Effects of dietary protein content and crystalline amino acid supplementation patterns in low protein diets on intestinal bacteria and their metabolites in weaned pigs raised under Different sanitary conditions

Abstract

The objective of this experiment was to investigate the effects of dietary crude protein (CP) content and crystalline amino acids (CAA) supplementation patterns in low CP (LCP) diets on intestinal bacteria and their metabolites in weaned pigs raised under clean (CSC) or unclean sanitary conditions (USC). One hundred forty-four piglets (6.35 ± 0.63 kg) were assigned to one of six treatments in a 3 × 2 factorial arrangement based on CP content and sanitary conditions in a randomized complete block design to give eight replicates with three pigs per pen over a 21-d period. Diets consisted of a high CP (HCP; 21%) and two LCP (18%) diets supplemented with 9 CAA (Lys, Met, Thr, Trp, Val, Ile, Leu, His, and Phe) or only six CAA (Lys, Met, Thr, Trp, Val, and Ile) to meet the requirements. The CSC room was washed weekly, whereas the USC room had sow manure spread in the pens from the beginning of the study and was not washed throughout the experiment. Jejunum and colon digesta were sampled on day 21. Both jejunum and colon digesta were analyzed for ammonia nitrogen, short-chain fatty acids, and biogenic amines but only colon digesta was analyzed for microbiome composition (16s rRNA sequencing on MiSeq). Data were analyzed using R software for 16S rRNA and the MIXED procedure of SAS for microbial metabolites. Sanitation, CP content, and CAA supplementation patterns did not affect the diversity of colonic bacterial composition in weaned pigs. Pigs raised under USC had greater (P < 0.05) jejunal ammonia nitrogen concentration than those raised under CSC. Pigs fed LCP diets had reduced (P < 0.05) jejunal ammonia nitrogen concentration compared to those fed the HCP diet. Interactions between sanitation and dietary CP content were observed (P < 0.05) for: (1) jejunal acetate and (2) colonic spermidine and spermine, whereby (1) acetate concentrations decreased from NCP to LCP in pigs raised under the CSC but those concentrations increased under the USC, and (2) spermidine and spermine concentrations increased in LCP diets compared to HCP diet under USC, unlike CSC which did not show any difference between HCP and LCP. In conclusion, reducing dietary CP lowered ammonia nitrogen content regardless of sanitation and increased microbial metabolites in weaned pigs raised under USC. However, LCP diets with different CAA supplementation patterns did not affect bacterial diversity in weaned pigs, regardless of the hygienic conditions where the animals were housed.

Lowering dietary protein has the potential to improve intestinal health in weaned pigs raised in commercial swine farms.

Introduction

Since in-feed antibiotics have been banned for growth promotion purposes in many jurisdictions, the importance of alternative strategies to promote health has increased considerably. High crude protein (HCP) concentration in the diet of weaned piglets stimulates protein fermentation as well as the proliferation of pathogenic bacteria in the gut. Thus, lowering dietary crude protein (CP) concentration by 3% to 4% units has been used as one of the strategies to improve the gut health and growth performance of weanling pigs (Nyachoti et al., 2006; Opapeju et al., 2009). This is because feeding a low CP (LCP) diet with crystalline amino acids (CAA) supplementation can reduce the amount of undigested protein which becomes available for fermentation. Reducing protein fermentation decreases the production of harmful metabolites such as biogenic amines and ammonia N, leading to decreased incidence of diarrhea (Heo et al., 2008, 2010; Opapeju et al., 2009). The metabolites derived from protein fermentation are detrimental not only to the gut health of host animals but also to the environment (Le et al., 2007, 2009). Because of the urease activity of the fecal microbiome, urea is easily converted to ammonia, which causes a concern such as aerial pollution and acidification of the soil. For example, feeding higher concentrations of nutrients especially high CP concentrations in diets has a greater potential for environmental contamination than low concentrations of nutrients in the diets due to increased aerial pollutants such as ammonia nitrogen derived from undigested nutrients (Hernández et al., 2011). Moreover, excessive ammonia negatively affects the immune system. Qin et al. (2022) and Li et al. (2023) found disrupted immune homeostasis and increased oxidative stress and lung injuries when pigs were exposed to aerial ammonia for 30 d.

The use of CAA is necessary for LCP diets to meet the indispensable amino acids (AA) requirements. However, LCP diets using CAA in pharmaceutical grades, such as histidine or phenylalanine, cannot be adapted in the industry due to their current high cost. Therefore, the effects of different CAA supplementation patterns between LCP diets supplemented with nine indispensable AA except for Arg as CAA or only six indispensable CAA (Lys, Met, Thr, Trp, Val, and Ile) were investigated in the current study. Williams et al. (2018) demonstrated that different CAA supplementation patterns alter the fecal microbial community. Moreover, sanitary conditions have been shown to change the gut microbiome in nursery pigs (Cho et al., 2020; te Pas et al., 2020). Unclean sanitary conditions (USC) may have detrimental effects on growth performance, which may be due to differences in nutrient metabolism in the gut, especially AA metabolism (te Pas et al., 2020). Sanitary conditions are one of the critical factors to be considered for AA requirements determination because increased AA requirements were observed in pigs kept under unclean sanitation due to the shift of use of AA from growth towards immune response (Jayaraman et al., 2017). te Pas et al. (2020) also reported that pigs raised under clean sanitary conditions (CSC) showed higher digestibility and a higher abundance of bacteria in the colon. Gut bacteria are one of the major regulatory factors underlying nutrient digestion and fermentation, mostly placed at the hindgut of animals (Pluske et al., 2018). A clean environment can modify the gut health of animals for both gut microbial composition and metabolism in a positive way, which in turn affects higher nutrient digestibility, and leads to improved growth performance of pigs (van der Meer et al., 2016; te Pas et al., 2020). However, there is a scarcity of information about the effects of sanitary conditions and CAA supplementation patterns on the gut microbial composition and their metabolites in weaned pigs. Thus, the current study was conducted to investigate the effects of dietary CP content and CAA supplementation patterns on bacteria composition and their metabolites in the intestine of weaned pigs raised under CSC or USC. This study hypothesized that: (1) USC may negatively modulate microbial structure in weaned pigs, and (2) feeding LCP diets would positively modulate microbial structure by increasing beneficial microbiota in the gut of weaned pigs under unclean sanitary conditions.

