Animals, diet, and abomasal catheterization surgery

The experimental design was previously described by Trotta [23]. Briefly, 24 Holstein steers [initial body weight (BW) = 250 ± 23.1 kg; 10–12 months of age] were initially housed in individual pens (3 m × 3 m) in the Intensive Research Building of the University of Kentucky C. Oran Little Agricultural Research Center in Versailles, KY. Before surgery, steers were deprived of feed (36 h) and water (12 h), fitted with temporary jugular vein catheters (Mila International Inc., Florence, KY), and received a subcutaneous injection of ceftiofur (6.6 mg/kg of BW) and intravenous injection of flunixin meglumine (1 mg/kg of BW). Steers were sedated by intravenous injection of xylazine (0.088 mg/kg of BW) with ketamine hydrochloride (1.76 mg/kg of BW), intubated, and anesthesia was maintained with isoflurane in O₂ using a ventilator in left lateral recumbency. The depth of anesthesia was monitored through observations of eye reflex and heart rate throughout the surgery. Steers were surgically fitted with infusion catheters in the abomasum that were constructed of Tygon tubing (6.35-mm internal diameter; Saint-Gobain North America, Malvern, PA) [30]. Flunixin meglumine was administered intravenously 24-h post-surgery and steers were monitored for abnormal feed intake, fecal consistency, behavior, and mobility for 7 d, and rectal temperatures were monitored for 3 d post-surgery.

Steers were limit-fed 5.65 ± 0.468 kg of an alfalfa hay cube-based diet (82.9% alfalfa hay cubes, 16.2% finely ground corn, 0.45% trace mineral premix, 0.45% vitamin premix; 40.7% neutral detergent fiber, 33.3% acid detergent fiber, 15.4% crude protein; 13.8% total starch, 1.32 Mcal/kg net energy for maintenance) on a dry matter (DM) basis. The diet was formulated to supply 1.33 times the net energy required for maintenance and exceed requirements for vitamins, minerals, ruminally degradable protein, and metabolizable protein for a steer gaining 0.46 kg/d [31]. Rations were provided twice daily at 0800 and 2000 h in equal portions. Steers were adapted to the basal diet for at least 7 d before the start of the treatment period.

Experimental design

Steers were transferred to metabolism stalls (1.2 m × 2.4 m) for the 7-d treatment period. Due to a limited number of stalls and infusion apparatus, steers were stratified by BW into 6 replicate blocks. Steers were then randomly assigned to one of four treatments within each replicate block. The experimental design was a randomized complete block design with a 2 × 2 factorial arrangement of treatments. All steers were abomasally infused with 3.94 ± 0.245 g raw corn starch · kg BW^-1 · d^-1. Along with raw corn starch, steers were abomasally infused with either water or 10% (wt:wt) sodium caseinate solution to provide 0 or 1.30 ± 0.299 g sodium caseinate · kg BW^-1 · d^-1. Additionally, steers received subcutaneous injection treatments in two equal portions daily at 0800 and 2000 h. Subcutaneous injection treatments were either vehicle (5 g/L bovine serum albumin (BSA) diluted in 9 g/L NaCl) or 100 μg GLP-2 in vehicle · kg BW^-1 · d^-1. This resulted in four treatments: 1) abomasal water + vehicle injection (control), 2) abomasal water + GLP-2 injection (GLP-2), 3) abomasal 10% sodium caseinate solution + vehicle injection (casein), and 4) abomasal 10% sodium caseinate solution + GLP-2 injection (casein + GLP-2). All steers received abomasal infusion and subcutaneous injection treatments for 7 d. The 7-d treatment period was chosen because postruminal casein infusion has been shown to increase small intestinal starch disappearance within 6 d [12] and acute (1 d) or chronic (10 d) exposure to GLP-2 at the same dosage level achieved pharmacological plasma concentrations of GLP-2 in cattle [32].

Glucagon-like peptide 2 preparation and administration

Glucagon-like peptide 2 was synthesized (Alan Scientific Inc., Gaithersburg, MD) based on the native bovine sequence [33, 34]. The purity (96%) was quantified by the manufacturer using high-performance liquid chromatography. The GLP-2 solution was prepared to a final concentration of 5 mg/mL by dissolving GLP-2 in 5 g/L BSA in 9 g/L NaCl solution. Aliquots were frozen in 10 mL screw-cap tubes at -20°C until use. Intestinotrophic effects of GLP-2 are greater when GLP-2 was administered subcutaneously compared with intramuscular or intraperitoneal injections [35]. Therefore, treatments were administered subcutaneously in the neck region in front of the shoulder. Left and right lateral administration sites were alternated for morning and evening injections, respectively.

Nutrient infusion preparation and techniques

Due to large inter- and intra-animal variability in postruminal starch flows when high-starch diets are fed to cattle [3, 36], postruminal starch flow was controlled via continuous abomasal infusion of starch. Raw corn starch (Clinton 185 Corn Starch; Archer Daniels Midland Company, Decatur, IL) was chosen as the starch source to facilitate the greatest limitations in small intestinal starch digestibility, as the action of pancreatic α-amylase and small intestinal α-glucosidases are required for complete hydrolysis to glucose. The chosen amount of infused raw corn starch was predicted to limit the extent of small intestinal starch disappearance [37] and mimic natural amounts of postruminal starch flow in cattle fed high-grain diets [38]. Additionally, infusion of 986 g of raw corn starch per day equates to 41.1 g/h, which does not cause diarrhea and appreciable amounts (10 to 15 g/h) of infused carbohydrate were expected to pass the ileocecal junction [39]. Sodium caseinate (AMCO Proteins, Burlington, NJ) was chosen as the protein source because of its relatively high solubility in water and because approximately 70–85% of luminal amino acids are transported across the apical membrane as small peptides [40]. The amount of casein was chosen relative to the amount of corn starch infused (1 part casein: 3 parts corn starch) and because previous experiments have demonstrated increased small intestinal starch disappearance [11, 15] and pancreatic α-amylase secretion [7] at similar levels.

