Current knowledge on the control of onset and cessation of colostrogenesis in swine

Introduction

Colostrum is the elixir for life in newborn piglets. If the prepartum peak of prolactin is inhibited, thereby preventing the onset of lactogenesis, piglets undoubtedly die (Farmer et al., 1998). A minimum amount of 250 g of colostrum must be ingested by an average size neonatal piglet (1.4 kg) during the first 24 h after birth in order to acquire immune protection and sustain body growth (Quesnel et al., 2012). However, this does not occur in all litters. It was estimated that approximately one-third of sows cannot produce enough colostrum to fully support their litters (Quesnel et al., 2012). This is most important in the current context where hyperprolific sow lines are being used on a regular basis, thereby increasing the demand for colostrum production. Furthermore, it was demonstrated that colostrum yield is not affected by litter size (Devillers et al., 2005) so that it is not greater in very large litters. It is therefore imperative to attempt to augment the amount of colostrum available to piglets. Numerous reviews have described the nutritional and endocrine factors affecting colostrum yield and composition in swine (Devillers et al., 2007; Farmer and Quesnel, 2009; Quesnel et al., 2012; Quesnel and Farmer, 2019). The goal of the present review is to focus specifically on the duration of colostrogenesis and to summarize our current knowledge of the mechanisms of action involved in the control of its onset and cessation. Such information would be most pertinent to develop novel management strategies in peripartal sows to maximize colostrum availability to piglets.

Changes in Composition of Lacteal Secretions Over Time

The importance of colostrum for piglets is due to the drastic changes in the composition of lacteal secretions occurring right after farrowing and continuing more gradually thereafter. These changes in major components are given in Table 1. Colostrum is characterized as the first lacteal secretions (present for the 24 h following farrowing) that are rich in protein, immunoglobulins, some microminerals and vitamins, and hormones and growth factors (Hurley, 2015). There is then a transition period (until 72 h postpartum) during which concentrations of protein and immunoglobulins decrease compared to those of colostrum, but fat, energy, and lactose contents increase (Theil et al., 2014). The composition of mature milk (after days 7 to 10) is then relatively stable for the remainder of lactation, with very low concentrations of immunoglobulins and high lactose and fat contents compared to colostrum (Table 1).

Table 1.

Composition of lacteal secretions from colostrum, to transition milk, and milk¹

	Colostrum			Transition milk		Milk
Composition	0 h	12 h	24 h	36 h	72 h	17 d
Protein, %	17.7	12.2	8.6	7.3	6.1	4.7
IgG, mg/mL	64.4	34.7	10.3	–	3.1	1.0
Dry matter, %	27.3	22.4	20.6	21.4	21.2	18.9
Fat, %	5.1	5.3	6.9	9.1	9.8	8.2
Lactose, %	3.5	4.0	4.4	4.6	4.8	5.1
Energy2, kJ/100g	260	276	346	435	468	409

¹From Theil et al. (2014) and Hurley et al. (2015).

²Calculated energy derived from lactose and fat contents (energy in protein not included because the roles of proteins differ in colostrum [immunity] and milk [growth] and hence are not normally being oxidized to a great extent).

Normal Onset of Colostrogenesis

The initiation of lactogenesis in mammals is divided into two phases. The first is the secretory differentiation phase. It occurs in late gestation when mammary epithelial cells differentiate into lactocytes that have the capacity to synthesize unique milk constituents, such as lactose and casein, that are secreted in limited quantity. The second phase is a secretory activation phase that occurs a few days after parturition and is characterized by the initiation of copious milk secretions associated with major changes in the concentrations of many milk constituents (Pang and Hartmann, 2007). The process of colostrogenesis englobes the synthesis of milk-specific constituents and the transfer of immunoglobulin (Ig) G into lacteal secretions and therefore occurs after the onset of the differentiation phase and stops before the secretory activation phase. In swine, the first lipid droplets are visible at around 105 d of gestation (Kensinger et al., 1986a) and the synthesis of lactose starts approximately 4 d before parturition (Hartmann et al., 1984).

