Effect of antibiotics and low-crude protein diets on growth performance, health, immune response, and fecal microbiota of growing pigs

Abstract

This study aimed to investigate the effects of diets with and without antibiotics supplementation and diets with 18.5% and 13.0% crude protein (CP) on growth performance, carcass characteristics, disease incidence, fecal microbiota, immune response, and antioxidant capacity of growing pigs. One hundred and eighty pigs (59-day-old; 18.5 ± 2.5 kg) were distributed in a randomized complete block design in a 2 × 2 factorial arrangement, nine replicates, and five pigs per pen. The factors were CP (18.5% or 13.0%) and antibiotics (none or 100 mg/kg tiamulin + 506 mg/kg oxytetracycline). Medicated diets were fed from days 59 to 73. After that, all pigs were fed their respective CP diets from 73 to 87 days. Data were analyzed using the Mixed procedure in SAS version 9.4. From days 59 to 73, pigs fed antibiotics diets had higher (P < 0.05) average daily feed intake (ADFI), average daily weight gain (ADG), gain to feed ratio (G:F), compared to the diets without antibiotics. From days 73 to 87 (postmedicated period), any previous supplementation of antibiotics did not affect pig growth performance. Overall (days 59 to 87), pigs-fed antibiotics diets had higher (P < 0.05) G:F compared to pigs-fed diets without antibiotics. In all periods evaluated, pigs fed 18.5% CP diets had higher (P < 0.05) ADG and G:F compared to pigs fed 13.0% CP. Pigs fed the 13.0% CP diets had lower (P < 0.05) fecal score and diarrhea incidence than those fed 18.5% CP. Pigs fed 18.5% CP diets had improved (P < 0.05) loin area compared to pigs-fed diets with 13.0% CP. At 66 days of age, pigs-fed antibiotics diets had lower (P < 0.05) alpha diversity estimated with Shannon and Simpson compared to the pig-fed diets without antibiotics. At family level, pigs fed 18.5% CP diets had higher (P < 0.05) relative abundance of Streptococcaceae, and lower (P < 0.05) relative abundance of Clostridiaceae at days 66 and 87 compared with pigs fed 13.0% CP. Pigs-fed antibiotics diets had lower (P < 0.05) immunoglobulin G and protein carbonyl concentrations at day 66 compared to the pigs-fed diets without antibiotics. The reduction of dietary CP from 18.5% to 13.0% reduced the growth performance and loin muscle area of growing pigs, although it was effective to reduce diarrhea incidence. Antibiotics improved growth performance, lowered diarrhea incidence, improved components of the humoral immune response, and reduced microbiota diversity. However, in the postmedicated period, we found no residual effect on the general health of the animals, and considering the overall period, only G:F was improved by the use of antibiotics.

The reduction of dietary crude protein from 18.5% to 13.0% reduces the growth performance of growing pigs, although it was effective to reduce diarrhea incidence. Antibiotics improved growth performance and health of animals; however, during the postmedicated period, only gain to feed ratio was improved by the use of antibiotics.

Introduction

Dietary antibiotics have been administered at therapeutic levels to mitigate the negative aspects of adapting to a new production phase and prevent disease, consequently improving animal growth performance (Ju et al., 2021). At the beginning of the growing–finishing phase, transporting and mixing animals are common practice on pig farms (Camp Montoro et al., 2022). These practices associated with less ­environmental control technology, which may expose animals to greater temperature fluctuations and increased pathogen exposure, coupled with the utilization of lower quality ingredients, in comparison to the initial nursery diets, are stressful to the pigs and can cause a negative impact on productive performance (Gomes et al., 2022).

Studies have shown that the beneficial effect of adding antibiotics to the diet on growth performance can be lost after discontinuing their use (Konz et al., 2011; Nitikanchana et al., 2012; Gomes et al., 2022). After antibiotic withdrawal, the dysbiosis of gut microbiota, suppression of innate immune defenses, propagation of opportunistic pathogens, and multidrug-resistant bacteria may occur resulting in increased susceptibility to diseases and lower growth performance (Zhang et al., 2020a). The use of antibiotics in production animals is considered a global health challenge, due to its association with selection of resistance in zoonotic bacteria (Diana et al., 2019; Ju et al., 2021).

Another negative impact of pig farming that has gained attention is related to environmental pollution due to the excretion of nitrogenous compounds (Ju et al., 2021; Rocha et al., 2022). Reducing dietary crude protein (CP) content has become a goal in the pig feed industry due to the limited availability and high cost of dietary protein sources, as well as the aim of enhancing gut health in pigs (Wang et al., 2018; Zhang et al., 2020b; Ju et al., 2021). Studies have reported that low CP diets effectively benefit gut health by reducing fermentation of nitrogenous compounds, increasing beneficial bacteria counts, and enhancing intestinal immune responses (Zhang et al., 2017; Wang et al., 2018).

However, there are limitations on the use of low CP diets, due to the lower content of dietary non-essential amino acids (NEAA), total nitrogen, intact protein, and/or bioactive compounds such as peptides and isoflavones, which may impair pig growth performance (Rocha et al., 2022). Indeed, there are divergences in the literature regarding the effect of low CP diets on the growth performance of pigs, with some works demonstrating no difference (Morales et al., 2015; Zhang et al., 2017; Qiu et al., 2018), while others demonstrating reduced growth performance (Morazán et al., 2015; Peng et al., 2016; Che et al., 2017; Li et al., 2018), even though diets are balanced for all essential amino acids. Based on a meta-analysis, Rocha et al. (2022) suggested that low CP diets should be formulated by assuming a minimum CP level to avoid the limitation of other nutrients that may be deficient when reducing the CP below a certain level.

Previous research has shown that in-feed antibiotics can increase the digestibility of nutrients, including CP (Yoon et al., 2012) and amino acids, and reduce the substrate of bacterial fermentation in the colon (Zhang et al., 2017). Furthermore, in-feed antibiotic can mitigate the negative effects of under-supplying total N in low CP diets by improving immune status, which may reduce NEAA expenditure to support immune response, improving the amino acid availability for muscle deposition (Gaskins et al., 2002; Peng et al., 2016).

Based on evidence from previous studies, we hypothesized that antibiotics only cause transient alterations in growing pigs performance and health. In addition, lowering dietary CP could compromise growth performance, although it improves intestinal health. Thus, the aim of this study was to investigate the effects of diets with and without antibiotics supplementation and diets with 18.5% and 13.0% CP on growth performance, carcass characteristics, cough, fecal consistency, diarrhea incidence, relative abundance and alpha diversity of fecal microbiota, immune response, and antioxidant capacity of growing pigs.

Material and Methods

All methods involving the handling of pigs followed the ethical principles of animal research (CONCEA) and were approved by the Commission of Ethics in the Use of Production Animals (CEUA), protocol 08/2022.

Animals, housing, experimental design, and diets

One hundred and eighty 59-day-old pigs (AGPIC 415 × Camborough) with initial body weight (BW) of 18.5 ± 2.5 kg were distributed in a randomized complete block design (three BW blocks) to four experimental diets in a factorial arrangement. The factors were CP (18.5% or 13.0%) and antibiotics (none or 100 mg/kg tiamulin + 506 mg/kg oxytetracycline). Medicated diets were fed from day 59 to 73. After that, all pigs were fed their respective CP diets from day 73 to 87 (post-medication period). There were nine replicated pens per treatment and five pigs per pen with the same number of barrows and females over entire treatment group. Each pen (2.50 × 1.50 m) was equipped with a nipple drinker and a dry feeder, with 50 cm wide and 25 cm deep feed pan. Pigs had free access to feed and water throughout the 28-d trial. Before the experimental period, in the last diet of the nursery phase, all pigs were fed the same diet, without antibiotics.

The antibiotics partially displaced corn and were used in therapeutic doses at the manufacturer’s recommended rate for pigs and in accordance with a valid prescription. Diets (Table 1) were formulated according to the nutritional recommendations of NRC (2012).