This was a series of studies with the effects of dietary protein content and CAA supplementation patterns on growth performance, intestinal histomorphology, and immune response in weaned pigs raised under different sanitary conditions (Lee et al., 2022).

Materials and Methods

Animals, housing, and experimental design

All methods performed in this study followed the Institutional Animal Care and Use Committee guidelines and regulations at the University of Manitoba. The pigs were handled according to the Canadian Council on Animal Care (2009). The experimental protocol for this study was approved by the University of Manitoba Animal Care Committee (AC11406).

A total of 144 piglets (TN70 × TN Tempo; Topigs Norsvin, Winnipeg, MB, Canada) were weaned at 21 ± 2 d of age and obtained from Glenlea Research Station at the University of Manitoba with an average initial body weight (BW) of 6.35 ± 0.63 kg. Pigs were assigned to a 3 × 2 factorial arrangement based on protein contents and sanitary conditions by using a randomized complete block design with eight replications and three pigs per pen (1.8 m × 1.2 m). The sex of pigs was equally balanced within the replicate. The experiment was conducted over a 21-d period.

The different sanitary conditions were set as described by Jayaraman et al. (2017) and van der Meer et al. (2020). A CSC room was disinfected using Virkon S before the experiment. Feces was pushed down underneath the flooring every day and the pens in the CSC room were washed once a week throughout the experiment. New masks and newly cleaned and washed coveralls and boots were placed in front of the clean room with disinfectant. However, the USC room was maintained without disinfection and kept without washing the pens throughout the experiment. Swine manure from the sow herd from the same facility (Glenlea Research Station at the University of Manitoba) but not giving birth to the piglets used in this study was obtained and kept in a freezer until use. At the beginning of the study, the manure (5 kg per pen) was thawed and mixed with distilled water and then spread on the pen floor and walls in the USC room. The ventilation system was turned off in the USC room to create dust and fine particles in the air of the room. Nylon bags containing dried pig manure were mounted on pens in the USC room to mimic the contamination in the commercial environment. Regular daily animal monitoring was always conducted with the clean room first and then the unclean room to avoid cross-contamination. Each pen had a plastic-covered expanded metal floor, a stainless-steel feeder, and a low-pressure nipple drinker. Room temperature was maintained at 29 ± 1 °C during week 1 and reduced to 1 °C for each following week.

Experimental diets

Three experimental diets were formulated with corn, wheat, and soybean meal, and CAA was supplemented to meet the indispensable AA to lysine ratio recommended by AMINOPig 1.0 (Evonik, 2011) for 5 to 10 kg (Table 1). The AA concentrations in corn, wheat, and soybean meal were analyzed before diet mixing, and those values were used for diet ­formulation. Diets contained two concentrations of CP as one HCP (21%) and two LCP (18%) diets, and those two LCP diets were supplemented with: (i) nine CAA (Lys, Met, Thr, Trp, Val, Ile, Leu, His, and Phe) to meet all indispensable AA requirements (LCP 1) or (ii) only six CAA (Lys, Met, Thr, Trp, Val, and Ile) to meet indispensable AA requirements except for Leu, His, and Phe (LCP 2). All experimental diets were formulated to meet or exceed the NRC (2012) requirements of minerals and vitamins for 7 to 11 kg of BW. Pigs were fed for 21 d and had free access to feed and water throughout the experiment. The calculated and analyzed nutrient composition of experimental diets is shown in Table 2.

Table 1.

Ingredient composition of experimental diets (as-fed basis)

Item	Diets1
	HCP	LCP 1	LCP 2
Ingredients, %
Corn	32.88	40.93	40.60
Wheat	25.00	25.00	25.00
Soybean meal	23.64	17.56	18.00
Whey powder	10.00	10.00	10.00
Spray-dried animal plasma	2.24	–	–
Vegetable oil	2.81	1.77	1.88
Limestone	1.29	1.29	1.29
Monocalcium phosphate	0.85	0.98	0.97
Salt	0.40	0.40	0.40
Vitamin-mineral premix2	0.15	0.15	0.15
L-Lysine HCl	0.39	0.76	0.74
DL-Methionine	0.18	0.30	0.30
L -Threonine	0.11	0.29	0.28
L -Tryptophan	0.06	0.12	0.12
L -Valine	–	0.21	0.20
L -Isoleucine	–	0.08	0.07
L -Leucine	–	0.05	–
L -Histidine	–	0.04	–
L -Phenylalanine	–	0.08	–

¹HCP, high crude protein; LCP 1, low crude protein supplemented with nine indispensable crystalline amino acids except for Arg; LCP 2, low crude protein supplemented with only six indispensable crystalline amino acids (Lys, Met, Thr, Trp, Val, and Ile).

