Providing domperidone throughout lactation enhances sow lactation performance

Abstract

The goal of this project was to determine the effects of domperidone given throughout lactation on hormonal and metabolic status, lactational performance, and gene expression in mammary epithelial cells of sows. Second parity sows were divided in two treatment groups: 1) daily intramuscular injections with canola oil (Control, CTL, n = 24), or 2) daily intramuscular injections with 0.5 mg/kg body weight (BW) of domperidone (DOMP, n = 23). Injections were given at 08h05 starting the day after farrowing until weaning. Over the first 4 d of treatment, DOMP sows also received 0.5 mg/kg BW of domperidone per os twice daily, whereas CTL sows were fed the vehicle. Litter size was standardized to 11 ± 1 within 24 h of birth and piglets were weighed at birth, 24 h postpartum, and on days 7, 22 (weaning on day 23), 35, and 56. Sow feed intake was recorded daily. Representative milk samples were obtained aseptically on day 21 of lactation from 15 sows per treatment for compositional analyses and milk fat globules were used to measure mRNA abundances of various genes. Jugular blood samples were obtained from all sows on days 2, 8, 16, and 23 of lactation to measure concentrations of prolactin, insulin-like growth factor-1 (IGF-1), leptin, adiponectin, insulin, glucose, urea, and free fatty acids (FFA). Concentrations of prolactin (P < 0.001) and FFA (P < 0.01) were increased in DOMP compared with CTL sows, whereas concentrations of insulin were decreased (P < 0.05). Urea concentrations were increased by treatment (P < 0.05) on days 16 and 23 of lactation, and those of IGF-1 were increased (P < 0.01) on day 16. Piglets from DOMP sows were heavier than those from CTL sows on day 22 (P < 0.01). Milk composition was unaffected by treatment. The mRNA abundance in milk fat globules for casein beta and whey acidic protein were lower (P ≤ 0.05) in DOMP than CTL sows. The long form of the prolactin receptor and the signal transducer and activator of transcription 5A mRNA abundances tended to be lower (P < 0.10) in DOMP than CTL sows. In conclusion, hyperprolactinemia induced by domperidone during lactation affected the endocrine and metabolite status of sows and stimulated growth of their suckling piglets.

Introduction

Starting at approximately 8 to 10 d of lactation, sows cannot produce enough milk to sustain optimal growth of their piglets (Harrell et al., 1993) and this problem was exacerbated with the current use of hyperprolific sow lines. Hence, it is most important to develop strategies to increase sow milk production. Nutrition of sows undoubtedly affects milking potential so that one must maximize sow feed intake during lactation (Gauthier et al., 2019). Yet, nutrient intake is most often insufficient to meet the requirements of lactating sows (Noblet et al., 1990), and this is most apparent in primiparous animals. Another avenue to stimulate sow milk yield is through better understanding of the hormonal control of lactation. One example being the demonstration of the essential role of prolactin for sow lactation. Indeed prolactin was shown to be essential not only for the initiation but also for the maintenance of lactation, and inhibition of prolactin secretion led to a drastic decrease in sow milk yield during any week of a 4-wk lactation (Farmer et al., 1998). However, when 15 mg of recombinant porcine prolactin were injected to sows thrice daily from days 2 to 23 of lactation, neither milk yield nor mammary development at the end of lactation were enhanced (Farmer et al., 1999). This absence of beneficial effects was suggested to be due to mammary gland receptors for prolactin being virtually all saturated even in control animals, so that exogenous prolactin could exert no biological action. In another study, 15 mg of porcine prolactin was injected twice daily to sows from day 102 of gestation until weaning, and milk yield was decreased in treated sows because of a premature onset of lactogenesis (King et al., 1996). Prolactin is mainly under negative regulation via dopamine, so that dopamine antagonists have been used to increase endogenous secretion of prolactin in sows (review by Farmer, 2016). When the dopamine antagonist domperidone was given to primiparous sows from either days 90 to 96 or days 90 to 109 of gestation, there were no beneficial effects on amount or composition of mammary parenchymal tissue (Caron et al., 2020). The 20-d treatment increased mRNA abundances of casein beta (CSN2) and whey acidic protein (WAP) in mammary parenchyma and decreased that of the long form of the prolactin receptor (PRLR-LF), but metabolic activity of mammary tissue was reduced most likely because of early involution. On the other hand, in a previous study, the provision of domperidone to late-pregnant sows stimulated mammary cell differentiation and subsequent milk yield (VanKlompenberg et al., 2013), hence suggesting that differentiation and not number of mammary cells may be crucial for milk production. Nevertheless, the impact of domperidone in lactating sows was never studied. The objective of the current project was to determine the effects of domperidone given throughout lactation on hormonal and metabolite status, lactational performance, and gene expression in milk fat globules of sows.

Materials and Methods

Animals were cared for according to a recommended code of practice (CCAC, 2009) and procedures were approved by the institutional animal care committee of the Sherbrooke Research and Development Centre of Agriculture and Agri-Food Canada.

