Early Solid Diet Supplementation Influences the Proteomics of Rumen Epithelium in Goat Kids

Simple Summary The rumen, as a unique digestive organ of ruminants, is vitally important to their growth and production, while the rumen of young ruminants is not fully developed. Previous studies have proven that early solid diet supplementation can significantly promote the development of rumen and improve their growth performance. However, the changes in the expressed proteome and related metabolism in rumen epithelium in response to a supplemented solid diet remains unclear. In this study, we confirmed that a solid diet significantly promoted the growth performance, rumen fermentation, and rumen epithelial development of goat kids. According to proteomic analysis, we further identified the proteins and pathways related to cell growth and volatile acid metabolism which were significantly changed via solid diet supplementation. This study can support the optimal breeding strategy to improve the performance and the growth potential of young ruminants. Abstract It is well known that solid diet supplementation in early life can significantly promote rumen development and metabolic function in young ruminants. However, the changes in the expressed proteome and related metabolism in rumen epithelium in response to a supplemented solid diet remain unclear. In this study, rumen epithelial tissue from goats in three diet regimes including milk replacer only (MRO), milk replacer supplemented concentrate (MRC), and milk replacer supplemented concentrate plus alfalfa pellets (MCA) were collected for measurement of the expression of epithelial proteins using proteomic technology (six per group). The results showed that solid diet significantly improved the growth performance of goats, enhanced the ability of rumen fermentation, and promoted the development of epithelial papilla (p < 0.05). Proteome analysis revealed the distinct difference in the expressed protein in the MRC and MCA group compared with the MRO group (42 upregulated proteins and 79 downregulated proteins in MRC; 38 upregulated proteins and 73 downregulated proteins in MCA). Functional analysis showed that solid diet supplementation activated a variety of molecular functions in the epithelium, including protein binding, ATP binding, structural constituent of muscle, etc., in the MRC and MCA groups. Meanwhile, the expression of proteins related to fatty acid metabolism, the PPAR signaling pathway, valine, leucine, and isoleucine degradation, and butanoate metabolism were upregulated, being stimulated by solid feed. In contrast, the proteins associated with carbohydrate digestion and absorption and glycosaminoglycan degradation were downregulated. In addition, the protein expression of enzymes involved in ketone body synthesis in the rumen was generally activated, which was caused by solid feed. In summary, solid feed promoted the development of rumen epithelium by changing the expression of proteins related to fatty acid metabolism, energy synthesis, and signal transduction. The ketone body synthesis pathway might be the most important activated pathway, and provides energy for rumen development.


Introduction
Rumen plays a key role in ruminants' performance and production [1]. Goat kids are born with immature rumen that have no physiological or metabolic functions [2]. Physical and metabolic development of the rumen is essential for its smooth transition from non-ruminant to mature ruminant state, and for improving the growth performance of young ruminants [3]. As we know, the introduction of solid diets in early life is a vital driver for the development of rumen epithelium due to the nutrient components in solid diet [4]. Previous studies found that supplementing solid feed during the pre-weaning period can effectively increase rumen weight and papillae size, enhance the physical barrier of rumen to harmful substances, and, ultimately, have a positive effect on the health and growth of young ruminants [5][6][7]. Other studies have reported that early feeding starter can influence the gene expression of rumen epithelium in lambs and sheep, for example, downregulates genes including IL-6, IL-10, and IFN-γ [8], and upregulates genes containing MCT1, MCT4, and NHE3 [9]. Experiments with goats confirmed that highgrain diets simultaneously increased the volatile fatty acids (VFA) production and the expression of genes involved in VFA absorption and cell proliferation in rumen [10][11][12]. In calf studies, the implementation of similar methods significantly increased the expression of fat and muscle tissue synthesis genes [13]. Recently, studies on the effects of different diets on rumen epithelial mRNA gene expression have received more attention. However, few studies have investigated the expressed proteins in rumen epithelium as affected by diet. In sheep and dairy cattle, a study reported the differential expression of selected proteins related to material transport and metabolism due to changes in diet and rumen environment [14,15]. Therefore, it is necessary to perform systemic cognition on rumen epithelial protein expressions of goats that are fed a supplemented solid diet, which can help us understand the molecular mechanism of rumen development and further improve the feeding strategy for young ruminants. Isobaric tags for relative and absolute quantitation (iTRAQ) are used in quantitative proteomics due to their high sensitivity and conveniences [16]. In this study, we applied proteomic analysis based on iTRAQ [17] to evaluate changes in the protein expression of rumen epithelium in early supplementation goats compared to a control group without a solid diet. Through the exploration based on the molecular level, we learned how solid diet drives proteome changes in rumen epithelium. We hypothesized that solid feed supplementation can promote the growth of rumen epithelium via stimulating the expression of proteins related to cell development and volatile acid metabolism. This work can support the optimal breeding strategy to improve the performance and the growth potential of young ruminants.

