Mammary development during lactation

Mammary gland development does not stop at the end of gestation but it continues during lactation. Average mammary weight of suckled glands increases linearly (57% increase) from day 5 until day 21 of lactation (Kim et al., Reference Kim, Hurley, Han and Easter1999). The increase in mammary gland weight during first lactation is due to both cellular hyperplasia and hypertrophy. Mammary gland growth during lactation is greater in the five most anterior glands compared with the posterior glands (Kim et al., Reference Kim, Hurley, Han and Easter2000), and there are indications that it may be positively related to the intensity of the post-ejection massage (Thodberg and Sorensen, Reference Thodberg and Sorensen2006). Mammary development during lactation is also affected by parity because mammary gland wet weight increased by 70%, 20% and 30% between day 113 of gestation and day 26 of lactation, for sows of parity 1, 2, and 3, respectively (Beyer et al., Reference Beyer, Jentsch, Hoffmann, Schiemann and Klein1994).

Hurley (Reference Hurley2001) stated that suckling intensity, therefore extent of milk removal, is the major stimulator of mammary gland growth during lactation. The importance of suckling intensity on a mammary gland on its development during lactation was suggested by a positive correlation between the size of a mammary gland (either in terms of weight or DNA) and the growth of the piglet suckling that gland (Kim et al., Reference Kim, Hurley, Han and Easter2000; Nielsen et al., Reference Nielsen, Pederson and Sorensen2001). Furthermore, mammary parenchymal tissue of sows that produce more milk (i.e. 5.25 v. 4.46 kg average piglet weight gain from days 2 to 21 of lactation) was shown to contain more DNA and more RNA per gland at the end of lactation than that of sows with a lower milk yield (Farmer et al., Reference Farmer, Palin and Hovey2010). However, it is not known whether mammary development is solely influenced by suckling intensity or if piglets suckling larger mammary glands have a greater body weight gain.

Mammary involution at weaning

When the piglets are weaned, there is an abrupt cessation of milk removal that leads to involution of the mammary glands. This process of involution is characterized by rapid regression of the mammary parenchyma during the first 7 days post-weaning (Ford et al., Reference Ford, Kim, Rodriguez-Zas and Hurley2003). Changes in mammary tissue are quite dramatic. Wet weight of parenchymal tissue was found to decrease by 68.8% in those first 7 days and parenchymal DNA decreased by 66.8%. During the first 2 days post-weaning, the mammary gland undergoes dramatic decreases in cross-sectional area, wet weight per gland and parenchymal DNA. From days 2 to 4, regression of the mammary gland is minimal followed by additional significant declines until day 7 post-weaning (Ford et al., Reference Ford, Kim, Rodriguez-Zas and Hurley2003). The proportion of DNA per tissue mass was not altered after weaning leading these authors to suggest that the number of cells per milligram of wet tissue did not change during the involution process. However, other compositional changes were observed in mammary parenchymal tissue after weaning, namely, a reduction in protein percentage and an increase in fat percentage. This suggests that involution of mammary glands led to an increase of lipid within the tissue, which could very well reflect a transitory accumulation of milk lipid (Ford et al., Reference Ford, Kim, Rodriguez-Zas and Hurley2003). An earlier histological study also showed that weaned mammary glands become engorged in the first few days after weaning and that milk in the lumen then seems to be reabsorbed (Cross et al., Reference Cross, Goodwin and Silver1958).

Ford et al. (Reference Ford, Kim, Rodriguez-Zas and Hurley2003) suggested that mammary gland involution in weaned sows is achieved in three phases, namely (1) from weaning until day 2, (2) from day 2 until day 4 or 5 post-weaning, and (3) from day 4 or 5 until day 7 post-weaning. Succinctly, in the initial phase, the sudden absence of milk removal leads to milk stasis and an inhibition of milk secretion via accumulation of an autocrine feedback inhibitor of lactation in the alveolar lumen (Wilde et al., Reference Wilde, Addey, Boddy and Peaker1995). A drastic loss of tissue fluid occurs during that phase alongside with an increase in tissue fat content and a considerable cell loss due to apoptosis. The second phase is characterized by more limited changes in mammary gland component mass that is consistent with the limited alterations in milk metabolites (Atwood and Hartmann, Reference Atwood and Hartmann1995). In the last phase of involution, only limited mammary secretions can be collected and they are very viscous (Atwood and Hartmann, Reference Atwood and Hartmann1995). There are very few alveolar structures remaining at this stage (Cross et al., Reference Cross, Goodwin and Silver1958) and there is a final reduction in mammary parenchymal tissue and DNA (Ford et al., Reference Ford, Kim, Rodriguez-Zas and Hurley2003). The process of mammary involution is affected by the duration of lactation. When comparing lactation lengths of 22 and 44 days, there was a greater Na/K ratio in milk at 44 days indicating that mammary epithelial cell tight junctions became leaky as lactation advanced (Farmer et al., Reference Farmer, Knight and Flint2007). The disruption in mammary epithelial integrity coupled with an increase in plasma lactose concentrations suggests that the process of involution started before weaning on day 44 of lactation. The onset of involution could be related to a reduced suckling intensity because piglets that were weaned at 44 days had access to creep feed as of day 22 of lactation.

