The animal feeding and handling protocol for this experiment were approved by the Animal Welfare Officer of the University of Veterinary Medicine Hannover, Foundation (Hannover, Germany) and complied with the German Animal Welfare Law (permit number: 33.19–42,502–04-19/3076; Lower Saxony State Office for Consumer Protection and Food Safety (LAVES); 15 March 2019).

For this molecular study, organs and blood samples were obtained from animals of a previous study (23) to avoid unnecessary repetition of animal model experiments. In brief, 28 male colored German goats from a commercial goat farm were selected for this experiment. This breed ensured uniform characteristics in terms of age, place of birth and genetic background, thus ensuring the reliability and reproducibility of the results. In addition, the same breed of goat was used in previous studies at this institute, which enables a direct comparison and continuity of the research results. The goats were initially fed a commercial milk replacer for 4 weeks, which was then slowly replaced by the control diet. After this adaptation period, the animals were divided into four different feeding groups and received (1) a control diet (16.48% crude protein (CP), 0.48% P, 1.3% Ca), (2) an N-reduced diet (8.35% CP, 0.51% P, 1.2% Ca), (3) a P-reduced diet (16.86% CP, 0.11% P, 1.2% Ca) or (4) an N- and P-reduced diet (8.1% CP, 0.11% P, 1.2% Ca). The level of urea was chosen to be underneath the threshold concentration at which the upregulation of rumen transepithelial urea transport begins. According to a previous study from our institute (24), the protein content was reduced to achieve a urea content of approximately 1 mmol/L in the blood. The amount of P in the diet was reduced to such extend that the P_i concentration in blood was approximately 1 mmol/L (25). The maximum amount of Ca in the diet for goats is suggested to be 1.5% of diet dry matter (DM) (26). In accordance with this recommendation, the Ca content in this study was adjusted to a maximum of 1.43% of DM. Individual feeding took place daily at 08.00 and 16.00 for 6 to 8 weeks. At the start of the experiment, the animals were 10 weeks old with an initial body weight (BW) of 19.04 kg (SEM 0.08) (23). A feed manufacturer specializing in feed production for animal research (ssniff Spezialdiäten GmbH, Soest, Germany) manufactured the pellets for the diets. All diets were isoenergetic, and contained approximately 12.7 MJ ME/kg DM. For further details on ingredients and chemical composition of the diets, see Weber et al. (23). Each feeding group consisted of seven animals housed together and received water ad libitum. Each animal received the amount of 55 g/kg^0.75 pelleted concentrate twice a day. In addition, 25% of the concentrate weight was fed as wheat straw. To calculate the average intake of nutrients and minerals per animal, all feed offered and refused were recorded daily. The BW of the animals was measured weekly.

Shortly before euthanasia, the goats underwent exsanguination following a standard abattoir captive bolt stunning procedure (Annex IV Directive 2010/63/EU). Blood samples were collected from the Vena jugularis using lithium-heparin syringes and serum syringes (Sarstedt AG & Co. KG, Nümbrecht, Germany). To mitigate diurnal variations, samples were consistently taken from fasting animals at the same morning time. Plasma and serum samples were centrifuged (2,000 g at room temperature, 15 min) for separation and subsequently stored at −20°C until further analysis. For technical reasons (Ussing chamber experiments), two goats per day were euthanized. To minimize potential time-related effects, animals from two feeding groups were slaughtered alternately. Within 5 min postmortem, the kidneys were removed, and tissue samples were taken from the cortex. The rib bones were dissected, the flesh was removed and then cut into cortical and medullary sections. The kidney cortices and bone sections were then rinsed with ice-cold saline (0.9% NaCl, w/v), immediately frozen in liquid N₂ and stored at −80°C until further analysis.