Table 1.

Ingredients and calculated nutritional composition of the diets (as-fed basis)

Ingredients, g/kg	18.5% CP	13.0% CP
Corn	689.2	847.7
Soybean meal	277.1	109.4
Dicalcium phosphate	11.5	13.5
Soybean oil	9.31	3.47
Limestone	7.39	7.60
Salt	2.03	2.04
Mineral–vitamin premix1	1.30	1.30
L-lysine, 78 %	1.79	7.03
DL-methionine	0.26	1.79
L-threonine	0.0	2.23
L-valine	0.0	1.55
L-isoleucine	0.0	1.05
L-tryptophan	0.0	0.65
L-histidine	0.0	0.56
BHT	0.10	0.10
Calculated and analyzed2 composition
Metabolized energy, kcal/kg	3300	3300
Crude protein, %	18.5 (18.1)	13.0 (12.8)
SID3 lysine, %	0.98 (1.05)	0.98 (0.98)
SID Met + Cys, %	0.55 (0.53)	0.55 (0.52)
SID threonine, %	0.60 (0.70)	0.59 (0.64)
SID tryptophan, %	0.20 (0.21)	0.17 (0.17)
SID valine, %	0.77 (0.94)	0.64 (0.74)
SID isoleucine, %	0.77 (0.77)	0.51 (0.59)
SID histine, %	0.45 (0.46)	0.34 (0.32)
Sodium, %	0.10	0.10
Calcium, %	0.66	0.66
Available phosphorus, %	0.33	0.31

¹Mineral–vitamin premix provided the following per kilogram of complete diet: copper—10 mg; iodine—1.5 mg; iron—100 mg; manganese—40 mg; selenium 0.3 mg; zinc—100mg; vitamin A—12,000 IU; vitamin D3—2,250 IU; vitamin E—65 IU; vitamin K—3 mg; thiamine—2.25; riboflavin—6 mg; pyridoxine—2.25 mg; vitamin B12—27 mcg; folic acid—400 mcg; biotin—150 mcg; pantothenic acid—22.5 mg; niacin—45 mg.

²Analyzed total amino acids and crude protein included in the parenthesis.

³Standardized ileal digestible.

Growth performance

The feed was weighed before feeding throughout the trial, and feed wastage was collected and weighed daily to determine the average daily feed intake (ADFI). At 59 (beginning), 66, 73, 80, and 87 days of age (end of the trial), pigs were individually weighed to determine BW, Gain to feed ratio (G:F), was calculated from average daily gain (ADG) and ADFI.

Disease score measurements

A trained technician checked pigs daily for signs of disease. Each day, the same technician recorded the number of pigs in each pen showing signs of cough using the method described by Puls et al. (2019). At the same time, the fecal consistency of each pig was visually assessed using the method described by Liu et al. (2010). Fresh excreta were ranked using the following scale: 0 = solid; 1 = semi-solid; 2 = semi-liquid; and 3 = liquid. The occurrence of diarrhea was defined as production of feces at level 2 or 3 for 2 continuous days. Diarrhea incidence (%) = the number of pigs with diarrhea in each pen × diarrhea days ÷ (5 pigs × 14 d) × 100%. Observations were made at approximately the same time each day throughout the experimental period.

Ultrasound measurements

At 73 and 87 days of age, one pig with a BW closest to the average weight of the pigs within its respective pen was individually scanned using a 3.5-MHz linear array transducer ultrasound equipment (Aloka SSD 500V, Hitachi-Aloka Medical Ltd., Tokyo-Japan). Images were collected between the 10th and 11th ribs, as recommended by the National Swine Improvement Federation Guidelines. Later, from the images collected, loin area and depth (Longissimus dorsi), and backfat thickness were calculated using the computer program Biosoft software (BioSoft Toolbox for Swine, Biotronics Inc., Ames, IA, USA).

Microbiome analyses

Feces collection and DNA extraction

At 66 and 87 days of age, fresh feces were collected from one pig, with BW closest to the pen average. Genomic DNA was extracted using mechanical disruption of cells, and phenol/chloroform as described by Stevenson and Weimer (2007).

Sequencing data processing

All sequencing of the feces microbial 16S rRNA gene was performed by Novogene Corporation Inc, Sacramento, CA. The microbial genomic DNA extracted from the samples of the feces was qualified and quantified, and the V4 hypervariable region of the 16S rRNA gene was then amplified and purified using the PCR primers 515F (5’-GTGCCAGCMGCCGCGGTAA-3’) and 806R (5’-GGACTACHVGGGTWTCTAAT-3’). The sequencing library was quantified by Qubit and real-time PCR, and the barcoded V4 PCR amplicons were sequenced using an Illumina HiSeq 2500 PE250 platform. The gene sequences used in the subsequent analysis (effective tags) were obtained by successively splicing raw sequence reads using FLASH (version 1.2.7) software (Magoč and Salzberg, 2011), and filtering raw tags using QIIME (version 1.7.0) software (Caporaso et al., 2010) according to the split.libraries-fastq.py script with the parameters (-q 19; -p 0.75). The tags were compared with the reference database (SILVA138) using UCHIME algorithm (Edgar et al., 2011) to detect chimera sequences. Any chimera sequences were removed from further analysis (Haas et al., 2011).

OTU cluster and taxonomic annotation

Gene sequences analysis were performed by UPARSE software (version 7.0.1090) (Edgar, 2013) using all the effective tags. Sequences with ≥97% similarity were assigned to the same OTUs. Representative sequence for each OTU was screened for further annotation. For each ­representative sequence per OTU, Qiime (version 1.7.0) (Altschul et al., 1990) in Mothur method was performed against the SSUrRNA database (SILVA138) (Wang et al., 2007) for species annotation at each taxonomic rank (Threshold:0.8~1) (Quast et al., 2012). To construct the phylogenetic relationships of the representative sequences, all the representative sequences were aligned using MUSCLE (version 3.8.31) software (Edgar, 2004). The OTUs abundance information were normalized using a standard of sequence number corresponding to the sample with the least sequences. Subsequent analysis of alpha diversity was all performed basing on this output normalized data. Prior to data analysis, the top 10 most abundant bacteria were selected, and the least abundant were combined and analyzed as “Others”.

Blood collection

At 66 and 87 days of age, after 12 h of fasting, blood was collected from one pig with a BW closest to the average weight of the pigs within its respective pen. Blood was collected by orbital sinus puncture into 10-mL uncoated vacuum tubes, and samples were centrifuged (3,500 × g for 10 min) to separate the serum. Serum was stored at −20°C until analysis.

Serum analysis

Serum urea nitrogen and humoral immune status

Serum samples were sent to the Viçosa Clinical Laboratory (Viçosa, MG, Brazil) for determination of serum urea nitrogen (SUN, mg/dL) concentration (Ureal Cobas C311, Linklab, software PNCQ, Roche Diagnostics, Indianapolis, IN), immunoglobulin G (IgG, mg/dL) (Atellica CH IgG_2, CH Analyzer, Siemens Healthineers, Erlangen, Germany), immunoglobulin A (IgA, mg/dL) (9D98-21 Reagent Kit, Architect cSystems, Abbott Laboratories, Chicago, IL), and Creatinine (Creatinine, mg/dL, WS-Kovalent, kinetic method, BS-380, Mindray, Shenzhen, China).

Antioxidant activity

Catalase activity (CAT, U/ml) was determined according to (Aebi, 1984) using H₂O₂ as substrate. Briefly, 1 ml of H₂O₂ was added to the reaction mixture containing 10 µL of sample in 1.0 ml of potassium phosphate buffer (50 mmol/L, pH 7.0). After addition, the reaction mixture was monitored at 240 nm for 1 min at an interval of 30 s. An extinction coefficient of ɛ₂₄₀ = 0.036 mmol/L cm was used for calculations. One unit of CAT activity was defined as the amount of enzyme that decomposes one mmol H₂O₂ for 1 min.

