Abstract

Background

   There is an emerging body of evidence that current poultry feed is
   formulated in excess for phosphorus (P), which results in unnecessarily
   high P excretions. Sustainable concepts for agricultural P flows should
   trigger animal-intrinsic mechanisms for efficient P utilization. In the
   current study, Lohmann Brown (LB) and Lohmann Selected Leghorn (LSL)
   laying hens were fed either a high P diet (P+) with 1 g/kg mineral P
   supplement or a low P diet (P-) with 0 g/kg mineral P supplement for a
   period of 4 weeks prior to sampling. Before and after onset of laying,
   i.e., at 19 and 24 weeks of life, kidney and plasma samples were
   collected to investigate the endogenous P utilization in response to
   restricted dietary P, laying hen strain, and sexual maturation.

Results

   Plasma analyses of minerals and metabolites confirmed the response to a
   low P diet, which was characterized by a significant reduction in
   plasma P levels at week 19 in both strains. The plasma calcium (Ca)
   levels were tightly regulated throughout the entire experimental
   period. Notably, there was a numerical trend of increased plasma
   calcitriol levels in P- fed birds of both strains compared to the
   P + group, which might have mediated a substantial role regarding the
   adaptive responses to low P supply. At week 19, RNA sequencing of
   kidney identified 1,114 and 556 differentially expressed genes (DEGs)
   unique to the LB and LSL strains, respectively. The number of DEGs
   declined with increasing maturity of the hens culminating in 90 and 146
   DEGs for LB and LSL strains at week 24. Analyses revealed an enrichment
   of pathways related to energy metabolism and cell cycle, particularly
   at week 19 in both strains. The diet-specific expression of target
   genes involved in P homeostasis highlighted transcripts related to
   active (SLC34A1, SLC20A2) and passive mineral transport (CLDN14,
   CLDN16), Ca utilization (STC1, CALB1), and acid-base balance (CA2,
   SLC4A1).

Conclusions

   Results suggest that both laying hen strains adapted to the lack of
   mineral P supplements and achieved a physiological Ca: P-ratio in body
   compartments through endogenous regulation as evidenced via the
   endocrine profile.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12864-025-11294-6.

   Keywords: Kidney, Laying period, Mineral homeostasis, Poultry,
   Transcriptomics

Background

   Egg production has reached an industrial scale through the use of
   high-performance laying hen breeds such as Lohmann Brown (LB) and
   Lohmann Selected Leghorn (LSL). Dietary mineral management,
   particularly of phosphorus (P), is crucial for performance and animal
   health in these laying hens [[38]1]. P is biologically available to
   laying hens in the form of non-phytate P, whereas about 60% of plant P
   in the form of phytate is not available for direct enteral absorption
   because of insufficient intestinal phytase activity [[39]1, [40]2].
   Therefore, the use of mineral P sources and phytases of microbial
   origin is very common to compensate for the low rate of digestible P
   obtained from cereals and other plant-based feed. Current dietary
   recommendations in laying hens suggest 2.2 g/kg non-phytate P (NPP)
   [[41]3], however, these levels exceed the required amount for achieving
   optimal performance and jeopardize efforts to reduce the environmental
   P emissions [[42]4]. Current approaches have monitored experimental hen
   populations for intestinal phytate cleavage and resulting myo-inositol
   and inositol phosphate isomer levels in the small intestine following
   two variable P diets (5.3 vs. 4.7 g/kg total P) and suggested that a
   20% reduction in currently recommended P levels could be achieved in
   commercial laying hens [[43]5]. A feasible potential to reduce dietary
   mineral P levels might still be larger, as a reduction in available P
   from 491 to 168 mg/hen/d in hens’ feed did not affect the egg
   production and health parameters but lowered the amount of excreted P
   in laying hens [[44]1].

   The adaptability of poultry to variations in dietary P levels has been
   attributed to their body P homeostasis network [[45]6], maintained
   through a complex interaction of multiple tissues (intestines, kidney,
   liver, bone), active and passive mineral transport, as well as feedback
   mechanisms due to endocrine regulators [[46]7–[47]9]. These findings
   were evident in broiler chickens and laying hens, where recent reports
   on low-P diets found regulatory changes at the level of the endocrine
   system and gene expression in the small intestine [[48]5, [49]10,
   [50]11] and kidney [[51]12]. Sexual maturation (onset of laying),
   commencing around the 19th week of life and extending beyond the 24th
   week, is characterized by an increased mineral demand in laying hens
   due to eggshell production, which links the P metabolism particularly
   closely with the calcium (Ca) metabolism [[52]13, [53]14]. In this
   context, besides intestinal tissues, the contribution of
   post-absorptive tissues (i.e. kidney) to mineral utilization is of
   relevance.

