weight gain of the animals measured at 12 weeks of age.
2.5.5. Microbial Function Prediction
Metabolic functional capacities of gut bacteria were predicted from 16S rRNA data using the Tax4Fun R package [
39] and the Kyoto Encyclopedia of Genes and Genomes (KEGG). Tax4Fun transformed the SILVA-classified zOTUs into prokaryotic KEGG organisms and normalized them according to the 16S rRNA copy number. The output tables provided KEGG orthology (KO) numbers for gene annotations and Enzyme Commission (EC) numbers.
2.6. Statistical Analysis
The individual animal was considered the experimental unit, and data on body weight, weight gain, and blood biochemical and histological parameters were represented as means with standard deviations (SEM) by using SAS 2002 software’s GLM technique (SAS, Cary, NC, USA). A one-way ANOVA was used to compare the means, and the differences were considered significant at p < 0.05.
4. Discussion
The body weight and weight values in NM and GF rabbits were similar to the findings (NM, 1169.4 vs. GF, 1327.7) of our previous work [
40]. Moreover, we already reported that GF rabbits were significantly heavier and had a higher relative growth rate as compared to NM rabbits. NM is a rabbit breed established in three governorates in Middle Egypt by the Animal Production Research Institute (APRI) [
41] and, according to the general classification used in the European Rabbit Breed Standards Book [
42], belongs to the small-sized breed, while GF belongs to the medium-sized breed.
Total proteins and globulins were significantly affected by the breed, as documented in previous studies in four breeds [
43] and three breeds [
44] under Egyptian environmental conditions. High globulin concentrations in GF rabbits could be attributed to increasing the gamma globulin fraction, which indicates immunity status [
45]. Previous studies have indicated that an altered serum component profile could reflect differences in the gut microbiome of rabbits [
31], which is consistent with our findings as shown in
Figure 4. This study suggested that host rabbit breeds can shape the gut microbiome and serum metabolome, including the interactions among the host-gut microbiome and serum metabolome that are important. Thus, total proteins and globulins may be indicators of the health status and production traits of meat rabbits.
The villi on the jejunum and cecum were higher in GF than NM rabbits in addition, the villi of the cecum in MN were less dense and their tips were marked by irregular edges, as shown in the SEM examination. These observations may be due to some kind of toxins produced in the intestine and cecum [
46] by bacteria that were in close contact only with epithelial cells that had lost their brush border [
47]. In addition, the fermentation operated by bacteria could produce hydrogen ions that can induce damage in the cecum and jejunum mucosa [
46].
The microbiota composition of the samples was not significantly different when comparing the two rabbit breeds (NM and GF) within the same intestinal tract. In fact, among the major constituents of the microbiota (relative abundances > 1%), none of them were found to be significantly diverse. On the other hand, differences were more evident when comparing the jejunum vs. cecum microbiota, in particular in the GF breed, where both alpha- and beta-diversity estimations were found to be significantly different. This was also reflected when considering the functional predictions deriving from the microbial profiles, with jejunum and cecal samples appearing very different from one another. The predominant phylum in microbiota samples for both intestinal tracts (jejunum and caecum) was
Firmicutes, followed by
Patescibacteria and
Bacteroidetes (in cecum) and
Proteobacteria and
Actinobacteria (in jejunum), in accordance with the fact that
Firmicutes was found to be the most dominant phylum in the rabbit microbiota, regardless of source, age, and season [
14]. Our results are also in harmony with the findings of Fu et al. [
48], who reported that Firmicutes was the most dominant phylum in the foregut and hindgut of rabbits, while the second most dominant one was Proteobacteria (in the foregut) or Bacteroidetes (in the hindgut). Another study reported that Firmicutes were the most abundant phylum in all of the sections of the gastro-intestinal tract examined (45.9%), such as the stomach, duodenum, jejunum, ileum, cecum, and colon, followed by
Bacteroidetes in the large intestine (38.9%),
Euryarchaeota (29.6%), and
Patescibacteria (13.8%) in the foregut, especially in jejunum [
13]. The abundance of members of the Firmicutes phylum was higher in GF than NM rabbits, both in the cecum (73.5% GF vs. 63.9% NM) and jejunum (44.0% GF vs. 40.2% NM) samples. Previous studies have shown that a high abundance of
Firmicutes, as shown in GF, can enhance intestinal mucosa and reduce oxidative stress in the intestinal tract in piglets [
49,
50]. This concept was further supported by histological examinations and scanning electron microscopy (SEM) in the cecum and jejunum villi of GF rabbits.