The superoxide dismutase (SOD, U/ml) activity was assessed by the pyrogallol autoxidation method, based on the ability of this enzyme to catalyze the dismutation of superoxide (O₂−) into O₂ and H₂O₂ (Marklund and Marklund, 1974). An aliquot of serum (10 μL) was incubated with 170 μL of sodium phosphate buffer (50 mM, pH 7.8) in a polystyrene microplate. The reaction was initiated by adding 20 μL of pyrogallol (100 μmol/L). The rate of increase in the absorbance at 320 nm was measured in a microplate spectrophotometer (Multiskan GO, Thermo Scientific) after 30 min. One enzyme unit of SOD is defined as the amount that causes 50% inhibition of pyrogallol auto-oxidation.

Reduced glutathione (GSH, U/ml) levels were evaluated according to the protocol of Akerboom and Sies (1981, modified by Welker et al., 2012). The serum (10 µl) was precipitated in 0.5 ml of 10% trichloroacetic acid and then centrifuged at 15,000g for 10 min at 4°C. The supernatant (0.2 ml) was mixed with 2.6 ml of potassium phosphate buffer (0.2 mol/L, pH 7.0). The reaction was initiated by the addition of 0.2 ml 5, 5ʹ-dithiobis-2-nitrobenzoic acid (DTNB), and the absorbance was monitored at 412 nm. The concentration of GSH was determined by using the standard curve of known concentrations of GSH.

The ferric reducing antioxidant plasma (FRAP, µmol/ml) was evaluated based on the reduction of Fe³⁺ to Fe²⁺ in the plasma by nonenzymatic antioxidants and subsequent complexation of Fe²⁺ with 2,4,6-tri(2-pyridyl)-s-triazine to form the Fe²⁺ 2,4,6-tri(2-pyridyl)-s-triazine (2-pyridyl)-s-triazine chromophore. The maximum absorbance of the chromophores was measured spectrophotometrically at 595 nm (Benzie and Strain, 1996).

Oxidative/nitrosative damage

Protein carbonyl (PC, nmol/ml) was determined according to Levine et al. (1990). Briefly, 2,4-dinitrophenylhydrazine dye binds to damaged residues of protein forming hydrazone, which presents a maximum absorbance at 370 nm. Total carbonylated protein was calculated based on the molar extinction coefficient of ɛ370 = 22,000 mM/cm.

Nitric oxide (NO, nmol/mL) was measured based on nitrite (NO₂⁻) metabolite according to the Griess method described by Guevara et al. (1998). The Griess reagent allows for the detection of nitrite, used as an indicator of NO synthesis. An aliquot of serum (50 μL) was added to microplates with 100 μL of Griess reagent (1% sulfanilamide and 0.1% naphthyl ethylene diamine in 2.5% H₃PO₄). The microplates were incubated, and the absorbance was determined in a microplate reader (λ = 540 nm). The NO level of the samples was determined using the standard curve with known concentrations of sodium nitrite.

Malondialdehyde (MDA, nmol/ml) was determined according to the method described by Jentzsch et al. (1996). Initially, an aliquot of serum (10 μL) was added to a trichloroacetic acid (15%)/thiobarbituric acid (0.375%)/hydrochloric acid (0.6%) solution (1 mL). The solution was kept in a water bath at 90°C for 40 min. After cooling on ice, 500 μL of butyl alcohol was added, and tubes were vortexed vigorously for 2 min. The samples were then centrifuged (10 min at 9,000 × g) at room temperature, and MDA levels were measured in the upper phase (λ = 535 nm) in a microplate reader.

Statistical analysis

Data were analyzed based on a randomized complete block design in a factorial arrangement. The main effects were factors (CP and antibiotic) and their interaction, which were handled as fixed effects. Blocks were based on initial BW, which were handled random effects. Data were analyzed using the Mixed procedure in SAS version 9.4 (SAS Inc., Cary, NC, USA). The experimental unit for data of growth performance and disease incidence were based on the pen, whereas for ultrasound measurements, blood, and microbiota were based on the individual pig with BW closest to the pen average. The incidence of disease data was averaged for each period before analysis. In addition, the data were tested for normal distribution with the UNIVARIATE (Shapiro–Wilk test). The means were separated using the LSMEANS statement in SAS. When an interaction between the factors was significant, a pairwise comparison was made using the PDIFF option in SAS. Statistical differences were considered significant with P < 0.05. Tendency was considered when 0.05 ≤ P < 0.10.

Results

Growth performance

There was no interaction between CP level and antibiotics supplementation for any performance variables, in any evaluated periods (Table 2). Pigs-fed antibiotics diets had increased (P < 0.05) BW at days 66 and 73, and tended to have increased (P = 0.085) BW at day 87 compared to the diets without antibiotics. From days 59 to 66 and from days 59 to 73, pigs fed antibiotics diets had increased (P < 0.05) ADG, ADFI, and G:F compared to the diets without antibiotics. From days 73 to 87 (post-medicated period), antibiotics did not affect ADG, ADFI, and G:F. Overall (days 59 to 87), pigs fed antibiotics diets had increased (P < 0.05) G:F and tended to have increased (P = 0.077) ADG compared to the diets without antibiotics, while there was no effect on ADFI.

Table 2.

Effects of antibiotics and low-crude protein diets on pig growth performance¹

Item	ATB–2		ATB+3		SEM	P-value
	18.5% CP4	13.0% CP5	18.5% CP4	13.0% CP5		ATB	CP	CP × ATB
BW, kg
Initial	18.5	18.5	18.5	18.5	0.02	0.610	0.610	1.000
66 d of age	21.1	20.7	21.7	21.2	0.15	<0.001	0.004	0.705
73 d of age	25.9	25.3	26.9	26.7	0.21	<0.001	0.044	0.425
87 d of age	39.9	37.5	40.2	38.8	0.43	0.085	<0.001	0.287
ADG, g
59 to 66 d of age	367	308	458	385	19.8	<0.001	0.002	0.716
59 to 73 d of age	529	487	604	582	14.3	<0.001	0.033	0.479
73 to 87 d of age	998	872	944	864	19.7	0.123	<0.001	0.251
Overall	763	679	774	723	15.0	0.077	<0.001	0.278
ADFI, g
59 to 66 d of age	1004	1026	1058	1074	17.6	0.007	0.303	0.873
59 to 73 d of age	1169	1194	1226	1274	17.7	<0.001	0.047	0.541
73 to 87 d of age	1886	1810	1821	1819	27.9	0.321	0.168	0.194
Overall	1527	1502	1524	1546	21.0	0.345	0.949	0.263
G:F
59 to 66 d of age	366	299	434	354	15.4	<0.001	<0.001	0.686
59 to 73 d of age	452	407	493	455	7.6	<0.001	<0.001	0.672
73 to 87 d of age	531	482	519	475	6.2	0.136	<0.001	0.775
Overall	500	452	509	467	5.0	0.030	<0.001	0.586

¹Experimental medicated diets (ATB+) were fed from 59 to 73 days old. In the postmedicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 days of age.

²ATB−, diet without antibiotics.

³ATB+, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁴18.5% CP, diet with 18.5% of CP.

⁵13.0% CP, diet with 13.0% of CP.

Pigs fed 18.5% CP diets had increased (P < 0.05) BW at days 66, 73, and 87 compared to pigs fed 13.0% CP diets. From days 59 to 66, pigs fed 18.5% CP diets had increased (P < 0.05) ADG and G:F compared to pigs fed 13.0% CP diets, while dietary CP did not affect ADFI. From days 59 to 73, pigs fed 18.5% CP diets had increased (P < 0.05) ADG, ADFI, and G:F compared to pigs fed 13.0% CP diets. From days 73 to 87 and during the entire study period (days 59 to 87), pigs fed 18.5% CP diets had increased (P < 0.05) ADG and G:F compared to pigs fed 13.0% CP diets, while there was no effect on ADFI.