   The renal mineral transport includes para- and transcellular processes,
   e.g. facilitated by SLC34A1, which encodes the type 2a sodium-phosphate
   cotransporter (NPt2a) in chickens [[54]12, [55]15], components of tight
   junctions such as claudin 14 (CLDN14), and intracellular carriers such
   as calbindin (CALB1). Approximately 80% of the absorption of P occurs
   in the proximal renal tubules through apically localized Na-dependent P
   cotransporters [[56]16]. Moreover, proximal tubular Na/P cotransport is
   enhanced by stanniocalcin 1 (STC1) [[57]17]. In mammals, parathyroid
   hormone (PTH) regulates Ca homeostasis by altering the expression of
   the renal Ca channel TRPV5 in the distal convoluted tubule and
   collecting duct [[58]18]. In addition, it causes a decreased abundance
   of renal NPt2a in the apical membrane of the proximal tubule [[59]16].
   Increased blood Ca levels suppress the synthesis and secretion of PTH
   from the parathyroid glands by binding to and activating the Ca-sensing
   receptor (CaSR). CaSR antagonizes PTH effects in proximal cells by
   inhibiting cAMP production and promotes P reabsorption by NaPt2 in the
   proximal tubule. Furthermore, PTH acts on renal 1-α hydroxylase to form
   calcitriol (active form of vitamin D3; 1,25-(OH)2D3) from calcidiol
   (storage form of vitamin D3; 25(OH)D3) in the kidney [[60]19]. These
   interactive mechanisms modulate bone resorption, Ca and P absorption in
   the intestine and reabsorption in the kidney to maintain physiological
   Ca and P concentrations in the circulation [[61]8, [62]20].

   High-performance laying hen strains of LB and LSL genotypes are
   characterized by a considerable genetic differentiation [[63]21,
   [64]22], which culminates in distinct differences in phenotypic traits
   such as body weight and egg quality traits, including egg weight and
   eggshell weight. In fact, the underlying cause for the observed
   variations can be attributed to divergent breeding histories [[65]23].
   Subsequent breeding programs for both LB and LSL strains primarily
   focused on traits such as feed efficiency and laying performance,
   resulting in a similarly high egg production capacity of the two
   strains. According to recent findings, LB and LSL exhibit significantly
   differences in endogenous responses towards improved mineral
   utilization via enhanced intestinal capacity [[66]5], jejunal microbial
   composition [[67]24], jejunal gene expression [[68]25], immune traits
   [[69]26], and endocrine drivers [[70]11]. Therefore, we hypothesize
   that high-yielding laying hens, when fed a low P diet at the onset of
   egg production, initiate endogenous adaptive mechanisms in
   post-absorptive tissues such as the kidneys to maintain physiological
   mineral equilibrium. To address this, the effects of a low P diet
   induced by a 4-week removal of mineral P sources from the diet compared
   to a diet containing mineral P on renal gene expression of LB and LSL
   laying hens before (week 19) and after the onset of laying (week 24)
   were investigated. Furthermore, the study examined the impact of diet,
   strain, and maturation on plasma metabolites relevant to mineral
   utilization.

Materials and methods

Ethics and consent to participate

   This study was part of the interdisciplinary Research Unit P-Fowl:
   Inositol phosphates and myo-inositol in the domestic fowl: Exploring
   the interface of genetics, physiology, microbiome, and nutrition
   ([71]https://p-fowl.uni-hohenheim.de/). Newly hatched female chickens
   were retrieved from a breeding company (Lohmann Breeders GmbH,
   Cuxhaven, Germany). The informed consent of the owner(s) for the use of
   the animals in the study has been provided. The animal trial was
   performed at the Agricultural Experimental Station of the University of
   Hohenheim, Germany. The experimental protocol was in strict compliance
   with the German Animal Welfare Legislation and approved by regional
   Ethics committee, i.e., the Regierungspräsidium Tubingen, Germany, and
   the Animal Welfare Committee of the University of Hohenheim (Project
   no. HOH67-21TE).