Animals and treatments

Forty-seven second parity Yorkshire × Landrace sows (bred using pools of semen from Landrace boars) were divided in two treatment groups at farrowing, ensuring that mean body weights (BW) on day 110 of gestation were similar for both groups. Treatment groups were: 1) intramuscular injections with 3.75 mL of canola oil (Control, CTL, n = 24), or 2) intramuscular injections with 0.5 mg/kg BW of the dopamine receptor antagonist domperidone (DOMP, n = 23). Injections were given once daily at 08:05, starting the day after farrowing and lasting throughout lactation. The domperidone (Glentham Life Sciences, Corsham, Wiltshire, UK) was resuspended in canola oil and injected as an emulsion that was prepared twice weekly. For emulsification, the domperidone and oil mixture (33.33 mg of domperidone for 1.0 mL of oil) was left 30 min at room temperature while mixed by hand sporadically, it was then put in a sonicator bath with 4°C water for 2 min and was vortexed for 15 to 20 s. This suspension was mixed by hand just before filling each syringe on order to ensure homogeneity. Over the first 4 d of treatment, DOMP sows also received 0.5 mg/kg BW of domperidone (TEVA-Domperidone®, TEVA Canada, Toronto, Canada) per os twice daily (08:00 and 20:00) to allow for a rapid increase in prolactin, as previously demonstrated (Farmer et al., 2019). The domperidone was dissolved in 8 mL of table syrup and control sows were also fed 8 mL of table syrup twice daily during those 4 d.

Throughout lactation, sows were housed in a 1.5 × 2.1 m pen and were fed a commercial lactation diet (18.5% CP, 13.8 MJ/kg DE and 1.07% lysine) in 2 equal meals at a rate of 1.6 kg on the day of farrowing (day 1) and then ad libitum for the remainder of lactation. Sows were weighed and had their backfat thickness measured ultrasonically at P2 of the last rib (WED-3000, Schenzhen Well D Medical Electronics Co., Guangdong, China) on days 2 and 23 of lactation. Litter size was recorded at birth and was standardized to 11 ± 1 piglets (within treatment group) in most instances within 24 h of birth, and piglets were weighed at birth, 24 h postpartum (after standardization of litter size), and on days 7, 22 (weaning on day 23), 35, and 56. Piglets had no access to dry feed while suckling and sow feeders were equipped with lids so that piglet weight gain provided an estimate of milk yield. Representative milk samples were obtained aseptically on day 21 of lactation from 15 sows per treatment, by collecting milk from 3 functional glands (anterior, middle, and posterior) encompassing both sides of the udder after an intravenous injection of 1.0 mL of oxytocin (20 IU/mL; P.V.U. Victoriaville, QC, Canada) was given. Collected mammary glands were emptied to ensure uniform milk composition. Pigs were separated from their dam for 40 min before oxytocin was injected. Before giving the oxytocin, the udder was washed with warm soapy water and then with RNaseZap® (Thermo Fisher Scientific Baltics, Vilnius, Lithuania). Milk was collected through a sterile gauze. Mortality rate was recorded.

At weaning, litters were transferred to 1.9 × 1.9 m pens and pigs were fed, consecutively, three commercial diets ad libitum. The first diet, containing 17.05% CP, 14.82 MJ/kg of DE, and 1.28% of total lysine, was provided to piglets until they had received an individual average of 1.5 kg of diet. The second diet, containing 19.56% CP, 14.60 MJ/kg of DE, and 1.35% of total lysine, was then given to piglets until they had each received on average 5 kg of this diet. The last diet, containing 19.50% CP, 14.45 MJ/kg DE, and 1.29% of total lysine, was given to piglets until 56 d of age.

Jugular blood samples were obtained from all sows at 07:00 (before the morning meal) on days 2, 8, 16, and 23 of lactation to measure concentrations of prolactin, insulin-like growth factor-1 (IGF-1), leptin, adiponectin, insulin, glucose, urea, and free fatty acids (FFA).

Blood handling and assays

Blood samples for prolactin, leptin, and urea (30 mL) were collected into Vacutainer tubes without anticoagulant (Becton Dickinson, Franklin Lakes, NJ) and left at room temperature for 3 h, stored overnight at 4°C, centrifuged for 12 min at 1,800 × g at 4°C the following day, and serum was then harvested. Blood samples for IGF-1, insulin, and FFA assays (20 mL) were collected in EDTA-tubes (Becton Dickinson and Cie, Rutherford, NJ), put on ice and centrifuged within 20 min for 12 min at 1,800 × g at 4°C, and plasma was immediately recovered. Lastly, blood samples for glucose analyses (6 mL) were collected into tubes containing 12 mg of potassium oxalate and 15 mg of sodium fluoride to inhibit glycolysis, were put on ice and centrifuged within 20 min at 1,800 × g for 12 min at 4°C, and plasma was immediately recovered. Serum and plasma samples were frozen at -20°C until assayed. A previously described radioimmunoassay (RIA) was used to determine concentrations of prolactin (Robert et al., 1989). The radioinert prolactin and the first antibody to porcine prolactin were purchased from A.F. Parlow (U.S. National Hormone and Peptide Program, Harbor UCLA Medical Centre, Torrance, CA). Parallelism of a serum pool from lactating sows was 96.20%. Average recovery calculated by addition of various doses of radioinert prolactin to 50 μL of a pooled sample was 98.36%. Sensitivity of the assay was 1.5 ng/mL. The intra- and interassay CV were 2.31% and 4.20%, respectively. Concentrations of IGF-1 were measured with a commercial RIA kit for humans (ALPCO Diagnostics, Salem, NH) with small modifications as detailed previously (Plante et al., 2011). Assay validation showed parallelism of 101.23% and average recovery of 101.68%. Sensitivity of the assay was 0.10 ng/mL. The intra- and interassay CVs were 4.46% and 6.12%, respectively. Leptin was measured with a multi-species commercial RIA kit (EMD Millipore Corp., St. Louis, MO, USA). Assay validation showed parallelism of 99.39% and average recovery of 93.30%. Sensitivity of the assay was 1.0 ng/mL, and intra- and interassay CVs were 6.40% and 3.81%, respectively. Adiponectin was assayed using a commercial porcine adiponectin ADP ELISA kit (Cedarlane, Burlington, ON). Assay validation showed parallelism of 98.14% and average recovery of 100.57%. Sensitivity of the assay was 1.88 µg/mL, and intra- and interassay CVs were 3.86% and 7.89%, respectively. Insulin was measured with a porcine insulin commercial RIA kit (EMD Millipore Corp., St-Louis, MO, USA) after validation. Assay validation showed parallelism of 99.43% and average recovery of 92.04%. Sensitivity of the assay was 1.61 µU/mL and intra- and interassay CVs were 3.49% and 4.46%, respectively. Glucose was measured by an enzymatic colorimetric method with a commercial kit (Wako Chemicals, Richmond, VA, USA). Intra- and interassay CVs were 2.22% and 4.69%, respectively. Urea was measured by colorimetric analysis using an autoanalyzer (Auto-Analyser 3; Technicon Instruments Inc., Tarrytown, NY, USA) according to the method of Huntington (1984). Intra- and interassay CVs were 0.71% and 1.29%, respectively. Concentrations of FFA were also measured by colorimetry with a commercial kit (Wako Chemicals, Richmond, VA, USA). Intra- and interassay CVs were 1.87% and 3.48%, respectively.