Ethics Statement
The study was conducted at the Green Sheep Valley Farm in Haimen City, Jiangsu Province. Procedures for breeding and slaughtering were implemented in accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals promulgated by the Ministry of Science and Technology, China, revised version, March 2017. The trial was also reviewed and approved by the Animal Ethics Committee of the Chinese Academy of Agricultural Sciences (AEC-CAAS-FRI-CAAS20180305).

Animals and Diets
Thirty-six pairs of twin goats with an average weight of 4.53 ± 0.52 kg were separated from their dams at 20 days of age and randomly divided into three groups. One group was fed with milk replacer only (MRO), which was provided by Beijing Precision Animal Nutrition Research Center, China, one group was fed with milk replacer supplemented with concentrate (MRC), and one group was fed with milk replacer supplemented concentrate plus alfalfa pellets (MCA). Each group had six replicates and four kids per pen as a replicate. During the trial, all goat kids had ad libitum access to water, milk replacer (MR), concentrate, and alfalfa pellets. Nutritional levels of MR, concentrate, and alfalfa pellets are shown in Table S1. At 60 days of age, six goats (healthy and BW close to the average of the corresponding groups) were chosen from each group, and slaughtered for rumen sample collection. The rumen epithelial tissue in ventral sac was quickly harvested and snap frozen in a liquid nitrogen tank for total protein extraction.

Determination of Rumen Fermentation Parameters and Morphology
The rumen content samples were thawed at 4 • C and then centrifuged at 2500× g at room temperature. Next, 1 mL of the supernatant per sample was separated and transferred into a 1.5 mL centrifuge tube which contained 0.2 mL of metaphosphoric acid solution (25% w/v). Then, the mixture was centrifuged at 10,000× g at 4 • C after placing in a water bath for 30 min. The collected supernatant was stored at 4 • C for the subsequent analysis. The VFA concentration was detected using gas chromatography (GC-6800, Beijing Beifen Tianpu instrument Technology, Co., Ltd., Beijing, China). The determination of enzyme activity (Pepsin, A-amylase, Lipase, Carboxymethyl Cellulase) was mainly carried out according to the operation procedure of the corresponding kits.
A 2 cm × 2 cm section of rumen epithelium tissue collected from each goat kid was directly washed with physiological saline and fixed in a 250 mL jar containing 10% neutral formalin solution after slaughter. The samples were dehydrated by different concentrations of ethanol, embedded in paraffin sections, and cut into 6 µM sections. The rumen papilla structure was observed under a light microscope at a magnification of 4 × 10 times (Olympus BX-51; Olympus Corporation, Tokyo, Japan) after staining with Yi-hong-hematoxylin (H.E.). The image-pro express image analysis processing system (Im-age-Pro Plus 6.0, Media Cybernetics, Silver Spring, MD, USA) was used to observe and measure the rumen papilla length, papilla width, lamina propria thickness, and epithelial thickness.