Disease score measurements

From days 59 to 73, there was an interaction (P < 0.05) between CP level and antibiotics supplementation for fecal score and diarrhea incidence (Table 3). The highest (P < 0.05) fecal score was observed in pigs fed 18.5% CP diets without antibiotics. The lowest (P < 0.05) fecal score was observed in pigs fed 13.0% CP diets with antibiotics. Pigs fed 13% CP without antibiotics and 18.5% CP with antibiotics presented intermediate results. Furthermore, pigs fed 18.5% CP diets without antibiotics presented the highest (P < 0.05) diarrhea incidence. In the same period, there was no interaction between CP level and antibiotics supplementation for cough. Pigs fed antibiotics diets tended to have reduced (P = 0.058) cough compared to the pigs fed diets without antibiotics, while CP did not affect cough.

Table 3.

Effects of antibiotics and low-crude protein diets on the incidence of disease¹

Item	ATB–2		ATB+3		SEM	P-value
	18.5% CP4	13.0% CP5	18.5% CP4	13.0% CP5		ATB	CP	CP × ATB
59 to 73 day of age
Cough6	0.167s	0.230	0.016	0.095	0.073	0.058	0.333	0.913
Fecal score7	1.240a	0.433b	0.465b	0.114c	0.070	<0.001	<0.001	0.002
D. Incidence (%)8	29.21a	2.39b	5.40b	0.32b	2.439	<0.001	<0.001	<0.001
73 to 87 d of age
Cough6	0.095	0.079	0.032	0.097	0.043	0.605	0.576	0.362
Fecal score7	1.830	0.635	1.590	0.425	0.120	0.070	<0.001	0.902
D. Incidence (%)8	57.62	9.36	50.00	2.70	5.265	0.184	<0.001	0.925

¹Experimental medicated diets (ATB+) were fed from 59 to 73 days old. In the postmedicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 days of age.

²ATB–, diet without antibiotics.

³ATB +, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁴18.5% CP, diet with 18.5% of CP.

⁵13.0% CP, diet with 13.0% of CP.

⁶Cough, total no. of observations/pen.

⁷Fecal consistency recorded as 0 = solid; 1 = semi-solid; 2 = semi-liquid; and 3 = liquid.

⁸The occurrence of diarrhea was defined as production of feces at level 2 or 3 for 2 continuous days. Diarrhea incidence (%) was defined as the number of pigs with diarrhea in each pen × diarrhea days ÷ (5 pigs × 14 d) × 100%.

^a–

^bMeans with different superscripts within a row differ (P < 0.05).

From days 73 to 87 (post-medicated period), there was no interaction between CP level and antibiotics supplementation for any disease incidence. Pigs fed antibiotics diets tended to have lower (P = 0.070) fecal score compared to the pigs fed diets without antibiotics, while antibiotics did not affect cough and diarrhea incidence. Pigs fed the 13.0% CP diets had lower (P < 0.05) fecal score and diarrhea incidence than those pigs fed 18.5% CP diets, while dietary CP did not affect cough.

Ultrasound measurements

There was no interaction between CP and antibiotics for backfat thickness, loin area, and depth, in any evaluated periods (Table 4). Antibiotic diets did not affect backfat thickness, loin area, and depth, at days 73 and 87. Pigs fed 18.5% CP diets had improved (P < 0.05) loin area at days 73 and 87, improved (P < 0.05) loin depth at d 73, tended to have lower (P = 0.089) backfat thickness at day 73 and tended to have improved (P = 0.073) loin depth at day 87, compared to the 13.0% CP. Dietary CP did not affect backfat thickness at day 87.

Table 4.

Effects of antibiotics and low crude protein diets on backfat thickness, loin area, and depth¹.

Item	ATB−2		ATB+3		SEM	P-value
	18.5% CP4	13.0% CP5	18.5% CP4	13.0% CP5		ATB	CP	CP × ATB
73 day of age
Backfat thickness	5.30	5.39	5.22	5.59	0.130	0.641	0.089	0.293
Loin area	11.23	10.92	11.80	10.76	0.329	0.547	0.048	0.273
Loin depth	23.77	22.87	24.68	22.83	0.590	0.462	0.027	0.429
87 days of age
Backfat thickness	6.03	6.20	6.08	6.02	0.167	0.689	0.738	0.511
Loin area	16.12	14.54	16.02	14.32	0.554	0.773	0.006	0.912
Loin depth	29.48	28.88	30.43	28.46	0.682	0.706	0.073	0.330

¹Experimental medicated diets (ATB +) were fed from 59 to 73 d old. In the post-medicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 d of age.

²ATB -, diet without antibiotics.

³ATB +, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁴18.5% CP, diet with 18.5% of CP.

⁵13.0% CP, diet with 13.0% of CP.

Alpha diversity of microbiota

At 66 days of age, there was no interaction between CP level and antibiotics supplementation for alpha diversity of fecal microbiota estimated with Chao1, Shannon, and Simpson. Pigs-fed antibiotics diets had lower (P < 0.05) alpha diversity estimated with Shannon (Figure 1B) and Simpson (Figure 1C) compared to the diets without antibiotics, while antibiotics did not affect alpha diversity estimated with Chao1 (Figure 1C). Crude protein did not affect alpha diversity estimated with Chao1, Shannon, and Simpson.

Figure 1.

Alpha diversity of fecal associated microbiota estimated with Chao1 richness (A), Shannon diversity (B), and Simpson diversity (C) at 66 days of age. ATB−, diet without antibiotics; ATB+, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline; 18.5% CP, diet with 18.5% of CP; 13.0% CP, diet with 13.0% of CP.

At 87 days of age, there was interaction (P < 0.05) between CP and antibiotics for alpha diversity estimated with Chao1 (Figure 2A); however, no significant differences were observed between the possible contrasts. Antibiotics did not affect the alpha diversity estimated with Shannon and Simpson. Pigs fed the 18.5% CP diets tended to have improved (P = 0.056) alpha diversity estimated with Simpson (Figure 2C) compared to the pigs fed 13.0% CP diets. However, CP did not affect alpha diversity estimated with Shannon (Figure 2B).

Figure 2.

Alpha diversity of fecal associated microbiota estimated with Chao1 richness (A), Shannon diversity (B), and Simpson diversity (C) at 87 days of age. ATB−, diet without antibiotics; ATB+, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline; 18.5% CP, diet with 18.5% of CP; 13.0% CP, diet with 13.0% of CP.

Relative abundance of microbiota

At phylum level, Firmicutes and Bacteroidota were the most abundant phylum, accounting for 90~95% of all microbiota in feces at day 66 and day 87 (Table 5). At phylum level, there was no interaction between CP levels and antibiotics inclusion in the diet for any relative abundance microbiota, in any evaluated periods.

Table 5.

Effects of antibiotics and low crude protein diets on the relative abundance of fecal associated microbiota at the phylum level in pigs¹.

Item	ATB−2		ATB+3		SEM	P-value
	18.5% CP4	13.0% CP5	18.5% CP4	13.0% CP5		ATB	CP	CP × ATB
66 days of age
Firmicutes	68.4	65.1	70.9	69.5	3.342	0.100	0.253	0.626
Bacteroidota	22.8	25.1	20.7	20.7	2.653	0.104	0.567	0.555
Proteobacteria	5.44	7.01	4.35	6.63	1.596	0.498	0.084	0.746
Euryarchaeota	1.63	0.96	1.75	2.05	0.676	0.358	0.781	0.459
Actinobacteriota	0.54	0.66	0.54	0.36	0.114	0.096	0.763	0.088
Others	1.11	1.12	1.76	0.75	0.542	0.790	0.338	0.334
87 days of age
Firmicutes	76	77.6	74.5	76.3	2.834	0.462	0.358	0.956
Bacteroidota	19.6	15	19.5	19	2.468	0.300	0.175	0.265
Proteobacteria	1.94	4.59	2.93	2.1	1.113	0.447	0.382	0.089
Euryarchaeota	0.79	0.69	0.7	0.84	0.347	0.925	0.965	0.725
Actinobacteriota	0.52	0.57	1.35	0.71	0.405	0.192	0.427	0.355
Others	1.14	1.52	1.04	1.12	0.234	0.270	0.317	0.511

¹Experimental medicated diets (ATB +) were fed from 59 to 73 d old. In the post-medicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 d of age.