²Provided the following nutrients (per kg of air-dry diet): vitamins: A, 2,000 IU, D₃ 200 IU, E, 40 mg, K, 2 mg, B₁, 1.5 mg, B₂, 7 mg, B₆, 2.5 mg, B₁₂, 25 µg, calcium pantothenate, 14 mg, folic acid, 1 mg, niacin, 21 mg, biotin, 70 µg. Minerals: Cu, 10 mg (as copper sulphate), iodine, 0.4 mg (as potassium iodine), iron, 120 mg (as ferrous sulphate), Mn, 10 mg (as manganous oxide), Se, 0.3 mg (as sodium selenite), Zn, 110 mg (as zinc oxide).

Table 2.

Calculated and analyzed nutrient composition of experimental diets (as-fed basis)

	Diets1
Item, %	HCP	LCP 1	LCP 2
Calculated nutrient composition
Crude protein	22.3	19.3	19.3
Total amino acids
Arg	1.33	1.04	1.05
His	0.56	0.48	0.44
Ile	0.92	0.82	0.83
Leu	1.78	1.51	1.48
Lys	1.49	1.47	1.47
Met	0.49	0.57	0.57
Met + Cys	0.90	0.89	0.89
Phe	1.05	0.91	0.85
Thr	0.98	0.96	0.96
Trp	0.32	0.32	0.32
Val	1.05	1.03	1.03
SID2 amino acids
Arg	1.23	0.96	0.97
His	0.50	0.43	0.39
Ile	0.82	0.74	0.74
Leu	1.59	1.35	1.32
Lys	1.35	1.35	1.35
Met	0.46	0.54	0.54
Met + Cys	0.81	0.81	0.81
Phe	0.93	0.81	0.74
Thr	0.85	0.85	0.85
Trp	0.30	0.30	0.30
Val	0.92	0.92	0.92
Analyzed nutrient composition
Dry matter	89.35	88.42	88.30
Crude protein	21.42	18.48	18.12
Total amino acids
Arg	1.25	1.01	0.98
His	0.52	0.47	0.41
Ile	0.86	0.80	0.78
Leu	1.67	1.45	1.37
Lys	1.40	1.41	1.43
Met	0.47	0.52	0.55
Met + Cys	0.85	0.82	0.85
Phe	0.99	0.89	0.79
Thr	0.95	0.87	0.94
Trp	0.32	0.32	0.32
Val	0.99	0.99	0.98

¹HCP, high crude protein; LCP 1, low crude protein supplemented with nine indispensable crystalline amino acids except for Arg; LCP 2, low crude protein supplemented with only six indispensable crystalline amino acids (Lys, Met, Thr, Trp, Val, and Ile).

²SID = standardized ileal digestible.

Chemical analyses

Diets were analyzed for dry matter, CP, and AA compositions. The dry matter content was measured according to AOAC (2006) (method 934.01; 2006), and nitrogen content was determined by the combustion method (method 990.03; AOAC, 2006) using the LECO N analyzer (model CNS-2000; LECO Corp., St. Joseph, MI, United States) to calculate CP (nitrogen × 6.25). Amino acid contents in the diets were determined by ion-exchange chromatography with postcolumn derivatization with ninhydrin. Amino acids were hydrolyzed with 6 N HCL for 24 h at 110 °C, and samples for sulfur-containing AA analysis had an additional oxidization step before hydrolyzation using performic acid (Llames and Fontaine, 1994; Commission Directive, 1998). Amino acids were quantified with the standard internal method by measuring the absorption of reaction products with ninhydrin at 570 nm. Trp was determined using HPLC with fluorescence detection (extinction 280 nm, emission 356 nm) after alkaline hydrolysis with barium hydroxide octahydrate for 20 h at 110 °C (Commission Directive, 2000). Tyr was not determined.

Sample preparation

One pig with the closest BW to the mean BW of each pen was selected at the beginning of the experiment and euthanized on day 21 to collect jejunum and colon digesta samples for ammonia nitrogen, volatile fatty acids (VFA), and biogenic amines analyses. Also, colonic digesta was sampled for microbial analysis. Jejunum and colon digesta were taken 2 m away from the ileocecal junction and 20 cm away from the cecum, respectively, using premeasured strings. Samples were immediately snap-frozen in liquid nitrogen and transferred to a ‒80 °C freezer. The ammonia nitrogen concentrations in the jejunum and colon digesta were determined using the method described by Novozamsky et al. (1974). A 1.5 mL of a reagent was obtained from a mixture of 200 mL of 0.05% sodium nitroprusside and 10 mL of 4% ethylenediaminetetraacetic acid and added to 50 µL of digesta fluid. A 2.5 mL solution containing 10% sodium hypochlorite was then added to the previous mixture. Test tubes containing the final mixture were incubated in the dark for 30 min, and then the optical density of the mixture was immediately read at 630 nm using a spectrophotometer (SoftMax Pro Software; Molecular Devices, San Jose, CA, United States). The VFA and branched-chain fatty acids (BCFA) concentrations were determined using the method described by Erwin et al. (1961) using gas ­chromatography (Varian Chromatography System, model Star 3400; Varian Medical Systems, Palo Alto, CA). Briefly, 1 mL of 25% metaphosphoric acid was mixed with 5 mL of digesta fluid. The mixture was neutralized with 0.4 mL of 25% NaOH and then 0.65 mL of 0.3 M oxalic acid was added. The samples were then centrifuged for 20 min at 3,000 × g at 4 °C, and 2 mL of the supernatant was transferred to a gas chromatography vial. The biogenic amines were analyzed by liquid chromatography according to the method described by Smělá et al. (2003).