Mammary involution during lactation

Suckling of a mammary gland is essential for its functionality to be maintained. Indeed, unsuckled mammary glands in early lactation regress at a similar rate and follow a similar pattern of regression than mammary glands that regress post-weaning (Kim et al., Reference Kim, Easter and Hurley2001). Mammary gland tissue weight decreased by 2/3 within the first 7 to 10 days of lactation when the gland was not suckled whereas the rate of regression was much slower thereafter (Ford et al., Reference Ford, Kim, Rodriguez-Zas and Hurley2003). In accordance, mammary glands that are not suckled during lactation do not show further loss of parenchymal tissue after weaning (Ford et al., Reference Ford, Kim, Rodriguez-Zas and Hurley2003). When comparing litter sizes of six to 12 piglets, Kim et al. (Reference Kim, Easter and Hurley2001) reported no differences in composition or size of unsuckled mammary glands at day 21 of lactation. This does not rule out the possible effect of litter size on regression rate of unsuckled glands earlier in lactation, however, it is clear that no such effect is present on day 21. On the other hand, one must consider that some glands can be ‘partly used’ in the case of smaller litters where a piglet will suckle 2 glands, one likely being a secondary gland that is only partially emptied (Illmann et al., Reference Illmann, Pokorna and Spinka2007). In such occurrences, it is not known what the impact of a ‘partly used’ gland will be on its regression rate during and after lactation. The rate of regression of unsuckled mammary glands is affected by dietary nutrient level during lactation. Kim et al. (Reference Kim, Easter and Hurley2001) reported that wet weight of unsuckled glands was 91% greater on day 5 of lactation in sows fed a high energy (17.5 v. 12 Mcal ME/day) high-protein (65 v. 32 g of lysine/day) diet compared with sows fed a low energy-low protein diet. These data indicate that the rate of regression of unsuckled glands is slowest under dietary conditions that promote mammary growth. Nevertheless, the impact of the extent and rapidity of regression of unsuckled mammary glands during lactation on their future development and milk yield is not known.

The degree to which mammary gland involution is reversible is particularly important in swine because of the common practice of cross-fostering piglets. Results from Kim et al. (Reference Kim, Easter and Hurley2001) showing extensive loss of wet weight and DNA in unsuckled glands suggest that the loss of lactation function would not be reversible after the initial several days of lactation. In 2005, Theil et al. performed a study to specifically address that question. In their experiment, teats were blinded for either 24 or 72 h post-farrowing to prevent suckling by piglets. After removing the tape, glands blinded for 24 h were suckled until weaning, whereas glands blinded for 72 h were refused by piglets and remained unsuckled. Mammary development (estimated by visual scoring) throughout lactation of glands blinded for 24 h was similar to that of regularly suckled glands, whereas that of teats blinded for 72 h was lesser as of day 6 of lactation. Regression of unsuckled mammary glands during early lactation was found to be reversible (gland rescue) during the first 24 h but to be irreversible after 3 days (Theil et al., Reference Theil, Labouriau, Sejrsen, Thomsen and Sorensen2005). Milk stasis in rescued glands (blinded for 24 h) and regressing glands (blinded for 72 h) led to lower percentages of cellular proliferation, lower expression of α-lactalbumin and prolactin receptor mRNA, and greater expression of insulin-like growth factor binding protein-5 (IGFBP-5) mRNA. This corroborates the fact that milk production from the rescued glands that were unsuckled during 24 h remained lower throughout lactation. The authors mentioned a potential role of suckling intensity for this lower milk yield because, even though body weight of piglets from all groups were similar at the onset of treatment, piglets suckling rescued glands were lighter in weight than control piglets on days 4 and 6 of lactation. Nevertheless, regardless of the cause, it is apparent that piglets cross-fostered later than 24 h after the foster sow farrowed weigh 900 g less at weaning than control piglets (Thorup, Reference Thorup1998).

In a later trial, Theil et al. (Reference Theil, Sejrsen, Hurley, Labouriau, Thomsen and Sorensen2006) showed the importance of suckling intensity for mammary regression. They compared no suckling, transient suckling (until 12 to 14 h postpartum) or regular suckling of mammary glands and reported that, based on visual scoring, regularly suckled glands maintained lactation whereas transiently suckled and non-suckled glands regressed during lactation. Mammary biopsies were obtained from 5 days before until 6 days after farrowing to evaluate cell turnover and to determine expression of specific genes. Cell proliferation was greatest prepartum, declined after farrowing and then either remained low (in unsuckled glands) or increased (in regularly and transiently suckled glands) between days 1 and 6 postpartum. Suckling during the first 12 to 14 h postpartum was insufficient to initiate and maintain lactation until 24 to 36 h postpartum but could induce mammary cell proliferation for at least 6 days postpartum. Alterations in gene transcription were also recorded with prolactin receptor mRNA abundance decreasing and that of IGFBP-5 increasing after farrowing in unsuckled and transiently suckled glands compared with regularly suckled glands. The authors stated that a high rate of prolactin receptor transcription and a low rate of IGFBP-5 transcription seem important for maintaining mammary glands functional during lactation.