Liquid Chromatography-Tandem Mass Spectrometry (LC/MS) Analysis
Proteins were extracted by using lysis buffer 3 (Configuration method: 1. Urea 210 g, Thiourea 76 g, SDS 1 g, Tris 1.2 g, put in a beaker. 2. Add Milli-Q H2O 250 mL and put it on a magnetic agitator overnight (or completely dissolve). 3. The concentrated HCL is adjusted to pH 8.0-8.5. The volume of Milli-Q H2O is fixed to 500 mL, and then packed separately for reserve.) and two magnetic beads. The mixtures were placed into a TissueLyser for 2 min at 50 Hz to release proteins. After centrifugating, the supernatant was transferred into a new tube, and reduced with 10 mM dithiothreitol (DTT) at 56 • C for 1 h and alkylated by 55 mM iodoacetamide (IAM) in the dark at room temperature for 45 min. After centrifugation (25,000× g, 4 • C, 20 min), the supernatant containing proteins was quantified by Bradford assay. We mixed 15-30 µg proteins with loading buffer in centrifuge tube and heated them at 95 • C for 5 min. Then, the supernatant was centrifuged at 25,000× g for 5 min and loaded to sample holes in 12% polyacrylamide gel. The SDS-PAGE at a constant voltage of 120 V for 120 min was performed to detect proteins quality. Once finished, we stained the gel with Coomassie Blue for 2 h, then added destaining solution (40% ethanol and 10% acetic acid) and placed it on a shaker (exchange destaining solution 3~5 times, 30 min a time). The protein solution (100 µg) with 8 M urea was diluted 4 times with 100 mM Tetraethylammonium bromide (TEAB). Then, the proteins were digested at 37 • C overnight by Trypsin Gold (Promega, Madison, WI, USA) in a ratio of protein: trypsin = 40:1. After trypsin digestion, the peptides were desalted using Strata X C18 column (Phenomenex) and vacuum-dried according to the manufacturer's protocol. The peptides were dissolved in 30 µL 0.5M TEAB with vortexing. After the iTRAQ labeling reagents were recovered to ambient temperature, they were transferred and combined with proper samples. Peptide labeling was performed by iTRAQ Reagent 8-plex Kit (8-plex iTRAQ reagent Multiplex kit, ABSciex, Framingham, MA, USA) according to the manufacturer's protocol. The labeled peptides with different reagents were combined (Table S2), desalted with a Strata X C18 column (Phenomenex), and vacuum-dried according to the manufacturer's protocol. The separation of peptides was carried out on a Shimadzu LC-20AB HPLC Pump system coupled with a high pH RP column. The peptides were reconstituted with buffer A (5% ACN, 95% H 2 O, adjust pH to 9.8 with ammonia) to 2 mL and loaded onto a column containing 5 µm particles (Phenomenex). The peptides were separated at a flow rate of 1 mL/min with a gradient of 5% buffer B (5% H 2 O, 95% ACN, adjusted pH to 9.8 with ammonia) for 10 min, 5-35% buffer B for 40 min, and 35-95% buffer B for 1 min. The system was then maintained in 95% buffer B for 3 min and decreased to 5% within 1 min before equilibrating with 5% buffer B for 10 min. Elution was monitored by measuring absorbance at 214 nm, and fractions were collected per 1 min. The eluted peptides were pooled into 20 fractions and vacuum-dried. Each fraction was resuspended in buffer A (2% ACN, 0.1% FA) and centrifuged at 20,000× g for 10 min. The supernatant was loaded onto a Thermo Scientific™ UltiMate™ 3000 UHPLC system (Thermo Scientific, Sunnyvale, CA, USA) equipped with a trap and an analytical column. The samples were loaded on a trap column at 5 µL/min for 8 min, and then eluted into the homemade nanocapillary C18 column (ID 75 µm × 25 cm, 3 µm particles) at a flow rate 300 nl/min. The gradient of buffer B (98% ACN, 0.1% FA) was increased from 5% to 25% in 40 min, and then increased to 35% in 5 min, followed by 2 min linear gradient to 80%, then maintenance at 80% B for 2 min, and, finally, returned to 5% in 1 min and equilibrated for 6 min. The peptides separated from nanoHPLC were subjected into the tandem mass spectrometry Q EXACTIVE HF X (Thermo Fisher Scientific, San Jose, CA, USA) for DDA (data-dependent acquisition) detection by nano-electrospray ionization. The parameters for MS analysis were listed as follows: electrospray voltage: 2.0 kV; precursor scan range: 350-1500 m/z at a resolution of 60,000 in Orbitrap; MS/MS fragment scan range: >100 m/z at a resolution of 15,000 in HCD mode; normalized collision energy setting: 30%; dynamic Exclusion time: 30 s; automatic gain control (AGC) for full MS target and MS2 target: 3 × 10 6 and 1 × 10 5 , respectively. The MS/MS scan numbers following one MS scan: 20 most abundant precursor ions above a threshold ion count of 10,000.