To determine the N, P_i and Ca status of the animals, the blood plasma concentrations of urea (commercial kit, R-Biopharm AG, Darmstadt, Germany), P_i, and Ca were measured colorimetrically using standard spectrometric methods (27, 28) [interassay CV 3.79% (N), 1.88% (Ca) and 5.05% (P_i); intraassay CV 1.2% (N), 2.85% (Ca) and 1.42% (P_i)]. For an indirect assessment of PTH levels, plasma cyclic adenosine monophosphate (cAMP) was determined using competitive ELISA (Enzo Life Sciences GmbH, Lörrach, Germany). Serum concentrations of calcidiol and calcitriol were determined commercially either by HPLC or competitive ELISA (Immundiagnostik AG, Bensheim, Germany). Plasma concentration of 24,25(OH)₂D₃ was measured using liquid chromatography–tandem mass spectrometry at the Institute for Agricultural and Food Sciences at the Martin-Luther-University Halle-Wittenberg, Halle, Germany as described elsewhere (29). To analyze plasma growth hormone (GH), serum IGF1, plasma triiodothyronine (T3) and serum magnesium (Mg) concentrations, assays were conducted at the Clinic for Cattle, Laboratory of Endocrinology, University of Veterinary Medicine Hannover, Foundation. Techniques included in-house enzyme-linked immunosorbent assay, radioimmunoassays (Beckman Coulter GmbH, Krefeld, Germany) (30) or a standardized spectrometric technique. The concentration of blood glucose was measured in whole blood samples using the mutant Q-GDH-based blood glucose monitor method with an Accu-Chek Performa blood glucose meter (Roche Diagnostics GmbH, Mannheim, Germany).

Renal cortical tissue RNA was isolated using the RNeasy Plus Mini-Kit (Qiagen GmbH, Hilden, Germany), while bone tissue RNA was isolated using the RNeasy Fibrous Tissue Mini-Kit (Qiagen GmbH). Both kits contained spin columns to eliminate genomic DNA and were used in accordance with the manufacturer’s protocol. The RNA concentrations were measured spectrophotometrically using NanoDrop One (Thermo Fisher Scientific Inc., Waltham, MA, USA). Additionally, the quality and integrity of the RNA were assessed using an RNA 6,000 nanoassay on an Agilent 2,100 Bioanalyzer (Agilent Technologies Deutschland GmbH, Böblingen, Germany). Reverse transcription of 200 ng isolated RNA was conducted using random hexamers, oligo-dT primers, and TaqMan™ Reverse Transcription Reagent (Thermo Fisher Scientific Inc.) in accordance with the manufacturer’s protocol.

To evaluate mRNA expression of NaPiIIa in renal cortical tissue, caprine gene specific TaqMan primers and probes were synthesized by TIB Molbiol Syntheselabor GmbH (Berlin, Germany; Table 1). Each reaction mixture of 20 μL contained TaqMan™ Gene Expression Master Mix (Thermo Fisher Scientific Inc.), 16 ng reverse transcript complementary DNA (cDNA), 300 nmol/L specific primers, and 100 nmol/L 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 analyzed on a real-time PCR cycler (CFX96™, Bio-Rad Laboratories GmbH, Feldkirchen, Germany). Expression levels of CYP24A1, CYP27B1, ERK1, ERK2, FGFR1c, α-Klotho, sodium-potassium adenosine triphosphatase (Na⁺/K⁺-ATPase), NaPiIIc, PiT1, PiT2 and xenotropic polytropic retrovirus receptor 1 (XPR1) in renal cortical tissue were quantified using SYBR Green® PCR assays with specific primers (Table 2). The expression of FGF23 in bone tissue was also quantified using SYBR Green® PCR assays with specific primers (Table 2). Reaction mixtures of 20 μL contained SensiFAST™ SYBR No-Rox Mix (BioCat GmbH, Heidelberg, Germany), 200 nmol/L specific primers, and 16 ng reverse transcribed cDNA. A real-time PCR cycler (Bio-Rad Laboratories GmbH) was used for amplification (3 min at 95°C; 40 cycles of 10 s at 95°C and 30 s at 60°C) and detection of PCR products. To determine the melting curve, the thermal profile began with an incubation of 10 min at 55°C with a gradual increase in temperature (0.5°C per 10 s) up to 95°C. Absolute copy numbers were determined using calibration curves generated with cloned PCR fragment standards (35). Specificity of amplicons was verified by sequencing (Microsynth Seqlab GmbH, Göttingen, Germany) and using NCBI Blast¹. For renal cortical tissue, the reference gene 18S rRNA was synthesized by TIB MOLBIOL (Table 2). Additionally, reference genes beta-2-microglobulin (B2M), ribosomal protein L19 (RPL19), and ribosomal protein S9 (RPS9) were quantified with SYBR Green® PCR assays (Table 2). Among these, RPL19 was identified as the most suitable reference gene for normalization. In bone tissue, the reference genes 18S, RPL19 and RPS9 were synthesized as described. In bone tissue, 18S rRNA was identified as the optimal reference gene for normalization. Gene expression was quantified with SYBR Green® PCR assay as described above. NormFinder software² was used to determine the best reference gene for normalization. Each reaction was run twice and included water as a no-template control to ensure accuracy and reliability.

