To protect the environment and its resources, reducing the nitrogen (N) and phosphorus (P) content in the diet of ruminants is a much-discussed option. Ruminants can compensate for reduced dietary N-intake quite well due to the ruminohepatic cycle that allows the use of urea(Reference Sarraseca, Milne and Metcalf1–Reference Harmeyer and Martens3). Ruminants are also able to maintain endogenous phosphate (Pi) homoeostasis during P-restriction because they can recycle Pi (Reference Horst4,Reference Reynolds, Huntington and Tyrrell5) . N and P are both essential for the survival of the micro-organisms in the rumen of the host animal and must therefore be available in sufficient quantities. N is an important source of protein synthesis in ruminants, and Pi is essential for energy metabolism, cell signalling and bone mineralisation(Reference Hansen, Felix and Bisaz6–Reference Krebs and Beavo8). Despite the ability of ruminants to compensate for reduced N- and P-intake, previous studies have shown that severe changes in mineral homoeostasis occurred. N-restriction resulted in lower blood concentration of Ca and calcitriol, the active form of vitamin D(Reference Muscher and Huber9,Reference Firmenich, Elfers and Wilkens10) , while P-restriction resulted in increased blood concentration of Ca with concomitantly reduced blood Pi concentration(Reference Schröder, Pfeffer and Failing11,Reference Breves, Ross and Höller12) . Most of the body’s Ca and Pi are stored in the bones, where they form the two main components of the bone mineral hydroxyapatite, which strengthens the bone matrix(Reference Hansen, Felix and Bisaz6). Changes in the Ca and Pi concentration in the blood are balanced by the interaction of various hormones that regulate reabsorption and excretion in the intestine and kidney. Additionally, these changes in concentration are regulated by a release of Ca and Pi from bone and are stored in the bone matrix. The bone-derived hormone fibroblast growth factor 23 (FGF23) is the most important regulator of Pi homoeostasis in monogastric species(Reference Burnett, Gunawardene and Bringhurst13,Reference Miedlich, Zhu and Sabbagh14) and ruminants(Reference Köhler, Grunberg and Schnepel15). Extracellular Pi acts as a signalling molecule on the bone and directly regulates the release of FGF23 from the bone(Reference Khoshniat, Bourgine and Julien16). It is assumed that Pi sensing in bone is mediated by the two Na-dependent Pi transporters 1 and 2 (PiT1, SLC20A1 and PiT2, SLC20A2)(Reference Bon, Couasnay and Bourgine17). While PiT2 is essential for the regulation of Pi-dependent FGF23 secretion in mice(Reference Bon, Frangi and Sourice18), PiT1 is essential for bone mineralisation in rats(Reference Yoshiko, Candeliere and Maeda19). In monogastric species, low levels of Pi in the blood reduce the excretion of FGF23. Consequently, the expression of renal Na-dependent Pi transporters remains unchanged at lower FGF23 levels, allowing the reabsorption of Pi via the kidneys(Reference Andrukhova, Zeitz and Goetz20). Apart from Pi, the synthesis of FGF23 in bone is also regulated by other factors. Phosphate-regulating endopeptidase X-linked (PHEX) that is expressed in osteoblasts and osteocytes is necessary for an increase in FGF23 gene transcription(Reference Liu, Zhou and Tang21). Hence, a reduction of PHEX is expected with the down-regulation of FGF23. Another inhibitory pathway of FGF23 secretion is induced by insulin and insulin-like growth factor 1 (IGF1) through binding of the insulin receptor (INSR) and the insulin-like growth factor 1 receptor (IGF1R). This activates the phosphatidylinositol-3-kinase (PI3K)/serine/threonine kinase (Akt) pathway. Its activation inhibits Forkhead box 1 (FOXO1), which leads to the inhibition of FGF23(Reference Agoro, Ni and Noonan22). An FGF23-inducing pathway in bone is that of Janus kinase 2 (JAK2)/signal transducer and activator of transcription 3 (STAT3) signalling(Reference Tsuji, Maeda and Kawane23). This pathway is activated by leptin and growth hormone (GH) when binding to their receptors’ leptin receptor (LEPR)(Reference Upadhyay, Farr and Mantzoros24) and growth hormone receptor (GHR), respectively(Reference Herrington and Carter-Su25,Reference Haussler, Whitfield and Kaneko26) . In addition to FGF23, mineral homoeostasis is also regulated by parathyroid hormone (PTH) which is secreted by the parathyroid gland(Reference Bergwitz and Jüppner27). Once secreted from the parathyroid gland, PTH binds to its receptor parathyroid hormone 1 receptor (PTH1R) on osteoblasts regulating bone resorption(Reference Ishizuya, Yokose and Hori28). By binding to PTH1R, PTH induces bone resorption through stimulation of receptor activator of NF-κB ligand (RANKL)(Reference Fu, Jilka and Manolagas29). Simultaneously, osteoprotegerin (OPG) is inhibited by PTH, resulting in the release of Ca and Pi from bone(Reference Fu, Jilka and Manolagas29). OPG acts as a decoy receptor of RANKL because it binds RANKL and therefore blocks the binding of RANKL to its receptor activator of NF-κB (RANK)(Reference Boyle, Simonet and Lacey30). For storing Ca and Pi in bone, osteoblasts are activated initiating bone formation(Reference Ko, Martins and Reddy31,Reference Yamaguchi, Chattopadhyay and Kifor32) . Osteoblastic maturation is induced by runt-related transcription factor 2 (RUNX2), alkaline phosphatase (ALP) and osteocalcin (OC), among others(Reference Ko, Martins and Reddy31,Reference Gordeladze and Reseland33) . From previous studies on ruminants, it is known that changes in mineral homoeostasis due to reduced N- and P-intake impaired bone metabolism(Reference Köhler, Grunberg and Schnepel15,Reference Elfers, Liesegang and Wilkens34) . As a next step, the changes which are caused in the bone by a reduction of dietary N and/or P will be characterised molecularly.