DNA extraction and sequencing

To assess the diversity of the microbial community, DNA was extracted from colon digesta using QIAamp PowerFecal Pro DNA Kit (Cat. No./ID: 51804; Qiagen, Germantown, MD, United States) according to the manufacturer’s instructions. The extracted DNA samples were sent to LGC Genomics GmbH (Berlin, Germany) for microbial sequencing using the 16S rRNA technique. The V4 region of the bacterial 16S rRNA gene was amplified from the total extracted DNA. The DNA library was demultiplexed for each sequencing lane using Illumina’s bcl2fastq version 2.20 software (Illumina; San Diego, CA, United States), and the reads were sorted by amplicon inline barcodes.

Microbial community analysis

The obtained raw sequences were preprocessed with the Mothur version 1.35.1 software program for the operational taxonomic unit (OTU) cluster analysis (Schloss et al., 2009) as follows: (1) the sequences containing ambiguous bases were removed with homopolymer stretches of more than eight bases or with an average Phred quality score below 33; (2) sequence alignments were performed against the 16S Mothur-Silva SEED r119 reference alignment; (3) short alignments and chimera were filtered, and sequencing error was reduced by preclustering; (4) taxonomical classification of the sequences was performed against the Silva reference classification and sequences from other domains of life were removed; (5) OTU was picked by clustering at the 97% identity level; and (6) OTU consensus taxonomical calling, integrating the taxonomical classification of the cluster member sequences was performed.

The OTU diversity was analyzed by using Quantitative Insights Into Microbial Ecology (QIIME) version 1.9.0 (Caporaso et al., 2010). Microbial diversity was assessed within samples (alpha-diversity) or between samples (beta-diversity). Alpha diversity (observed OTUs) was calculated through rarefaction with ten iterations. Beta diversity was calculated on the sequence reads based on weighted and unweighted UniFrac distance matrices, and principal coordinate analysis (PCoA) plots were used for visualization of the results for beta-diversity. All P-values are corrected with a false discovery rate according to the Benjamini–Hochberg method.

Statistical analyses

Data for microbial metabolites (volatile fatty acids, biogenic amines, and ammonia nitrogen) were analyzed using the MIXED procedure of SAS (version 9.4; SAS Institute Inc., Cary, NC). The model included the main effects of different CP contents and sanitary conditions and their interaction, and replicate was included as random effects. There was no sex effect, therefore, sex was removed from the model. Means were separated using specific orthogonal contrasts to compare HCP with the combination of LCP 1 and LCP 2 (HCP vs. LCP 1 and LCP 2) for the main effect of CP contents and to compare LCP 1 with LCP 2 (LCP 1 vs. LCP 2) for the main effect of CAA supplementation patterns. Microbiome data were analyzed using R version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria). Analyses were based on packages “vegan” and “phyloseq”. If not otherwise indicated data were analyzed with the Kruskal–Wallis test and Dunn’s test for post hoc analysis. An individual pig was used as the experimental unit, and significant differences were considered at P < 0.05.

Results and Discussion

A contrast in sanitary conditions was generated by differences in hygiene with housing and management of the rooms. All pigs were clinically healthy throughout the trial.

Diversity of microbial community in the colon

No difference was found (P > 0.05) in both alpha- and beta-diversity of the microbial community in colon digesta between the different sanitary conditions (Figure 1), CP levels (Figure 2), or both dietary and environmental treatments (Figure 3). A sanitary status model was applied in this study to stimulate a low-grade immune response (compared to other immune challenge models such as enterotoxigenic Escherichia coli or Salmonella) and measure aspects of microbial diversity and microbial metabolites observed in pigs under different sanitary conditions. It was expected that sanitary conditions might affect the microbiota diversity in colon digesta based on the results of previous studies (Kubasova et al., 2018; Cho et al., 2020; te Pas et al., 2020). High diversity of the microbial community is often found in pigs raised in CSC (te Pas et al., 2020). However, there was no difference in microbial diversity between the different sanitary conditions in the present study. The sow manure seemed to have a low effect on microbiota modulation in the USC group in this study. One important thing to be considered is the time of sampling (duration of the experiment) for microbial analysis. The exposure to different sanitary conditions might not be enough to change the microbial composition in the gut of pigs in this study. Kubasova et al. (2018) showed the microbiota composition did not differ during the nursery phase (from weaning until 4 wk after weaning), however, the complexity of pig fecal microbiota increased with age (up to 25 wk of age). In contrast, pigs used in Cho et al. (2020) showed a difference in microbial diversity on day 7 of the trial. Another possible explanation for this result together with the growth performance data in the previous publication (Lee et al., 2022) could be that microbial compositions could be changed by transplanting fecal microbiota from sow feces to weaned pigs raised under USC. Several studies showed that fecal microbiota transplantation has beneficial effects on growth performance and immune function in young piglets (Hu et al., 2018; Niederwerder et al., 2018; Wan et al., 2019). Hu et al. (2018) reported that Prevotella and Oscillospira were increased on day 12 in piglets that received daily oral inoculation of a fecal microbiota suspension from healthy adult pigs from days 1 to 11, which are beneficial bacteria associated with the digestion of carbohydrates (Wu et al., 2011), and butyrate production (Konikoff and Gophna, 2016), respectively. It was possible that the sow feces spread in USC beneficially affected pig health, including growth performance in this study. Indeed, pigs raised under USC showed reduced daily gain and feed efficiency in week 2, however, the growth performance of pigs raised in different sanitary conditions was not different in week 3, which is the time when colon digesta was collected for microbial diversity analysis (Lee et al., 2022). Intestinal microbial shifts in pigs are correlated with the growth performance of pigs (Kim and Isaacson, 2015), and the body weights of weaned pigs on day 21 did not differ between different sanitary conditions in this study, thus, no significant difference was observed in microbial diversity.