Two containers of infusate suspensions per steer were prepared daily, immediately before infusion, for use over 12-h intervals. Infusate suspensions were maintained through continuous stirring via a stir bar with stir plate. Infusate suspensions (3.26 ± 0.109 kg) were prepared using the appropriate amount of raw corn starch, 5% (wt:wt) CrEDTA solution as an indigestible flow marker [41], and tap water every 12 h for each steer. A 10% (wt:wt) solution of sodium caseinate [42] was prepared with warm tap water in an insulated kettle (LEC-40; Legion Industries Inc., Waynesville, GA) using an industrial electric mixer (CDP3330; Baldor Electric Co., Ft. Smith, AR). The solution was mixed for at least 2 h and then was allowed to rest for 8 h to allow for air bubbles to escape [43]. The final solution was stored in 18.9-L buckets at -20°C until use. For casein treatments, the sodium caseinate solution replaced tap water in the infusate suspension. Corn starch and casein solute concentrations and abomasal infusion rates were similar to those reported by Kreikemeier [39] and Richards [15]. Abomasal infusion treatments were continuously infused (271 ± 9.08 g infusate/h) through Tygon tubing (3.18-mm internal diameter) using a multi-channel peristaltic pump (205U/CA; Watson-Marlow, Falmouth, United Kingdom). Infusion tubing was inserted into abomasal infusion catheters and secured with hose clamps. The amount infused was determined by the weight of the residual infusate after each 12-h period. The infusate containers were placed on top of a portable shelf (> 2 m height) to aid in the pumping of the suspension. Measures of fecal flow and postruminal starch disappearance from steers used in the current study were previously reported [23].

Tissue collection

Upon completion of the 7-d infusion period, the infusion tubing was disconnected from abomasal infusion catheters and steers were transported to the University of Kentucky Meats Laboratory within 1 h. Steers were humanely stunned via captive bolt and exsanguinated. After slaughter, the viscera were rapidly removed and separated for individual weights and subsample collection. The pyloric and ileocecal junctions were cut to separate the small intestine from the abomasum and cecum, respectively, and the mesentery was cut to separate the entire small intestine from the viscera. The small intestine was measured in length by looping the intestine between pegs at either end of a 2.43-m board [44]. The small intestine was separated into the duodenum (0.1 to 1.1 m caudal to the pyloric sphincter), jejunum (proximal half of non-duodenal small intestine), and ileum (distal half of non-duodenal small intestine) [45]. After the small intestinal length was measured, digesta was gently removed from the lumen of the small intestine. Each intestinal section (duodenum, jejunum, ileum) was weighed and sampled (1 m) from the midpoint. Each 1-m segment was cut into three equidistant subsamples, everted, and rinsed in ice-cold saline. The first subsample was cut laterally, scraped with a glass microscope slide, and 200 mg of mucosa was weighed and preserved with RNAlater stabilization solution (Thermo Scientific Inc., Waltham, MA) in RNase-free tubes. Tubes were initially stored at 4°C for 24 h to allow for tissue penetration and then were transferred to -80°C storage, as recommended by the manufacturer. The second subsample was scraped as described previously and the mucosa was flash-frozen in liquid nitrogen in an aluminum pouch and stored at -80°C for cellularity and enzymatic analyses. The third subsample was weighed, mucosa scraped from the entire section, and then the isolated mucosa was weighed. Measures of small intestinal length and mass, cellularity, and enzyme activity were previously published [23].

RNA isolation, library preparation, and sequencing

The jejunal mucosa was selected for RNA isolation and sequencing because small intestinal α-glucosidase activity (maltase, isomaltase, glucoamylase), Na⁺-dependent glucose uptake activity, and mRNA expression of intestinal glucose transporters (SLC5A1 and SLC2A2) is greatest in the jejunum of cattle [44, 46, 47]. Additionally, relative mRNA expression of the GLP-2 receptor (GLP2R) is greater in the jejunum than other gastrointestinal tissues [48]. Approximately 200 mg of jejunal mucosal tissue was shipped for RNA extraction, library preparation, and RNA sequencing by Novogene Corporation Inc. (Sacramento, CA). Total RNA was extracted from jejunal mucosal tissue using TRIzol reagent (Thermo Fisher Scientific Inc., Waltham, MA). RNA concentration (1,003 ± 863 ng/μL) was measured at 260 nm using a microvolume UV-VIS spectrophotometer (NanoDrop; Thermo Fisher Scientific Inc., Waltham, MA) and purity was assessed using the 260/280 nm ratio (range = 1.8 to 2.2) for protein contamination and 260/230 nm ratio (≥ 1.8) for nucleic acid contamination. Samples were assessed for RNA integrity (mean ± standard deviation = 6.59 ± 1.30; range = 4.2 to 9; median = 7) using the RNA 6000 Nano Kit (Agilent Technologies Inc., Santa Clara, CA) and automated electrophoresis (2100 Bioanalyzer Instrument; Agilent Technologies Inc., Santa Clara, CA). One sample in the control treatment was not prepared for RNA sequencing due to low RNA integrity (RIN < 4).

Strand-specific cDNA library preparation was conducted by Novogene Corporation Inc. (Sacramento, CA). Messenger RNA was isolated from total RNA using poly-T oligo-attached magnetic beads. Fragmentation was facilitated using divalent cations under elevated temperature in First Strand Synthesis Reaction Buffer (5X). First-strand cDNA was synthesized using random hexamer primer and Moloney murine leukemia virus reverse transcriptase and second-strand cDNA was synthesized using DNA polymerase I and RNAse H. Remaining overhangs were converted to blunt ends using exonuclease and polymerase activities. Three-prime ends of DNA fragments were adenylated and adaptors with hairpin loop structures were ligated for hybridization preparation. Library fragments were purified with AMPure XP System (Beckman Coulter Inc., Brea, CA) and 370–420 bp cDNA fragments were selected for PCR amplification. Polymerase chain reaction products were purified and assessed with the 2100 Bioanalyzer Instrument (Agilent Technologies Inc.). Strand-specific RNA sequencing was conducted NovaSeq 6000 Sequencing System (Illumina Inc., San Diego, CA) to generate 150-bp paired-end reads at an average depth of 55,666,406 reads per sample.