Protein Quantification and Data Analysis
The raw MS/MS data were converted into MGF format, and the MGF files were searched by the local Mascot server against the database (Table S3). Besides, quality control was performed to determine if a reanalysis step was needed. An automated software, called IQuant, was applied to the quantification of proteins. All proteins with a false discovery rate (FDR) less than 1% proceeded to downstream analysis. Proteomic data are available by contacting the corresponding author.

Statistics
One-way ANOVA in SPSS 19.0 (SPSS Inc., Chicago, IL, USA) was used to compare the differences in rumen fermentation parameters and rumen epithelial morphology. The expression level of proteins involved in rumen ketogenesis was compared using Wilcoxon rank-sum test.

Growth Performance and Rumen Development
In this study, we observed that solid feed supplementation could improve the growth performance of goat kids. Specifically, solid feed significantly increased the average daily gain (ADG), dry matter intake (DMI), and feed conversion rate (FCR) of goat kids in the MRC and MCA group (p < 0.05) (Table S4). In addition, compared with the MRO group, solid feed also significantly promoted the development of rumen epithelium in the MRC and MCA group, including an increased epithelium thickness and an increased height and width of the rumen papillae (p < 0.05) ( Figure 1A, Table S5). The concentration of rumen VFA (mainly acetate, butyrate, and propionate) showed similar conditions (p < 0.05) ( Figure 1B) [1,10]. generated by the R package ggplot2 (https://github.com/tidyverse/ggplot2; accessed on 9 December 2022).

Growth Performance and Rumen Development
In this study, we observed that solid feed supplementation could improve the growth performance of goat kids. Specifically, solid feed significantly increased the average daily gain (ADG), dry matter intake (DMI), and feed conversion rate (FCR) of goat kids in the MRC and MCA group (p < 0.05) (Table S4). In addition, compared with the MRO group, solid feed also significantly promoted the development of rumen epithelium in the MRC and MCA group, including an increased epithelium thickness and an increased height and width of the rumen papillae (p < 0.05) ( Figure 1A, Table S5). The concentration of rumen VFA (mainly acetate, butyrate, and propionate) showed similar conditions (p < 0.05) ( Figure 1B) [1,10].

The Identification of Differently Expressed Proteins (DEPs)
Based on analysis of iTRAQ data, a total of 114,401 (94,287 unique) spectra were identified. Of the 26,793 peptides, 24,609 peptides were unique. A total of 6003 proteins were identified with 1% FDR during the process of analysis (Table S6), and more than half of the proteins contained at least two peptides. The sequence coverage of most of the identified proteins was less than 20%, and approximately 60% of the proteins in the protein mass distribution were between 10 and 70 kDa.

The Identification of Differently Expressed Proteins (DEPs)
Based on analysis of iTRAQ data, a total of 114,401 (94,287 unique) spectra were identified. Of the 26,793 peptides, 24,609 peptides were unique. A total of 6003 proteins were identified with 1% FDR during the process of analysis (Table S6), and more than half of the proteins contained at least two peptides. The sequence coverage of most of the identified proteins was less than 20%, and approximately 60% of the proteins in the protein mass distribution were between 10 and 70 kDa.
For this experimental design with more than one replicate, proteins with 1.8-fold change (mean value of all comparison groups) or 0.56-fold change and a p-value (t-test of all comparison groups) less than 0.05 were defined as DEPs. In this experiment, in the MRC vs. MRO group, 121 kinds of DEPs were identified, of which 42 were upregulated and 79 were downregulated ( Figure 2A). Further, 111 kinds of DEPs were detected in the MCA vs. MRO group, including 38 upregulated proteins and 73 downregulated proteins ( Figure 2B). In addition, In the MCA vs. MRC group, a total of 35 DEPs were observed, including 15 upregulated proteins and 20 downregulated proteins. ( Figure 2C). As shown in Figure 2D, 56 DEPs were shared with the MCA vs. MRO group and MRC vs. MRO group. The Heatmap also showed that, compared with the MCA vs. MRC group, there were more similar protein expression characteristics between the MCA vs. MRO group and MRC vs. MRO group ( Figure 2E). all comparison groups) less than 0.05 were defined as DEPs. In this experiment, in MRC vs. MRO group, 121 kinds of DEPs were identified, of which 42 were upregul and 79 were downregulated ( Figure 2A). Further, 111 kinds of DEPs were detected in MCA vs. MRO group, including 38 upregulated proteins and 73 downregulated pro ( Figure 2B). In addition, In the MCA vs. MRC group, a total of 35 DEPs were obser including 15 upregulated proteins and 20 downregulated proteins. ( Figure 2C). As sh in Figure 2D, 56 DEPs were shared with the MCA vs. MRO group and MRC vs. M group. The Heatmap also showed that, compared with the MCA vs. MRC group, t were more similar protein expression characteristics between the MCA vs. MRO g and MRC vs. MRO group ( Figure 2E). To be specific, the DEPs of Shisa family member 4 (SHISA4) and Glutathione S-tr ferase (GSTA3) were all upregulated in both the MRC vs. MRO and MCA vs. MRO gro In the MRC vs. MRO group, Semaphorin 3A (SEMA3A) and Tyrosine-protein ki (FGR) were upregulated and emaphorin 3D (SEMA3D) and the NLR family pyrin dom containing 12 (NLRP12) were downregulated. In addition, Thyroid hormone receptor To be specific, the DEPs of Shisa family member 4 (SHISA4) and Glutathione S-transferase (GSTA3) were all upregulated in both the MRC vs. MRO and MCA vs. MRO groups. In the MRC vs. MRO group, Semaphorin 3A (SEMA3A) and Tyrosineprotein kinase (FGR) were upregulated and emaphorin 3D (SEMA3D) and the NLR family pyrin domain containing 12 (NLRP12) were downregulated. In addition, Thyroid hormone receptor beta (THRB) and Semaphorin 3D (SEMA3D) were the most up-or down-regulated DEPs in the MCA vs. MRO group (Tables 1 and 2).