Hormonal changes in late pregnancy

It is largely recognized that the onset of lactogenesis is under endocrine regulation, with the reproductive hormones estrogens, progesterone, and prolactin, and some metabolic hormones, notably glucocorticoids and insulin, playing key roles (Pang and Hartmann, 2007). In swine, the classic study of DeHoff et al. (1986) described the temporal changes of numerous hormones during mammogenesis and lactogenesis. Total estrogens concentrations start to increase between days 60 and 75 of gestation, reach a peak around days 110 to 112 of gestation and then decrease drastically at farrowing. Kensinger et al. (1986b) further demonstrated that the important increase in metabolic activity of the mammary glands occurring in late gestation is associated with greater estrogen concentrations originating from the fetuses. Mammary DNA was found to be related to circulating concentrations of estrogen in sows on day 110 of gestation. In DeHoff et al.’s (1986) study, progesterone concentrations were highest on day 45 of gestation, declined until day 60, and then remained at approximately 25 ng/mL until the rapid preparturient decline. Concentrations of prolactin remained stable at approximately 15 ng/mL throughout pregnancy until the rapid prepartum rise around day 110 of gestation (DeHoff et al., 1986). This corroborated earlier results showing a prolactin surge associated with parturition in sows about 2 d prepartum through several days postpartum (Dusza and Krzymowska, 1981). Circulating concentrations of somatotropin were also shown to double in the preparturient period and to decrease afterward (DeHoff et al., 1986). Associated with these alterations in endocrine status, the mammary glands undergo major histological changes.

Between days 75 and 112 of gestation, the adipose and stromal tissues are extensively replaced by lobuloalveolar tissue (Hacker and Hill, 1972; Kensinger et al., 1982; Ji et al., 2006). There is also a shift in mammary gland composition going from a high lipid content to a high protein content during the last third of gestation (Ji et al., 2006). Both histological changes and differences in DNA concentrations in mammary tissues from gilts indicate increased epithelial cell division between days 75 and 90 of gestation, with maximum cell concentrations being present by approximately day 90. Then, between days 90 and 105, there is an increase in cellular organelles associated with functional differentiation of the epithelium and abundant accumulation of secretion in the alveoli, indicating the onset of the lactogenic process (Kensinger et al., 1982, 1986a). Prolactin is not only involved in the onset of lactogenesis but also in the development of mammary glands taking place in late pregnancy. Feeding 10 mg of the dopamine agonist bromocriptine thrice daily to gilts from 90 to 109 d of gestation decreased total parenchymal mass by 46% on day 110 of gestation (Farmer and Petitclerc, 2003).

Endocrine status and colostrogenesis

In 1945, Nordskog and Clark (1945) provided the first demonstration that consumption of ergotized barley by late-pregnant sows can lead to agalactia. It was later shown that this effect was likely due to a decrease in prolactin concentrations, as bacterial endotoxins decreased prolactin secretion in postparturient sows and significantly depressed growth of neonatal piglets, being indicative of lower milk production (Smith and Wagner, 1984). This relation between toxins and prolactin was later corroborated by Kopinski et al. (2007) who fed sorghum ergot to prepartum sows and reduced milk production with a concomitant decrease in prolactin concentrations. The essential role of prolactin for the onset of lactation in swine was clearly demonstrated in 1998 when the prepartum peak of prolactin was inhibited with the dopamine agonist bromocriptine and led to agalactia (Farmer et al., 1998). Based on those findings, it can be postulated that the prepartum peak of prolactin is required for colostrum production. This is supported by the observation that the normal swelling of the udder in the week preceding lactation was not present when milk production was inhibited with sorghum ergot (Kopinski et al., 2007).