Bioinformatic and statistical analyses

Raw reads were filtered to remove sequencing adaptors, low-complexity reads, and reads containing low-quality bases using fastp software [49]. Quality control scores (Q20 = 96.8 ± 0.370%; Q30 = 91.41 ± 0.831%) and GC content (47.9 ± 1.25%) were calculated. Paired-end clean reads were aligned to the Bos taurus reference genome (ARS-UCD1.2) using HISAT 2 (version 2.0.5) [50]. Mapped reads of each sample were assembled using StringTie (version 1.3.3b) [51] in a reference-based approach. There was an average of 40,644,458 uniquely mapped reads per sample (73.0% of total reads). Reads per gene were counted using featureCounts (version 1.5.0-p3) [52]. Quantification of each gene was calculated as the number of fragments per kilobase of transcript sequence per million base pairs sequenced (FPKM). Visualization of box plots, violin plots, and FPKM density plots confirmed that gene expression was distributed equally across each steer and that the FPKM density conformed to a negative binomial distribution. Differential gene expression (DEG) analysis was quantified using the DESeq2 R-package (version 1.20) [53].

The DEG analysis used the negative binomial generalized linear model to fit gene expression level as a negative binomial distribution and Wald statistics to perform hypothesis testing. Multiple testing adjustment of the P-values was performed using the Benjamini-Hochberg procedure for controlling the false discovery rate (FDR). Genes were identified as differentially expressed if the FDR-adjusted P-value (padj) was ≤ 0.05. Gene ontology and KEGG pathway enrichment analyses of DEGs were conducted using the clusterProfiler R-package [54]. Enrichment analyses were considered significant if the FDR-adjusted P-value was ≤ 0.05. Tendencies were declared when 0.05 < FDR-adjusted P ≤ 0.10. Changes in gene expression were classified as upregulated or downregulated based on the sign of the log2 fold change.

Initially, pairwise comparisons between each of the four treatments were assessed. However, casein did not generate DEG that collectively affected KEGG pathway or gene ontology enrichment compared with control. In addition, there were minimal effects that resembled a casein × GLP-2 treatment interaction when GLP-2 vs. casein + GLP-2 treatments were compared. Therefore, all subsequent data analyses were conducted to compare the main effects of postruminal casein infusion (n = 11 for water, n = 12 for casein) and exogenous GLP-2 administration (n = 11 for BSA, n = 12 for GLP-2).

Summary of results

A summary of the effects of exogenous GLP-2 administration on differentially expressed genes, KEGG pathways, and gene ontologies is presented in Table 1. Exogenous GLP-2 administration upregulated (padj < 0.05) 667 DEGs, 26 KEGG pathways, 198 biological processes, 56 cellular components, and 60 molecular functions. Exogenous GLP-2 downregulated (padj < 0.05) 1,101 DEGs, 14 KEGG pathways, 270 biological processes, 105 cellular components, and 46 molecular functions. Postruminal casein infusion did not result (padj > 0.05) in any enriched KEGG pathways or gene ontologies of the jejunal mucosa.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Summary of the effects of exogenous GLP-2 administration on the number of differentially expressed genes, KEGG pathways, and gene ontologies of the jejunal mucosa.

https://doi.org/10.1371/journal.pone.0308983.t001

Differentially expressed genes

Exogenous GLP-2 administration resulted in 1,768 DEGs (padj < 0.05) with 667 upregulated genes and 1,101 downregulated genes (Fig 1). Postruminal casein infusion resulted in 7 DEGs (padj < 0.05) with 0 upregulated genes and 7 downregulated genes in the jejunal mucosa. Transcripts of the jejunal mucosa that were downregulated (padj < 0.05) with postruminal casein infusion include ENSBTAG00000047783 (log2 fold change = −4.00), ENSBTAG00000040367 (log2 fold change = −6.07), ENSBTAG00000047029 (log2 fold change = −1.13), CCR7 (C-C motif chemokine receptor 7; log2 fold change = −2.41), SFRP5 (secreted frizzled related protein 5; log2 fold change = −4.11), FCER2 (FC epsilon receptor II; log2 fold change = −3.48), and C1QTNF2 (C1q and TNF related 2; log2 fold change = −5.41) (S4 File).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Volcano plot of jejunal mucosal genes influenced by exogenous GLP-2 administration.

Data points highlighted in red were differentially expressed (padj < 0.05).

https://doi.org/10.1371/journal.pone.0308983.g001

Gene ontologies

The top 10 biological processes that were upregulated (padj < 0.05) with exogenous GLP-2 administration were the alcohol metabolic process, sphingolipid metabolic process, organic hydroxy compound metabolic process, anion transport, cellular lipid catabolic process, monocarboxylic acid metabolic process, carbohydrate metabolic process, transition metal ion transport, small molecule catabolic process, and lipid catabolic process (Fig 2). The top 10 cellular components that were upregulated (padj < 0.05) with exogenous GLP-2 administration were apical junction complex, apical part of the cell, brush border, Golgi subcompartment, Golgi apparatus part, apical plasma membrane, the cluster of actin-based cell projections, cell-cell junction, occluding junction, and Golgi membrane. The top 10 molecular functions that were upregulated (padj < 0.05) with exogenous GLP-2 administration were hydrolase activity (hydrolyzing O-glycosyl compounds), hydrolase activity (hydrolyzing glycosyl bonds), anion transmembrane transporter activity, coenzyme binding, transition metal ion transmembrane transporter activity, lipid transporter activity, organic anion transmembrane transporter activity, active transmembrane transporter activity, UDP-glycosyltransferase activity, and cofactor binding.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Top enriched gene ontology terms of the jejunal mucosa that were upregulated with exogenous GLP-2 administration.

