The experimental protocol of this study was following the Guide for the Care and Use of Laboratory Animals prepared by the Institutional Animal Care and Use Committee of China Agricultural University. This experiment was conducted in the FengNing Swine Research Unit of China Agricultural University (Academician Workstation in Chengde Jiuyun Agricultural and Livestock Co., Ltd., Hebei, China). The product α-GML was provided by Zhejiang Libiduo Biotechnology Co. Ltd and contained 85% α-GML.
Animals and experimental design
Fifth or sixth parity sows (n = 80; Landrace × Large White) with similar backfat (BF) thickness were assigned randomly to one of four treatments: control diet (corn-soybean basal diet, n = 20), 500 mg/kg α-GML diet (basal diet + 500 mg/kg α-GML, n = 20), 1000 mg/kg α-GML diet (basal diet + 1000 mg/kg α-GML, n = 20), and 2000 mg/kg α-GML diet (basal diet + 2000 mg/kg α-GML, n = 20). Each treatment (Table 1) consisted of a gestation and a lactation diet and were formulated to meet nutritional requirements of late gestation (day 85 of gestation to parturition) and lactation sows according to the recommendations by the National Research Council [17]. Sows were housed individually in stalls from day 85 to day 107 of gestation on partially slatted concrete floors and ingested a total of 2.5 kg diet daily (9:00 and 15:00). Then, sows were moved to farrowing stalls on day 107 of gestation and ingested 2.76 kg of daily diet (9:00 and 15:00) from day 107 of gestation to parturition. Sows were offered approximately 0.5 kg on the first day after farrowing and then fed daily feed allotment was increased 1.0 kg more every day until maximum feed intake was reached. When maximum feed intake was achieved, sows were allowed ad libitum access to feed. Throughout the experiment, sows and piglets were allowed ad libitum access to water. On the day of farrowing (day = 0), litter size was adjusted to 12 ± 1 piglets by cross-fostering within dietary treatment. The ambient temperature in the farrowing room for sows was maintained at 20-23℃. Heat lamps were used to provided heat to neonatal pigs. The suckling piglets underwent routine tooth clipping, tail removal, subcutaneous iron dextran injections, and immunization. During the entire experiment, sow's milk was the sole food source for suckling piglets. Feed intake of the sows was recorded daily to allow calculation of average daily feed intake (ADFI) during the entire lactation period, and piglets were weighed at birth and day 21 of lactation. The growth performance and survival rates of sucking piglets were determined. Backfat thickness (P2, 6 cm from the midline at the head of the last rib) of sows at day 110 of gestation and day 21 of lactation was measured using an ultrasonic device (Piglog105; SFK Technology A/S, Herlev, Denmark) to assess the body condition of sows.
Collection of colostrum and milk from sows
Six sows from each group were randomly selected by injection of 10 IU oxytocin via the ear vein to induce milk letdown into the teat canal on day 0 (colostrum) and 21 (milk) of lactation. Milk samples were collected from different mammary glands (front, middle, and rear), and equal volumes (15 mL) of milk from three glands of a sow were mixed, and the samples were stored at –20℃ frige for subsequent analysis.
Collection of blood samples from piglets
Blood samples (2 mL) were obtained from six randomly piglets each group via the jugular vein using non-heparinized vacutainer tubes on day 21 of lactation. Serum samples were harvested for each blood sample after centrifugation at 3,000 × g for 15 min, and were stored at –20℃ frige for subsequent analysis.
Collection of fecal samples from piglets
Six piglets each group were used to collect fresh feces by the rectal palpation method on day 21 of lactation. Fecal samples were immediately frozen in liquid nitrogen and placed at -80℃ frige for subsequent analysis.
Analysis of immunoglobulin in colostrum and milk
The concentrations of immunoglobulin A (IgA), immunoglobulin G (IgG), and immunoglobulin M (IgM) in colostrum and milk were determined according to Che et al. [18]. Briefly, the optical density values of IgA, IgG and IgM standards were measured at 340 nm, 700 nm, and 340 nm using a UV-2401PC (Shimadzu Co., Japan UV-VIS recording spectrophotometer), and the concentration of each immunoglobulin was calculated by standard curve.