Milk composition

Whole milk was analyzed for dry matter, protein, fat, lactose, and leptin contents as well as sodium (Na) and potassium (K) concentrations. Dry matter was measured using forced air oven drying (method 925.23; AOAC, 2005). Protein content was determined in duplicates with the micro-Kjeldahl method (Kjeltec Auto System; Tecator AB, Hoganas, Sweden) according to AOAC Method 991.20 (AOAC, 2005), and fat was extracted using an established ether extraction method (method 905.02; AOAC, 2005). Lactose was measured by a colorimetric method using a commercial kit (Megazyme International Ireland Ltd., Bray, Co. Wicklow, Ireland) with 30 µL of galactosidase and a final incubation time of 60 min. The concentrations of Na and K were measured with an atomic absorption spectrometer (PerkinElmer AAnalyst 300, Normalk, CT, USA). Validation for Na showed a parallelism of 99.54% and average mass recovery of 101.05%. Validation for K showed a parallelism of 96.03% and average mass recovery of 100.34%. Intra- and interassay CV for all above milk analyses were below 2.3%. Leptin was measured with a multi-species commercial RIA kit (EMD Millipore Corp., St. Louis, MO, USA). Assay validation showed parallelism of 100.15% and average recovery of 99.49%. Sensitivity of the assay was 1.0 ng/mL, and intra-assay CV was 7.18%; all samples being assayed at once.

Total RNA isolation from milk fat globules and complementary DNA synthesis

Total RNA was isolated from milk fat globules as a non-invasive procedure to evaluate the expression of genes in mammary epithelial cells (Brenaut et al., 2012). The extraction protocol was adapted from Toledo et al. (2020) as described by Farmer et al. (2021). Extracted RNA concentration and integrity were determined with the NanoDrop Spectophotometer ND-1000 (NanoDrop Thechnologies Inc., Wilmington, DE, USA) and by nucleic acids electrophoresis using the Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA). The cDNA synthesis was performed with the Superscript IV Reverse Transcriptase (200 U/mL; Fisher Scientific) and oligo(dT) 20 primers.

Relative mRNA abundance of selected genes in milk fat globules

The relative mRNA abundances were determined using real-time PCR analyses. Studied genes included CSN2, dopamine receptor D2 (DRD2), epidermal growth factor (EGF), leucine rich alpha-2-glycoprotein 1 (LRG1), PRLR-LF, secreted phosphoprotein 1 (SPP1), signal transducer and activator of transcription 5A and 5B (STAT5A and STAT5B), and WAP (Table 1). Some of these genes are involved in the prolactin signaling pathway (CSN2, PRLR-LF, STAT5A, STAT5B, and WAP) (Hennighausen et al., 1997) and others (EGF, SPP1, and LRG1) present differential mRNA abundance in the mammary parenchyma of prepubertal and late gestating gilts treated or not with domperidone (Farmer et al., 2021; M. F. Palin, unpublished results). The qPCR reactions and cycling conditions were as previously described (Farmer et al., 2021). Amplifications were performed in triplicate using an ABI 7500 Fast Real-Time PCR System (PE Applied Biosystems) and specificity of primers was verified with the melting curve analysis. Standard curves were made with serial dilutions of cDNA pools (Labrecque et al., 2009) and amplified in duplicate to obtain relative mRNA abundance using the standard curve method (Applied Biosystems, 1997). For each gene, a standard curve (in duplicate) was included in each 96-well plate to account for experimental differences between plates. Four reference genes [beta-actin (ACTB), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), peptidylpropyl isomerase A (PPIA) and hypoxanthine phosphoribosyltransferase 1 (HPRT1)] were also amplified (Table 1). Based on the NormFinder algorithm (Andersen et al., 2004), the PPIA reference gene was the least affected by treatments. Hence, for each gene of interest, the relative quantity ratios were calculated by dividing the relative quantity units of studied genes by those of PPIA. Mean values from triplicates were used for statistical analyses.