GeneRatio is the ratio of upregulated DEGs to all genes in a gene ontology. Padj is the probability value for hypergeometric test adjusted for Benjamini-Hochberg FDR correction.

https://doi.org/10.1371/journal.pone.0308983.g002

The top 10 biological processes that were downregulated (padj < 0.05) with exogenous GLP-2 administration were ribonucleoprotein complex biogenesis, ribosome biogenesis, rRNA metabolic process, rRNA processing, ncRNA processing, ribosomal large subunit biogenesis, ribonucleoprotein complex assembly, ribosomal small subunit biogenesis, and ribonucleoprotein complex subunit organization (Fig 3). The top 10 cellular components that were downregulated (padj < 0.05) with exogenous GLP-2 administration were ribosomal subunit, cytosolic ribosome, ribosome, cytosolic part, cytosolic large ribosomal subunit, preribosome, large ribosomal subunit, cytosolic small ribosomal subunit, and nucleolar part. The top 10 molecular functions that were downregulated (padj < 0.05) with exogenous GLP-2 administration were the structural constituent of ribosome, rRNA binding, translation factor activity (RNA binding), catalytic activity (acting on RNA), translation initiation factor activity, catalytic activity (acting on DNA), chromatin binding, snoRNA binding, histone binding, and helicase activity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Top enriched gene ontology terms of the jejunal mucosa that were downregulated with exogenous GLP-2 administration.

GeneRatio is the ratio of upregulated DEGs to all genes in a gene ontology. Padj is the probability value for hypergeometric test adjusted for Benjamini-Hochberg FDR correction.

https://doi.org/10.1371/journal.pone.0308983.g003

KEGG pathways

The KEGG pathways that were upregulated (padj < 0.05) with exogenous GLP-2 administration were sphingolipid metabolism, adherens junction, bile secretion, galactose metabolism, vitamin digestion and absorption, mineral absorption, peroxisome, carbon metabolism, tight junction, other glycan degradation, starch and sucrose metabolism, retinol metabolism, lysosome, insulin resistance, ascorbate and aldarate metabolism, N-glycan biosynthesis, valine leucine and isoleucine degradation, glycerolipid metabolism, PPAR signaling pathway, and lysine degradation, protein digestion and absorption, AMPK signaling pathway, fat digestion and absorption, propanoate metabolism, gastric cancer, and various types of N-glycan biosynthesis (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Enriched KEGG pathways of the jejunal mucosa that were upregulated with exogenous GLP-2 administration¹.

https://doi.org/10.1371/journal.pone.0308983.t002

The KEGG pathways that tended to be upregulated (padj < 0.10) with exogenous GLP-2 administration were citrate cycle (TCA cycle), amino sugar and nucleotide sugar metabolism, tryptophan metabolism, arginine biosynthesis, serotonergic synapse, pentose and glucoronate interconversions, fructose and mannose metabolism, glycolysis/gluconeogenesis, Hippo signaling pathway, ether lipid metabolism, ABC transporters, drug metabolism–cytochrome P450, pyruvate metabolism, porphyrin and chlorophyll metabolism, and protein processing in endoplasmic reticulum (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Enriched KEGG pathways of the jejunal mucosa that tended to be upregulated with exogenous GLP-2 administration¹.

https://doi.org/10.1371/journal.pone.0308983.t003

Downregulated KEGG pathways (padj < 0.05) in response to exogenous GLP-2 administration include ribosome, RNA transport, DNA replication, ribosome biogenesis in eukaryotes, spliceosome, cell cycle, base excision repair, mismatch repair, nucleotide excision repair, primary immunodeficiency, RNA polymerase, Fanconi anemia pathway, B cell receptor signaling pathway, and leishmaniasis (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Enriched KEGG pathways of the jejunal mucosa that were downregulated with exogenous GLP-2 administration¹.

https://doi.org/10.1371/journal.pone.0308983.t004

mRNA expression of jejunal carbohydrases, hexose transporters, and transcription factors

Exogenous GLP-2 administration did not influence (padj = 0.17) SI mRNA expression in the jejunum but tended to increase (padj = 0.07) jejunal MGAM mRNA expression (Table 5). Jejunal LCT and TREH mRNA expression were increased (padj < 0.05) with exogenous GLP-2 administration. Exogenous GLP-2 administration increased (padj = 0.01) the jejunal mRNA expression of SLC5A1 and tended to increase (padj = 0.09) the jejunal mRNA expression of SLC2A2. Jejunal SLC2A5 and CDX2 mRNA expression were not influenced (padj > 0.15) by exogenous GLP-2 administration. Exogenous GLP-2 increased (padj < 0.01) or tended to increase (padj = 0.08) the mRNA expression of genes encoding the transcriptional coactivators, HNF1A and GATA4. Jejunal KAT2B mRNA expression was upregulated (padj = 0.01) by exogenous GLP-2 administration.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Effects of exogenous GLP-2 administration on mRNA expression of genes encoding brush border carbohydrases, intestinal glucose transporters, and transcriptional regulators of carbohydrate-responsive genes¹.

https://doi.org/10.1371/journal.pone.0308983.t005

Influence of postruminal casein infusion on the jejunal transcriptome

To our knowledge, this is the first study that has evaluated the effects of postruminal casein infusion on the jejunal mucosal transcriptome in ruminants. Several plausible explanations exist as to why postruminal casein infusion for 7 d did not affect the jejunal transcriptome in the current study. The basal diet was formulated to exceed net energy and metabolizable protein requirements, whereas effects of postruminal casein infusion might have been detected if dietary energy or protein was limiting. For example, small intestinal starch disappearance increased by 37.6 percentage units (from 54.9% to 92.5%) when postruminal casein was added to a protein-free diet [55]. By comparison, postruminal casein infusion only increased small intestinal starch disappearance by a maximum of 10.2 percentage units when dietary metabolizable protein requirements were met [12]. Supporting this concept, a recent review summarized the effects of postruminal casein infusion from 51 studies and concluded that casein increased DM intake when metabolizable protein supply was deficient and that casein had no effect or decreased DM intake when metabolizable protein supply was in excess [56]. Therefore, it is possible that the lack of transcriptomic response to postruminal casein infusion was due to the positive energy and metabolizable protein balance of steers in the current study.