The Enrichment Analysis of DEPs According to Gene Ontology (GO)
To further explore the specific functional characteristics of the DEPs, we performed a GO enrichment analysis of the DEPs using David Bioinformatics Resources [18]. In general, in the MRC vs. MRO, the DEPs were significantly enriched in the 77 terms (p < 0.05), in which 23 terms were annotated with molecular function (MF), 15 terms were annotated with cellular component (CC), and 39 terms were annotated with biological process (BP) ( Table S7). In the MCA vs. MRO, a total of 73 enriched terms were detected (p < 0.05), including 23 terms in MF, 18 terms in CC, and 32 terms in BP (Table S8). In the MCA vs. MRC group, 117 terms were enriched significantly (p < 0.05), consisting of 17 terms in MF, 21 terms in CC, and 79 terms in BP (Table S9)

The KEGG Enrichment Analysis of DEPs
To understand the effects of the DEPs on the related pathways after the introduction of solid feed, we performed a KEGG pathway enrichment analysis of the DEPs. In the MRC vs. MRO group, the DEPs were enriched in 32 pathways, most of which belonged to the categories of organismal systems and metabolism. The network cluster analysis showed that the pathways with a high enrichment ratio, including Carbohydrate digestion and absorption (chx04973), Synthesis and degradation of ketone bodies (chx00072), and Butanoate metabolism (chx00650), were the critical nodes connecting the respective clusters ( Figure 4A,D). In the MCA vs. MRO group, the DEPs were enriched in 29 pathways. Similarly, organismal systems and metabolism were the main categories. The pathways of carbohydrate digestion and absorption (chx04973), other glycan degradation (chx00511), Glycosaminoglycan degradation (chx00531), Fatty acid elongation (chx00062), Fatty acid degradation (chx00071), and PPAR signaling (chx03320) were determined to be important nodes, as shown by their high enrichment ratio ( Figure 4B,E). In addition, the pathways enriched by DEPs in the MCA vs. MRC group were mainly enriched in the pathways of Nucleotide excision repair (chx03420) and Proximal tubule bicarbonate reclamation (chx04964) (Figure 4C,F).  . Each row represents an enriched function, and the length of the bar represents the enrichment ratio, which is calculated as "input gene number"/"background gene number." The color of the bar is the same as the color in the circular network above, which represents different clusters. For each cluster, if there are more than five terms, the top five with the highest enrichment ratio will be displayed. (D-F) Each bubble represents an enriched function, and the size of the bubble from small to large: The color of the bar is the same as the color in the circular network, which represents different clusters. For each cluster, if there are more than five terms, the top five with the highest enrichment ratio will be displayed.