To quantify the expression of NaPiIIa and PiT2, brush border membranes (BBM) were prepared in accordance with the method of Wilkens et al. (35). Cytosolic preparation was required for quantification of ERK1/2, whereas crude membranes (CM) were isolated for quantification of Na⁺/K⁺-ATPase, both in accordance with the method of Wilkens et al. (36). Total cell lysate preparation was required to determine CYP27B1 and FGFR1 protein expression in accordance with the method of Weber et al. (23). Protein concentrations of all cell fractions were measured using a commercial Bradford assay (Serva Elektrophoresis GmbH, Heidelberg, Germany). For detection of NaPiIIa protein expression, 15 μg of BBM was separated by 8.5% SDS-PAGE and transferred to nitrocellulose membranes (GE Healthcare Europe GmbH, Freiburg, Germany). The protein was incubated overnight with anti-NaPiIIa polyclonal antibody (Alpha Diagnostic Intl. Inc., San Antonio, TX, USA), diluted 1: 300 in phosphate-buffered saline (PBS) with 0.1% Tween 20 (PBST) and 2.5% non-fat milk powder at 4°C overnight. Immunoreactivity of the primary antibody was detected with an HRP-coupled secondary anti-rabbit antibody (diluted 1: 20,000). The specificity of the NaPiIIa antibody was successfully demonstrated by preincubation of the primary antibody with the corresponding antigenic peptide (data not shown). For detection of PiT2 protein expression, 25 μg of BBM was separated by 8.5% SDS-PAGE and transferred to nitrocellulose membranes (GE Healthcare Europe GmbH). Overnight incubation with a polyclonal anti-PiT2 antibody (self-made, Davids Biotechnologie GmbH, Regensburg, Germany, gift from Prof. V. Sorribas, Laboratory of Molecular Toxicology, University of Zaragoza, Spain) was performed, diluted 1: 1,000 in PBST. An HRP-coupled secondary anti-rabbit antibody (diluted 1: 2,000) was used to detect immunoreactivity of the primary antibody. To quantify ERK1/2 protein expression, 10 μg of cytosol was separated by 8.5% SDS-PAGE and transferred to nitrocellulose membranes (GE Healthcare Europe GmbH). To determine the level of phosphorylation, a primary anti-ERK1/2 (Cell Signaling Technology Europe B. V., Frankfurt am Main, Germany) was first incubated overnight and diluted 1: 2,000 in 5% bovine serum albumin (BSA) buffer and tris-buffered saline (TBS) with 0.1% Tween 20 (TBST). Immunoreactivity of the phosphospecific antibody was detected using an anti-rabbit antibody (diluted 1:2,500). The antibody was then removed by incubating the membrane with stripping buffer (pH 2.0). The membrane was blocked and incubated again overnight with the non-phosphospecific primary anti-ERK1/2 