P-restriction is expected to result in reduced FGF23 expression, and thus it is hypothesised that there is an increase in the inhibitory PI3K/Akt pathway with a simultaneous reduction in the stimulatory JAK2/STAT3 pathway. N-restriction disrupted the somatotropic axis in goats with lower IGF1 concentrations(Reference Elfers, Wilkens and Breves35). Therefore, an increase in FGF23 was hypothesised in the goats in this study due to reduced PI3K/Akt activation. Finally, the bone mineral density and the bone mineral content will be examined to see whether these postulated molecular changes also influence the bone macroscopically.
Material and methods
Animals and feeding regimens
Information about the animals and the feeding regimen has already been published in a previous study(Reference Weber, Hustedt and Schnepel36). To avoid unnecessary repetition of animal model experiments, organs not used in the study of Weber et al.(Reference Weber, Hustedt and Schnepel36) were taken in this study. Twenty-eight male Coloured German Goats from a commercial goat farm received (1) a control diet (16·48 % crude protein (CP), 0·48 % P, 1·30 % Ca), (2) an N-reduced diet (8·35 % CP, 0·51 % P, 1·20 % Ca), (3) a P-reduced diet (16·86 % CP, 0·11 % P, 1·20 % Ca) or (4) an N- and P-reduced diet (8·10 % CP, 0·11 % P, 1·20 % Ca) for six to eight weeks. At the beginning of the experiment, the animals had an initial body weight of 19·04 (sd 2·2) at an age of 10 weeks(Reference Weber, Hustedt and Schnepel36). The pellets were produced by a specialised feed manufacturer for animal research feed production (ssniff Spezialdiäten GmbH). The diets had approximately 12·7 MJ ME/kg DM and were isoenergetic. For further details on the ingredients and chemical composition of the diets, see Table 1 and Weber et al.(Reference Weber, Hustedt and Schnepel36).
Components and composition of wheat straw and pelleted concentrate diets*
N+/P+, control diet; N-/P+, N-restricted diet; N+/P-, P-restricted diet; N-/P-, N- and P-restricted diet; ADFom, acid-detergent fibre expressed exclusive of residual ash; aNDFom, neutral-detergent fibre assayed with heat stable amylase expressed exclusive of residual ash; CP, crude protein; BDL, below detection level; ME, metabolisable energy; DCAD, dietary cation–anion difference.
* Parts of the data already published in Weber et al.(Reference Weber, Hustedt and Schnepel36).
† Mineral–vitamin premix per kg: 0·2 g P; 12·1 g Ca; 1·7 g Na; 2·2 g Mg; 1 200 000 IU vitamin A; 120 000 IU vitamin D; 10 000 mg vitamin E; 675 mg vitamin K; 4960 mg iron; 6336 mg Zn; 501 mg Cu; 3000 mg Mn; 201 mg Co; 15 mg Se; 202 mg I.
‡ Sipernat 22S (Evonik Industries AG) – a fine particle silica that can be used as highly absorbent carrier substance, flow regulator, anti-caking and anti-dusting agent in the food and feed industry.
§ Composition analysed by the Association of German Agricultural Investigation and Research Centre (VDLUFA).
|| DCAD (meq/kg DM) = (meq Na + meq K) - (meq Cl + meq S).
Seven goats per feeding group were housed together and water was available ad libitum. Each animal was fed 55 g/kg0·75 pelleted concentrate per animal twice a day. Additionally, the goats received 25 % of the concentrate weight as wheat straw. To ensure documentation of the average intake of nutrients for each animal, the goats received their diet individually and the amount eaten was documented daily. The body weight of the goats was measured once a week.
Body fluid and tissue sampling
Blood samples were taken from the vena jugularis with lithium heparinate-coated syringes and serum syringes (Sarstedt) shortly before euthanasia using a standard abattoir captive bolt stunning procedure (Annex IV Directive 2010/63/EU). Blood samples were always taken at the same time in the morning from the fasting animal to avoid diurnal effects. Plasma and serum samples were centrifuged (2000 g at room temperature, 15 min) for separation and stored at −20°C until further analysis.