Within the
Firmicutes phylum, the most abundant order was
Clostridiales and, among genera, those from the
Ruminococcaceae and the
Lachnospiraceae families. As highlighted in the co-abundance groups’ analysis, the abundance of the members of these two families was highly correlated, establishing a sort of “core” microbiota. A high abundance of
Clostridiales,
uncultured Clostridiales vadinBB60 group, and
Ruminiclostridium 5, such as that highlighted in the jejunum of GF, may reduce the effect of
Escherichia coli as a pathogenic agent through remodeling the signaling pathway [
51]. The high abundance of
Bacteroides and
Ruminococcus in the cecum of GF breed rabbits could be related to a healthy gut, in accordance with the findings of previous studies on rabbits [
31].
In cecum samples, members of the
Eubacteriaceae family occupied a central position in the interaction network, with positive correlations observed between
Lachnospiraceae,
Ruminiclostridium, and
Eubacteriaceae, possibly due to their functional potential in the specific metabolic pathways.
Eubacteriaceae and
Lachnospiraceae bacteria exhibited metabolic specificities for pyruvate and carbohydrate degradation [
24]. Members of
Ruminococcaceae seemed to be highly specialized in pyruvate-to-lactate fermentation [
52,
53], as happens in the degradation of plant material such as pectin and cellulose in the colonic fermentation of dietary fibers in mammals [
54,
55]. These results could be confirmed by Ye et al. [
31], who reported that families
Ruminococcus and
Lachnospiraceae could be considered biomarkers for improving the health and production performance of meat rabbits. Moreover, other low-abundance microorganisms, such as the
Akkermansia genus (
Verrucomicrobiales order), could play a key role in the hydrolysis of diverse ingested polysaccharides and contribute to a more complete digestion of dietary cellulose [
51,
56]. The significant decrease in the proportion of
Verrucomicrobia in the jejunum of the NM breed as compared to the cecum suggests less optimal jejunum health as well as a more pro-inflammatory state, as already reported in mice [
57]. The positive correlations among
Ruminiclostridium,
Lachnospiraceae, and
Eubacteriaceae that were observed in the cecum with increasing
Verrucomicrobia in the jejunum of the GF could explain the significant difference in the microbiota composition between organs in the GF breed.
The abundance of the
Escherichia-Shigella genus in the cecum of NM breed rabbits was positively correlated to body weight gain, as were Enterococcus and both body weight at 12 weeks and body weight gain in the jejunum of the same animals.
Escherichia-Shigella should be considered a pathogen in hosts [
9]; meanwhile,
Enterococcus could be considered a beneficial bacteria that could produce bacteriocins active against bacteria such as Listeria and indigenous clostridia in the gut of rabbits [
58]. On the other hand, in the cecum of the GF rabbits, there were significant correlations between Lactobacillus and both the body weight at 12 weeks and the body weight gain.
Lactobacillus can promote the fermentation of carbohydrates into lactic acid and intestinal health [
51], increase the concentration of short-chain fatty acids (SCFAs) in the intestines of mice, promote the growth of intestinal epithelial cells [
59], and have a biological antagonistic effect on pathogenic bacteria such as
E. coli. These results confirmed that the cecum is the main organ harboring the microbial fermentation processes in the gut of the rabbit [
16], and hosting
Lactobacillus, as occurred in the GF breed, could cause increased growth.
The functional potential of cecum samples was higher than that of jejunum samples, in particular for the secretion system, transporters, amino acids and carbohydrate metabolism, ribosome biogenesis, and translation. This could be attributed to the fact that cecum is the richest and most diverse microbial community in the rabbit gut [
14,
60]. The significantly different carbohydrates and lipid metabolic activity between the cecum and jejunum we observed could be due to the difference in microbial communities and their capacity to ferment to obtain metabolic energy [
9].
The limited number of replicates per condition (n = 10) and the high variability observed in the microbial profiles of the samples could have negatively influenced a better characterization of the differences in the microbiota composition inherent to the two breeds. On the other hand, we were able to highlight the different bacterial communities inhabiting the jejunum and cecum tracts of both rabbit breeds and gain insight into the specific functions they preside over.