Table 1.

Primer sequences used for real-time PCR amplifications of studied genes

Genes1	Primer sequences, 5′-3′	GenBank accession no.	Product size, bp	Primers (F/R) concentration, nM	Amplification efficiency, %a
Studied genes
CSN2	(F)AAGCCTTTCAAGCAGTGAGGAA (R)TCTGGCGTTCATTCTCTGTTTG	NM_214434	101	300/300	97.96
DRD2	(F)CTTTGTCACTCTGGACGTCAT (R)AGCTGTAGCGTGTGTTGTAG	NM_001244253	116	300/300	93.27
EGF	(F)ACGGTGGTGTGTGTATGTATATTG (R)CCATTTCAAGTCTCTGTGCTGAC	NM_214020	101	300/300	102.39
LRG1	(F)TCTTGGAGCCCAGAAGGAA (R)CCTTGGCTGAGACCACAAATAG	XM_003123071	98	300/300	104.36
PRLR-LF	(F)GCTTTGAAGGGCTATAGCATGGT (R)GCTCTTCGGACTTGCCTTTCT	NM_001001868	106	600/600	101.97
SPP1	(F)TCCTAGCGCCACAGAATACTATTTC (R)GCTCAGGGCTTTCGTTGGA	NM_214023	90	300/300	101.44
STAT5A	(F)ATCTCATCTATGTGTTTCCCGACC (R)CGGAGCGAGCACAGGAGT	NM_214290	74	300/300	103.20
STAT5B	(F)GACTCTGAAATTGGTGGCATCA (R)GATTCCAAAACATTCTTTCCTGAGA	NM_214168	67	300/300	95.08
WAP	(F)CCAGGGACGACCAGTGTAGG (R)GCCTCTGTGTCAGGGTCCAG	NM_213841	82	300/300	110.09
Reference genes
ACTB	(F)CATCACCATCGGCAACGA (R)GGATGTCGACGTCGCACTT	XM_003124280	128	300/300	97.52
GAPDH	(F)CCCCAACGTGTCGGTTGT (R)CTCGGACGCCTGCTTCAC	NM_001206359	91	300/300	99.48
HPRT1	(F)GACCAGACTTTGTTGGATTTGAAA (R)CAAACATGATTCAAGTCCCTGAAG	NM_001032376	94	300/300	99.26
PPIA	(F)GGTCCTGGCATCTTGTCCAT (R)TCATGCCCTCTTTCACTTTGC	NM_214353	130	300/300	102.99

¹ACTB, actin beta, CSN2, casein beta; DRD2, dopamine receptor D2; EGF, epidermal growth factor; F, forward primers; GAPDH, Glyceraldehyde-3-Phosphate Dehydrogenase; HPRT1, Hypoxanthine Phosphoribosyltransferase 1; LRG1, Leucine Rich Alpha-2-Glycoprotein 1; PPIA, Peptidylprolyl Isomerase A; PRLR-LF, long form of the prolactin receptor; R, reverse primers; SPP1, secreted phosphoprotein 1; STAT5A and STAT5B, signal transducer and activator of transcription 5A and 5B; WAP, whey acidic protein.

^aAmplification efficiency (E) was calculated with E = 10^[-1/slope].

Statistical analyses

The MIXED procedure of SAS (SAS Inst. Inc., Cary, NC) was used for statistical analyses. The univariate model used for milk composition and mammary gene expression data included the effect of treatment, with the residual error being the error term used to test main effects of treatment. Repeated measures ANOVA with the factors treatment (the error term being sow within treatment) and day or week (the residual error being the error term) and the treatment by day or week interaction were carried out on BW, backfat and blood data of sows as well as litter BW. Separate analyses of variance for each day were also carried out on these data. Data in Tables are presented as least squares means ± maximal SEM, except for insulin where back-transformed data (analyses were done using the logarithmic transformation) are shown with the lower and upper confidence intervals. A logistic analysis (using odds ratio) was used to look at the effect of treatment on the incidence of mortality in suckling piglets.

Results

Sow and piglet performance

The BW and backfat thickness of sows are shown in Table 2. Sow BW was similar for DOMP and CTL sows on days 2 or 23 of lactation (P > 0.10), but lactation BW loss tended to differ, being 4.2 kg less (21.3% lower) for DOMP than CTL sows (P = 0.10). There were no treatment differences in backfat thickness of sows on either days 2 or 23 of lactation, nor was backfat loss over lactation affected by domperidone treatment (P > 0.10). Average daily sow feed intake for the 3 wk of lactation is shown in Table 2. Sows from both treatment groups consumed the same amount of feed on weeks 1 and 3, but DOMP sows tended to consume more feed on week 2 of lactation (7.8% increase, P = 0.10). Piglet BW and BW gain over various time periods are shown in Table 2. Mean piglet BW were similar across treatments at birth and after standardization of litter size at 24 h (P > 0.10). However, DOMP piglets tended to be heavier than CTL piglets on day 7 (P ≤ 0.10) and were heavier on day 22 (P < 0.01). Accordingly, BW gain tended to be greater for DOMP than CTL piglets between 24 h and day 7 of lactation (P ≤ 0.10) and was greater over the rest of lactation (days 7 to 22, P ≤ 0.01). There were no differences in incidence of mortality during lactation between CTL and DOMP litters (P > 0.10). In the post-weaning period, piglet BW tended to be heavier for DOMP than CTL litters on day 35 (P ≤ 0.10) but did not differ on day 56 (P > 0.10). The BW gains from days 22 to 35 or days 35 to 56 were also similar for piglets of both treatments (P > 0.10).