The method of nutrient administration, time to collect jejunal samples after infusion termination, and approach used to measure gene expression could have influenced the outcome of the current experiment. In the current study, casein was delivered continuously via postruminal infusion for 7 d. It is possible that the continuous postruminal infusion of casein masked effects that would be observed with a single pulse dose or multiple pulse doses per day. Previous research demonstrated that the timing of sampling in relation to luminal nutrient flow is important for detecting DEGs in the intestinal mucosa of cattle [57]. Li [57] found that the number of DEGs in duodenal mucosa decreased from 1490 DEGs to 482 DEGs 24 h after terminating the postruminal infusion of partially hydrolyzed starch. Small intestinal transit time is approximately 3 h in cattle [12] and all steers in the current study were slaughtered within a maximum of 3 to 4 h after termination of the 7-d postruminal infusion. Therefore, it is possible that the final portion of luminal casein had passed the jejunum by the time samples were collected for analysis. Because the intestinal mucosa is a heterogeneous tissue with numerous cell types [24], approaches using single-cell RNA sequencing technology could potentially detect DEGs of lowly expressed cell populations, such as enteroendocrine cells, that were likely underrepresented in the current study [58].

A common finding between the current study and Li [57] is that 7-d postruminal infusion of partially hydrolyzed starch or raw corn starch with casein does not influence many functions of the intestinal epithelium. Although previous research detected 1490 DEGs, the enrichment analyses showed that DEGs found after 7-d of postruminal partially hydrolyzed starch infusion only affected a few, generic gene ontologies of the duodenal mucosa (such as biological process, cellular process, molecular function, binding) [57]. Likewise, postruminal infusion of casein did not result in any enriched KEGG pathways or gene ontologies in the jejunal mucosa in the current study. These findings collectively suggest that continuous postruminal infusion of starch or starch with casein for 7 d does not influence the transcriptome of the intestinal mucosa in cattle.

Increased small intestinal starch disappearance, pancreatic mass, and α-amylase activity have been consistent responses to postruminal casein infusion in ruminants [5–15, 17, 23, 55, 59]. Notably, the lack of jejunal transcriptomic response in the current study indicates that increasing pancreatic α-amylase activity via 7-d postruminal casein infusion does not increase the mRNA expression of jejunal brush border carbohydrases. These findings support the previously proposed concept that carbohydrases of the bovine pancreas and small intestinal mucosa do not coordinate a response to luminal nutrient flows [23]. Factors mediating the positive response of increased small intestinal starch digestion to postruminal casein may include increasing pancreatic α-amylase activity [23], increasing the digestion of other nutrients [7, 13, 16], stimulating microbial activity [60–62], changing passage rate [62], or fostering chemical interactions with starch in the small intestinal lumen [63, 64].

Effects of exogenous GLP-2 on transcriptional mechanisms associated with jejunal mucosal growth

A schematic of the functional and transcriptomic effects of exogenous GLP-2 on the jejunum in cattle is summarized (Fig 4). Exogenous GLP-2 administration for 7 d increased jejunal mucosal mass, mucosal DNA content, and mucosal protein content of steers used in the current study by 45.6%, 32.6%, and 42.8%, respectively [23]. Exogenous GLP-2 administration also increases jejunal villus height, crypt depth, and crypt cell proliferation in cattle [32]. After GLP-2 binds to its receptor, it is generally thought that independent actions of insulin-like growth factor 1 (IGF-1) and epidermal growth factor (EGF) are critical downstream mediators of increased cellular proliferation [65–68]. These mediators are thought to lead to activation of Wnt/β-catenin and PI3K/Akt signaling pathways to increase cellular proliferation and differentiation [65, 69]. Currently, there are no annotations of GLP-2 receptor signaling pathways in the KEGG pathway or gene ontology databases. Nevertheless, several transcripts associated with EGF receptor signaling (ERBB2, ERBB3), insulin signaling (INSR, IRS2), Wnt/β-catenin signaling (WNT1, WNT8B, WNT11, FZD3, FZD5, DVL1, GSK3B, CTNNB1, TCF7L2) and PI3K/Akt signaling (PDPK1) pathways were upregulated with exogenous GLP-2 administration. These data suggest that part of the increase in jejunal mucosal growth of steers exposed to exogenous GLP-2 in the current study might have occurred through the activation of GLP-2 receptor signaling pathways.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Schematic of the selected functional and transcriptomic effects of exogenous GLP-2 administration on jejunal mucosal growth and small intestinal starch assimilation capacity.

Functional effects were summarized from studies in cattle [23, 32]. Transcriptomic effects were summarized from the current study. Figure was created with BioRender.com.

https://doi.org/10.1371/journal.pone.0308983.g004

Environmental and cellular cues between conserved signaling pathways (Wnt/β-catenin, TGF-β, Notch, Hedgehog) can lead to cross-talk influencing the activation of the Hippo signaling pathway [70]. The Hippo signaling pathway tended to be upregulated in response to exogenous GLP-2 administration in the current study. In addition to the upregulated transcripts associated with cross-talk from Wnt/β-catenin signaling, key transcriptional regulators of the Hippo signaling pathway (MOB1B, YAP1, SMAD3, SMAD9) controlling cellular proliferation and apoptosis were upregulated in response to exogenous GLP-2 administration. Protein kinases of the Hippo signaling pathway can be regulated by several upstream inputs from intracellular, extracellular, and noncellular components including G protein-coupled receptors, metabolism, mechanical stimuli, cell polarity and density, and stress [70, 71]. Notably, upregulation of transcripts encoding for cadherin 1 (CDH1), an adherens junction protein, and its mammalian ajuba family member (LIMD1) suggest that protein kinase activity (MST1/2, LATS1/2) regulating the Hippo signaling pathway was inhibited, allowing YAP1 translocation to the nucleus to function as a transcriptional co-activator of pro-proliferation and anti-apoptosis genes [72–74].