Analysis of fatty acid profiles in colostrum and milk
For analysis of fatty acid composition, milk samples (1 g) were transferred to a 25 mL Teflon-lined tube and neutralized with 4 mL of n-Hexane:isopropanol (3:2) and 2 mL of sodium sulfate (6.67%). After centrifugation at 5,000 × g for 10 min, all supernatants were transferred to a 20 mL hydrolysis tube, 200 µL of C11:0 internal standard was added, and dry with mixed nitrogen. Add 4 mL of hydrochloric acid methanol (3 mol/L) solution to the hydrolysis tube and tighten the cap, then reflux in a water bath at 80℃ for 2 h. After the water bath, 5 mL of 7% K2CO3 and 3 mL of hexane were added to the hydrolytic tube, mix by vortexing, centrifuged at 1,000 × g for 1 min, and about 1 mL of the upper liquid was filtered into a 1.5 mL glass bottle with a filter (filter pore size of 0.22 μm). Finally, the fatty acid methyl ester dissolved in the supernatant was analyzed by gas chromatography-mass spectrometry using Agilent 7890 B (Agilent Technologies, Palo Alto, CA) gas chromatograph and Agilent Technologies (60 m × 250 μm × 0.25 μm) column. Fatty acids were expressed as the proportion of each individual fatty acid to the total amount of all fatty acids in the sample. n-3 polyunsaturated fatty acid (PUFA), n-6 PUFA, and n-6: n-3 PUFA ratio were calculated.
Analysis of immunoglobulin in serum
Concentrations of IgA, IgG, and IgM were analyzed using ELISA kits validated for swine (CUSABIO Biotech Company, Wuhan, China) according to methods described [6].
Fecal microbial flora composition
Total feces bacterial DNA (n = 6) was extracted according to the manufacturer's instructions of the Qiagen QIAmp DNA stool extraction kit (Qiagen). The V3-V4 hypervariable regions of bacterial 16S rRNA were amplified by a PCR system using universal primers 338F (5'-ACTCCTACGGGAGGCAGCAG-3') and 806R (5'-GGACTACHVGGGTWTCTAAT-3'). The PCR amplification procedures were set in ABI GeneAmp® 9700 system (ABI, USA) as follows: predenaturation (95℃, 3 min), amplification (95℃, 30 s; 55℃, 30 s; 72℃, 45 s; a total of 27 cycles), extended (72℃, 10 min) [19]. The paired-end reads of pooled purified amplicons were sequenced on the Illumina MiSeq PE300 platform (Illumina, San Diego, CA). The original sequences were demultiplexed, quality filtered, trimmed, and denoised using Trimmomatic and merged according to the overlapping relationship by FLASH software (v1.2.11, http://ccb.jhu. Edu/software /FLASH/index.shtml) [20]. The operational taxonomic units (OTUs) with 97% similarity threshold were clustered Using UPARSE (version 7.1, http://drive5.com/uparse/) and their representative sequence categorized and analyzed by the RDP Classifier (http://rdp. cme.msu.edu/) against the Silva (SSU128) 16S rRNA database with a confidence threshold of 70% [17]. The composition and structure of fecal microbiota were analyzed according to the standardized OUT with the QIIME software (version 1.8.0) [21]. The linear discriminant analysis (LDA) effect size (LEfSe) algorithm was applied to identify specific taxa from phylum to genus level among each group of samples.
Statistical analysis
All data were analyzed using the one-way ANOVA procedure of SPSS statistical software (SPSS 26.0, IBM SPSS Company, Chicago, USA). Each sow (litter) or piglet was considered as the experimental unit. Data were evaluated for normality and homoscedasticity by the Shapiro-Wilk and Levene's tests, respectively. The linear and quadratic effects of different α-GML levels were determined using orthogonal polynomials for reproductive performance, concentrations of immunoglobulin and fatty acid composition of sows in milk, and serum antioxidant capacity and immunoglobulin levels of piglets. The bacterial community at the phylum, genus, and species level were analyzed by non-parametric factorial Kruskal-Wallis test and unpaired Wilcoxon Comparison test. The data were expressed as mean ± SEM, and differences were considered significant at P < 0.05, and the tendency was declared with 0.05 ≤ P < 0.10.