Table 2.

Body weight (BW) and backfat thickness of sows and BW of their piglets. Sows were either injected with canola oil (controls, CTL, n = 24) or with 0.5 mg/kg BW of domperidone (DOMP, n = 23) once daily from days 2 to 22 of lactation. Treated sows also received 0.5 mg/kg of domperidone per os twice daily on days 2, 3, and 4 of lactation

	Treatment
Variable measured2	CTL	DOMP	SEM1
Sow BW3, kg
Day 2 of lactation	255.45 [249.62, 261.29]	253.83 [247.87, 259.78]	2.96
Day 23 of lactation	235.70 [229.32, 242.08]	238.28 [231.76, 244.80]	3.24
BW loss in lactation	19.75c [16.15, 23.35]	15.54d [11.86, 19.23]	1.83
Sow Backfat thickness3, mm
Day 2 of lactation	19.6 [18.1, 21.1]	19.0 [17.4, 20.5]	0.8
Day 23 of lactation	16.0 [14.7, 17.3]	15.2 [13.9, 16.6]	0.7
Backfat loss in lactation	3.6 [2.8, 4.5]	3.7 [2.9, 4.6]	0.4
Sow weekly feed intake3, kg/day
Week 1	4.23 [3.88, 4.58]	4.52 [4.17, 1.88]	0.18
Week 2	6.06c [5.64, 6.49]	6.53d [6.09, 6.96]	0.22
Week 3	6.88 [6.47, 7.28]	7.14 [6.72, 7.55]	0.21
Piglet BW4, kg
Number of piglets in parenthesis
Birth	1.41 (16.2) [1.33, 1.48]	1.43 (16.4) [1.36, 1.51]	0.04
24 h	1.58 (12.0) [1.52 1.64]	1.63 (11.7) [1.57, 1.69]	0.03
Day 7	2.72c (12.0) [2.63, 2.81]	2.83d (11.7) [2.74, 2.93]	0.05
Day 22	7.34a (11.8) [7.14, 7.54]	7.75b (11.6) [7.55, 7.95]	0.10
Day 35	10.35c (11.7) [10.04, 10.65]	10.72d (11.4) [10.40, 11.03]	0.16
Day 56	23.83 (11.6) [23.21, 24.45]	24.30 (11.4) [23.67, 24.93]	0.31
Piglet BW gain, kg
Lactation:   week 1 (24 h to day 7)	1.14c [1.10, 1.19]	1.20d [1.15, 1.25]	0.02
weeks 2 and 3 (days 7 to 22)	4.61a [4.45, 4.77]	4.91b [4.74, 5.07]	0.08
Post-weaning:   days 22 to 35	2.99 [2.73, 3.26]	2.96 [2.69, 3.23]	0.14
days 35 to 56	13.49 [13.05, 13.93]	13.57 [13.11, 14.02]	0.22

¹Maximum value for the standard error of the mean (SEM).

²Means are presented with 95% lower and upper limits of confidence interval.

³Day or week effect (P < 0.001).

⁴Treatment x day effect (P < 0.10).

^a,

^bMeans within a row with different superscripts differ (P ≤ 0.01).

^c,

^dMeans within a row with different superscripts tend to differ (P ≤ 0.10).

Hormonal and metabolic data in sows

Concentrations of various hormones and metabolites in sow blood are shown in Table 3. Prolactin concentrations were affected (P < 0.001) by both treatment and day of lactation, with values decreasing as lactation advanced and DOMP sows having greater prolactin concentrations than CTL sows. There was also a tendency (P < 0.10) for a treatment × day interaction. Separate analyses per day showed that prolactin concentrations tended to be greater (P < 0.10) in DOMP than CTL sows on day 2, and were greater (P < 0.001) on days 8, 16, and 23 of lactation. There was a treatment × day interaction (P < 0.01) on IGF-1 concentrations. Values were greater (P = 0.01) in DOMP than CTL sows on day 16 of lactation and also tended to be greater (P = 0.07) on day 23. Leptin and adiponectin concentrations were only affected by day of lactation (P < 0.001). Leptin concentrations were greatest on day 2, whereas concentrations of adiponectin were lowest on day 2. Concentrations of insulin were affected by treatment (P < 0.05) and day (P < 0.01); they were lower in DOMP than in CTL sows and were highest on day 8 of lactation. Analyses on each day showed a tendency for DOMP sows to have lower concentrations of insulin than CTL sows on days 2 and 23 of lactation (P < 0.10). Glucose concentrations were only affected by day of lactation (P = 0.01), with values being lowest on day 23. There was a treatment × day interaction (P < 0.05) on concentrations of urea. Treated sows had greater circulating urea than CTL sows on days 16 and 23 of lactation and lowest values were seen on day 2 of lactation in both treatment groups. Concentrations of FFA were affected by both treatment (P < 0.01) and day (P < 0.05). Values were greater (P = 0.001) for DOMP than CTL sows on day 23 of lactation and tended to be greater (P < 0.10) also on days 2 and 16. Concentrations of FFA were greater on day 16 compared with day 23 of lactation.

Table 3.