antibody (cell signaling), diluted 1:1,000 in 5% BSA and TBST. Immunoreactivity of the primary antibody was detected with anti-rabbit antibody (diluted 1: 2,500). The monoclonal ERK1/2 antibody recognizes bovine ERK1/2. For Na⁺/K⁺-ATPase, 10 μg of CM were separated by 8.5% SDS-PAGE and transferred onto nitrocellulose membranes (GE Healthcare Europe GmbH). To detect the Na⁺/K⁺-ATPase protein, the membranes were incubated overnight at 4°C with an anti- Na⁺/K⁺-ATPase antibody (Enzo Life Sciences GmbH) diluted 1: 10,000 in TBST and 5% non-fat milk powder. Immunoreactivity of the primary antibody was detected with an HRP-coupled anti-mouse antibody (diluted 1: 20,000). The monoclonal Na⁺/K⁺-ATPase antibody recognizes sheep and bovine Na⁺/K⁺ ATPase. For the quantification of CYP27B1, 10 μg of cell lysate was used. Membranes were incubated overnight with caprine anti-CYP27B1 (self-made, Davids Biotechnologie GmbH), diluted 1:500 in TBST and 2.5% non-fat milk powder. An HRP-coupled anti-rabbit antibody (diluted 1: 20,000) was used to detect the primary antibody. The specificity of the goat CYP27B1 antibody was successfully demonstrated by preincubation of the primary antibody with the corresponding antigenic peptide (data not shown). For the determination of FGFR1 protein, 25 μg of cell lysate was separated by 8.5% SDS-PAGE and transferred onto nitrocellulose membranes (GE Healthcare Europe GmbH). The protein was incubated overnight with a polyclonal anti-FGFR1 antibody (Thermo Fisher Scientific Inc.) 1:1,000 in TBST and 5% BSA. The primary antibody was detected with an HRP-coupled anti-rabbit antibody (diluted 1:2,500). The primary antibodies used were either specific to the particular caprine protein or exhibited species cross-reactivity for caprine, ovine or bovine target tissue. The protein sequences were verified for their homology with NCBI Blast (see text footnote 1). Immunoblot assays detecting the expression of ERK1/2, NaPi IIa and Na⁺/K⁺ ATPase proteins in caprine tissue were performed as described elsewhere (5, 37). Membranes were incubated solely with the secondary antibody to ensure exclusive detection of specific signals. For the detection of the specific proteins, a chemiluminescence system and the ChemiDoc MP (Bio-Rad Laboratories GmbH) were used. Densitometric measurements of proteins were performed using Image Lab 5.2.1 software (Bio-Rad Laboratories GmbH). For semi-quantification, the values of the quantified specific proteins were normalized to the amount of total protein stained with Indian ink. This normalization process aids in providing relative measurements of protein expression levels.

