In animals and humans, a number of studies have indicated that dietary composition plays an important role in the immune system [31, 32]. Increased attention is now being focused on the effects of nutritional factors during pregnancy on immune competence and health of the offspring. The present study demonstrates for the first time that both low and high protein:carbohydrate ratios in the diets fed to primiparous sows throughout pregnancy affect baseline immune parameters in neonatal offspring and modify immune responses to challenges later in life.
In late pregnancy, the LP diet caused lower protein concentrations in serum compared with AP and HP sows, an effect which is manifested by protein malnutrition [33, 34]. As also shown in our investigation, concentrations of immunoglobulins are often unaffected by this type of malnutrition . Moreover, the changes in IgG, IgM and IgA serum levels we observed in sows at different stages of pregnancy had similar patterns to those previously described by Klobasa et al. .
Because of the structure of the epitheliochorial porcine placenta, neonatal piglets acquire maternal immunity soon after birth by absorbing intact immunoglobulins from ingested colostrum for humoral immune protection. The amount of immunoglobulin uptake by the suckling piglet mainly depends on the concentration of immunoglobulins in colostrum, the intestinal capacity to absorb macromolecules prior to gut closure, the time to first suckle, and the colostrum intake [36, 37]. In the present study, piglets born to sows receiving limited or excess dietary protein during pregnancy displayed lower serum immunoglobulin levels on the first day of life compared with those from mothers fed an adequate protein diet. Indeed, we did not find significant differences in colostral immunoglobulin concentrations between the sows fed different dietary protein levels . The majority of prenatal stress research indicates that physiological alterations in prenatally stressed offspring can be attributed to the action of maternal glucocorticoids [2, 38]. Previous studies in pigs have shown that prenatal stress during the late phase of gestation also decreased the circulating IgG levels in neonates [8, 39]. An increase in maternal cortisol during late gestation may accelerate foetal gut maturation and thereby impair the acquisition of colostral immunoglobulins after birth . Interestingly, although cortisol levels were significantly increased only in LP sows at the end of pregnancy, we found strong inverse relationships between the plasma cortisol levels in both LP and HP sows on D108 of gestation and the serum levels of total protein and IgA in their offspring on the first day of life. Thus, an altered permeability of the gut in neonates of sows fed diets with inadequate protein levels could explain the decreased immunoglobulin concentrations in the piglets. In a previous experiment using the same dietary treatments as in this study, we found that both LP and HP diets alter materno-foetal HPA regulation . Thus, it appears that maternal glucocorticoids are involved in mediating the effects of imbalanced nutrition during pregnancy in pigs on the immunity of offspring. Furthermore, a suboptimal intake of colostrum may result in an inadequate transfer of maternal immunoglobulins to the newborn and thereby also contribute to pre-weaning mortality . Although piglets of the LP and HP groups showed lower serum immunoglobulin levels in the early postnatal period, we detected a higher mortality rate only in the LP group during the suckling period. As discussed by Le Dividich et al. , reduced immunoglobulin concentrations do not always lead to an increased risk of mortality if piglets receive an adequate intake of colostrum. While early postnatal vitality was not measured in our study, LP piglets exhibited the lowest muscle mass at birth , and we found that their early weight gain was significantly lower than that of AP and HP piglets. From these results, we hypothesise that LP piglets may have consumed less colostrum over the first hours after birth, which, together with their low birth weight, decreased vitality and increased pre-weaning mortality risk. However, besides immunoglobulins, it is known that several other components of colostrum and milk such as hormones, growth factors and cytokines may play an important role in the physiological development and immune protection of neonatal piglets [36, 43, 44]. A major finding from the present study is that both limited and excess protein in the maternal diet during pregnancy caused reduced humoral immunity in neonatal piglets that was not evident in older pigs. Indeed, these transient early modifications could also alter the immune-reactivity towards infections later in life .