Two goats per d were killed due to technical reasons (Ussing chamber experiments). Animals from two feeding groups were slaughtered alternately each day to avoid significant time effects.
Shortly after euthanasia, a metacarpal bone and a rib were dissected. For this purpose, the flesh was removed from the bones, then immediately frozen in liquid N2 and stored at −80°C until further preparation. The metacarpal bone was examined using µCT, and the mineral content was then analysed. The rib bone was separated into cortical and medullary sections, and the cortical portion was used for quantitative PCR (qPCR) analysis.
Biochemical and immunological assays
To determine the N, Pi and Ca status of the animals, blood plasma concentrations of urea (commercial kit, R-Biopharm AG), Pi and Ca were measured colorimetrically with standard spectrometric methods(Reference Ray Sarkar and Chauhan37,Reference Kruse-Jarres38) (interassay CV 3·79 % (urea), 1·88 % (Ca) and 5·05 % (Pi); intra-assay CV 1·2 % (urea), 2·85 % (Ca) and 1·42 % (Pi)). Plasma concentration of bone-specific alkaline phosphatase was measured by a competitive ELISA (Quidel Corporation). Plasma OC and CrossLaps (CTX) were also measured by competitive ELISA (TECOmedical Group and Immundiagnostik AG). The serum concentrations of leptin were measured with a bovine-specific competitive enzyme immunoassay(Reference Sauerwein, Heintges and Hennies39). Plasma GH and insulin as well as serum IGF1 were measured in the Clinical-Endocrinological Laboratory at the University of Veterinary Medicine Hannover, Foundation. Plasma GH was measured using an in-house ELISA. Plasma insulin and serum IGF1 were measured using RIA (Beckman Coulter GmbH)(Reference Piechotta, Kedves and Araujo40).
Determination of mineral content and density, ash, Ca, P, Mg and Zn content of the metacarpus
Mineral content and density of the metacarpus were measured by quantitative computer tomography (Stratec XCT bone scanner, Stratec Medizinaltechnik GmbH). Total distal and trabecular content and density were measured at 10 % of the total length of the metatarsi. To determine the total medial content and density as well as cortical medial content and density, the bones were scanned at 50 % of the total length, in the middle of the diaphysis. To determine the mineral content, the bones were sawn into small pieces. Subsequently, the content of almost fat-free DM of the bones was determined by lypophilisation for about 48 h, followed by defatting with petroleum ether by using the SOXTHERM (C. Gerhardt GmbH & Co. KG) and drying at 80°C. According to Rieger et al.(Reference Rieger, Ratert and Wendt41) and in accordance with the official methods of the VDLUFA (Association of German Agricultural Analytic and Research Institutes), the bones were then ground in a hammer mill and analysed for crude ash, crude fat (to determine fat-free DM), Ca, P, Mg and Zn.
Quantitative reverse transcription real-time PCR
Total RNA from bone rib tissue was isolated using the RNeasy Fibrous Tissue Mini-Kit (Qiagen GmbH) following the manufacturer’s protocol. The RNA concentrations were measured spectrophotometrically using NanoDrop One (Thermo Fisher Scientific Inc. Waltham). The quality and integrity of the RNA were assessed using an RNA 6000 nanoassay for an Agilent 2100 Bioanalyzer (Agilent Technologies Deutschland GmbH). The RNA integrity number of all bone rib tissue samples was on average 8·69 (sem 0·12). Reverse transcription of 200 ng of isolated RNA for real-time qPCR was performed using random hexamers, oligo-dT primers and TaqMan™ Reverse Transcription Reagent (Thermo Fisher Scientific Inc.) by the manufacturer’s protocol.
To determine the mRNA abundance of Akt1, Akt2, ALP, calcitonin receptor (CALCR), FGF23, FOXO1, GHR, IGF1R, INSR, JAK2, LEPR, OC, OPG, PHEX, PiT1, PiT2, PTH1R, RANK, RANKL, RUNX2, STAT3 and vitamin D receptor (VDR), SYBR Green® PCR assays with specific primers (Table 2) were used. Reaction mixtures of 20 µl contained SensiFASTTM SYBR No-Rox Mix (BioCat GmbH), 200 nmol/l of specific primers and 16 ng of reverse-transcribed complementary DNA. For amplification (3 min at 95°C; 40 cycles of 10s at 95°C and 30s at 60°C) and detection of the PCR products, a real-time PCR cycler (CFX96TM, Bio-Rad Laboratories GmbH) was used. To determine the melting curve, the thermal profile began with an incubation period of 10 min at 55°C with a gradual increase in temperature (0·5°C per 10s) up to 95°C. To evaluate the rRNA expression of 18S in bone rib tissue as a reference gene, caprine gene-specific TaqMan primers and probes were synthesised by TIB Molbiol Syntheselabor GmbH (Table 3). The reaction mixture of 20 µl each contained TaqMan™ Gene Expression Master Mix (Thermo Fisher Scientific Inc.), 16 ng of reverse transcript complementary DNA, 300 nmol/l of specific primers and 100 nmol/l of specific probe. PCR products were amplified (50°C, 2 min; 95°C, 10 min; 40 cycles of 95°C, 15 s and 60°C, 1 min) and analysed on a real-time PCR cycler (CFX96TM, Bio-Rad Laboratories GmbH). Absolute copy numbers were determined using calibration curves generated with cloned PCR fragment standards(Reference Wilkens, Kunert-Keil and Brinkmeier50). The specificity of amplicons was verified by sequencing (Microsynth Seqlab GmbH) and using NCBI Blast (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The expressions of the reference genes ribosomal protein L19 (RPL19) and ribosomal protein S9 (RPS9) were quantified with SYBR Green® PCR assays (Table 2). The best consideration as a reference gene for normalisation was 18S rRNA, calculated by NormFinder software (https://www.moma.dk/normfinder-software). Reactions were performed twice and included water as a no-template control.