Net tissue accretion can be due to a combination of increasing cellular proliferation and survival and decreasing cellular apoptosis and proteolysis [19, 21, 32, 35, 75–77] and activation of these pathways by GLP-2 could vary based on physiological state, the presence of luminal nutrient supply, and with mucosal inflammation/injury [19–21]. Because increased mucosal mass can occur via multiple mechanisms of increasing proliferation and decreasing apoptosis, there can be multiple mediators of GLP-2-induced mucosal growth [65]. In the current study, many KEGG pathways associated with genetic information processing were downregulated with exogenous GLP-2 administration including ribosome, RNA transport, DNA replication, ribosome biogenesis in eukaryotes, spliceosome, cell cycle, base excision repair, mismatch repair, nucleotide excision repair, and RNA polymerase. In addition, several transcripts associated with ubiquitin-mediated proteolysis (30/166) were downregulated with exogenous GLP-2 administration, which suggests that part of the increase in jejunal mucosal growth with exogenous GLP-2 administration occurred through decreased proteolysis in the current study. Steers received subcutaneous GLP-2 injections twice daily for 7 d in the current study, and tissues were collected on day 8. Therefore, jejunal mucosa was collected approximately 14–15 h after the last GLP-2 injection was given on day 7. Glucagon-like peptide 2 has a short half-life (7 min) in plasma in humans [78] but, subcutaneous GLP-2 administration can sustain elevated pharmacological plasma GLP-2 concentrations for more than 7 h in cattle [32]. It is possible that changes in the mRNA expression of genes in genetic information processing pathways are transient and that desensitization of the GLP-2 receptor could have occurred with chronic GLP-2 treatment for 7 d [32]. It is possible that the effects of exogenous GLP-2 for 7 d on genetic information processing pathways occurred because of decreased cellular apoptosis and proteolysis, chronic exposure causing GLP-2 receptor desensitization, timing of sampling, or because the pharmacological dose given could potentially differ from responses that would occur under normal physiological conditions.

Effects of exogenous GLP-2 on transcriptional mechanisms associated with jejunal starch assimilation

Epithelial cells are the predominant cell types (> 80%) of the jejunal mucosa [79] and the brush border membrane is estimated to contain approximately 10% to 11% of the total protein content of epithelial cells [80]. Sucrase-isomaltase and maltase-glucoamylase comprise 8.2% and 2.7% of the total protein content of the brush border membrane in humans [81]. Therefore, increasing the growth of the jejunal mucosa via the administration of intestinotrophic hormones, like GLP-2, could potentially result in increased expression of genes encoding brush border membrane proteins [82] (hydrolases, channels and transporters, cytoskeletal components, myosin motor proteins, and adhesion proteins) that are important for adaptation of the intestinal mucosa to high-grain diets [24]. In addition to the brush border of absorptive epithelial cells, the jejunal mucosa sampled in the current study for high-throughput RNA sequencing contains multiple cell types (goblet cells, immune cells, and enteroendocrine cells) that could have specialized functions in response to changes in luminal nutrient flows and exogenous GLP-2 administration [24].

Several prior studies spanning different animal models have demonstrated that exogenous GLP-2 can increase the mRNA expression and enzyme activity of brush border carbohydrases [22, 23, 76, 83, 84]. Brubaker [76] was the first to demonstrate that exogenous GLP-2 increased duodenal maltase activity in mice, but did not influence sucrase or lactase activities. In parenterally-fed rats, intravenous administration of GLP-2 increased jejunal and ileal sucrase-isomaltase mRNA expression [83]. Exogenous GLP-2 administration has been shown to increase sucrase-isomaltase and maltase-glucoamylase mRNA expression and activity in pre-term parenterally-fed neonatal piglets [22]. However, effects of exogenous GLP-2 on sucrase-isomaltase mRNA expression and activity were not observed when administered to full-term parenterally-fed neonatal piglets [84] and enteral nutrition decreased sucrase, maltase, and lactase activities compared with parenteral nutrition [22]. These studies indicate that effects of exogenous GLP-2 on brush border carbohydrase expression and activity are influenced by species, age, and luminal nutrient flow.

We demonstrated that exogenous GLP-2 administration increased jejunal maltase, isomaltase, and glucoamylase activities in the current study [23]. In accordance with increased jejunal α-glucosidase activity, genes associated with brush-border carbohydrate digestion including maltase-glucoamylase, lactase, and trehalase (MGAM, LCT, TREH) were upregulated or tended to be upregulated in the jejunal mucosa of steers treated with exogenous GLP-2 in the current study. It had been suggested that GLP-2-induced suppression of proteolysis and apoptosis may contribute to increased MGAM expression and activity but, more specific mechanisms may be involved [22]. Caudal type homeobox 2 (CDX2) has long been thought to be an important transcription factor contributing to the upregulation of MGAM [22, 83]. However, other authors did not find a strong association between CDX2 and MGAM mRNA expression in their studies [22, 83] and we did not find evidence of this association in our current study, as the mRNA expression of CDX2 was not influenced by exogenous GLP-2 administration. Recent experiments have demonstrated that carbohydrate-responsive genes (SI, MGAM, SLC5A1) are transcriptionally regulated by histone H3 modifications at lysine residues (acetylation/methylation) and increased coactivator binding (KAT2B, CDX2, HNF1, GATA4) in the promoter/enhancer and transcribed regions of those genes in rodents fed high-starch diets [85–89]. In accordance with these findings in rodents, we found that genes associated with transcriptional regulation of MGAM, SI, and SLC5A1 (KAT2B, HNF1, GATA4) were upregulated by exogenous GLP-2 administration. Our data suggests that exogenous GLP-2 administration increased MGAM transcription by increasing histone H3 acetylation and increasing interactions of transcriptional coactivators in the promoter and transcribed mRNA regions [90].