²ATB−, diet without antibiotics.

³ATB+, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁴18.5% CP, diet with 18.5% of CP.

⁵13.0% CP, diet with 13.0% of CP.

Pigs-fed antibiotics diets tended to have lower (P = 0.096) relative abundance of Actinobacteriota at day 66 compared to the pigs fed diets without antibiotics. Antibiotics did not affect the relative abundance of Firmicutes, Bacteroidota, Proteobacteria, Euryarchaeota, and Others at day 66. At 87 days of age, antibiotics diets did not affect any relative abundance microbiota at the phylum level.

Pigs fed 18.5% CP tended to have lower (P = 0.084) relative abundance of Proteobacteria at day 66 compared to the pigs fed 13.0% CP diets. However, dietary CP did not affect the relative abundance of Firmicutes, Bacteroidota, Euryarchaeota, Actinobacteriota, and Others at day 66. At 87 days of age, CP diets did not affect any relative abundance microbiota at the phylum level.

At family level, there was no interaction between CP level and antibiotics supplementation for any relative abundance microbiota, in any evaluated periods (Table 6).

Table 6.

Effects of antibiotics and low-crude protein diets on the relative abundance of fecal associated microbiota at the family level in pigs¹

Item	ATB−2		ATB+3		SEM	P-value
	18.5% CP4	13.0% CP5	18.5% CP4	13.0% CP5		ATB	CP	CP × ATB
66 days of age
Prevotellaceae	17.2	19.7	14.6	14.3	2.383	0.076	0.629	0.523
Lachnospiraceae	12.5	13.3	14.6	15.2	0.902	0.027	0.437	0.969
Streptococcaceae	12.4	6.14	14.3	7.99	1.787	0.174	<0.001	0.984
Ruminococcaceae	7.66	6.62	8.14	8.26	1.007	0.215	0.582	0.494
Lactobacillaceae	7.80	6.96	3.79	5.89	1.528	0.031	0.580	0.201
Veillonellaceae	5.20	5.26	4.50	4.96	1.477	0.497	0.720	0.788
Oscillospiraceae	5.14	6.08	6.73	6.20	1.005	0.157	0.736	0.225
Selenomonadaceae	2.36	4.69	2.38	2.70	0.598	0.096	0.028	0.090
Muribaculaceae	3.71	3.52	4.12	4.45	0.672	0.281	0.908	0.679
Clostridiaceae	1.22	1.85	1.47	2.56	0.550	0.205	0.028	0.546
Others	24.7	25.9	25.3	27.6	1.825	0.515	0.324	0.751
87 days of age
Lachnospiraceae	15.4	16.1	14.9	15.6	0.861	0.453	0.258	0.983
Prevotellaceae	14.5	10.8	15.5	14.4	2.687	0.239	0.221	0.510
Streptococcaceae	11.0	5.01	9.43	4.73	1.219	0.356	<0.001	0.514
Ruminococcaceae	9.17	9.33	9.20	8.48	0.563	0.458	0.608	0.421
Lactobacillaceae	7.49	11.9	8.02	9.14	1.705	0.458	0.075	0.279
Veillonellaceae	7.19	6.94	7.52	9.09	0.970	0.165	0.454	0.305
Oscillospiraceae	6.13	6.17	5.40	6.37	0.569	0.635	0.365	0.405
Selenomonadaceae	4.17	4.19	3.94	6.36	1.136	0.326	0.221	0.228
Muribaculaceae	3.84	3.02	2.85	3.40	0.531	0.553	0.784	0.187
Clostridiaceae	2.58	4.66	2.53	3.57	0.601	0.331	0.011	0.377
Others	18.5	21.8	20.7	18.8	1.584	0.781	0.653	0.096

¹Experimental medicated diets (ATB+) were fed from 59 to 73 days old. In the postmedicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 days of age.

²ATB−, diet without antibiotics.

³ATB+, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁴18.5% CP, diet with 18.5% of CP.

⁵13.0% CP, diet with 13.0% of CP.

Pigs-fed antibiotics diets had higher (P < 0.05) relative abundance of Lachnospiraceae, lower (P < 0.05) relative abundance of Lactobacillaceae, and tended to have lower; relative abundance of Prevotellaceae (P = 0.076) and Selenomonadaceae (P = 0.096), at day 66 compared to the pigs-fed diets without antibiotics. Antibiotics did not affect the relative abundance of Streptococcaceae, Ruminococcaceae, Veillonellaceae, Oscillospiraceae, Muribaculaceae, Clostridiaceae, and Others. At 87 days of age, antibiotics did not affect any relative abundance microbiota at the family level.

Pigs fed 18.5% CP diets had higher (P < 0.05) relative abundance of Streptococcaceae, and lower (P < 0.05) relative abundance of Clostridiaceae at days 66 and 87 compared to the pigs fed 13.0% CP diets. Moreover, pigs fed 18.5% CP diets had lower (P < 0.05) relative abundance of Selenomonadaceae at day 66 and tended to have lower (P = 0.075) relative abundance of Lactobacillaceae at day 87 compared to the pigs fed 13.0% CP diets. Crude protein did not affect the relative abundance of Prevotellaceae, Lachnospiraceae, Ruminococcaceae, Lactobacillaceae, Veillonellaceae, Oscillospiraceae, Muribaculaceae, and Others at day 66. Moreover, CP did not affect the relative abundance of Lachnospiraceae, Prevotellaceae, Ruminococcaceae, Veillonellaceae, Oscillospiraceae, Selenomonadaceae, Muribaculaceae, and Others at day 87.

At species level, there was an interaction (P < 0.05) between CP and antibiotics for relative abundance of Prevotellaceae bacterium at 66 days of age (Table 7). Pigs fed 13.0% CP diets had higher (P < 0.05) relative abundance of Prevotellaceae bacterium only when fed diets without antibiotic.

Table 7.

Effects of antibiotics and low-crude protein diets on the relative abundance of fecal associated microbiota at the species level in pigs¹

Item2	ATB−3		ATB+4		SEM	P-value
	18.5% CP5	13.0% CP6	18.5% CP5	13.0% CP6		ATB	CP	CP × ATB
66 days of age
Streptococcus gallolyticus	12.2	6.01	14.1	7.69	1.773	0.181	<0.001	0.923
Megasphaera elsdenii	4.43	4.27	3.44	3.89	1.236	0.298	0.821	0.643
Prevotella copri	4.66	3.86	2.59	2.44	0.821	0.034	0.547	0.682
Lactobacillus johnsonii	3.30	1.86	1.30	2.09	0.749	0.178	0.618	0.094
Prevotellaceae bacterium	2.17b	4.73a	2.38b	1.85b	0.756	0.075	0.170	0.041
Lactobacillus amylovorus	1.90	2.22	0.96	1.53	0.539	0.033	0.235	0.741
Faecalibacterium prausnitzii	1.47	1.23	1.52	2.09	0.503	0.338	0.729	0.387
Ralstonia pickettii	1.41	1.74	0.73	1.12	0.829	0.287	0.554	0.958
Lactobacillus reuteri	1.22	1.49	0.75	1.10	0.338	0.030	0.115	0.830
Eubacterium rectale	0.41	0.36	0.52	0.47	0.093	0.181	0.566	0.999
Others	66.8	72.2	71.7	75.7	3.745	0.084	0.053	0.763
87 days of age
Streptococcus gallolyticus	10.9	4.84	9.24	4.60	1.210	0.358	<0.001	0.495
Megasphaera elsdenii	5.59	5.00	5.30	6.74	0.823	0.329	0.564	0.174
Lactobacillus amylovorus	3.26	6.64	3.91	5.11	1.162	0.679	0.037	0.309
Prevotella copri	2.96	1.93	3.80	2.66	0.806	0.174	0.063	0.911
Prevotellaceae bacterium	2.45	1.85	3.09	2.89	0.705	0.116	0.453	0.699
Faecalibacterium prausnitzii	1.69	1.68	1.76	1.47	0.191	0.712	0.422	0.452
Lactobacillus reuteri	1.35	1.86	1.30	1.37	0.227	0.212	0.190	0.307
Lactobacillus johnsonii	1.15	1.24	1.03	0.87	0.250	0.276	0.869	0.582
Eubacterium rectale	1.00	0.85	0.87	0.78	0.160	0.457	0.354	0.802
Ralstonia pickettii	0.08	0.27	0.13	0.03	0.091	0.272	0.570	0.112
Others	69.6	73.8	69.6	73.4	2.470	0.928	0.085	0.937