Primers used for SYBR Green assays in compact bone rib tissue from young goats
* Akt1, Akt serine/threonine kinase 1; Akt2, Akt serine/threonine kinase 2; ALP, alkaline phosphatase; CALCR, calcitonin receptor; FGF23, fibroblast growth factor 23; FOXO1, Forkhead box O1; GHR, growth hormone receptor; IGF1R, insulin-like growth factor 1 receptor; INSR, insulin receptor; JAK2, Janus kinase 2; LEPR, leptin receptor; OC, osteocalcin; OPG, osteoprotegerin; PHEX, phosphate-regulating endopeptidase X-linked; PiT1, pituitary-specific transcription factor 1; PiT2, pituitary-specific transcription factor 2; PTH1R, parathyroid hormone 1 receptor; RANK, receptor activator of NF-κB; RANKL, receptor activator of NF-κB ligand; RPL19, ribosomal protein L19; RPS9, ribosomal protein S9; RUNX2, runt-related transcription factor 2; STAT3, signal transducer and activator of transcription 3; VDR, vitamin D receptor.
Primers and probes used for TaqManTM assays in bone rib tissue from young goats
* 18S rRNA, 18S ribosomal RNA.
Statistical analysis
Sample size (n 7/group) was determined based on metabolic data from a previous study(Reference Firmenich, Elfers and Wilkens10) with a statistical power of 0·8 and α error of 0·05. All data are given as means with their standard errors of the mean (sem) unless otherwise stated. GraphPad Prism version 9.3 (GraphPad Software) was used for data analysis.
All data were tested for normal distribution using Kolmogorov–Smirnov test. The results confirmed that the parameters were normally distributed, so the data are represented as measured means and sem. A two-way ANOVA was conducted to evaluate the main effects of N-intake and P-intake. Post hoc tests were not performed due to the absence of significant interactions between N-intake and P-intake factors.
A significance level of P < 0·05 was considered statistically significant, while P < 0·1 was used to define trends.
Potential linear relationships between mRNA expression levels and the correlation between gene expression with blood parameters were calculated using a simple correlation analysis with Pearson’s correlation coefficient. Potential outliers were tested with the ROUT method.
Results
Feed intake and growth performance
All animals were clinically healthy throughout the study. The clinical health of the animals was ensured through monitoring and care by a qualified veterinarian who examined the animals daily and closely observed their behaviour and physical condition. The data on the goats’ feed intake, daily weight gain and body weight have already been published by Weber et al.(Reference Weber, Hustedt and Schnepel36).
Blood metabolites
Blood metabolites are shown in Table 4. Some of the blood metabolites (Ca, GH, IGF1, Pi and urea) have already been published in Weber et al.(Reference Weber, Hustedt and Schnepel36) and Zillinger et al.(Reference Zillinger, Hustedt and Schnepel51). The plasma concentration of CTX was significantly increased in the animals receiving the N-restricted diet compared with the control group. Blood plasma concentration of OC was significantly lower in the P-restricted feeding groups and increased in the N-restricted animals compared with the control group. The blood plasma concentration of bone-specific alkaline phosphatase increased significantly with P-restriction compared with the control group. The OC concentration in plasma correlated negatively with the blood concentration of Ca (r = –0·62; P < 0·001) and positively with the blood concentration of Pi (r = 0·72; P < 0·001) and IGF1 (r = 0·58; P = 0·002). The blood concentrations of total protein, insulin and leptin were not affected by the different diets.
Blood metabolites and hormones of young goats fed an N- and/or P-reduced diet (Mean values with their standard errors of the mean (SEM); seven animals per group)
BALP, bone-specific alkaline phosphatase; CTX, CrossLaps; OC, osteocalcin; N, nitrogen; P, phosphorus.
For all parameters, P-value of NxP ≥ 0·112.
* n 6; due to the outlier test performed.