Of note, exogenous GLP-2 administration did not increase jejunal SI mRNA expression in the current study. In humans and mice, approximately 80% of the apparent maltase activity is derived from sucrase-isomaltase and the remaining 20% is derived from maltase-glucoamylase [91, 92]. In humans, maltase-glucoamylase is the primary protein exhibiting α-glucosidase activity at low substrate concentrations while sucrase-isomaltase is the primary protein exhibiting α-glucosidase activity at high substrate concentrations [93]. Others have suggested that the opposite response may occur in ruminants, because of the lower K_m value for isomaltase compared to maltase in jejunal homogenates from cattle [94]. The upregulation of jejunal MGAM mRNA expression and activity in the current study may support this concept. Although ruminants do not contain sucrase activity [37], it should be acknowledged that small intestinal SI mRNA expression can be transcriptionally regulated by luminal nutrient flows in cattle, as increasing dietary fructose consumption in milk replacer decreased small intestinal SI mRNA expression in calves [95]. The absence of intestinal sucrase activity [96], missing stalk region of sucrase-isomaltase [37], lack of transcriptional response of SI to exogenous GLP-2, and lack of effect of GLP-2 on postruminal starch disappearance [23] leaves many questions about the regulation of α-glucosidase activity in ruminants to be answered. Limitations in the extent of small intestinal starch digestion in ruminants might be similar to humans with phenotype V of congenital sucrase-isomaltase deficiency [37] and may not be completely overcome by exogenous GLP-2 administration.

Exogenous GLP-2 administration has been shown to increase the mRNA expression, protein abundance, and activity of intestinal glucose transporters in rodents [97–100] and net portal glucose absorption in parenterally-fed piglets [101, 102]. A prior study had demonstrated that intravenous GLP-2 infusion increased hepatic portal plasma flow but, did not influence net portal glucose absorption in cattle fed a 50:50 alfalfa hay and calf starter diet and fasted for 12 h prior to measurements [103]. To date, the effects of exogenous GLP-2 administration on portal glucose absorption in fed animals has not been evaluated. Although portal glucose absorption was not measured in the current study, we found that exogenous GLP-2 administration increased or tended to increase the expression of genes encoding SGLT1 (SLC5A1) and GLUT2 (SLC2A2) in cattle postruminally infused with starch, suggesting that the capacity for apical and basolateral glucose transport might have increased with exogenous GLP-2 administration. Like MGAM, SLC5A1 mRNA expression is transcriptionally regulated by histone H3 acetylation and co-activator binding (HNF1, GATA4, and CDX2) in both the promoter and transcribed regions [87–89, 104–107] and we found evidence of these associations in the current study. It has also been demonstrated that activation of the sweet taste receptor complex (T1R2-T1R3) of enteroendocrine L-cells with artificial sweeteners results in increased villus height and crypt depth, GLP-2 secretion, maltase activity, SGLT1 protein abundance, and glucose uptake activity in small intestinal tissue from ruminants [108]. Further investigations using GLP-2 receptor knockout mice have revealed that GLP-2-induced increases in SGLT1 expression and glucose uptake activity are dependent on GLP-2 receptor binding [109]. After GLP-2 binds to its receptor, vasoactive intestinal peptide or pituitary adenylate cyclase-activating polypeptide bind to VIPR1, increasing intracellular cAMP, resulting in increased SGLT1 expression [109]. Notably, we found that exogenous GLP-2 administration increased jejunal VIPR1 mRNA expression in the current study, which may indicate a role of vasoactive intestinal peptide in the downstream regulation of SLC5A1 mRNA expression by GLP-2 in cattle.

Interestingly, exogenous GLP-2 administration has been shown to promote the translocation of GLUT2 from the basolateral membrane to the apical membrane in the rat jejunum to provide a low affinity, high capacity hexose transport mechanism when luminal hexose concentrations exceed transport capacity by the high affinity, low capacity glucose transporter, SGLT1 [100]. The apical GLUT2 translocation hypothesis [110] is highly controversial [111, 112] largely due to the methodology used to assess its validity under normal physiological conditions and because of contamination that occurs during the isolation of brush border membranes. Some evidence for GLUT2 localization on the apical membrane in lactating dairy cows and neonatal calves exists [113–115]. In the current study, it is difficult to speculate whether a tendency to increase SLC2A2 mRNA expression resulted in greater GLUT2 protein abundance or activity or the amount of GLUT2 protein localized to each membrane. Differentially expressed genes associated with positive regulation of protein exit from the endoplasmic reticulum and localization to the plasma membrane were significantly enriched with exogenous GLP-2 administration. Sphingolipids contribute to lipid raft formation and intracellular translocation which aid in the trafficking of proteins to the membrane and numerous gene ontology terms related to sphingolipid metabolism and signaling were upregulated with exogenous GLP-2 administration. In the current study, exogenous GLP-2 upregulated the AMPK signaling KEGG pathway, and its activation has been associated with GLUT2 insertion into the apical membrane [116, 117]. In contrast, the expression of genes encoding proteins associated with suspected mechanisms of apical GLUT2 translocation (CACNA1D, TAS1R2, TAS1R3, PRKCB) [117–120] were not upregulated by exogenous GLP-2 administration in the current study. Conflicting points of evidence from the current study provide justification for further investigation of the apical GLUT2 translocation hypothesis in cattle.

Tissues of the portal-drained viscera, including the small intestine, utilize large amounts of glucose to maintain metabolic homeostasis [39, 121–123]. Increasing luminal starch flow in the small intestine results in increased glucose utilization by the portal-drained viscera [124]. With this knowledge from prior studies and the likely increases in α-glucoside hydrolysis and intestinal glucose uptake with exogenous GLP-2 administration in the current study, it would be expected that exogenous GLP-2 administration would increase glucose utilization by the jejunal mucosa. As expected, exogenous GLP-2 upregulated several genes associated with starch and sucrose metabolism and galactose metabolism KEGG pathways. In addition, exogenous GLP-2 upregulated several gene ontologies related to carbohydrate digestion, absorption, and metabolism including carbohydrate metabolic process, hydrolase activity, glucosidase activity, monosaccharide metabolic process, and hexose metabolic process. Glycolysis/gluconeogenesis (HKDC1, HK2, PGM1, PCK2, PDHA1) and the tricarboxylic acid cycle (ACO1, IDH1, SDHA, PC, DLD, SUCLG1, SUCLG2) KEGG pathways tended to upregulate with exogenous GLP-2 administration. These findings suggest that glucose utilization by the jejunal mucosa increased with exogenous GLP-2 administration. Overall, the current study provides a transcriptional overview of the mechanisms associated with increased jejunal starch assimilation capacity in response to exogenous GLP-2 administration.