¹Experimental medicated diets (ATB +) were fed from 59 to 73 d old. In the post-medicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 d of age.

²IgG, immunoglobulin G; IgA, immunoglobulin A.

³ATB -, diet without antibiotics.

⁴ATB +, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁵18.5% CP, diet with 18.5% of CP.

⁶13.0% CP, diet with 13.0% of CP.l.

Pigs-fed antibiotics diets had lower (P < 0.05) relative abundance of Prevotella copri, Lactobacillus amylovorus, and Lactobacillus reuteri and tended to have higher (P = 0.084) relative abundance of Others at day 66 compared to the pigs fed diets without antibiotics. Antibiotics did not affect the relative abundance of Streptococcus gallolyticus, Megasphaera elsdenii, Lactobacillus johnsonii, Faecalibacterium prausnitzii, Ralstonia pickettii, and Eubacterium rectale.

Pigs fed 18.5% CP diets had higher (P < 0.05) relative abundance of Streptococcus gallolyticus and tended to have lower (P = 0.053) relative abundance of Others at day 66, compared to the pigs fed 13.0% CP diets. Dietary CP did not affect the relative abundance of Megasphaera elsdenii, Prevotella copri, Lactobacillus johnsonii, Prevotellaceae bacterium, Lactobacillus amylovorus, Faecalibacterium prausnitzii, Ralstonia pickettii, Lactobacillus reuteri, and Eubacterium rectale.

At species level, there was no interaction between CP and antibiotics for any relative abundance microbiota at 87 days of age. At 87 days of age, antibiotics did not affect any relative abundance microbiota at the species level. Pigs fed 18.5% CP diets had higher (P < 0.05) relative abundance of Streptococcus gallolyticus had lower (P < 0.05) relative abundance of Lactobacillus amylovorus tended to have higher (P = 0.063) relative abundance of Prevotella copri, and lower (P = 0.085) relative abundance of Others at day 87, compared to the 13.0% CP diets. The CP level did not affect the relative abundance of Megasphaera elsdenii, Prevotellaceae bacterium, Faecalibacterium prausnitzii, Lactobacillus reuteri, Lactobacillus johnsonii, Eubacterium rectale, and Ralstonia pickettii.

SUN, creatinine, and humoral immune status

There was no interaction between CP level and antibiotics supplementation for SUN, creatinine, IgG, and IgA, in any evaluated periods (Table 8). Pigs-fed antibiotics diets had lower (P < 0.05) IgG at d 66, lower (P < 0.05) creatinine at day 87 and tended to have lower (P = 0.098) IgA at day 87 compared to the pigs fed diets without antibiotics. Inclusion of antibiotics did not affect SUN, creatinine, and IgA at d 66, and SUN and IgG at day 87.

Table 8.

Effects of antibiotics and low-crude protein diets on pig humoral immune status, creatinine, and serum urea nitrogen¹

Item2	ATB−3		ATB+4		SEM	P-value
	18.5% CP5	13.0% CP6	18.5% CP5	13.0% CP6		ATB	CP	CP × ATB
66 d of age
SUN, mg/dL	22.7	13.8	23.2	13.2	1.01	1.000	<0.001	0.585
Creatinine, mg/dL	0.97	0.94	0.92	0.93	0.024	0.267	0.731	0.480
IgG, mg/dL	290	306	204	225	34.5	0.021	0.593	0.948
IgA, mg/dL	11.0	10.6	10.9	10.3	0.81	0.799	0.537	0.965
87 d of age
SUN, mg/dL	21.9	13.4	20.3	13.8	1.05	0.564	<0.001	0.374
Creatinine, mg/dL	1.03	0.95	0.99	0.89	0.020	0.022	<0.001	0.647
IgG, mg/dL	419	417	410	386	17.9	0.272	0.477	0.552
IgA, mg/dL	17.7	12.2	11.6	11.5	2.05	0.098	0.173	0.187

¹Experimental medicated diets (ATB +) were fed from 59 to 73 d old. In the post-medicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 d of age.

²IgG, immunoglobulin G; IgA, immunoglobulin A; SUN, serum urea nitrogen.

³ATB -, diet without antibiotics.

⁴ATB +, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁵18.5% CP, diet with 18.5% of CP.

⁶13.0% CP, diet with 13.0% of CP.

Pigs fed 18.5% CP diets had higher (P < 0.05) SUN at days 66 and 87 and had higher (P < 0.05) creatinine at day 87, compared to the pigs fed 13.0% CP diets. The level of CP did not affect IgG and IgA at days 66 and 87, and creatinine at day 66.

Antioxidant activity and oxidative/nitrosative damage

At 66 days of age, there was no interaction (P > 0.05) between CP level and antibiotics supplementation for any analysis of antioxidant activity and oxidative/nitrosative damage (Table 9). Pigs-fed antibiotics diets had lower (P < 0.05) PC compared to the diets without antibiotics. While antibiotics did not affect CAT, SOD, GSH, FRAP, NO, and MDA. At 66 days of age, CP diets did not affect any analysis of antioxidant activity and oxidative/nitrosative damage.

Table 9.

Effects of antibiotics and amino acids on pig antioxidant activity and oxidative/nitrosative damage¹

Item2	ATB−3		ATB+4		SEM	P-value
	18.5% CP5	13.0% CP6	18.5% CP5	13.0% CP6		ATB	CP	CP × ATB
66 d of age
CAT, U/ml	275.6	275.5	273.0	277.0	7.85	0.947	0.816	0.804
SOD, U/ml	161.7	167.7	161.0	169.0	7.64	0.964	0.364	0.894
GSH, U/ml	12.09	11.90	12.00	12.16	1.03	0.936	0.988	0.873
FRAP, µmol/ml	1.68	1.64	1.69	1.59	0.06	0.761	0.290	0.634
PC, nmol/ml	9.75	12.00	8.84	8.39	0.63	0.004	0.213	0.065
NO, nmol/ml	6.05	4.35	4.55	3.54	0.87	0.235	0.163	0.724
MDA, nmol/ml	2.20	2.16	2.12	2.11	0.06	0.379	0.692	0.815
87 d of age
CAT, U/ml	306.3	291.6	308.9	295.3	11.51	0.785	0.228	0.960
SOD, U/ml	183.9	184.3	173.1	182.4	9.34	0.520	0.619	0.650
GSH, U/ml	12.32	9.91	10.23	12.22	0.83	0.893	0.802	0.012
FRAP, µmol/ml	1.83a	1.84a	1.90a	1.75b	0.03	0.853	0.035	0.014
PC, nmol/ml	7.79	7.87	8.38	7.24	0.56	0.967	0.390	0.326
NO, nmol/ml	3.83	3.61	3.60	2.91	0.68	0.648	0.664	0.618
MDA, nmol/ml	1.72	1.80	1.75	1.67	0.06	0.474	0.946	0.273

¹Experimental medicated diets (ATB +) were fed from 59 to 73 d old. In the post-medicated period, all pigs were fed their respective crude protein (CP) control diets from 73 to 87 d of age.