Mineral content and density, ash, Ca, P, Mg and Zn content of the metacarpus
Results of the measurements by computer tomography are summarised in Table 5. The results of the mineral analysis of the metacarpal bone are summarised in Table 6. Total distal content and density, trabecular content and density as well as total medial content were significantly lower with P-restriction. Total medial density and cortical medial density and content were not affected by dietary modulation. The ash and Mg content were significantly lower with P-restriction, while the P content of the bone was reduced by trend with P-restriction. The Mg content increased by trend (P = 0·091) with N-restriction. No significant effects were observed for Zn with either N-reduction or P-reduction. However, an interaction between N and P was detected for Zn (data not shown; P = 0·049), whereas all other interactions had P-values ≥ 0·359.
Metacarpal mineral content (mg/cm) and density (mg/cm3) of young goats fed an N- and/or P-reduced diet (Mean values with their standard errors of the mean (SEM); seven animals per group)
N, nitrogen; P, phosphorus.
For all parameters, P-value of NxP ≥ 0·466.
Effects of an N- and/or P-reduced diet on ash, Ca, P, Mg and Zn of the metacarpus of young goats (g/kg fat-free DM; mean values with their standard errors of the mean (SEM); seven animals per group)
Effects of a nitrogen- and/or phosphorus-reduced diet on mRNA expression of selected target genes in the rib bones
The results of the mRNA expression of this study are shown in Table 7. The gene expression of ALP, PHEX, PiT2, RANK and VDR remained unchanged in all feeding groups. A significant increase in the mRNA expression of Akt1, Akt2, CALCR, GHR, IGF1R, INSR, JAK2, LEPR, PiT1, PTH1R and STAT3 was demonstrated in P-restricted feeding compared with the control animals. RUNX2 showed a strong tendency towards increased gene expression with P-restricted feeding (P = 0·052) as well as with simultaneous N- and P-restricted feeding (P = 0·075). A significant increase in mRNA expression of Akt1, FOXO1, OC, RANKL and STAT3 was also detected with N-restriction. The ratio of RANKL to OPG increased significantly with P-restriction, with an interaction also observed (data not shown; P = 0·049). For all other interaction parameters, P-value of NxP ≥ 0·075. The gene expression of INSR (P = 0·052) and PTH1R (P = 0·061) showed a tendency towards increased gene expression with N-restriction. INSR gene expression correlated positively with Akt1 (r = 0·92; P < 0·001; Fig. 1(a)) and Akt2 gene expression (r = 0·82; P < 0·001; Fig. 1(b)). PTH1R mRNA expression correlated positively with plasma CTX (r = 0·45, P = 0·016) and RANKL gene expression (r = 0·52; P = 0·004; Fig. 2). P-restricted feeding significantly lowered the mRNA expression of FGF23, OC and OPG. OC mRNA correlated negatively with plasma Ca and positively with plasma Pi (r = –0·49, P = 0·008 and r = 0·48, P = 0·010). FGF23 mRNA expression correlated positively with plasma Pi (r = 0·81; P < 0·001) and negatively with PTH1R mRNA (r = –0·37; P = 0·05).
Relative mRNA expression levels (normalised to 18S rRNA) of various genes in compact bone rib tissue of young goats fed an N- and/or P-reduced diet (Mean values with their standard errors of the mean (SEM); seven animals per group)
18S rRNA, 18S ribosomal RNA; N, nitrogen; P, phosphorus; Akt1, Akt serine/threonine kinase 1; Akt2, Akt serine/threonine kinase 2; ALP, alkaline phosphatase; CALCR, calcitonin receptor; FGF23, fibroblast growth factor 23; FOXO1, Forkhead box O1; GHR, growth hormone receptor; IGF1R, insulin-like growth factor 1 receptor; INSR, insulin receptor; JAK2, Janus kinase 2; LEPR, leptin receptor; OC, osteocalcin; OPG, osteoprotegerin; PHEX, phosphate-regulating endopeptidase X-linked; PiT1, pituitary-specific transcription factor 1; PiT2, pituitary-specific transcription factor 2; PTH1R, parathyroid hormone 1 receptor; RANK, receptor activator of NF-κB; RANKL, receptor activator of NF-κB ligand; RUNX2, runt-related transcription factor 2; STAT3, signal transducer and activator of transcription 3; VDR, vitamin D receptor.
Linear relationship between insulin receptor (INSR) mRNA abundance and (a) Akt serine/threonine kinase 1 (Akt1) mRNA abundance in the bone cortex (r = 0·92, P < 0·001) and (b) Akt serine/threonine kinase 2 (Akt2) mRNA abundance in the bone cortex (r = 0·82, P < 0·001). The level of significance with Pearson’s correlation coefficient was set at P = 0·05.
Linear relationship between mRNA abundance of parathyroid hormone 1 receptor (PTH1R) and mRNA abundance of receptor activator of NF-κB ligand (RANKL) in the bone cortex (r = 0·71, P < 0·001). The level of significance with Pearson’s correlation coefficient was set at P = 0·05.