Other hormones were also shown to be associated with the process of colostrogenesis in swine. A negative correlation was found between circulating concentrations of progesterone around parturition and lactose content in milk or colostrum (Holmes and Hartmann, 1993; Devillers et al., 2004). Piglets from sows that had greater circulating concentrations of progesterone immediately after farrowing also had a reduced growth rate and increased neonatal mortality when compared with piglets from sows having lower progesterone concentrations (de Passillé et al., 1993). Furthermore, a delay in the prepartum decrease in progesterone concentrations and prepartum increase in prolactin concentrations was associated with a low yield of colostrum characterized by leaky mammary epithelium and reduced lactose synthesis in sows (Foisnet et al., 2010a). There are indications that insulin-like growth factor-1 (IGF-1) could also play a role in colostrogenesis, as shown by the positive relation between colostrum yield and plasma concentrations of IGF-1 in high-producing sows before and during parturition (Foisnet et al., 2010a). The specific role of other reproductive hormones, such as estrogens and glucocorticoids, for colostrum production in sows is not known and it is evident that much work remains to be done in order to clarify the impact of hormonal status on colostrum yield and the duration of colostrogenesis.

Metabolic status and colostrogenesis

It is a known fact that during lactation sows become catabolic due to the large amount of nutrients required for milk synthesis. Milk production is a priority for sows to the expense of their body condition and future reproductive performance (Theil et al., 2012). However, very little is known about the impact of colostrogenesis and colostrum yield on the metabolic status of sows. Unlike dairy cows, peripartal sows do not generally develop ketosis. Even though dietary fat source can affect plasma concentrations of ketone bodies, concentrations are much less in sows than in dairy cows (Theil et al., 2013). Considering that the current hyperprolific sows need to produce more colostrum and that special efforts will be made to increase colostral yield, it may be important to study the potential effect of increased colostral yield on metabolic disorders. Nevertheless, taking into account that 1) colostrogenesis only lasts for a short period of time in early lactation, 2) the total amount of colostrum produced is much smaller than that of milk, and 3) body reserves will be affected first and temper any negative effects, it seems unlikely that increased colostrum production could be detrimental to later reproductive performance. On the other hand, if the colostral yield was increased over a number of parities, potential harmful effects may become more evident.

Early Onset of Colostrogenesis

Puberty

Increasing prolactin concentrations in pre-pubertal gilts stimulates mammary development but also induces premature lactogenesis because lacteal secretions are observed. When 2 mg/d of porcine prolactin were injected subcutaneously twice daily to 75 kg gilts for 29 d, lacteal secretions were present in mammary tissue at slaughter at the end of the treatment period (Farmer and Palin, 2005). This corroborated earlier findings of McLaughlin et al. (1997) who used the same treatment and noted that mammary glands of treated gilts were characterized by distended alveolar and ductal lumina, and the presence of secretory material. The impact of such a premature lactogenesis on future colostrum yield is, however, not known. When 2.3 g of the phytoestrogen genistein was fed to growing gilts from 90 until 183 d of age, mammary development was enhanced (as indicated by increased parenchymal DNA) but, in contrast to studies with prolactin, no lacteal secretions were observed (Farmer et al., 2010). Yet, circulating concentrations of estradiol were not affected so one cannot rule out its possible involvement in the early onset of lactogenesis.