Circulating concentrations of hormones and metabolites in sows injected with canola oil (controls, CTL, n = 24) or with 0.5 mg/kg BW of domperidone (DOMP, n = 23) once daily from days 2 to 22 of lactation. Treated sows also received 0.5 mg/kg of domperidone per os twice daily on days 2, 3, and 4 of lactation

	Treatment
Variable measured2	CTL	DOMP	SEM1
Prolactin3,4,5, ng/mL
Day 2 of lactation	37.27c [32.78, 41.75]	43.23d [38.65, 47.82]	2.27
Day 8 of lactation	26.68a [23.61, 29.74]	35.45b [32.32, 38.58]	1.55
Day 16 of lactation	18.50a [15.94, 21.05]	31.06b [28.45, 33.67]	1.30
Day 23 of lactation	14.24a [11.86, 16.63]	27.84b [25.35, 30.33]	1.24
IGF-16, ng/mL
Day 2 of lactation	59.35 [52.47, 66.22]	59.31 [52.28, 66.33]	3.49
Day 8 of lactation	79.90 [69.09, 90.72]	87.22 [76.18, 98.27]	5.48
Day 16 of lactation	90.43a [72.43, 108.42]	123.01b [104.62, 141.39]	9.13
Day 23 of lactation	95.31c [79.84, 110.78]	115.36d [99.56, 131.17]	7.85
Leptin4, ng/mL
Day 2 of lactation	5.57 [4.63, 6.51]	5.40 [4.45, 6.36]	0.48
Day 8 of lactation	4.67 [3.92, 5.43]	4.53 [3.75, 5.30]	0.38
Day 16 of lactation	3.79 [3.28, 4.31]	4.19 [3.66, 4.71]	0.26
Day 23 of lactation	4.18 [3.66, 4.70]	4.08 [3.55, 4.61]	0.26
Adiponectin4, µg/mL
Day 2 of lactation	3.49 [3.17, 3.81]	3.30 [2.97, 3.63]	0.16
Day 8 of lactation	3.80 [3.38, 4.23]	3.60 [3.16, 4.03]	0.22
Day 16 of lactation	3.76 [3.32, 4.20]	3.60 [3.15, 4.05]	0.22
Day 23 of lactation	3.80 [3.40, 4.20]	3.55 [3.14, 3.96]	0.20
Insulin3,4, µU/mL
Day 2 of lactation7	13.86c [10.57, 18.17]	9.48d [7.19, 12.50]
Day 8 of lactation7	20.03 [13.28, 30.21]	16.54 [10.87, 25.18]
Day 16 of lactation7	12.49 [8.48, 18.39]	9.94 [6.70, 14.76]
Day 23 of lactation7	11.58c [8.40, 15.97]	7.78d [5.60, 10.80]
Glucose4, mMol/L
Day 2 of lactation  Day 8 of lactation	4.24 [4.04, 4.44] 4.16 [3.81, 4.50]	4.05 [3.84, 4.26] 4.17 [3.82, 4.53]	0.10 0.18
Day 16 of lactation	4.10 [3.78, 4.42]	4.15 [3.82, 4.47]	0.16
Day 23 of lactation	3.92 [3.66, 4.17]	3.68 [3.42, 3.95]	0.13
Urea6, mMol/L
Day 2 of lactation	7.17 [6.37, 7.97]	7.17 [6.36, 7.99]	0.41
Day 8 of lactation	9.10 [8.15, 10.05]	9.06 [8.09, 10.03]	0.48
Day 16 of lactation	10.80a [9.93, 11.67]	9.54b [8.66, 10.43]	0.44
Day 23 of lactation	11.49a [10.67, 12.32]	9.67b [8.83, 10.52]	0.42
FFA3,4, µEq/L
Day 2 of lactation	290.24c [184.78, 395.70]	418.40d [310.68, 526,13]	53.49
Day 8 of lactation	338.61 [223.98, 453.23]	389.34 [272.25, 506.43]	58.14
Day 16 of lactation	350.06c [223.86, 476.27]	515.50d [386.58, 644.42]	64.01
Day 23 of lactation	195.32a [112.21, 278.44]	397.92b [313.02, 482.82]	42.15

¹Maximum value for the standard error of the mean (SEM).

²Means are presented with 95% lower and upper limits of confidence interval.

³Treatment effect (P < 0.01).

⁴Day effect (P < 0.05).

⁵Tendency for treatment x day interaction (P < 0.10).

⁶Treatment × day interaction (P < 0.05).

⁷Back-transformed values with lower and upper limits of confidence interval in brackets.

^a,

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

^c,

^dMeans within a row with different superscripts tend to differ (P < 0.10).

Milk composition and mammary gene expression

Sow milk composition is shown in Table 4. None of the measures of standard milk composition (dry matter, fat, protein, lactose) were affected by the domperidone treatment and the same was true for the Na/K ratio (P > 0.10). Leptin concentrations in milk were also not affected by treatment (P > 0.10). The expression levels of selected genes in milk fat globules are shown in Table 5. The mRNA abundances for CSN2 and WAP were lower (P ≤ 0.05) in milk from DOMP compared with that of CTL sows, whereas expression for LRG1 tended to be higher (P < 0.10) in DOMP sows. The PRLR-LF and STAT5A mRNA abundances tended to be lower (P < 0.10) in DOMP compared with CTL sows.

Table 4.