Discussion
In this study, the influence of N- and/or P-restriction on bone metabolism in young goats focuses on molecular-level changes. Our findings support previous research highlighting the critical role of P in bone mineralisation and structural integrity. The reduction in ash content observed with P-restriction underscores the importance of P-containing diets in maintaining hydroxyapatite crystals in bone. Concomitantly, the decrease in Mg content observed suggests a potential secondary effect, as Mg plays an important role in bone health, affecting both bone density and quality. The tendency for decreased bone P content under P-restriction emphasises the direct influence of P-containing diets on bone P levels, which is critical for the maintenance of hydroxyapatite crystals in bone. Interestingly, the tendency for increased Mg levels with N-restriction (P = 0·091) suggests a compensatory mechanism. N is essential for amino acid and protein synthesis, which are essential for bone matrix formation. N-restriction could potentially alter bone metabolism and lead to a compensatory increase in Mg uptake or retention as the body attempts to maintain bone homoeostasis under suboptimal nutrient conditions.
Furthermore, in our study, we observed that P-restriction not only affects bone mineralisation but also has significant effects on Ca homoeostasis. P is crucial for maintaining the balance of minerals in the body, and its restriction can lead to compensatory mechanisms that alter Ca metabolism. Specifically, P-restriction can lead to an increase in blood Ca levels as the body mobilises Ca from bones to maintain necessary Pi concentrations in the blood. This mobilisation process can potentially lead to an imbalance in mineral homoeostasis, which contributes to milk fever, especially in dairy cows. Milk fever, or hypocalcemia, is a metabolic disorder that often occurs in dairy cows around the time of calving and is characterised by low blood Ca levels. The disorder can be exacerbated by an imbalance of P and Ca in the diet. According to Keanthao et al.(Reference Keanthao, Goselink and Dijkstra52), P-restriction leads to a reduction in bone P content and can cause a compensatory increase in blood Ca levels. This imbalance highlights the need for carefully balanced diets to prevent metabolic disorders such as milk fever.
In monogastric species, the reduced plasma concentration of Pi led to increased bone resorption by osteoclasts to release Pi from bone while inhibiting bone mineralisation by osteoblasts(Reference Baylink, Wergedal and Stauffer53,Reference Harada and Rodan54) . In rats fed a low energy diet, the concentration of bone formation marker OC in the serum, which represents late osteoblastic maturation(Reference Cundy, Reid and Grey55), reduced with a simultaneous reduction in the blood concentration of Pi (Reference Ndiaye, Cournot and Pélissier56). In the young goats of the current study, the lower OC concentration and the associated mRNA abundance during P-reduction can be explained by the fact that bone formation and thus OC concentration are energy-dependent, as in monogastric animals(Reference Ndiaye, Cournot and Pélissier56). The positive correlation of plasma Pi with plasma OC and OC mRNA supports this assumption. An additional reason for the diminished OC gene expression could be a change in vitamin D metabolism during P-reduction, as the expression of the vitamin D activator, CYP27B1, in the kidney was reduced(Reference Zillinger, Hustedt and Schnepel51). OC synthesis is triggered by calcitriol, the active hormonal form of vitamin D, via the VDR and a specific vitamin D-responsive element in the promoter of the OC gene(Reference Morrison, Yeoman and Kelly57). In mice treated with calcitriol, this stimulated OC RNA abundance after binding to VDR because one of the effects of calcitriol on bone is to promote osteoblast differentiation(Reference Zhang, Ducy and Karsenty58).
Furthermore, it is assumed that the high Ca concentrations in the blood during a P-reduction lower OC levels because in humans with hypercalcaemia, the blood concentration of OC was significantly reduced and therefore reduced bone formation was hypothesised(Reference Delmas, Demiaux and Malaval59). This assumption is supported by the negative correlation of OC gene expression and OC in blood with Ca in blood.
An earlier study in N-restricted goats demonstrated increased bone turnover with an increase in the bone formation marker OC and a simultaneous increase in bone resorption marker CTX(Reference Elfers, Liesegang and Wilkens34). In the study by Elfers et al.(Reference Elfers, Liesegang and Wilkens34), the N-restricted goats developed reduced levels of plasma Ca that were thought to have caused the increase in CTX. At the same time, the blood calcidiol concentration increased, and it was hypothesised that this was the reason for the increase in OC concentration(Reference Elfers, Liesegang and Wilkens34), as a vitamin D dependency of OC was suspected in human bone cells(Reference Uchida, Ozono and Pike60,Reference Skjødt, Gallagher and Beresford61) . In the present project, the blood concentration of calcidiol did not change, indicating that OC levels in goats are probably not influenced by calcidiol. Rather it appears that the reduction in blood Ca concentration alone caused the increase in bone turnover, which is reflected in an increase in both CTX and OC. The lower Ca concentration in the blood induces the secretion of PTH from the parathyroid glands in monogastric species(Reference Kemper, Habener and Rich62) as well as cows(Reference Wilkens, Oberheide and Schröder63). Unfortunately, no specific assay for measuring PTH in goat blood is commercially available. By tending to increase PTH1R expression in bone during N-restriction in this study, PTH could induce bone formation by osteoblasts, which in turn would stimulate bone resorption by osteoclasts(Reference Kronenberg, Bringhurst, Nussbaum, Mundy and Martin64,Reference Lee, Deeds and Chiba65) , to release increased Ca from the bone.