Effects of exogenous GLP-2 on transcriptional mechanisms associated with intestinal barrier function

In addition to nutrient digestion and absorption, the gastrointestinal tract acts as a barrier, limiting permeability to potentially toxic or pathogenic organisms and compounds. Epithelial permeability of the intestinal mucosa can occur through transcellular transport associated with solute or water transport through a cell or paracellular transport in between cell-cell junctions [125]. In our study, exogenous GLP-2 administration tended to decrease postruminal fluid flow and increase fecal dry matter content, suggesting increased postruminal fluid absorption [23]. Exogenous GLP-2 administration has been shown to decrease jejunal conductance and ex vivo fluxes of Na⁺, CrEDTA, and horseradish peroxidase in mice [126]. Cameron and Perdue [127] found that exogenous GLP-2 administration decreased bacteria adhering to and penetrating the intestinal mucosa in stress-induced mice.

Others have found that exogenous GLP-2 or GLP-2 receptor agonist administration increased the mRNA expression of genes encoding tight junction proteins [128–131]. In the current study, exogenous GLP-2 administration upregulated adherens junction and tight junction KEGG pathways. In these KEGG pathways, 36% and 25% of the total number of transcripts in each KEGG pathway term were upregulated, respectively. Several gene ontologies including transcripts encoding tight junction proteins were upregulated with exogenous GLP-2 administration including apical junction complex, apical, basolateral, and lateral plasma membrane; bicellular tight, cell-cell, occluding junction; cell, cell-cell junction organization; epithelial cell differentiation and morphogenesis; apical junction, cell, cell-cell, and bicellular tight junction assembly; and anion and chloride transport. Also, connections of adherens and tight junction proteins to the actin cytoskeleton through adaptor proteins may have been improved, as indicated by an upregulated cluster of actin-based cell projections, anchoring junction, actin-based cell projections, actin cytoskeleton, and desmosome gene ontologies. Upregulation of gene ontologies related to phospholipid and sphingolipid metabolism and terms related to apical membrane suggest that exogenous GLP-2 increased the physical barrier (plasma membrane) of individual cells of the jejunal mucosa. These data suggest that exogenous GLP-2 administration enhanced the physical barrier function of the jejunal mucosa by decreasing transcellular (plasma membrane) and paracellular (tight and adherens junction) permeability. Increased postruminal fluid absorption coupled with increased mRNA expression of cell-cell junction proteins suggest that the capacity for absorption was increased and that selective permeability of the jejunal epithelium was improved.

In addition to the physical barrier of the small intestine, intestinal permeability can be affected by exogenous factors, epithelial cell turnover, cytokines, and immune cells [125]. Notably, intestinal barrier dysfunction is associated with increased intestinal inflammation and initiation of immunoregulatory processes [132]. In the current study, exogenous GLP-2 administration downregulated primary immunodeficiency and B cell receptor signaling KEGG pathways. Gene ontologies related to immune function that were downregulated with exogenous GLP-2 administration include B cell activation, regulation of B cell activation, immunoglobulin-mediated immune response, B cell-mediated immunity, immunoglobulin production, regulation of lymphocyte activation, adaptive immune response, and others. These data suggest that the adaptive immune function of the jejunal mucosa was improved with exogenous GLP-2 administration. Collectively, next-generation RNA sequencing findings of the current study suggest that increased jejunal mucosal growth with exogenous GLP-2 administration may be associated with improved intestinal barrier function, which was supported by changes in KEGG pathways and gene ontologies associated with decreased intestinal permeability, inflammation, and immune system activation.

Implications for animal growth and feed efficiency

To our knowledge, this is the first report that has investigated the effects of exogenous GLP-2 on the jejunal mucosal transcriptome using an untargeted RNA-sequencing approach in mammals. Although our primary focus was to describe transcriptional mechanisms affected by exogenous GLP-2 and its relationship with phenotypic responses (growth, starch assimilation, barrier function), this dataset could be useful for identifying novel functions of exogenous GLP-2 and lead to future investigations detailing mechanisms of GLP-2-mediated responses in the small intestinal mucosa. Comparing the findings of the current study with other RNA-sequencing datasets shows that DEGs and KEGG pathways affected by exogenous GLP-2 are similar to DEGs and KEGG pathways identified in cattle with divergent average daily gain [133, 134]. Notably, the mRNA expression of APOB, CLCA4, CUBN, CYP2B6, and SLC9A3 were differentially expressed in the jejunal mucosa in all 3 studies. Cattle with greater average daily gain exhibit 9 upregulated KEGG pathways of the jejunal mucosa [133] that were also found to be upregulated in the current study in response to exogenous GLP-2 administration. These pathways include vitamin digestion and absorption, galactose metabolism, mineral absorption, bile secretion, retinol metabolism, starch and sucrose metabolism, protein digestion and absorption, PPAR signaling pathway, and insulin resistance.

It has been suggested that increasing proliferation of the small intestinal epithelium could result in an increased capacity for nutrient absorption that would typically outweigh increased energy expenditure for maintaining additional tissue [24]. Fitzsimons [135] hypothesized that beef cattle with greater gain:feed have a greater capacity for nutrient absorption from the small intestine. Supporting this hypothesis, others have found that jejunal mucosal density is positively correlated with gain:feed [136]. Also, the number of cells in the duodenal and ileal mucosa is positively correlated with improvements in feed efficiency [137]. Those authors suggested that the benefits of increased metabolic activity in the small intestinal mucosa due to increased cellularity would be greater than the increased energetic cost of maintaining the tissue, leading to improved feed efficiency [137]. Our phenotypic [23] and transcriptomic data from the current study suggest that some of the metabolic pathways and molecular functions of the jejunal mucosa affected by exogenous GLP-2 administration may also underlie physiological differences in feed efficiency in beef cattle. These observations warrant future research to investigate if potential changes to small intestinal function with exogenous GLP-2 administration could lead to improved growth, health, and/or feed efficiency of beef cattle.