The PPI Analysis of DEPs
In order to further understand the interaction between DEPs and their related functional pathways, PPI network analysis was performed according to the STRING database and KEGG results ( Figure 5). In the MRC vs. MRO group, a total of 23 DEPs were identified as key nodes which showed various connections in the network, including 6 upregulated DEPs (SLC26A3, Interleukin 1 receptor type 2 (IL1R2), HMGCS2, HMGCL, ATP1A3, and ACAA2) and 16 downregulated DEPs (Tropomyosin 1 and 2 (TPM1 and 2), Sorbin and SH3 domain containing 1 (SORBS1), PRKCB, Cardiac phospholamban (PLN), Perilipin 4 (PLIN4), Beta-mannosidase (MANBA), Laminin subunit alpha 4 (LAMA4, G protein subunit alpha o1 (GNAO1), Chondroitinsulfatase (GALNS), Collagen type VI alpha 6 chain (COL6A6), Collagen type III alpha 1 chain (COL3A1), Collagen type I alpha 1 chain . Each row represents an enriched function, and the length of the bar represents the enrichment ratio, which is calculated as "input gene number"/"background gene number." The color of the bar is the same as the color in the circular network above, which represents different clusters. For each cluster, if there are more than five terms, the top five with the highest enrichment ratio will be displayed. (D-F) Each bubble represents an enriched function, and the size of the bubble from small to large: The color of the bar is the same as the color in the circular network, which represents different clusters. For each cluster, if there are more than five terms, the top five with the highest enrichment ratio will be displayed.

Rumen Ketogenesis in Response to Solid Feed Supplementation
In this study, considering the higher concentration of VFAs and the length of rumen papillae in response to solid feed supplementation, we identified the protein expression of key enzymes in the process of rumen ketogenesis to characterize the efficiency of energy production in the development of rumen ( Figure 6A). Compared with the MRO group, the protein expressions of the enzymes including Acetyl CoA synthetase (ACSS2), HMGCS2, HMGCL, and D-beta-hydroxybutyrate Dehydrogenase (BDH1) were significantly higher (p < 0.05) in the MRC group ( Figure 6B). Similarly, in the MCA group, the protein expressions of ACSS2, HMGCS2, and HMGCL were also higher than in the MRO group (p < 0.05) ( Figure 6C). the protein expressions of the enzymes including Acetyl CoA synthetase (ACSS HMGCS2, HMGCL, and D-beta-hydroxybutyrate Dehydrogenase (BDH1) were signi cantly higher (p < 0.05) in the MRC group ( Figure 6B). Similarly, in the MCA group, t protein expressions of ACSS2, HMGCS2, and HMGCL were also higher than in the MR group (p < 0.05) ( Figure 6C).

Rumen Fermentation Parameters and Papillae Development
In the study, the diet with solid feed supplantation increased the concentrations of VFAs in the rumen and promoted the growth of rumen papillae [1,4]. On the one hand, the introduction of solid feed provides rich fermentation substrates for nutrient-decomposing bacteria in the rumen [20,21]. The VFAs produced by rumen microbiota became the primary energy supply for the development of rumen epithelium [22,23]. Moreover, the physical friction of the solid feed against the rumen wall further stimulates the growth of the rumen papillae [7,9]. A larger rumen epithelial surface had higher metabolic efficiency for nutrients, including VFAs and microbial protein, to meet the more exuberant growth demands of the host [24]. In addition, the metabolic pattern of rumen epithelium is also regulated by solid feed. Thus, we further discussed the effect of solid feed supplementation on rumen epithelial proteins from the view of biochemistry and metabolism in detail.