Despite the reduction in OC, it was hypothesised that early osteoblastic maturation increased in the P-restricted goats due to the trend towards increased RUNX2 gene expression. In a study on P-restricted mice, RUNX2 mRNA also increased, while OC mRNA expression was lower. It was hypothesised that P-restriction prevented differentiation from immature to mature osteoblasts(Reference Ko, Martins and Reddy31). The increase in the enzyme bone-specific alkaline phosphatase, which also functions as a bone formation marker(Reference Cundy, Reid and Grey55), during P-restriction presumably occurred to generate free Pi which can be provided to the bone cells in times of reduced blood Pi concentration(Reference Penido and Alon66). A reduced bone mineralisation rate was also reflected in a lower bone mineral density and bone mineral content in the P-restricted goats as it was also seen in hens and pigs receiving a P-restricted diet(Reference Sohail and Roland67,Reference Pokharel, Regassa and Nyachoti68) .
The observed increase in blood Ca concentration with P-restriction may be responsible for the increase in CALCR mRNA expression, as a study with CALCR knockout mice demonstrated that the presence of CALCR protects against hypercalcaemia probably by inhibiting bone resorption activity(Reference Davey, Turner and McManus69). In cases of high concentrations of Ca in the blood, the hormone calcitonin that is secreted from the thyroid gland(Reference Foster, Baghdiantz and Kumar70) reduces serum Ca levels by decreasing osteoclastic bone resorption(Reference Aliapoulios, Goldhaber and Munson71,Reference Raisz and Niemann72) through stimulation of CALCR expression as demonstrated in rats(Reference Ikegame, Ejiri and Ozawa73). Besides, rat osteoclast cells were identified as the cells expressing the highest amount of CALCR(Reference Nicholson, Moseley and Sexton74,Reference Warshawsky, Goltzman and Rouleau75) , which is why the increase in CALCR mRNA expression in this study could reflect an increased number of osteoclasts. Increased extracellular Pi led to inhibition of RANK–RANKL-induced osteoclastic differentiation in the cell culture of a previous study(Reference Mozar, Haren and Chasseraud76). In the present study, neither RANK nor RANKL gene expression changed in the caprine rib bone, although plasma Pi was diminished. OPG binds RANKL, thus preventing RANKL from binding to RANK and protecting against excessive bone resorption(Reference Chen, Wang and Duan77). Therefore, the reduction in OPG mRNA abundance may indicate induced bone loss during P-reduction, as observed in glucocorticoid-induced bone loss in mice(Reference Piemontese, Xiong and Fujiwara78). This lower OPG mRNA expression is induced by the reduction in FGF23 mRNA expression during P-restriction, which leads to an increase in PTH secretion(Reference Ben-Dov, Galitzer and Lavi-Moshayoff79), and is reflected in a negative correlation of FGF23 mRNA with PTH1R mRNA in the present study. By binding to PTH1R on osteoblasts, PTH induces bone resorption through stimulation of RANKL expression and inhibition of OPG expression(Reference Fu, Jilka and Manolagas29).
This also supports the assumption that the elevated plasma Ca concentration was caused by bone resorption. Although RANKL gene expression did not change with P-restriction in this study, a positive correlation of PTH1R mRNA with RANKL mRNA (Fig. 2) was observed, supporting the hypothesis that bone resorbing activity increased. RANKL/OPG ratio shows the physiological ratio of bone formation and turnover increased with P-restriction in the present study. The higher this ratio, the more pronounced the bone resorption(Reference Jura-Półtorak, Szeremeta and Olczyk80).
Although neither RANK nor OPG gene expression nor the RANKL/OPG ratio changed in the goats during N-reduction, the increase in RANKL mRNA indicates an increase in osteoclast activation(Reference Lloyd, Yuan and Kostenuik81), and this is supported by the previously mentioned rise in CTX concentration in the blood of N-reduced goats. Furthermore, the influence of PTH1R on bone resorbing activity in the N-restricted-fed goats was demonstrated by a positive correlation of PTH1R mRNA with RANKL mRNA expression (Fig. 2) and with CTX in blood.
In monogastric species, a reduced blood Pi concentration led to reduced synthesis of FGF23 from the bone due to restricted P-intake(Reference Bergwitz and Jüppner27). This effect was also demonstrated in ruminants in a previous study on P-depleted sheep(Reference Barthel, Mathern and Whitfield82) as well as in the young goats of the present study receiving a P-reduction. The positive correlation of FGF23 mRNA with the blood concentration of Pi confirms the linear relationship. However, it is unclear which mechanisms the Pi regulates in FGF23 synthesis and secretion. In rat osteoblasts, an increase in PiT1 mRNA expression with reduced Pi was observed(Reference Zoidis, Ghirlanda-Keller and Gosteli83), as was the case in the P-restricted goats of this study, suggesting that PiT1 plays an essential role in Pi transport activity in bone in addition to PiT2.