The sample size (n 7/group) was determined based on metabolic data from a previous study (5) with a statistical power of 0.8 and α error of 0.05. All data are given as means with their standard errors (SEM) unless otherwise stated. For data analysis, GraphPad Prism version 9.3 (GraphPad Software, San Diego, CA, USA) was used. All data were tested for normal distribution and were analyzed by two-way ANOVA with Tukey’s multiple comparisons test. p < 0.05 was considered significantly different, and p < 0.1 was used to define trends. Potential linear relationships between mRNA and protein expression levels, as well as correlations between gene and protein expression with blood metabolites, were calculated, using a simple correlation analysis with Pearson’s correlation coefficient. Non-linear relationships were calculated using Spearman’s correlation coefficient. Furthermore, potential outliers were tested using the ROUT method (38).

All animals were clinically healthy throughout the study. The clinical health of the animals was ensured through monitoring and care by a qualified veterinarian that examined the animals daily and closely monitored their behavior and physical condition. The data on feed intake, daily weight gain and BW of the goats have been previously published (23). The data on feed intake are listed in Table 3. Daily BW gain was lower by 79% in the P-restricted group and by 64% in the N-/P-restricted group, resulting in final BW reductions of 20 and 16%, respectively, compared to the control group. The DM intake diminished by about 14% in the P- and 13% in the N-/P-groups. The concentrate intake diminished by about 14% in the P- and the N-/P- groups. The BW of goats in the N-restricted group remained unaffected. N-intake was lower by about 13% with P-restriction, 50% with N-restriction, and 56% with combined N- and P-restriction. Ca-intake was lower by 17% with P-restriction and 20% with N-/P-restriction. P-intake dropped by about 78% in both the P-restricted and N-/P- restricted groups.

Blood metabolites are shown in Table 4. Some of these blood metabolites have already been published in Weber et al. (23). In the N-restricted feeding groups, blood plasma concentrations of urea were lower by about 83%. The P-reduced feeding groups showed significantly reduced blood plasma concentrations of P_i by about 67%, with concomitant higher blood Ca concentrations by about 37%. The glucose concentration in whole blood and the T3 concentration in plasma were consistent across all feeding groups (23). Plasma concentrations of cAMP was lower with N-restriction by trend and remained unaffected by P-restriction. Serum levels of calcidiol and calcitriol were not affected by N-reduction or P-reduction alone, but there was a significant interaction between N and P on calcitriol concentration. Blood plasma concentrations of 24,25(OH)₂D₃ were lower with the P-reduced diet, whereas an N-reduction had no effect on them. Plasma concentrations of GH were not affected by diet, whereas serum IGF1 concentrations tended to be lower with N-reduction and were significantly lower with P-reduction. Serum concentrations of Mg were higher with P-reduction and diminished by N-reduction.

The average RNA integrity number (RIN) of the kidney cortex tissue samples was 9.14 (SEM 0.12). The results of all genes and corresponding proteins expressed in the renal cortex are shown in Tables 5, 6. CYP27B1 mRNA expression was significantly reduced due to both N-reduction and P-reduction, as well as the interaction between N and P. Protein expression of CYP27B1 was lower in the N-restricted feeding group compared to the other feeding groups. Furthermore, CYP27B1 mRNA expression was negatively correlated with blood Ca concentration (r = −0.47, p = 0.014; Figure 1). FGFR1c mRNA and protein expression were significantly lower in the P-reduced feeding groups. FGFR1 protein expression correlated positively with plasma concentration of P_i (r = 0.42; p = 0.025; Figure 2) and negatively with plasma concentration of Ca (r = −0.41; p = 0.031; Figure 3). Furthermore, Na⁺/K⁺-ATPase mRNA expression was significantly lower in both groups receiving an N-reduced diet, while its protein expression remained unchanged. Na⁺/K⁺-ATPase mRNA expression was positively correlated with its protein expression. Although, NaPiIIa gene expression was not affected, NaPiIIa protein expression was higher in the P-reduced animals and tended to be higher in the N-reduced animals. The gene expression of NaPiIIc was lower by trend with P-restriction. Protein expression of ERK1/2 did not change, while pERK1/2 and the ratio of pERK1/2 to ERK1/2 showed a trend toward diminished expression in the P-reduced animals. Neither ERK1 nor ERK2 gene expression changed with different diets. Protein expression of ERK1/2 positively correlated with gene expression of FGFR1c (r = 0.41; p = 0.029; Figure 4). Expression of PiT2 mRNA showed a trend toward higher expression in the N-restricted feeding groups. However, protein expression of PiT2 was not affected by any of the feeding groups. The mRNA expression of CYP24A1, α-Klotho, PiT1 and XPR1 mRNA did not differ between the feeding groups.

The average RIN of the bone tissue samples was 8.69 (SEM 0.12). The mRNA expression of FGF23 is presented in Table 7. The FGF23 mRNA expression was significantly lower with P-restricted feeding.