Introduction

The sufficient availability of high-quality agricultural products relies on the sustainable usage of resources. Phosphorus (P) is an irreplaceable component of life and, thus, widely used in all agricultural production systems. However, in pig husbandry the dietary P levels often exceed animal- and age-specific requirements. Pigs are considered to be major excretors of P from agricultural systems which lead to a severe environmental burden. In order to balance economic and environmental sustainability related to the uneven density of animal production, novel approaches of P management aim to slow down the rate of emissions through transformative or incremental system-wide processes (Kebreab 2013). This includes knowledge and measures to reduce P excretion targeting benefits for animal health and environment.

Decreased intake of P by farm animals results in lowered P excretion but also lowered P retention. At high P supply, P absorption and excretion will continue to increase, while P retention will reach a plateau at certain amounts (Rodehutscord et al. 1999). Hence, an optimal P efficiency requires a high P absorption, a sufficient skeletal storage, and a low P excretion (Fig. 1) (Poulsen 1994). The addressed mechanisms are interlinked with dietary calcium levels (Berndt and Kumar 2009; Taylor and Bushinsky 2009).

Fig. 1
figure 1

The flow of calcium and P from diet to animal products and slurry

Feeding regimens containing variable calcium and P contents are known to affect related endocrine levels in mammals (Proszkowiec-Weglarz and Angel 2013; Oster et al. 2016), such as vitamin D (calcitriol) and parathyroid hormone (PTH). The serum levels of both hormones are interdependent due to sophisticated feedback mechanisms involving receptors and transporters localised in the small intestine, bone, and kidneys (Berndt and Kumar 2009). Hence, organismal requirements are ensured by affecting re-/absorbing and excreting tissues. Since PTH regulates the activation of calcitriol via 1α-hydroxylase (Friedman et al. 1996), a P-deficient diet was associated with increased renal mRNA abundances of the 1α-hydroxylase (Cyp27B1) (Alexander et al. 2008). Consequently, both activity and intracellular redistribution of sodium-dependent phosphate transporters (SLC34 family) rely on dietary P levels in porcine jejunum (Saddoris et al. 2010) and murine kidney cells (Lanaspa et al. 2013). Moreover, the mineralisation of the skeleton relies on dietary P supplies as indicated by trabecular bone characteristics such as bone mineral density and structure model index (Oster et al. 2016). Interestingly, recent studies suggested that both deficient and excessive P supplies may have negative effects on bone mineralisation and health (Sørensen 2012).

The characterisation of physiological and molecular processes enabling increased P efficiency in monogastric species is crucial towards a P-resilient livestock production. The study investigates the effects of wheat/barley-based diets varying in calcium and P contents on growing pigs. Six tissues and body compartments including serum, bone, duodenum, jejunum, colon, and kidney cortex were used to detect mechanisms of adaptation. We monitored a number of traits, including (i) performance and growth data, (ii) calcium and P mineral homoeostasis, (iii) bone characteristics, and (iv) expression of candidate genes which represent the receptors of relevant hormones, their in-/activating enzymes and P transporters.

Materials and methods

Animals and diets

The study was approved by the Scientific Committee of the FBN, and the experimental setup was generally licensed by the ethics committee of the federal state of Mecklenburg-Western Pomerania, Germany (Landesamt für Landwirtschaft, Lebensmittelsicherheit und Fischerei; LALLF M-V/TSD/7221.3-1-053-15). Twenty-one German Landrace piglets obtained from four litters were randomly assigned to one of three wheat/barley/soybean-based diets (Table S1). Offspring were sired by four boars. The study comprised three to four piglets per sex and dietary group. From weaning (day 28) until slaughter (day 64), piglets received a diet containing low (L, calcium: 0.79%; P: 0.56%), medium (M; calcium: 1.27%; P: 0.84%), or high calcium and P levels (H; calcium: 1.69%; P: 1.02%). The dietary P content in compound feed (group M) corresponded to current recommendations (GfE 2006). Dietary compositions resulted in only slight variation of the calcium:P-ratios (L: 1.41; M: 1.51; H: 1.65). The achieved dietary levels of digestible P were 0.31% (L), 0.49% (M), and 0.60% (H). Neither phytase nor other phosphatases were added. Piglets were individually reared in cages on flat decks in environmentally controlled rooms. The animals had ad libitum access to pelleted feed and water. Piglets were weighed weekly. Feed intake was quantified and feed conversion ratio (FCR) was calculated.