Furthermore, in mice, FGF23 expression in bone was lowered by reduced blood concentration of calcitriol via reduced VDR expression(Reference Masuyama, Stockmans and Torrekens84). As mentioned above, reduced CYP27B1 gene expression in the kidney(Reference Zillinger, Hustedt and Schnepel51) could be another reason for the reduction in FGF23 mRNA expression.
One signalling pathway that inhibits the expression of FGF23 in bone is the PI3K/Akt pathway. It is induced by insulin and IGF1 via their receptors’ INSR and IGF1R. Activation of the PI3K/Akt signalling pathway inhibits the transcription factor FOXO1 through phosphorylation, which leads to the down-regulation of FGF23 gene expression(Reference Bär, Feger and Fajol85). The increased expression of INSR, IGF1R, Akt1 and Akt2 in bone indicates activation of the PI3K/Akt signalling pathway by stimulated phosphorylation of FOXO1 during P-reduction in young goats, which resulted in lowered FGF23 mRNA expression. The positive correlation of INSR mRNA expression with Akt1 (Fig. 1(a)) and Akt2 (Fig. 1(b)) mRNA expression in this study indicates that an increase in INSR mRNA upregulates the expression of Akt1 and Akt2 mRNA in goats.
One pathway activating FGF23 expression is that of the JAK2/STAT3 signalling pathway. It is activated by leptin and GH through their receptors’ LEPR and GHR in bone(Reference Herrington and Carter-Su25,Reference Haussler, Whitfield and Kaneko26) . During P-restriction, it cannot be excluded that the increased LEPR expression and GHR expression could induce the JAK2/STAT3 signalling pathway despite lowered FGF23 mRNA. In a study on mice, leptin was found to reduce the expression of CYP27B1 in the kidney. This reduction was mediated by LEPR in the murine renal proximal tubules to normalise elevated serum Ca and Pi concentrations(Reference Tsuji, Maeda and Kawane23).
In summary, the prominent lower blood Pi concentration in the P-restricted goats reduced FGF23 mRNA expression in bone. Furthermore, an arrest of immature to mature osteoblastic differentiation was hypothesised during P-reduction indicating reduced bone formation. The potential reduction in bone formation was also reflected in a lower bone formation marker OC as well as a lower bone mineral content and bone mineral density. At the same time, higher OPG mRNA expression with P-restriction may indicate less inhibition of RANKL–RANK binding. Consequently, osteoclastic differentiation may increase which may result in bone resorption due to the lower Pi blood concentration. The potential increase in bone resorption was also reflected by an increase in the RANKL/OPG ratio. The increase in PTH1R mRNA expression with P-restriction may also reflect osteoblastic bone resorption. The increase in CALCR mRNA expression due to the high Ca concentrations in the P-restricted goats suggests an increase in osteoclast numbers, supporting the hypothesis of increased bone resorption. With N-restriction, blood Ca concentration was lower, which probably led to increased PTH1R gene expression in bone. This could reflect an increase in bone formation, which in turn might have stimulated bone resorption to release Ca from the bone. The increase in RANKL gene expression suggests the activation of osteoclasts. This hypothesised increase in bone turnover was also reflected in an increase in OC and CTX in the blood.
In addition to the molecular changes observed, it is important to address potential confounding factors that may have influenced our results. Future studies using pair feeding or more controlled feeding conditions are needed to differentiate the direct effects of treatment from those of reduced DM intake. IGF1 is a critical mediator in bone metabolism, influenced by nutrient intake including DM intake. The relationship between DM intake and IGF1 levels could confound the interpretation of the effects of N- and P-restrictions on bone health. This approach would help to delineate the specific contributions of nutrient restrictions v. overall dietary intake on bone health in young goats.
In conclusion, while our study highlights the critical roles of P and N in bone metabolism and the molecular mechanisms involved, it is essential to consider the potential confounding effects of DM intake and the role of IGF1 and other blood markers. Addressing these factors in future research through controlled feeding conditions will enhance our understanding of nutrient regulation of bone health and help to develop more effective dietary strategies for maintaining bone integrity in livestock.
Acknowledgements
The authors wish to thank F. Sherwood-Brock for proofreading the manuscript, Dr J. Behrens for performing the animal experiment and Y. Armbrecht and M. Rhode for taking care of the animals.
The author(s) received no financial support for the research, authorship and/or publication of this article.
Conceptualisation, funding acquisition, project administration and supervision: A.S.M.-B.; experiments: L. S. Z., A. L., K. H., N. S., M. S., C. S. and H. S.; investigation and formal analysis: L. S. Z.; validation: A. S. M-B., K. H. and N. S.; visualisation, writing – original draft preparation: L. S. Z.; data curation: A. S. M-B.; review and editing: A. S. M-B. All authors have read and agreed to the published version of the manuscript.
The authors have declared no conflict of interest.
The data were not deposited in an official repository. The data and models supporting the results of this study are available from the authors upon request.
Open access