²FRAP, ferric reducing antioxidant plasma; CAT, catalase; GSH, glutathione; SOD, superoxide dismutase; PC, protein carbonyl; NO, nitric oxide; MDA, malondialdehyde.

³ATB -, diet without antibiotics.

⁴ATB +, diet with 100 mg/kg tiamulin and 506 mg/kg oxytetracycline.

⁵18.5% CP, diet with 18.5% of CP.

⁶13.0

% CP, diet with 13.0% of CP.

^a‐

^bMeans with different superscripts within a row differ (P < 0.05).

At 87 days of age, there was an interaction (P < 0.05) between CP level and antibiotics supplementation for GSH and FRAP. In the GSH no significant differences were observed between the possible contrasts. Concerning FRAP, pigs fed 13.0% CP diets with antibiotics presented lower (P < 0.05) FRAP compared with pigs within other treatments. At 87 days of age, antibiotics and CP levels in the diets did not affect CAT, SOD, PC, NO, and MDA.

Discussion

The stress caused by changing production site and diet at the beginning of the grower–finisher can have a considerable effect on increased susceptibility to disease and reduced growth efficiency (Camp Montoro et al., 2022). The reduction of CP and the addition of antibiotics to the diet of pigs are strategies used to overcome this issue.

In the present study, as demonstrated by the improvement of growth performance variables in the medicated period, antibiotics effectively mitigated the negative aspects of moving pigs from the nursery to the growing–finishing phase. When the post-medicated period (days 73 to 87) is analyzed alone, ADG is similar implying that the immediate beneficial effect of antibiotics was lost. However, overall period G:F was improved and ADG tended to be improved thus, in the long term, the use of antibiotics presented a beneficial effect on growth performance. Our results are consistent with other studies that demonstrate the positive impact of antibiotic diets on the growth performance of pigs during use and in the long term (Ma et al., 2019; Puls et al., 2019; Dang et al., 2021).

Reducing the CP level in diets has been recognized as an effective strategy for the swine industry saving protein ingredients without compromising growth performance, improving efficiency of nitrogen use, and promoting gut health in pigs (Wang et al., 2018; Rocha et al., 2022; Zhu et al., 2022). In fact, the last two were confirmed in this study by the lower fecal score, incidence of diarrhea, and circulating urea level associated with lowering dietary CP. However, in the present study, the 13.0% CP diets (with addition of threonine, valine, isoleucine, tryptophan, and histidine) resulted in reduced ­performance variables throughout the entire period evaluated. Our results are consistent with other studies that demonstrated that a low CP diet reduces the growth performance of pigs even when the essential amino acid requirement is met by feed grade amino acid supplementation (Peng et al., 2016; Che et al., 2017; Li et al., 2018; Ju et al., 2021).

In the present study, feeding the 18.5% CP diets resulted in higher levels of SUN, reflecting greater amino acid catabolism (Valente Júnior et al., 2021). There are some explanations in the literature for the negative effects on growth performance when low CP diets are fed, such as the decrease in total dietary NEAA (Hou et al., 2017; Rocha et al., 2022). Peng et al. (2016) proposed that total N deficiency in low CP diet would become a limiting factor for the generation of NEAA from the essential amino acid. In addition, some amino acids that are usually considered as NEAA may become conditionally essential when the dietary CP is below a certain level (Che et al., 2017; Wang et al., 2018).

Moreover, lower CP diets supplemented with feed grade amino acid leads to a reduction in the supplementation of protein sources, such as soybean meal. For this reason, low CP diets may have a deficiency of intact proteins and the related release of bioactive compounds such as peptides and isoflavones, which may impair the growth performance of pigs (Rocha et al., 2022). Bioactive peptides confer biological functions beyond their nutritional value, they can exert actions at the level of the small intestine, and in the whole body after passage in the blood (Hou et al., 2017). Among the benefits of the bioactive compounds cited in the ­literature are antimicrobial, antioxidant, and immunomodulatory activities (Hou et al., 2017; Sun et al., 2020). Thus, reducing the protein sources to achieve a lower CP level may affect animal growth performance. Altogether, in the present work, the 13.0% CP level may have affected animal growth performance due to the deficiency of NEAA, intact protein, or bioactive compounds.

Consistently with reduction of worsening growth performance, our results indicated that animals that consumed 13.0% CP diets had decreased loin area and depth. Pork production is measured by the efficiency of muscle tissue deposition (Liao et al., 2015). The protein accretion is accompanied by water deposition in lean gain, so the actual advantage of lean accretion is greater than 4:1 over fat (Patience et al., 2015). Thus, knowledge about the growth and development of muscle is fundamentally important from either a technical or an economic standpoint (Liao et al., 2015).

Insulin-like growth factor-1 (IGF-1) is one of the best-characterized growth factors and has been shown to modulate skeletal muscle protein synthesis (Yoshida and Delafontaine, 2020). Ju et al. (2021) showed that protein restriction, from 20.0% to 14.0%, significantly decreased the abundance of mRNA of growth hormone receptor (GH-R) and IGF-1 in the liver of growing pigs. The reduction of GH-R inhibits the synthesis of liver IGF-1, which decreases the synthesis of protein in other peripheral tissues, such as skeletal muscle. This effect could be more significant in loins, due to being a leaner cut, and for this reason, it is usually the evaluated muscle (Tejeda et al., 2020).

One of the main concerns when pigs are fed low-CP diets is increased backfat thickness, which has been associated with higher net energy available for fat deposition. This occurs due to the reduction in caloric increment, since expenditure on catabolism and urinary excretion of excess N from the diet is reduced (Morazán et al., 2015; Wang et al., 2019; Upadhaya et al., 2021). In addition, high CP diets act on the mechanism of downregulation of triglyceride synthesis in adipose tissue, mainly by inhibition of lipogenic gene expression (Zhao et al., 2010). However, in the present study, the CP level did not influence backfat thickness. This result can be attributed to the greater efficiency of muscle deposition compared to fat deposition during the growing phase (Irshad et al., 2013), when there is a higher protein deposition rate (NRC, 2012).

Respiratory diseases and diarrhea cause economic losses in the swine industry worldwide (Diana et al., 2019; Pessoa et al., 2021). Prophylactic treatments with the use of antibiotics in the diet, coincide with moments of moving pigs to the growing–finishing barns (Burch et al., 2008). This has been helpful to mitigate the severity of respiratory diseases and diarrhea (Gaskins et al., 2002; Roberts et al., 2011; Diana et al., 2019; Puls et al., 2019). In the present study, the cough, fecal score, and diarrhea incidence were reduced in pigs-fed antibiotics diets during the medicated period. Moreover, pigs fed 13.0% CP diet presented lower diarrhea incidence only when fed without antibiotics. Thus, the diet with antibiotics was able to mitigate the increase in diarrhea caused by the higher level of CP. During the post-medicated period, there was no carryover effect of antibiotics on disease measurements, whereas pigs fed 13.0% CP diets presented lower diarrhea incidence and better fecal score than those fed 18.5% CP diets.

Excess dietary protein along with peptides and amino acids that escape digestion and absorption in the small intestine become available for bacterial fermentation in the colon, resulting in numerous irritating metabolites, and toxic byproducts (Wang et al., 2018; Pollock et al., 2019; Zhang et al., 2020b; Li et al., 2022). Above certain threshold levels, these metabolites can produce harmful effects. They can negatively affect growth and differentiation of intestinal epithelial cells, disturb colonocyte energy metabolism, lead to reduced sodium absorption in the colon and, therefore, reduced water absorption (Rist et al., 2013; Wang et al., 2018; Zhang et al., 2020b). Other studies have also demonstrated the effects of fermentable protein on gut health or overall health based on decreased stool consistency in swine (Heo et al., 2008, 2009; Lynegaard et al., 2021).