Milk composition on day 21 of lactation for sows injected with canola oil (controls, CTL, n = 15) or with 0.5 mg/kg BW of domperidone (DOMP, n = 14) once daily from days 2 to 22 of lactation. Treated sows also received 0.5 mg/kg of domperidone per os twice daily on days 2, 3, and 4 of lactation

	Treatment
Variable measured	CTL	DOMP	SEM1
Dry matter, %	18.0	18.4	0.2
Fat, %	6.6	7.1	0.2
Protein, %	4.9	4.8	0.9
Lactose, %	5.1	5.2	0.4
Na/K	0.33	0.32	0.09
Leptin, ng/mL	70.0	69.6	3.8

¹Maximum value for the standard error of the mean (SEM).

Table 5.

Studied genes mRNA abundance in milk fat globules for sows injected with canola oil (CTL, n = 13) or with 0.5 mg/kg BW of domperidone (DOMP, n = 12) once daily from days 2 to 22 of lactation. Treated sows also received 0.5 mg/kg of domperidone per os twice daily on days 2, 3, and 4 of lactation

Genes1	Treatments
	CTL	DOMP	SEM2	P-value
CSN2	1.69 [1.44, 1.93]	1.37 [1.17, 1.57]	0.12	0.047
DRD2	1.19 [0.79, 1.59]	0.93 [0.50, 1.36]	0.21	0.372
EGF	1.15 [0.98, 1.32]	1.06 [0.91, 1.21]	0.08	0.388
LRG1	0.70 [0.51, 0.88]	1.31 [0.63, 2.00]	0.33	0.085
PRLR-LF	1.11 [0.93, 1.28]	0.89 [0.74, 1.04]	0.09	0.071
SPP1	0.67 [0.00, 1.37]	1.05 [00, 2.16]	0.54	0.558
STAT5A	1.11 [0.99, 1.22]	0.97 [0.87, 1.08]	0.05	0.084
STAT5B	1.08 [0.94, 1.21]	1.04 [0.92, 1.16]	0.06	0.661
WAP	1.23 [1.01, 1.46]	0.93 [0.78, 1.09]	0.11	0.032

¹Values correspond to relative mRNA abundance as determined with the standard curve method, described in materials and methods. Means are presented with 95% lower and upper limits of confidence interval.

²Maximum value for the standard error of the mean (SEM).

Discussion

There are indications that hyperprolactinemia brought about by domperidone can stimulate milk yield in humans (Ingram et al., 2012; Khorana et al., 2021; Wada et al., 2019) and current findings provide the first demonstration that increasing prolactin concentrations during lactation can augment milk production in sows. In an early report, 15 mg of porcine prolactin was injected twice daily to sows from day 102 of gestation until weaning, and even though prolactin concentrations were increased throughout lactation milk yield was decreased (King et al., 1996). This decrease was more pronounced later in lactation (days 19 to 22) than earlier (days 5 to 8) and the negative effect of treatment was reported to be due to a premature onset of lactogenesis. Indeed, there was evidence of lacteal secretions in mammary biopsies from treated sows on day 110 of gestation, and plasma lactose concentrations of treated sows on day 105 of gestation were greater than those from control sows. This increase in lactose is indicative of leakage of milk components into the blood due to the absence of removal from the glands. Such an effect of increased prolactin concentrations on early onset of lactogenesis was also recently observed when sows received the dopamine antagonist domperidone from days 90 to 109 of gestation (Caron et al., 2020). Without the presence of piglets to remove lacteal secretions from the mammary glands, precocious mammary involution was induced. To the best of our knowledge, there is only one trial where prolactin concentrations in sows were increased in lactation only. Farmer et al. (1999) injected sows with 15 mg of recombinant porcine prolactin from days 2 to 23 of lactation. Sow milk yield and piglet growth were not affected by treatment and upon collection of mammary glands it was apparent that circulating concentrations of prolactin in control sows were 80-fold greater than receptor affinity in mammary tissue. This suggested that virtually all receptors were saturated with endogenous prolactin in the mammary parenchyma of control sows. Such a state could have prohibited the additional exogenous prolactin from having any biological effect on mammary tissue. In the current study, the sustained hyperprolactinemic state in lactation was induced through the use of a dopamine antagonist so that the sow synthesized her own prolactin instead of receiving it exogenously and should have adjusted its number of prolactin receptors accordingly. Such a hypothesis is supported by the relation between physiological state and gene expression for the prolactin receptor in mammary tissue, whereby pubertal gilts have much lower prolactin concentrations and mRNA abundance for the long form of the prolactin receptor than late-pregnant gilts (Farmer and Palin, 2021). This potential impact of increased endogenous prolactin on its receptor number is most likely the reason why positive effects on milk yield were observed in the current study. The fact that milk composition was unaltered by treatment indicates that the increased growth rate of piglets from DOMP sows was due to greater milk yield. Of interest is the fact that the increase in piglet BW gain was more pronounced in late (weeks 2 and 3) than in early (week 1) lactation. This could be due to various reasons, namely: 1) the treatment effect on prolactin concentrations was greater in later lactation, 2) there could be a lag period between the increased prolactin concentrations and the effect on mammary tissue, 3) the effect could be most apparent in the period of maximum milk yield, or 4) the effect could be greatest when piglets are heavier, hence stimulating the mammary glands to a greater extent. It would be interesting to do follow-up studies comparing the effect of sustained hyperprolactinemia at either different time periods during lactation or with varying suckling intensities.