Late pregnancy

When intramuscular injections of 15 mg of porcine prolactin were given twice daily to primiparous sows from day 102 of gestation until weaning, lactose concentrations in the plasma of sows were drastically increased on day 105 of gestation (32.4 vs. 6.2 mg/L in treated and control sows, respectively; King et al., 1996). This was associated with an early onset of lactogenesis, as lacteal secretions were observed on day 110 of pregnancy when biopsies were performed. The increased plasma lactose in sows was therefore due to the synthesis of lactose by mammary epithelial cells and leakage into the circulation. The composition of colostrum was also affected, with treated sows having more fat and less protein in their colostrum than control sows. On the other hand, there were no changes in concentrations of DNA or RNA in mammary tissue obtained from biopsies on day 108 of gestation indicating an absence of effect on mammogenesis. Administration of prolactin in late gestation and lactation also reduced litter growth rate in early lactation, namely, from days 0 to 5 of lactation (King et al., 1996), but the specific impact on colostrum yield is not known. It is important to mention, however, that this negative effect on piglet growth was thought to be due to the pharmacological concentrations of prolactin that were achieved (averaging 1745 ng/mL for the 8 h following the morning injection on days 105 of gestation and 7 and 20 of lactation). Indeed, in the study from Crenshaw et al. (1989), daily intramuscular injections of 30 mg of porcine prolactin were given from day 107 of gestation until parturition followed by 60 mg for 2 d postpartum, and circulating prolactin 4 h postinjection increased from 21 to 75 ng/mL on day 110 of gestation and from 98 to 167 ng/mL on day 1 of lactation with no effect on piglet growth. Accordingly, when hyperprolactinemia was brought about using the dopamine antagonist domperidone from 90 to 110 d of gestation, circulating concentrations of prolactin in gilts increased from approximately 20 to 70 ng/mL and growth rate of piglets was not hindered. In fact, piglets grew faster due to enhanced milk yield linked with greater alveolar volume in mammary tissue. Beneficial effects on the secretory activity of mammary tissue and on mammary epithelial cell differentiation were seen, but the specific impact on colostrum production was not studied (VanKlompenberg et al., 2013). Recent results demonstrated that the treatment of gilts with domperidone during that same late gestation period does lead to early onset of lactogenesis even though prolactin concentrations are not at pharmacological levels (Caron et al., 2019). Yet, piglet growth could not be measured in that latter study because gilts were slaughtered at the end of gestation.

Somatotropin is another hormone that can affect lactation in mammals but its lactogenic role in swine is not clear. An early study reported a 15% to 22% increase in milk yield (Harkins et al., 1989); however, those results could not be reproduced in later studies (Cromwell et al., 1992; Toner et al., 1996). Recent findings show that mammary development of late-pregnant gilts is greatly stimulated by a 4-fold increase in concentrations of IGF-1 without any effect on lactogenesis. Gilts were injected daily with 5 mg of porcine somatotropin (Reporcin) from days 90 to 109 of gestation and upon slaughter, on day 110, no lacteal secretions were present in mammary tissue (Farmer and Langendijk, 2019).

Farrowing induction

When farrowings were induced on day 113 of gestation with a prostaglandin analog (i.e., one day before the average date of parturition), colostrum yield was not affected even though transient increases in prolactin and cortisol concentrations were observed 24 h prepartum (Foisnet et al., 2011). On the other hand, induced sows had greater lactose concentrations in colostrum at the onset of farrowing compared with control sows, which may indicate an earlier onset of colostrogenesis. It is important to note, however, that permeability of the mammary epithelium was not affected by treatment and that average daily gain of piglets between days 1 and 21 of lactation was lower in induced sows, most likely due to a smaller piglet body weight on day 1 (Foisnet et al., 2011). In another study where prostaglandins were used alone or in combination (24 h later) with an oxytocin-like molecule (carbetocin) to induce farrowings at day 114 of gestation, colostrum onset was not studied per se but colostrum yield was reduced when both prostaglandins and carbetocin were used (Boonraungrod et al., 2018). This was thought to be due to possible prolonged uterine contractions that could have adversely affected newborn piglets. Hence, providing oxytocin to prepartum sows is not recommended in order to achieve optimal piglet performance.