Figure 1.

The effect of sanitary conditions on colonic bacterial diversity in weaned pigs (n = 8). (a) The relative abundance of bacterial composition at genus level in colon digesta of weaned pigs raised under clean or unclean sanitary conditions (P > 0.05). (b) PCoA plots of the microbial communities in colon digesta of weaned pigs raised under clean or unclean sanitary conditions (P > 0.05). Clean, clean sanitary conditions; Unclean, unclean sanitary conditions.

Figure 2.

The effect of protein content on colonic bacterial diversity in weaned pigs (n = 8). (a) The relative abundance of bacterial composition at genus level in colon digesta of weaned pigs fed the diets containing different protein contents (P > 0.05). (b) PCoA plots of the microbial communities in colon digesta of weaned pigs fed the diets containing different protein contents (P > 0.05). HCP, high crude protein; LCP 1, low crude protein supplemented with nine indispensable amino acids except Arg; LCP 2, low crude protein supplemented with only six indispensable amino acids (Lys, Met, Thr, Trp, Val, and Ile).

Figure 3.

Colonic bacterial diversity among treatment groups was evaluated at the genus level (n = 8). (a) The relative abundance of bacterial composition at genus level in colon digesta of weaned pigs (P > 0.05). (b) PCoA plots of the microbial communities in colon digesta of weaned pigs. Clean_HCP, high crude protein under clean sanitary conditions (P > 0.05); Clean_LCP 1, low crude protein supplemented with nine indispensable amino acids except for Arg under clean sanitary conditions; Clean_LCP 2, low crude protein supplemented with only six indispensable amino acids (Lys, Met, Thr, Trp, Val, and Ile) under clean sanitary conditions; Unclean_HCP, high crude protein under unclean sanitary conditions; Unclean_LCP 1, low crude protein supplemented with nine indispensable amino acids except for Arg under unclean sanitary conditions; Unclean_LCP 2, low crude protein supplemented with only six indispensable amino acids (Lys, Met, Thr, Trp, Val, and Ile) under unclean sanitary conditions.

Differences in the diversity of gut microbiota were also expected between pigs fed diets containing different CP levels because a major reason for the complexity of microbial community in the gut could be the number of available substrates for bacteria fermentation (Blaut and Clavel, 2007). Dietary carbohydrates and protein can be used as primary substrates for bacteria fermentation. These substrates include indigestible dietary components and endogenous components, such as mucus and sloughed epithelial cells (Blaut and Clavel, 2007). Although diet is the primary factor that can modulate gut microbiota, dietary CP content had no significant effects on microbial diversity in the current study. Zhang et al. (2016) demonstrated that feeding an LCP diet caused significant modulation of some intestinal bacterial groups in pigs, including reducing E. coli and increasing Clostridium cluster XIVa and Clostridium cluster IV groups. Also, E. coli is one of the major proteolytic fermenters, and its abundance is decreased by lowering CP levels in the diet, whereas Clostridium cluster XIVa and Clostridium cluster IV groups, butyrate producers, increased as carbohydrase fermentation increased. Peng et al. (2017) also reported decreased E. coli in growing pigs when the CP level was reduced from 20.0% to 15.3%. According to a review paper by Zhang et al. (2020), however, the diversity of the microbial population was not affected by dietary CP levels in most of the studies, although dietary CP content sometimes modulated protein fermentation by bacteria. A similar result was revealed by Yu et al. (2019), who reported that colonic bacterial community was not changed in weaned pigs fed diets containing 20, 17, and 14% CP, which was probably due to a low level of fermentable carbohydrates in the diets. Thus, the results of the current study showing no differences in colonic bacterial community between pigs fed HCP and LCP diets could be explained by insufficient fermentable carbohydrate sources in these diets.

Different bacterial community structures were observed when different CAA supplementation patterns were used in piglet diets (Williams et al., 2018; Spring et al., 2020), however, the diversity of the microbial community in the current study did not differ between LCP 1 (nine indispensable CAA except for Arg) and LCP 2 (only six CAA). Williams et al. (2018) demonstrated that pigs fed diets with fewer CAA inclusion showed an increased abundance of Paraprevotellaceae, Lactobacilliaceae, and Ruminococcaceae than those fed diets with more CAA inclusion. Diets with fewer CAA supplementation had more intact protein sources than those with more CAA supplementation, which might affect the microbial composition due to the different amounts of available substrates for microbial fermentation. However, the amounts of intact protein and free AA were similar between LCP 1 and LCP 2 diets in the current study, which might explain the lack of significant difference in the obtained results. The results of the present study were also supported by Zhao et al. (2020), who observed no differences in alpha-diversity of colonic microbiota in fattening pigs fed diets with different dietary CP levels or CAA providing patterns.

Microbial metabolites in the jejunum and colon

A greater (P < 0.05) jejunal ammonia nitrogen concentration was found in pigs housed in USC than those in CSC (Table 3). Pigs fed LCP diets had reduced (P < 0.05) jejunal ammonia nitrogen compared to those fed the HCP diet, however, no difference was found in colonic ammonia nitrogen in weaned pigs. Decreased ammonia nitrogen concentrations in the pig intestine due to reduced dietary protein levels have been demonstrated in previous studies (Nyachoti et al., 2006; Heo et al., 2008; Opapeju et al., 2009). The relatively high variation (SEM) among different treatments might explain the lack of differences in colonic ammonia nitrogen in the present study.