An important point to mention is the potential effect of IGF-1 on the greater growth rate of suckling piglets from DOMP sows observed in the current study. It was previously demonstrated that increasing concentrations of IGF-1 in late pregnancy (days 90 to 109 of gestation) stimulates mammary development in gilts (Farmer and Langendijk, 2019), thereby suggesting that a positive effect of IGF-1 on mammary tissue may also be observed during lactation. The growth factor IGF-1 is the mediator of action of somatotropin and when somatotropin was injected to sows during lactation, some studies showed an increase in milk yield (Boyd et al., 1983; Harkins et al., 1989), whereas others did not (Crenshaw et al., 1989; Cromwell et al., 1992). Differences in the dose and activity of the somatotropin used could have accounted for those discrepancies but concentrations of IGF-1 were not measured, hence making comparison difficult. Transgenic over-expression of IGF-1 was achieved in swine and found not to affect sow milk yield (Monaco et al., 2005). Nevertheless, the role of IGF-1 in stimulating milk production in the present study cannot be discounted. The fact that the domperidone treatment led to increases in IGF-1 concentrations in the later part of lactation, which was also the time during which the greatest piglet weight gain was observed, could suggest a role for IGF-1. It was also on days 16 and 23 of lactation that significant decreases in urea were observed in DOMP sows, indicating less catabolism and more efficient use of protein sources (Gourley et al., 2020), which are generally associated with increased IGF-1 concentrations (Farmer et al., 1992; Farmer and Langendijk, 2019). On the other hand, it is also in later lactation that the greatest increases in prolactin concentrations between DOMP and CTL sows were seen, being 32.87%, 67.89%, and 95.51% for days 8, 16, and 23 of lactation, respectively.

Some, but not all, of the changes encountered in gene expression analyses support a role for prolactin on mammary tissue of lactating sows. A lower mRNA abundance of PRLR-LF was also reported previously in mammary parenchyma from sows receiving domperidone for 20 d in late gestation (Caron et al., 2020) and could indicate a downregulation of increased prolactin on the expression of its own receptor, as was also seen in rabbits (Djiane et al., 1982). The tendency for lower expression of the prolactin-related gene STAT5A could, therefore, be linked with such a downregulation of domperidone on the prolactin receptor. The glycoprotein LRG1 has been reported to be involved in apoptosis, angiogenesis and cell migration (Jemmerson et al., 2021; Zhang et al., 2016; Zhong et al., 2015). In mice mammary tissue, LRG1 gene expression is low during gestation and lactation, and is up-regulated during the onset of mammary involution (Hughes et al., 2012). The increase in LRG1 expression observed during mammary involution was also significantly reduced in Stat3 KO mice (Hughes et al., 2012), thus suggesting a role for LRG1 in mammary apoptosis (Jemmerson et al., 2021). As mentioned earlier, a sustained hyperprolactinemia, resulting from injections of the dopamine antagonist domperidone, induced precocious mammary involution in late gestating gilts (Caron et al., 2020). Based on the abovementioned studies, one might consider that the observed tendency for increased expression of LRG1 in DOMP sows from the current study could be indicative of initiation of the involution process. This would also tie in with the lower mRNA abundances of CSN2 and WAP in DOMP sows because early mammary involution is generally associated with a reduction in the expression of major milk proteins (Piantoni et al., 2010). On the other hand, the downregulation of CSN2 and WAP is surprising considering the positive effect of DOMP on milk yield (current study), the previously observed increases in CSN2 and WAP mRNA abundance in late-pregnant gilts that were injected with domperidone (Caron et al., 2020) and the fact that the suckling stimulus was maintained in the current study, as indicated by the BW gain of piglets. The mammary involution normally seen with advancing lactation (44 compared with 22 d) is associated with lower milk lactose content and decreased circulating prolactin concentrations in sows (Farmer et al., 2007), which was not the case in the present study. Furthermore, milk stasis leading to decreased circulating prolactin is known to be the major effector for the initiation of the involution in late lactation (Farmer et al., 2007). Hence, it is unlikely that involution was generated by treatment in the current experiment.

As expected, the increased milk production of DOMP sows was associated with a greater feed intake. The fact that the decreases in BW and backfat thickness seen during lactation were not enhanced by domperidone indicates that the greater feed intake of DOMP sows helped to compensate for the increased nutrient use towards milk synthesis. The similar backfat thicknesses across treatments at the end of lactation are in agreement with the unaltered circulating concentrations of leptin and adiponectin. Indeed, adiponectin and leptin are both secreted by adipocytes (see review by Palin et al., 2017) and adiponectin concentrations are generally inversely related with body fat (Lara-Castro et al., 2006). These relationships also corroborate the lowest adiponectin and greatest leptin concentrations seen on day 2 of lactation when backfat thickness was greatest. On the other hand, the increased FFA concentrations of DOMP sows compared with CTL sows in later lactation would indicate greater use of body fat for milk production (Miao et al., 2008), which is corroborated by the 5% lower backfat at the end of lactation in DOMP sows.

Current findings provide the first demonstration that increasing prolactin concentrations during lactation augments milk production in sows. Domperidone not only affected prolactin concentrations but also increased concentrations of IGF-1 and led to advantageous effects on the metabolism of sows. It is not known whether the beneficial effects of treatment were due solely to the increase in prolactin or were also brought about by the greater IGF-1 concentrations. This needs to be established in future studies.

Glossary

Abbreviations

BW

body weight

DOMP

domperidone

FFA

free fatty acids

IGF-1

insulin-like growth factor-1

Conflict of interest statement

The authors declare no actual or potential conflicts of interest that affect their ability to objectively present or review research or data.