Delay of Colostrogenesis

Use of an oral progestogen

The progestogen altrenogest (sold under various brand names including Regumate and Matrix) was given orally to sows (20 mg/d) from days 109 to 112 or 113 of gestation to see the potential effect of increased progesterone on lactogenesis (Foisnet et al., 2010b). Interestingly, treatment did not affect the onset of lactogenesis because the delay in farrowings also delayed the onset of the major hormonal changes seen at parturition without affecting their relative timing. Accordingly, the colostrum yield was not reduced. Altrenogest did not maintain the mammary tight junctions open for a longer period of time after farrowing and this may be due to a lack of full biological action of altrenogest on mammary tissue compared with endogenous progesterone (Foisnet et al., 2010b). There was also a trend for a reduced transfer of IgG into colostrum when altrenogest was given until day 113 of gestation instead of day 112. This was thought to be due to reduced concentrations of estradiol.

Early Cessation of Colostrogenesis

The period of colostrogenesis is generally recognized as lasting until approximately 24 h following the onset of farrowing, as indicated by the changes in the composition of colostrum to become transient milk between 24 and 36 h postpartum (Theil et al., 2014). Recent data using transcriptional profiling of swine mammary glands showed differential expression of a large number of genes involved in lactogenesis close to and at farrowing, with the expression of genes associated with the synthesis of milk components (i.e., protein, fat, and lactose) being increased (Palombo et al., 2018). At 1 d postpartum, there were marked upregulations of genes involved in the synthesis of casein and whey acidic protein as well as the uptake of leucine by mammary glands, indicating the cessation of colostrogenesis.

The potential relationships of sow and litter characteristics with the cessation of colostrogenesis and onset of copious milk secretion were investigated (Vadmand et al., 2015). A total of 79 independent variables were studied. They included traits related with sow nutrition, litter characteristics, farrowing characteristics, and composition of lacteal secretions. The onset of copious milk secretion was quite variable among sows but it occurred at approximately 31 h postpartum and it was only affected by one of the studied variables, namely, litter standardization. The onset of copious milk secretion tended to happen earlier (by approximately 2 h) in litters that received cross-fostered piglets in order to standardize litter size compared with those that did not. It was suggested that the faster onset of lactation may have been due to more intense suckling from more vigorous piglets (Vadmand et al., 2015). It is of interest to note that the late onset of lactation positively affected the amount of transient milk produced but negatively affected the average milk yield during lactation. Time of onset of lactation appeared to have an impact on the shape of the lactation curve (Vadmand et al., 2015).

In a study looking at the effects of conjugated linoleic acid (CLA) on yield and composition of colostrum and timing for initiation of milk production, Krogh et al. (2012) noted that sows fed diets supplemented with CLA from day 108 of gestation until weaning produced less colostrum that tended to contain more fat. The amount of total solids transferred to the piglets was decreased so that neonatal piglet mortality increased (Krogh et al., 2012). The onset of milk production was not altered, remaining at 34 h after the birth of the first piglet in both groups, but milk yield was increased when feeding CLA. The authors did not recommend such treatment due to the negative impact on colostrum yield and piglet performance, and results further emphasized the importance of colostrum intake for newborn piglets. In fact, it is known that colostrum must be available as an energy source for at least 17 to 18 h after farrowing because of the poor glycogen depots of newborn piglets (Theil et al., 2011). In terms of piglet survival, it would therefore be more advantageous to try and prolong the colostral phase.

Prolongation of Colostrogenesis

The status of tight junctions between mammary epithelial cells is the major effector of changes in composition from colostrum to transient milk because it affects the passage of immunoglobulins and bioactive compounds from the dam circulation to lacteal secretions. After farrowing, as colostrogenesis progresses, mammary tight junctions become less permeable until they are closed and colostrum production is terminated while the amount of milk secreted increases drastically. The leakiness of mammary tight junctions is reflected by the Na/K ratio in milk, being more permeable when that ratio is high (Foisnet et al., 2010a). Manipulation of the sow endocrine status in the postpartum period can be used to alter the permeability of mammary tight junctions and affect the timing of the cessation of colostrogenesis.