Table 3.

Effects of crude protein level and sanitary conditions on ammonia N and volatile fatty acids (VFA) concentrations in jejunal and colonic digesta in weaned pigs¹

	Clean conditions			Unclean conditions			SEM	P-values2
Item	HCP	LCP 1	LCP 2	HCP	LCP 1	LCP 2		San	CP	AA	San × CP
Ammonia N, mg/L
Jejunum	50.0	24.8	22.4	64.4	33.5	31.0	5.9	0.034	0.001	0.676	0.583
Colon	63.8	61.6	94.6	75.4	85.0	122.8	27.7	0.348	0.366	0.196	0.762
Jejunal VFA, mmol/L
Acetic acid	4.60	2.71	3.21	3.83	5.43	3.95	0.675	0.113	0.513	0.467	0.041
Propionic acid	0.14	0.16	0.20	0.22	0.16	0.38	0.054	0.032	0.241	0.010	0.894
Butyric acid	0.36	0.31	0.33	0.26	0.27	0.27	0.030	0.006	0.437	0.669	0.298
Isobutyric acid	0.72	0.58	0.49	0.83	0.78	0.93	0.073	0.001	0.190	0.672	0.077
Isovaleric acid	0.02	0.01	0.03	0.12	0.10	0.14	0.023	0.001	0.869	0.226	0.858
Valeric acid	0.07	0.06	0.05	0.01	0.02	0.02	0.009	0.001	0.676	0.625	0.305
Total BCFA3	0.80	0.65	0.57	1.02	0.87	1.09	0.087	0.001	0.110	0.401	0.335
Total VFA	5.08	2.89	3.75	4.29	5.85	4.61	0.723	0.097	0.521	0.792	0.041
Colonic VFA, mmol/L
Acetic acid	60.0	50.2	57.3	56.7	52.1	53.3	2.91	0.450	0.043	0.170	0.649
Propionic acid	28.3	31.6	30.1	29.3	31.8	35.9	2.30	0.217	0.072	0.577	0.590
Butyric acid	21.6	18.6	29.0	18.1	21.4	22.2	4.32	0.483	0.427	0.211	0.844
Isobutyric acid	0.94	0.89	1.56	0.70	0.67	0.72	0.160	0.002	0.291	0.031	0.287
Isovaleric acid	1.14	1.13	1.95	0.79	0.82	0.92	0.209	0.002	0.191	0.038	0.376
Valeric acid	6.10	7.83	9.65	4.80	8.15	8.63	1.662	0.627	0.036	0.496	0.741
Total BCFA	8.25	9.85	13.06	6.29	9.64	10.28	1.756	0.256	0.029	0.285	0.881
Total VFA	109	99	107	104	105	112	5.8	0.719	0.917	0.242	0.331

¹HCP, high crude protein; LCP 1, low crude protein supplemented with nine indispensable crystalline amino acids except for Arg; LCP 2, low crude protein supplemented with only six indispensable crystalline amino acids (Lys, Met, Thr, Trp, Val, and Ile).

²San, main effect of sanitation; CP, main effect of crude protein levels (HCP vs. LCP 1 and LCP 2); AA, main effect of crystalline amino acids supplementation patterns (LCP 1 vs. LCP 2); San × CP, interactive effect of sanitation and crude protein levels (HCP vs. LCP 1 and LCP 2).

³Branched chain fatty acids = isobutyrate + isovalerate + valerate.