Previous studies in pigs have reported that gestational stress in sows may activate the HPA axis in offspring during novel or challenging situations [10, 11] and may also affect the immunity of neonatal pigs in response to stressful stimuli [8, 46]. In our study, piglets from sows fed diets with limited protein levels during pregnancy showed increased basal cortisol levels one day after weaning, whereas cortisol levels in HP piglets were only slightly increased or unaltered in AP piglets. After LPS challenge, there was no effect of inadequate dietary protein to carbohydrate ratios on cortisol release. This finding is in accordance with the results of de Groot et al.  and Couret et al. , where prenatal stress also did not affect cortisol levels of piglets in response to LPS. The differences in cortisol release after weaning and LPS administration in the present study may be due to the different nature of the stressors. It seems that the HPA axis of LP piglets is more sensitive to weaning stress with abrupt social, nutritional, and environmental changes, whereas cortisol release after LPS challenge reflects an immunological stress. Therefore, we conclude that elevated cortisol concentrations in LP sows at the end of pregnancy can affect HPA activity in offspring, especially in response to a multifactorial stress such as weaning.
Our results revealed a significant increase in ConA-stimulated cell proliferation after weaning in piglets of AP sows that may indicate a more effective cellular immune response [47, 48]. Together with the unaltered cortisol level, the higher in vitro lymphocyte function in AP piglets may be considered as a greater ability to cope with weaning stress . Furthermore, the percentage of CD4+ cells and thereby the CD4+/CD8+ ratios were significantly increased after the stress of weaning in HP piglets only. These findings are comparable with frequently reported changes in lymphocyte subpopulations after sustained stress. A long-term dexamethasone treatment of weanling piglets resulted in a higher percentage of CD4+ cells , and chronic stress in sheep and humans has been shown to increase the CD4+/CD8+ ratio in peripheral blood [51–53]. The rise in CD4+/CD8+ ratio may reflect recovery of immune competence in an attempt to restore homeostasis following a stressful event. Although some caution must be exercised in interpreting our results, the changes in distribution of CD4+ and CD8+ cells in HP piglets could indicate the extent of T-lymphocyte recovery due to the severity of weaning stress perceived by these piglets.
In the present study, pigs of all diet groups responded to intraperitoneal LPS administration with elevations in cortisol and cytokine levels in similar temporal patterns that were appropriate for acute gram-negative infection [54, 55]. However, the levels of circulating IL-6 were higher in LP and HP piglets compared to AP piglets, and LP piglets additionally showed a higher increase in IL-10. From previous research in pigs it is known that prenatal stress may also alter physiological responsiveness of offspring to an immune stimulation with LPS. De Groot et al.  suggested that the non-specific inflammatory response to an LPS challenge increased in piglets from sows treated with cortisol during pregnancy, and it was recently shown that maternal restraint stress during gestation in pigs enhanced the magnitude of the TNF-α and IL-6 responses to LPS in the offspring . In our study, the increased inflammatory cytokine responses in piglets exposed to inadequate maternal protein-to-carbohydrate levels in utero confirm these results and support the hypothesis that maternal stress or malnutrition during pregnancy can influence aspects of immune responses in the offspring to challenges.
In humans, elevated levels of IL-6, a predominantly pro-inflammatory cytokine, have been identified as a marker of disease risk and dysregulated inflammation [56, 57]. Focussing on the highly increased IL-6 levels in LP and HP offspring 3 h after LPS injection, these piglets appear to be more susceptible to bacterial endotoxin exposure with an increased risk of an acute systemic inflammatory response syndrome . This interpretation may be supported by the magnitude of TNF-α response observed in these pigs 1 h after LPS. Moreover, the higher release of peripheral IL-10 in LP and HP pigs at the same time as IL-6 may emphasise the crucial role of IL-10 in regulating inflammatory responses [59, 60]. Although the LPS challenge is a simplified inflammation model, the present results suggest that such alterations in cytokine release can be disadvantageous for recovery from frequently occurring bacterial infections of the gastrointestinal and respiratory system in pigs.