Cell Development of Rumen Epithelium
In this study, some GO terms were associated with the cell development of rumen epithelium, including extracellular matrix structural constituent conferring tensile strength, extracellular matrix organization, extracellular matrix structural constituent, collagencontaining extracellular matrix, plasma membrane, extracellular exosome, extracellular region, etc., which contained the DEPs of COL3A1, Collagen type XVI alpha 1 chain (COL16A1), COL6A6, COL1A1, LAMA4, FGR, formin-like 3 (FMNL3), etc.
The complete construction of rumen epithelial cells was a prerequisite for the rumen metabolism [25]. Previous studies have proven that COL3A1, COL16A1, COL6A6, and COL1A1 were involved in the regulation of blood vessel development and construction in the gut epithelium. LAMA4, encoding a secreted glycoprotein, was found to influence cell adhesion, which plays a key role in cell proliferation, the maintenance of activity, differentiation, and migration [26] In this study, these DEPs were decreased in response to solid feed. FMNL3 was identified as an effector of Rho GTPases, contributing to different cellular actin cytoskeleton structures by its ability to polymerize straight actin filaments at the barbed end [27]. FGR was a critical covalent modification and occurs in multicellular organisms as a result of intercellular communication during the maintenance of adult tissues [28]. In contrast, FMNL3 and FGR were increased with solid feed supplantation in this study. According to our findings, we thought solid feed introduction promoted the development of rumen epithelium by regulating the expression of proteins related to cell growth and integrity. As we know, solid feed is rich in nutrients including starch, fat, protein, etc. The complex and diverse microbial communities in the rumen can efficiently ferment them into VFAs and high-quality microbial proteins, which are energy substrates that can be directly absorbed and utilized by rumen epithelial cells [4]. Hence, we hypothesized that the supplement of solid feed provided these material bases for the rumen growth of goat kids, and promoted the faster development of rumen epithelial cells, including cell adhesion and frame construction. However, the internal mechanism of these regulations still needs to be confirmed in further study.

Metabolism of Rumen Epithelium
As we know, with an increase in age, the main digestive organs of young ruminants are transferred from the hindgut to the mature rumen [29,30]. VFAs also replace glucose as the main energy substance for their growth [21,31]. In this study, the metabolism pathways associated with VFAs, including fatty acid elongation, fatty acid degradation, butanoate metabolism and synthesis, and the degradation of ketone bodies, were the most significant, which could be attributed to the increasing concentration of butyrate after solid feed supplementation. The rumen epithelial mitochondria is the main site of butyric acid metabolism. The energy produced during the conversion of butyrate to ketone bodies is the main source of energy for epithelial growth [32]. Hence, the ability of the rumen to produce ketone bodies (mainly β-hydroxybutyrate acid (BHBA)) is considered to be an important marker of rumen development [33]. In terms of biochemical steps, the expression of enzyme proteins involved in VFA metabolism and ketone body synthesis showed a significant increase, including ACSS2, HMGCL, HMGCS2, ACAA2, and BDH1. Acetyl-CoA is an important pivotal substance in energy metabolism which is not only a carbon source for the synthesis of ketones or cholesterol, but also participates in the tricarboxylic acid cycle (TCA) for energy production [34]. As the synthetase of acetyl-CoA, the increased expression of ACSS2 protein meant the active performance of energy metabolism in the rumen epithelium. The significant GO term of ATP binding also corresponded to this result. HMGCS2 is the rate-limiting enzyme involved in the generation of ketone bodies from lipids in the rumen [35]. Recent studies have also reported that the accumulation of BHBA and the increase in HMGCS2 expression in ketogenesis strengthened the ability of gut epithelial cells to proliferate, differentiate, and maintain gut homeostasis [36,37]. BDH1 is the last enzyme of rumen ketogenesis, catalyzing the reversible reduction of acetoacetate (ACAC) to BHBA [38]. ACAA2 is an enzyme of the thiolase family that performs the function of mitochondrial fatty acid extension and degradation by catalyzing the last step of the β-oxidation pathway. ACAA2 often showed a high expression in tissues with active lipid metabolism such as mammary gland and liver [39]. In addition, some DEPs associated with the transport of fatty acids were also identified. The protein expressions of two transporters (SLC26A3 and SLC16A1) were increased with the introduction of solid feed. SLC26A3, as an anion exchanger, mediates apical Cl − /HCO 3 exchange in intestinal epithelia [40]. SLC26A3 was also observed in the rumen epithelium, where its main function is to transport short chain fatty acid (SCFA) [41]. The surface of rumen epithelium maintained an acidic environment due to the accumulation of a large number of VFAs, and SLC26A3 promotes the absorption of short-chain fatty acids (SCFA-) through SCFA-/HCO3-exchange [42]. SLC16A1 (MCT1) has been suggested to be a rumen transporter, mediating SCFA, ketone bodies, and lactic acid [43]. After SCFA have reached the mitochondrial-rich layer cells by the diffusion of functional syntheses, they undergo oxidative metabolism, and subsequent metabolites are transported to the blood by SLC16A1. Therefore, rumen epithelium utilized the high concentration of VFAs fermented from solid feed via increasing the expression of proteins related to VFA metabolism and transport. One study also showed that the direct infusion of sodium butyrate to neonatal lambs could promote the expression of genes related to VFA absorption and ketone body metabolism in rumen epithelium, which is consistent with our results [23]. These results suggested that butyrate fermented from solid feed is the most direct factor which promotes the development of rumen epithelium.
In contrast, we observed that the significant pathways of carbohydrate digestion and absorption were inactive, which was associated with two downregulated DEPs (AKT2 and PRKCB). Protein kinase C (PKC), encoded by the PRKCB1, is a family of serine-and threonine-specific protein kinases which could be involved in endothelial cell apoptosis, intestinal sugar absorption, etc. AKT2 is associated with the positive regulation of glucose import, and the knockout or decreased expression of Akt2 could inhibit the uptake and utilization of glucose by cells [44], enhance the resistance to fat accumulation, and improve the metabolic efficiency of lipids [45]. The energy metabolism process of ruminants changed from glucose-based into volatile acid-based with the increase in age and the development of rumen [46]. Combined with our results, the introduction of solid feed accelerated this process, and the high concentration of VFAs in the rumen dominated the metabolic pattern of rumen epithelium, which inhibited the metabolism of glucose and promoted the utilization of VFAs in the mitochondria via decreasing the expression of AKT2 and PRKCB.