Collection and preparation of serum samples

At day 64, pigs were narcotised by electrical stunning and sacrificed by exsanguination in the experimental slaughterhouse of FBN. Blood samples were collected from trunk blood. Serum was prepared and samples were stored at −80 °C until use. Tissue sampling comprised intestinal mucosa and kidney cortex which were immediately collected, frozen in liquid nitrogen, and stored at −80 °C. Intestinal samples of ~5 cm in length have been taken from the duodenum (10 cm distal from the pyloric junction), jejunum (1 m distal from the pyloric junction), and the anterior part of the colon (Patterson et al. 2008). Intestinal sections have been washed with PBS to ensuring removal of all feed residues.

Serum hormones and mineral measurements

Serum calcitriol (Immunodiagnostic Systems, Frankfurt am Main, Germany) and PTH (Immundiagnostik, Bensheim, Germany) were determined in duplicate using commercially available enzyme-linked immunosorbent assays (ELISA) according to manufacturer’s protocols. Moreover, serum cortisol and serum thyroid hormones triiodothyronine (T3) and thyroxin (T4) were determined in duplicate using a commercially available magnetic bead-based quantitative immunoassay (MAGPIX system) according to manufacturer’s protocols (Merck Millipore, Darmstadt, Germany). Mineral serum measurements (inorganic P, calcium) were analysed with commercial assays using Fuji DriChem 4000i (FujiFilm, Minato, Japan).

Chemical analyses of bones

The individual left femurs were collected at slaughter (day 64) and stored at −20 °C until use. The proximal part (30% of total length) of the bones was separated and lyophilised. Subsequently the bones were degreased. The bone tissue was comminuted by ball mill after embrittlement in liquid nitrogen. Wet chemical analyses according to common methods of VDLUFA were applied (Naumann et al. 1976). Contents of crude ash as well as calcium and P were determined.

RNA isolation

Total RNA was isolated using TRI Reagent per manufacturer’s directions (Sigma-Aldrich, Taufkirchen, Germany), then treated with DNase and purified with the column-based NucleoSpin RNA II-Kit (Macherey–Nagel, Düren, Germany). RNA integrity was determined by visualisation on a 1% agarose gel containing ethidium bromide and the concentration was measured using the NanoDrop ND-1000 spectrometer (PEQLAB, Erlangen, Germany). Absence of DNA contamination was verified by PCR amplification of the porcine RPL32 gene (forward primer: 5′-AGCCCAAGATCGTCAAAAAG-3′; reverse primer: 5′-TGTTGCTCCCATAACCAATG-3′). All RNA samples were stored at −80 °C. First-strand cDNA was synthesised from 2 μg of total RNA using random primers (Promega, Fitchburg, WI, USA) and oligo d(T) 13VN in the presence of Superscript III reverse transcriptase (Invitrogen, Karlsruhe, Germany). The final cDNA was diluted with Aqua dest to a total volume of 100 µl.

Quantitative real-time PCR (qRT-PCR)

On the transcriptional level, receptors of calcitriol, PTH, and thyroid hormones were analysed, i.e. VDR (vitamin D receptor), PTH1R (parathyroid hormone 1 receptor), and THRA (thyroid hormone receptor alpha). Furthermore, genes encoding calcitriol in-/activating enzymes were analysed, i.e. Cyp24A1 (vitamin D3 24-hydroxylase), Cyp27A1 (vitamin D3 25-hydroxylase), and Cyp27B1 (1α-hydroxylase). Moreover, analyses of P transporters comprise SLC34A1 (solute carrier family 34, member 1; NaPi2a), SLC34A2 (solute carrier family 34, member 2; NaPi2b), and SLC34A3 (solute carrier family 34, member 2; NaPi2c).

Transcript levels of selected target (VDR, Cyp24A1, Cyp27A1, Cyp27B1, PTH1R, THRA, SLC34A1, SLC34A2, SLC34A3) and reference genes (RPL32) were quantified by qRT-PCR. Individual mRNA samples (n = 21 per tissue) were analysed in duplicate on a LightCycler 480 system using the LightCycler 480 SYBR Green I Master (Roche, Mannheim, Germany) according to manufacturer’s instructions. Briefly, reactions were performed in a final volume of 10 µl using 5.0 µl of LightCycler 480 SYBR Green I Master (Roche), 0.5 µl (10 µM) of each primer (Table S2), 2 µl (40 ng) cDNA, and 2.0 µl of Aqua dest. The temperature profiles comprised an initial denaturation step at 95 °C for 10 min followed by 40 cycles consisting of denaturation at 95 °C for 15 s, annealing at 60 °C for 10 s, and extension/fluorescence acquisition at 72 °C for 15 s. Amplified products were subjected to melting curve analyses and gel electrophoresis to verify the absence of non-specific products. For all the assays, threshold cycles were converted to copy numbers using a standard curve generated by amplifying serial dilutions of a corresponding PCR standard (107–101 copies). Transcripts with a mean ≤5 copies per 10 ng RNA transcribed were considered as not detectable (nd).