Both dietary carbohydrates and proteins and host-derived proteins are utilized for microbial fermentation (Egert et al., 2006). Carbohydrates are the preferred substrates for bacterial fermentation, and protein fermentation occurs when fermentable carbohydrates have been used up (Ouwehand et al., 2005). Carbohydrate fermentation yields health-promoting metabolites such as acetate, propionate, and butyrate, whereas protein fermentation produces toxic components, including ammonia and phenolic and indolic compounds (Scheppach, 1994; Egert et al., 2006). The BCFA, which includes isobutyrate, isovalerate, and valerate, are metabolites produced by the deamination of branched-chain AA such as leucine, valine, and isoleucine (Rasmussen et al., 1988). Although there were no differences in the diversity of microbial composition, VFA and biogenic amine concentrations were influenced by different sanitary conditions, CP content, or CAA supplementation patterns in the current study. Similar results were found in previous work by van Zanten et al. (2014) in which dietary inclusion of symbiotics in humans did not affect microbiota diversity but increased BCFA concentrations in fecal samples. However, the reason for this result has not been clearly identified. The ratio of acetate:propionate:butyrate in the gut is reported to be 3:1:1 (Cummings, 1981; Scott et al., 2013), which is similar to the ratio found in jejunal and colonic digesta in this study. Different sanitary conditions affect jejunal VFA concentrations, whereas CP content and CAA supplementation patterns did not influence VFA concentrations in jejunal digesta in weaned pigs, except that propionic acid concentration was higher (P < 0.05) in pigs fed the LCP 2 diet than those fed LCP 1 diet. Pigs raised under USC had greater (P < 0.05) propionic acid concentrations than CSC, whereas pigs raised under CSC had greater (P < 0.05) butyric acid concentrations in jejunal digesta of weaned pigs. Cho et al. (2020) and te Pas et al. (2020) reported that butyrate concentrations decreased in the colon digesta or feces when pigs were raised in low sanitary conditions, which is consistent with the current result showing a higher butyric acid concentration in the pigs housed under the CSC. However, contrasting results were observed for the propionic acid concentration of pigs housed under the USC, which could be explained by the so-called “hygiene hypothesis” (Strachan, 1989). The “hygiene hypothesis” is a concept that lack of exposure to infectious agents may be the culprit for the increased immune-mediated disease prevalence, and microbial exposure helps animals to develop healthy immune systems and intestinal microbiota (Stiemsma et al., 2015). Montagne et al. (2012) postulated poor sanitary conditions led to ecosystem modification which could be considered more beneficial such as more Lactobacillus or increased VFA in the intestine of pigs. The BCFA concentrations were greater (P < 0.05) in piglets housed in USC compared to those in CSC, which was in accordance with a previous study by Cho et al. (2020). Intestinal concentrations of BCFA are used as indicators for the extent of protein fermentation (Macfarlane et al., 1992). The BCFA is considered one of the harmful metabolites derived from proteolytic fermentation together with ammonia, indoles, and phenols, thus, has been considered a predisposing factor for postweaning diarrhea (Gao et al., 2019). Therefore, greater BCFA concentrations indicate an adverse effect on the gut health of pigs. Interactions between sanitation and dietary CP content were detected (P < 0.05) for jejunal acetate as well as total VFA, whereby the acetate and total VFA concentrations decreased from HCP to LCP in pigs raised under the CSC but those concentrations increased under the USC. This indicates that LCP diets were effective in increasing acetic acid concentration under USC. The SCFA including acetate, propionate, and butyrate, play an important role in colonocytes as an energy source (Roediger, 1982), inflammatory reactions and immune parameters (Sweeney et al., 2012; Xu et al., 2016), and gut cell proliferation of the host (Daly and Shirazi-Beechey, 2006). For instance, Sweeney et al., (2012) observed a positive relationship between acetic acid concentration and pro-inflammatory cytokine in the colon of pigs fed β-glucan diets. Liu et al. (2017) reported that orally administrated acetic acid suppresses gastric apoptosis and promotes mucin production in mice. Acetic acid takes the highest concentrations among the SCFA, thus, the increased acetic acid concentrations of pigs especially those raised under the USC are more important to ameliorate the immune response derived from the USC.

Although feeding LCP diets increased acetic acid concentrations in the jejunum of pigs housed in USC, higher (P < 0.05) acetic acid concentrations in colon digesta were observed in pigs fed the HCP diet regardless of the sanitary conditions, which is in agreement with the previous studies by Hobbs et al. (1996) and Nyachoti et al. (2006). However, this finding contrasts with the previous report by Qiu et al. (2018) that LCP diets had higher acetate and butyrate concentrations in the ileal digesta of growing pigs because of increased amounts of digestible carbohydrates in LCP diets. Compared to the differences in the amount of corn (increased 12% from HCP to LCP) and soybean meal (decreased 12% from HCP to LCP) in the previous study by Qiu et al. (2018), the differences in the amount of corn (increased 8% from HCP to LCP diets) and soybean meal (decreased 6% from HCP to LCP diets) were less in the current study. Thus, this resulted in less difference in digestible carbohydrates between HCP and LCP diets and no increase in VFA concentrations in pigs fed the LCP diets. One possible reason for the increased acetic acid concentration in pigs fed the HCP diet might be the inclusion of spray-dried animal plasma in the HCP diet. Che et al. (2020) reported increased acetate, propionate, and butyrate concentrations, along with improved bacterial diversity in the colon of pigs fed a diet containing spray-dried animal plasma. Contrary to the results in jejunal digesta, the isobutyric and isovaleric acid concentrations were higher (P < 0.05) in CSC than at USC in colonic digesta. The reason for this discrepancy is not clear. Moreover, increased (P < 0.05) colonic isobutyric and isovaleric acid concentrations were measured in the LCP 2 diet compared to the LCP 1 diet. Suppressed utilization of branched-chain AA for microbial fermentation is expected in the LCP diets, thereby reducing BCFA concentrations (Yu et al., 2019). The BCFA concentrations were reduced in the colonic digesta or feces in pigs fed LCP diets compared to HCP diets in the previous studies (Hobbs et al., 1996; Yu et al., 2019), however, increased (P < 0.05) valeric acid, as well as total BCFA concentrations, were observed in the colon of pigs fed the LCP diets than the HCP diet in the current study. Isobutyric acid, isovaleric acid, and valeric acid are the end products of the deamination of Leu, Val, and Ile, respectively (Poston, 1976; Langer et al., 2000). The discrepancy between the results of this and previous studies might be due to suboptimal or deficiency of some AA in the LCP diets, which is supported by the plasma AA concentrations reported in a companion study (Lee et al., 2022). Higher plasma Lys, Met, Thr, Ile, and Val concentrations were found in pigs fed the LCP diets compared to those of the HCP diet, which could have resulted from the imbalance of ideal protein ratio in LCP diets, leading to increased deamination of AA not used for growth and muscle synthesis.