Signal Transduction of Rumen Epithelium
In this study, PPAR signaling pathways were significantly enriched with upregulated protein (HMGCS2) and downregulated protein (Perilipin 4 (PLIN4) in both the groups of MRC/MCA vs. MRO. The induction of HMGCS2 is mainly controlled by the positive regulation of peroxisome proliferator-activated receptors (PPARs) [47]. SCFAs, as efficient ligands, could effectively activate the PPAR signaling pathway and promote the expression of the target gene HMGCS2 [48]. PLIN4, as a lipid droplet protein (LDP), exists widely in oxidized tissue, and the inactivation of PLIN4 could promote the metabolism of lipids for preventing their accumulation [49]. Thus, we hypothesized that VFAs in the rumen as ligand could activate PPAR signaling pathways, which would regulate the expression of HMGCS2 and PLIN4 for promoting fatty acid metabolism.

Conclusions
In summary, solid feed as an initiating agent increased the growth performance of goat kids, enhanced the ability of rumen fermentation, and promoted the development of rumen epithelium. In addition, according to the previous studies on the transcriptional level, we further explored the effect of solid feed supplantation on the development of rumen epithelium at the protein level. As the results indicate, solid feed promoted the development of rumen epithelium by changing the expression of proteins related to cell construction, fatty acid metabolism, and PPAR signal transduction. The ketone body synthesis pathway might be one important pathway which is activated in response to solid feed supplementation, which could provide abundant energy for rumen development. Our findings broaden the theoretical knowledge of the intrinsic mechanisms of rumen development in young ruminants.
Supplementary Materials: Supplementary materials can be found at https://www.mdpi.com/ article/10.3390/biology12050684/s1, Table S1: Nutritional components of milk replacer, concentrate, and alfalfa pellets (dry matter basis) %; Table S2: iTRAQ labeling information; Table S3: Mascot Search Parameters; Table S4: Effect of solid feed supplantation on growth performance of goat kids; Table S5: Effects of solid feed supplantation on the rumen morphology in 20~60 days old of goat kids; Table S6: The identification of proteins in the study; Table S7: Gene ontology analysis of differently expressed proteins in the MRC vs. MRO group; Table S8: Gene ontology analysis of differently expressed proteins in the MCA vs. MRO group; Table S9: Gene ontology analysis of differently expressed proteins in the MCA vs. MRC group.
Author Contributions: Y.Z. contributed to animal trial, data collection and evaluation, laboratory and statistical analysis, and writing; K.C. contributed statistical analysis and critical manuscript review; J.C. and X.L. contributed to manuscript review and data visualization; N.Z. contributed to study design, feed formulation, and critical manuscript review. All authors have read and agreed to the published version of the manuscript. Informed Consent Statement: Not applicable.

Data Availability Statement:
The related data in this study can be available by contacting to the corresponding author.