Productivity and development

The first suggestion that suckling of a mammary gland in one lactation may have an impact on its productivity in the subsequent lactation was made by Fraser et al. (Reference Fraser, Thompson and Rushen1992). Indeed, these authors taped the rear teats of sows in first lactation and showed that, in second lactation, this accentuated the difference in body weight gain between piglets suckling the front teats, which are known to be heavier (Dyck et al., Reference Dyck, Swierstra, McKay and Mount1987), than piglets suckling the rear teats. Even though interesting, these results were somewhat biased by a confounding effect of treatment with teat location. This idea was brought up again by Ford et al. (Reference Ford, Kim, Rodriguez-Zas and Hurley2003) who noted that mammary glands that were suckled during lactation were larger at the end of the lactation process than non-suckled glands. This is because of the fact that glands which are suckled during lactation undergo substantial growth (approximately doubling in size) so that they start the post-weaning involution process at a larger size. These authors therefore again suggested a possible beneficial effect of teat use on redevelopment, and possibly milk yield, during the next gestation. This is particularly important in the current context of swine husbandry where hyperprolific sow lines are being used and producers need to decide on optimal litter size of their first-parity sows to ensure proper body condition and reproductive performance over the next parities.

The answer to that question was recently provided by Farmer et al. (Reference Farmer, Palin, Theil, Sorensen and Devillers2012) who demonstrated that non-suckling of a mammary gland in first parity impairs its development and milk yield in second parity. In that study, either the same teats or different teats were blinded with tape during the first and second lactation so that in the second lactation, piglets had access to teats that were either suckled or unsuckled previously. Differences between those teats were present both in terms of animal performance and in variables measured at the mammary gland level. With regard to piglet growth, piglets suckling previously used teats weighed 1.12 kg more at 56 days of age than piglets suckling previously unused teats. Interestingly, a difference in growth rate of piglets was seen as early as between days 2 and 4 of lactation, suggesting a potential beneficial effect on colostrum yield. Fraser et al. (Reference Fraser, Thompson and Rushen1992) also observed that, in second lactation, differences in body weight gain between piglets suckling teats that were used or not in the first lactation were most pronounced during the first week of lactation compared with the second or third week. It therefore seems that suckling of a teat in the previous lactation could have an impact on colostrum production, and this merits further investigation.

The composition of mammary glands on day 17 of lactation was also affected by use or non-use of a teat in the previous lactation. The composition of mammary glands suckled during the previous lactation reflected more parenchymal tissue per gland and greater parenchymal DNA and RNA content per gland on day 17 of the second lactation (Farmer et al., Reference Farmer, Palin, Theil, Sorensen and Devillers2012). Furthermore, abundance of mRNA for the prolactin receptor was also reduced in glands which were previously unsuckled. Results, therefore, indicate the occurrence of both hyperplasia (as indicated by greater parenchymal DNA) and increased metabolic activity (as indicated by greater parenchymal RNA) of parenchymal cells from glands that were previously used. It is not known, however, whether the observed carry-over effects from the first to the second lactation are solely because of differences in mammary cell proliferation, activity and mRNA abundance of specific genes, or if other factors such as increased vascularization could also come into play.

Nursing behaviour of piglets

Another interesting and novel aspect of the project by Farmer et al. (Reference Farmer, Palin, Theil, Sorensen and Devillers2012) was the finding that teat use or non-use in parity one affected the behaviour of suckling piglets during the next parity. On day 3 of lactation, piglets suckling teats that were not used previously showed increased aggressive behaviour, longer post-ejection phase, and a greater incidence of missed nursings than piglets suckling previously used teats (Farmer et al., Reference Farmer, Palin, Theil, Sorensen and Devillers2012). Such behaviours are indicative of piglets being more eager to obtain milk, a likely reflection that they are hungrier. Effects of previous teat use on behaviour of piglets were not seen later in lactation (i.e. day 10). Surprisingly, piglets in parity two could differentiate between teats that were suckled or not in parity one. Indeed, when some teats were blocked in lactation one but none of the teats were blocked in lactation two, piglets fought more often and for a longer period of time to get access to teats which were previously suckled.

It is, therefore, now evident that non-suckling of a mammary gland in first lactation will impair its productivity in the next lactation. Yet, numerous questions still remain. First, would the same effect be seen between the second and third lactation or is this only the case for primiparous sows? Second, would suckling of a teat for a short period of time in early lactation be enough to allow optimal development and productivity in the next lactation and, if that is the case, what is the minimum duration of suckling required? Third, what are the mechanisms of action of this effect?