Exogenous oxytocin

High doses of oxytocin were shown to alter the permeability of mammary tight junctions in goats (Linzell and Peaker, 1974), rodents (Nguyen and Neville, 1998), cattle (Jonsson et al., 2013; Herve et al., 2018), and swine (Farmer et al., 2017). In that last study, post-parturient sows were injected twice daily with 75 IU of oxytocin, starting 16.0 ± 2.4 h after the birth of the last piglet, to total four injections. Within 8 h of the first injection, changes were already present in the composition of lacteal secretions of treated vs. control sows. Milk from treated sows contained 29% more protein, 51% more IgA, 184% more IgG, 110% more IGF-1, 17% more energy, 14% more solids, and had a 71% greater Na/K ratio than that from control sows (IgG, IGF-1, and Na/K shown in Figure 1). It is not known if the increased energy content was mainly due to the greater immunoglobulin concentrations. The greater Na/K ratio was indicative of increased permeability of mammary tight junctions. Furthermore, treated sows were injected with ovalbumin emulsified in incomplete Freund’s adjuvant on day 80 of gestation and also received a booster shot on day 95 of gestation. This was performed to measure the transfer of this specific antibody into lacteal secretions. In the milk sample taken 8 h after the first oxytocin injection, antibody titers against ovalbumin were increased by 264% in the milk of treated sows compared to that of control sows (Figure 1). Taken together, these results clearly show a prolongation of the passive transfer of immunoglobulins, growth factors, and hormones from the sow’s circulation to the milk with only one injection of a supraphysiological dose of oxytocin in the early postpartum period.

Figure 1.

Changes in milk composition in sows that received four injections of saline (CTL, slashed line, n = 10) or 75 IU of oxytocin (OXY, full line, n =10) in early lactation. Injections were given twice daily (8:00 a.m. and 4:30 p.m.) starting on day 2 of lactation, i.e., between 12 and 20 h after the birth of the last piglet (adapted from Farmer et al., 2017).

Interestingly, there was no effect of the last three injections of oxytocin on mammary tight junctions, as suggested by similar Na/K ratios in milk. It is therefore apparent that there exists a specific time frame during which hormonal manipulations can be used to prolong the duration of colostrogenesis. Even though there was a tendency for piglets from treated dams to have a lower incidence of mortality in that study (Farmer et al., 2017), the sample size was small (n = 10) so that the potential beneficial effect of such a treatment on piglet survival needs to be further demonstrated on a larger scale.

Summary and Conclusions

The amount of colostrum ingested by newborn piglets is determinant for their survival and growth. It has long been recognized that many factors can influence the availability of colostrum to piglets (Fraser, 1984) but one aspect that has been overlooked is the duration of colostrogenesis. It is apparent that hormonal manipulations can impact the onset and cessation of the process of colostrogenesis. As early as puberty, mammary cells can start to produce lacteal secretions when stimulated with prolactin. However, the impact of such a premature lactogenesis on future colostrum yield is not known. Increasing prolactin concentrations in late pregnancy also induces early lactogenesis and the effect on piglet growth seems to be related to the concentrations of prolactin achieved, growth rate being inhibited when prolactin concentrations were at pharmacological levels. The duration of colostrogenesis can be altered via exogenous oxytocin. Injecting a supraphysiological dose of oxytocin (75 IU) 16 h after the birth of the last piglet alters the permeability of mammary tight junctions, as indicated by a greater Na/K ratio in milk, thereby delaying the occurrence of tightening of these tight junctions and leading to further passive transfer of immunoglobulins and bioactive components from the dam’s circulation to the milk. This prolongation of the colostral phase could be advantageous for the survival of newborn piglets because they are born immunodeficient and with very poor energy reserves.

Glossary

Abbreviations

CLA

conjugated linoleic acid

IGF-1

insulin-like growth factor-1

Conflict of interest statement

The authors declare no real or perceived conflicts of interest.