Data analyses

Data referring to gene expression, physiological traits, hormones, and bone measurements were analysed via variance analyses (PROC MIXED; SAS version 9.4; SAS Institute, Cary, NC, USA), including effects represented by dietary group, sex, and sire (V ijk = µ + dieti + sexj + sirek + errorijk). The retrieved LSmeans were compared using Tukey’s post hoc test. The level of significance was set at p < 0.05.

Results

Piglet performance and feed conversion ratio

High levels of dietary calcium and digestible P revealed decreased live weights compared to low- and medium-fed animals (Table 1). Moreover, cumulative daily feed intake and daily body weight gain were decreased in high-fed animals which resulted in an increased feed conversion ratio (FCR).

Table 1 Performance traits of pigs fed experimental diets with low, medium, and high calcium and digestible P contents

Diet-specific serum hormone levels and mineral measurements

As displayed in Table 2, serum calcitriol was increased in L animals but decreased in H animals (L > M > H). Regarding parathyroid hormone (PTH), dietary effects were observed between L and H samples (L < H). T3 was decreased in H animals (L > H; M > H). No significant dietary effects on serum T4 levels, cortisol, inorganic P, and calcium were observed.

Table 2 Serum measurements of pigs fed experimental diets with low, medium, and high calcium and digestible P contents

Bone characteristics

Femur length was unaffected by diet (Table 3). The fat-free dry matter of the femur (DMfat free) was decreased in L animals compared to M and H animals (L < M; L < H). Regarding crude ash, dietary effects were observed between L and H samples (L < H). No significant differences were observed for femoral calcium and P measures and calculated femoral calcium–P ratio.

Table 3 Femur characteristics of pigs fed experimental diets with low, medium, and high calcium and digestible P contents

Gene expression in re-/absorbing and excreting tissues

The dietary challenges revealed site-specific transcriptional responses in duodenum, jejunum, colon, and kidney (Table 4). In duodenum, mRNA abundances of Cyp24A1 (L > M; L > H) and SLC34A3 differed significantly between dietary groups (L < H). In jejunum, genes encoding for Cyp24A1 (L > M; L > H; M > H), Cyp27B1 (L < H), THRA (M < H), and SLC34A3 (L > H) were diet-dependently altered. In colon, THRA (L < H) and Cyp27B1 were higher expressed in H animals (L < H; M < H). Gene expression in kidney cortex revealed lower mRNA abundances of VDR in L animals (L < M; L < H). Renal expression of THRA was increased in H animals (L < H; M < H), whereas genes encoding Cyp27A1 and Cyp27B1 were higher expressed in L animals (L > M; L > H). Moreover, mRNA abundances of SLC34A3 differed between H and L animals (L > H). Genes encoding for SLC34A1 and SLC34A2 were not detectable in duodenum, jejunum, and colon, whereas SLC34A3 was expressed in all analysed tissues.

Table 4 Tissue-specific relative gene expression and copy numbers of selected transcripts in pigs fed experimental diets with low, medium, and high calcium and digestible P contents. Significant differences are displayed in bold. FC fold change, nd not detectable

Trait relationships

PTH showed negative linear correlations with calcitriol and T3, whereas the latter ones were positively correlated (Fig. 2). Cortisol and serum calcium levels showed positive linear correlations with T4. Serum P was not correlated to any tested trait. Live weight, daily feed intake, and daily BW gain were positively correlated with calcitriol and T3 but negatively correlated with PTH.

Fig. 2
figure 2

Heatmap displaying Pearson correlation coefficients comprise performance traits and serum measurements. The diagonals summarise significant alterations indicated in Tables 1 and 2

Discussion

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).

Conclusion

The responses to the diet containing low calcium and P levels were sufficient to maintain physiological calcium and P serum concentrations by recruiting the bone mineral storage. However, the diet containing high calcium and P levels revealed to be inappropriate for an adequate growth performance since a negative effect of high dietary calcium–P levels on feed intake was observed. Indeed, the dynamic influxes and effluxes of calcium and P among organs and tissues were reflected by the pronounced endocrine and transcriptional responses and might be of critical importance to produce P-resilient phenotypes. The intestinal mucosa and kidney cortex were highlighted as initial sites to maintain mineral homoeostasis. The observed responses contribute to implement feeding strategies to preserve global P resources and to reduce agricultural residues. Results clearly suggest that the usage of calcium and P should be monitored and better regulated within a framework of improved governance. Current feeding recommendations for livestock systems need to consider aspects for animal health as well as economic and environmental perspectives to reduce dietary mineral intake in growing pigs.