No significant differences were seen in the biogenic amine concentrations in jejunal digesta in weaned pigs, except for lower (P < 0.05) histamine in the LCP 2 diet compared to the LCP 1 diet, which is due to the lower inclusion rate of total histidine in LCP 2 diet (0.41%) than LCP 1 diet (0.47%; Table 4). The same result was found in colonic digesta, where lower (P < 0.05) histamine concentration was measured in LCP 2 diet compared to LCP 1 diet. However, higher (P < 0.05) His concentrations were observed in colonic digesta in pigs fed LCP diets compared to the HCP diet. Histamine is produced from histidine by decarboxylation (Jarisch et al., 2015). One possible reason for this result could be that histidine in LCP diets might not have been fully used for protein synthesis due to the imbalance of ideal protein in LCP diets. Interactions were present (P < 0.05) for sanitation and dietary CP content in spermidine and spermine concentrations in colonic digesta, whereby there was increased spermidine and spermine concentrations in pigs fed LCP diets than the HCP diet under the USC, whereas those concentrations did not differ between LCP and HCP diets under the CSC. Spermidine and spermine are polyamines derived from the decarboxylation of ornithine and N-rich AA such as glutamine, asparagine, and arginine (Ramani et al., 2014). The first step of polyamine biosynthesis is the decarboxylation of ornithine into putrescine, then putrescine is turned into spermidine by the addition of an aminopropyl group via spermidine synthase, and then spermine is derived from spermidine by spermine synthase (Ramani et al., 2014). Therefore, similar results were observed between spermidine and spermine concentrations. Polyamines, including spermidine and spermine, play an essential role in rapidly dividing immune cells such as the proliferation and differentiation of lymphocytes (Li et al., 2007) or promotion of T-cell development (Carriche et al., 2021). Increased spermidine and spermine concentrations in the colon of pigs housed in the USC room in this research might be due to activation of the immune system. However, studies by Cao et al. (2017) and Liu et al. (2020) showed that spermine supplementation alleviates inflammatory response, enhances the immune and ileal barrier function, and maintains large intestinal microbial homeostasis in piglets. Thus, an increment of those polyamines in pigs fed LCP diets under USC indicates that feeding LCP diets to weaned pigs may ameliorate inflammation induced by USC, which is supported by the results showing greater plasma anti-inflammatory cytokine (interleukin-10) concentrations in pigs fed the LCP diets than those fed the HCP diet under USC (Lee et al., 2022).

Table 4.

Effects of crude protein level and sanitary conditions on biogenic amine concentrations in jejunal and colonic digesta in weaned pigs¹

	Clean conditions			Unclean conditions			SEM	P-values2
Item, mg/g	HCP	LCP 1	LCP 2	HCP	LCP 1	LCP 2		San	CP	AA	San × CP
Jejunum
Putrescine	0.10	0.06	0.05	0.12	0.09	0.06	0.031	0.420	0.112	0.554	0.935
Histamine	0.05	0.12	0.04	0.08	0.12	0.03	0.020	0.608	0.419	0.001	0.360
Cadaverine	0.41	0.39	0.41	0.34	0.51	0.43	0.128	0.833	0.571	0.796	0.493
Spermidine	0.10	0.08	0.06	0.09	0.10	0.13	0.023	0.176	0.872	0.780	0.152
Tyramine	0.03	0.02	0.00	0.01	0.03	0.03	0.017	0.447	0.922	0.786	0.158
Spermine	0.05	0.05	0.07	0.04	0.05	0.06	0.009	0.528	0.274	0.291	0.637
Tryptamine	0.01	0.01	0.02	0.01	0.02	0.02	0.004	0.395	0.067	0.851	0.402
Colon
Putrescine	0.13	0.14	0.14	0.14	0.19	0.22	0.034	0.135	0.170	0.732	0.383
Histamine	0.04	0.07	0.04	0.02	0.11	0.05	0.017	0.339	0.026	0.008	0.112
Cadaverine	0.81	1.09	1.31	0.65	1.45	0.97	0.218	0.784	0.016	0.564	0.651
Spermidine	0.29ab	0.24ab	0.20b	0.31ab	0.37a	0.39a	0.035	0.001	0.952	0.774	0.031
Tyramine	0.09	0.05	0.02	0.04	0.05	0.03	0.027	0.522	0.300	0.306	0.210
Spermine	0.03bc	0.03c	0.03c	0.04bc	0.05ab	0.06a	0.005	0.001	0.085	0.221	0.007
Tryptamine	0.03	0.04	0.03	0.03	0.04	0.03	0.005	0.675	0.240	0.005	0.554

¹HCP, high crude protein; LCP 1, low crude protein supplemented with nine indispensable crystalline amino acids except for Arg; LCP 2, low crude protein supplemented with only six indispensable crystalline amino acids (Lys, Met, Thr, Trp, Val, and Ile).

²San, main effect of sanitation; CP, main effect of crude protein levels (HCP vs. LCP 1 and LCP 2); AA, main effect of crystalline amino acids supplementation patterns (LCP 1 vs. LCP 2); San × CP, interactive effect of sanitation and crude protein levels (HCP vs. LCP 1 and LCP 2).

^a,bWithin a row, means with different superscripts differ (P < 0.05).

Conclusion

In conclusion, feeding LCP diets reduced proteolytic fermentation regardless of the sanitation conditions, as indicated by lower ammonia nitrogen content in jejunal digesta compared to those fed the HCP diet. Increased acetic acid concentrations and spermidine and spermine concentrations in piglets fed LCP diets under USC indicate that LCP diets might be effective in increasing beneficial VFA and alleviating inflammatory response under USC. However, LCP diets with different CAA supplementation patterns did not affect bacterial diversity in weaned pigs, regardless of the hygienic conditions where the animals were housed.

Glossary

Abbreviations

AA

amino acids

ADFI

average daily feed intake

ADG

average daily gain

BCFA

branched-chain fatty acids

BW

body weight

CAA

crystalline amino acids

CP

crude protein

CSC

clean sanitary conditions

G:F

gain to feed ratio

HCP

high crude protein

LCP

low crude protein

OTU

operational taxonomic unit

PCoA

principal coordinate analysis

USC

unclean sanitary conditions

VFA

volatile fatty acids