Lower dietary phosphorus supply in pigs match both animal welfare aspects and resource efficiency

Reaching sustainability will require tracking the fate of P and to understand how we can decrease P losses from the agri-environment system and tighten the resource cycle. Indeed, diet and animals are considered the most important determining factors driving P efficiency. The P demand should match genetic and physiological requirements to avoid P excess in manure. In this context, the broad biodiversity of monogastric P utilisation has been described (Hittmeier et al. 2006; Alexander et al. 2008).

The higher calcium and P supply in H animals prompted a lower feed intake, lower body weight gain, and higher FCR. Consequently, the study pointed out molecular routes that are responsive to such dietary challenges and, therefore, represented molecules and genes related to P efficiency. In this study, calcitriol and PTH levels but not serum calcium and P levels were affected by diet. Indeed, endocrine factors such as calcitriol and PTH are known to coordinate serum calcium and P levels (Dusso et al. 2005; Talmage and Mobley 2008). Indeed, calcitriol and PTH were negatively correlated (Fig. 2) to balance enteral absorption, osseous mobilisation, and renal excretion rates as previously reviewed (Berndt and Kumar 2009). Obviously, the endocrine response to diets varying in calcium and P levels enabled serum mineral homoeostasis. According to previous studies (Engstrom et al. 1985; Sommerville et al. 1985; Riond et al. 2001; Oster et al. 2016), lowered PTH levels but increased calcitriol levels reflect the organismal effort to minimise urinary calcium and P losses and to enhance enteral calcium and P absorption in L samples. In contrast, H samples aimed to maximise renal calcium and P losses via lower calcitriol and higher PTH levels.

To regulate mineral homeostasis within the intra- and extracellular fluid, maturation and development provoke dynamic influxes and effluxes of calcium and P which is largely buffered by the osseous storage. Calcitriol and PTH also have an impact on bone tissue via control of diet-specific osteoblast- and osteoclast-mediated actions. It became obvious that calcitriol and PTH responses were sufficient to maintain physiological calcium and P serum concentrations by recruiting the bone mineral storage. Hence, results suggest an altered bone mineralisation when comparing L and H animals which is in agreement with previous studies (Ryan et al. 2011; Varley et al. 2011). Moreover, the skeleton is an important target tissue of thyroid hormones such as T₃, which controls bone turn-over and maintenance throughout life (Williams 2013; Bassett and Williams 2016). In our study, the reduced T₃ levels in H animals compared to L animals reflect the observed decline in feed intake (Wadden et al. 1990) and may account for impaired bone resorption and formation phases via reduced osteoblast differentiation and function. In fact, reduced T₃ levels have been associated with increased bone mineralisation and higher risk of fracture (Vestergaard et al. 2005; Tuchendler and Bolanowski 2014). Interestingly, the decreased endocrine T₃ levels observed in serum of H animals were accompanied by increased abundances of THRA encoding the thyroid receptor alpha. Since T₃ action relies on both hormone availability and mRNA copy number, results might suggest adaptation processes at the receptor level (Kenessey and Ojamaa 2004) to match, e.g. intestinal demands modulating cell proliferation rates (Plateroti et al. 2006).

However, since T₃ impacts on growth via anabolic and catabolic processes, the reduced T₃ levels reflect the lower body weight and BW gain as shown by correlation analyses (Fig. 2). It will be of great scientific interest to find out which molecular mechanisms are able to sense the high calcium and P content in the diet which lead to a lower feed intake but increased FCR. Indeed, it may be conceivable that differences for feed intake might have affected endocrine responses such as for T₃. Interestingly, the low P supply did not affect growth and feed efficiency; it even tended to improve performance, while mineral homeostasis was maintained. The animals were able to cope with lowered P supply at least over the period of time tested here.

The endocrine responses to both L and H diets indicate the interplay between intestine, bone, and kidney. Whereas mRNA abundances of PTH1R were unaltered in all analysed tissues, the higher abundances of Cyp27A1, and Cyp27B1 in kidney mirrored the increased calcitriol levels on L diet. These results are in accordance to previous reports when P-restricted diets have been applied to mice (Zhang et al. 2002). Additionally, Cyp27B1 expression was affected in jejunum and colon tissues, although the mRNA copy number was rather low. However, the tissue-specific increased abundances of Cyp27B1 in H animals might account for a local calcitriol synthesis in the intestine required for, e.g. immunological aspects (Dusso et al. 2005; Liu et al. 2006). In this context, local requirements for calcitriol might be balanced via significantly different Cyp24A1 expression in duodenum and jejunum as the encoded 24-hydroxylase catalyses the first step in the deactivation of calcitriol (Sakaki et al. 2005). Indeed, the dietary challenge in this study revealed a strong transcriptional response of Cyp24A1 at local tissue sites. Moreover, VDR was diet-specifically expressed in kidney but not in intestine. The higher renal abundances of VDR might reflect a compensatory regulation of L animals to achieve a mineral balance. However, VDR is known to initiate various effects in different tissues including intestine and kidney. Specifically, effects were mediated via the VDR–RXR receptor heterodimer binding to vitamin D response elements which are detected in the promoter region of a broad range of genes (Haussler et al. 2013). Despite unaffected VDR in intestine tissues, however, it is conceivable that dietary effects are mediated since calcitriol serum levels were very high in L animals.

Regarding the P transporter represented by the SLC34 family, intestinal SLC34A1 and SLC34A2 were at the lower detection limit, whereas renal SLC34A1 was highly abundant. In contrast, SLC34A2 was highlighted as relevant P transporter in the posterior parts of the small intestine in mice (Radanovic et al. 2005). In particular, it has been proposed that SLC34A2 is responsible for transcellular P uptake in jejunum and its mRNA expression is upregulated by decreased P levels in chickens (Li et al. 2012) and rats (Cao et al. 2016). Obviously, this does not reflect the porcine responses revealed in this study. In our study, H animals showed lower abundances of SLC34A3 in duodenum and jejunum but higher abundances in kidney when compared with L animals. This pattern might follow superior endocrine responses to minimise the P influx via intestine and primary urine. Correspondingly, it has been shown that SLC34A3 is specifically regulated in response to high P diets in rats (Segawa et al. 2005). However, it has been proposed that mechanisms involved in P transport worked independently of transcriptional events (Saddoris et al. 2010).