Furthermore, previous studies have reported that diets formulated with different levels of CP alter the composition of the intestinal microbiota in pigs (Pollock et al., 2019; Luise et al., 2021; Li et al., 2022). In the present study, changes in microbial composition at the family and species levels were observed. There was a reduction in the relative abundance of Streptococcaceae and Streptococcus gallolyticus in pigs fed 13.0% CP diets, in both evaluated periods. These results are corroborated by other authors (Yu et al., 2019; Duarte and Kim, 2021). This can be explained by the fact that Streptococcus is an amino acid fermenting, and in 13.0% CP diets, there was lower soybean meal substrate to stimulate the growth of protein fermenters. At the same time, Streptococcus has been associated with increased risk of infections and diseases (Zhang et al., 2020b). This helps to explain the improvement in fecal score and lower diarrhea incidence of animals that consumed 13.0% CP diets in the present study.

At family level, in both evaluated periods, animals that consumed 13.0% CP diets had higher relative abundance of Clostridiaceae. Although this family contains an opportunistic pathogen, being one of the main causes of diarrhea in pigs, it also contains other species (Clostridium butyricum) that are not harmful and can be beneficial to pigs (Duarte and Kim, 2022). In the present study, a decrease in fecal score and diarrhea incidence was observed in animals that consumed 13.0% CP diets. These results are consistent with Liu and Fan, (2023) that reported that the CP reduction (13.5%, 10.7%, and 8.0%) increased the abundance values of Clostridium sensu stricto 1 in the colonic microflora of finishing pigs. These results may be related to the fact that Clostridiaceae is a fiber-degrading bacteria family (Duarte and Kim, 2022), and as soybean meal was partially replaced by corn in 13.0% CP diets, there was a change in the fiber and starch substrate to stimulate Clostridium growth (Liu and Fan, 2023).

Normally, antibiotics are administered to reduce pathogenic bacteria. However, most antibiotics are not specific and, therefore, also eliminate a broad range of not pathogenic microorganisms that are often crucial for health (Neuman et al., 2018; Yang et al., 2021). In the present study, 100 mg/kg tiamulin and 506 mg/kg oxytetracycline were added to the antibiotic diets. Oxytetracycline belongs to the tetracycline family and is a broad-spectrum antimicrobial with inhibitory activity against gram-positive and gram-negative bacteria. Tiamulin is a derivative of pleuromutilin, being a narrow-spectrum antibiotic, with bacteriostatic action for gram-positive bacteria, anaerobic bacteria, and mycoplasma (Halling-Sørensen, 2001; Kumar et al., 2005). This may explain the decrease in diversity and richness after antibiotics treatment observed in the present and several studies (Li et al., 2020a; Jo et al., 2021; Huang et al., 2022). That selective pressure may lead to long-lasting changes in normal microbial homeostasis, making the gastrointestinal tract of pigs more susceptible to intrusions and colonization of pathogens (Looft et al., 2012). For these reasons, the loss of richness and diversity of the intestinal microbial is referred to as gut dysbiosis (Huang et al., 2022).

In the present study, the reduction in microbiota diversity (estimated with Shannon and Simpson) and the shifts in the relative abundance of individual taxa due to antibiotics were temporary rather than permanent. The fact that we did not find a residual effect over microbiota diversity can be explained by the dose and duration of administration of antibiotics, age of the animals, or characteristics of the antibiotics used such as spectrum of activity, formulation of the drug, pharmacokinetics, and pharmacodynamics (Zimmermann and Curtis, 2019; Yang et al., 2021).

The main serum immunoglobulins, IgG and IgA, are key components of the humoral immune response (Zhang et al., 2018). Gomes et al. (2022) found that animals-fed antibiotics diets presented lower IgA and IgG compared to the control diet, however, only during the medicated period. In the present study, lower IgG of pigs-fed antibiotics diets was also observed during the medicated period. This indicates that the immunoregulatory effect of dietary antibiotic, which culminated in an attenuation of the incidence of the disease and improvement of the performance variables is more evident during the medicated period.

Creatinine is a breakdown product of creatine phosphate in muscle (Patel et al., 2013). Serum creatinine concentration is a reliable surrogate marker of muscle mass (Schutte et al., 1981; Patel et al., 2013; Abonyi et al., 2022). In the present study, in the post-medicated period, the 18.5% CP diets resulted in greater serum creatinine. This is consistent with the improvement of performance variables and loin area and depth in pigs fed 18.5% CP diets.

At the same time, the antibiotics diets resulted in lower serum creatinine. Renal creatinine clearance is often used to estimate glomerular filtration rate, a biomarker for kidney injury (Patel et al., 2013; Stasi et al., 2021). However, creatinine is slow to reach diagnostic concentrations, and it may take several days to identify kidney damage (Laou et al., 2022). Sallustio et al. (2019) demonstrated that LPS injection led to an increase in serum creatinine, compared to a healthy group. These changes may be caused by the development of oxidative stress, which triggers cell membrane damage mechanisms (Gerunov et al., 2021). In the present study, the reduction of serum creatinine in animals consuming antibiotics is consistent with the lower oxidative stress indicated by the reduction in PC, in addition to attenuation of the incidence of the disease and improvement of the performance variables observed during the medicated period.

Protein carbonyl, the final product of the oxidation of protein, has been used as a biomarker of oxidative stress because of its formation during the initial period of oxidative stress and its high stability (Dalle-Donne et al., 2003; Mateos and Bravo, 2007; Silva-Guillen et al., 2020). Dietary supplementation with bacitracin reduced PC concentrations in challenged pigs, changed gut microbiota toward a less harmful milieu, and improved intestinal health, resulting in improved growth performance (Duarte et al., 2023). In the present study, the lower PC level of pigs fed antibiotics diets indicate lower oxidative stress, which helps to explain the better performance obtained during the medicated period.

The FRAP assay provides an index of the total antioxidant power of plasma (Long et al., 2021; Zhang and Piao, 2022). Pigs fed 13.0% CP diets, only when fed with antibiotics, presented lower FRAP in the post-medicated period. Lower growth performance in pigs fed 13.0% CP diets, combined with lower FRAP, may be associated with lower availability of dietary isoflavones (Smith and Dilger, 2018) combined with removal of antibiotics in the post-medicated period. Isoflavones are occurring flavonoid compounds found at high concentrations in the soybean (Smith and Dilger, 2018). These are deglycosylated and absorbed by the intestine, acting with multiple biological properties, including anti-inflammatory and antioxidant effects (Smith et al., 2019; Li et al., 2020b). The antioxidant activity of isoflavones is mainly through their inhibition of NF-κβ, a transcription regulator of many proinflammatory genes (Smith and Dilger, 2018).

Conclusion

The reduction of dietary CP from 18.5% to 13.0% reduced the growth performance and loin muscle area of growing pigs, although it was effective to reduce diarrhea incidence. During the medicated phase, antibiotics improved growth performance, lowered incidence of cough and diarrhea, improved components of the humoral immune response, and reduced microbiota diversity. However, during the post-medicated period, we found no residual effect on the general health of the animals, and considering the overall period, only G:F was improved by the use of antibiotics.

Glossary

Abbreviations

ADFI

average daily feed intake

ADG

average daily weight gain

ATB

antibiotics

BW

body weight

CAT

catalase

CP

crude protein

D. incidence

diarrhea incidence

G:F

gain to feed ratio

GSH

glutathione

FRAP

ferric reducing ability of plasma

IgA

immunoglobulin A

IgG

immunoglobulin G

MDA

malondialdehyde

NO

nitric oxide

PC

protein carbonyl

SID

standardized ileal digestible

SOD

superoxide dismutase

SUN

serum urea nitrogen

Conflict of interest statement

The authors declare no real or perceived conflict of interest.