Experimental setup and laying hens

   As reported previously, both LB (n = 40) and LSL (n = 40) hens were
   supplied with experimental maize-soybean meal-based diets having all
   required nutrients in recommended doses, except P [[72]27]. The high P
   diet (P+) was supplemented with 1 g P/kg. The total P concentration in
   the P + diet was 4.9 g/kg dry matter (DM) for developer feed, 5.0 g/kg
   DM for pre-layer feed, and 4.9 g/kg DM for layer feed. The experimental
   low P diet (P-) lacked any mineral P supplement and the respective
   total P content was 3.7 g/kg DM for developer feed, 3.8 g/kg DM for
   pre-layer feed, and 3.6 g/kg DM for layer feed. The respective
   non-phytate P (NPP) values in the P-supplemented group were calculated
   as 2.3 g/kg (developer feed), 2.4 g/kg (pre-laying feed), and 2.3 g/kg
   (layer feed) to match current NPP recommendations [[73]27, [74]28]. All
   diets lacked phytase supplements. Animals sampled at week 19 received
   developer feed (week 15–16), pre-layer feed (week 16–17), and layer
   feed (week 17–19) according to their affiliated dietary group; animals
   sampled at week 24 received respective layer diets from week 20–24
   (Table [75]S1). The hens were fed the experimental diets for a period
   of four weeks before slaughtering in week 19 (n = 40) and week 24
   (n = 40), respectively, where kidney (n = 80) and plasma samples
   (n = 80) were collected, i.e., 10 hens per strain per sampling week per
   diet. The feed composition remained consistent at both sampling stages
   to ensure that all sampled birds received layer feed. Hens were killed
   from 09:00 h to 15:00 h through individual stunning with a gas mixture
   (35% CO[2], 35% N[2], 30% O[2]) and decapitation. Trunk blood was
   collected in lithium-heparin tubes, centrifuged, and the prepared
   plasma samples were stored at − 80 °C until analyses. Whole kidneys
   were collected approximately 5 min postmortem. The tissue was chopped
   into small pieces and frozen at − 80 °C until analyses. The number of
   eggs laid until slaughter was documented for all animals. Until week
   24, the average number of eggs produced were 10.0 ± 5.8 for LB and
   12.0 ± 3.5 for LSL. A total of 3 laying hens from the LB strain
   affiliated with both P- and P + diet groups showed no egg production
   until week 24 (i.e. delayed sexual maturation or infertility) and were
   excluded from the subsequent plasma analyses.

Measurement of plasma parameters

   The plasma samples were analyzed for inorganic P, total Ca, magnesium
   (Mg), albumin, triacylglycerides, and alkaline phosphatase activity
   (ALP) using the Fuji DriChem 4000i commercial assays (FujiFilm, Minato,
   Japan). Commercially available enzyme-linked immunosorbent assays
   (ELISA) were used to quantify hormones in duplicate according to the
   manufacturer’s instructions. The measured hormones were 25(OH) vitamin
   D (EIA-5396, DRG, Marburg, Germany), 1,25(OH) vitamin D (AC-62F1,
   Immunodiagnostic Systems GmbH, Frankfurt am Main, Germany), and PTH
   (CSBE118880Ch, CusaBio, Houston, USA). The raw data were processed
   according to 4-parameter logistic curve analysis with MARS data
   analysis software (BMG Labtech, Ortenberg, Germany). For data analysis,
   a linear model was applied including sampling week, diet group, strain,
   and interactions as fixed effects (R, version 4.3.0, nmle package
   v3.1-160). The statistical model was structured as follows:

   y[ijk] = µ + strain[i] + week[j] + diet[k] + (strain × week ×
   diet)[ijk]+ ϵ[ijk].

   Where y[ijk] is the response variable, µ is the overall mean, strain[i]
   represents the effect of genotype (LB, LSL), week[j] represents the
   effect of week (wk19, wk24), and diet[k] represents the effect of
   dietary P level (P-, P+), and ϵ[ijk] is the residual error. The ANOVA
   and TukeyHSD functions were applied to analyze the differences among
   means between different experimental groups. The significance cut-off
   was set at a P-value < 0.05. The results were displayed in graphical
   form using R (ggplot2 package v3.3.6).

RNA extraction and processing of sequencing data

   Before RNA extraction, whole frozen kidney samples were ground with
   pestle and mortar while submerged in liquid nitrogen. About forty grams
   of each kidney sample were used for RNA isolation using TRI reagent
   (Sigma-Aldrich, Taufkirchen, Germany) and phenol-chloroform-based RNA
   extraction according to the TRI reagent protocol. To ensure the absence
   of DNA, the isolated RNA was treated with DNase I (Roche Diagnostics,
   Mannheim, Germany) and finally purified using the NucleoSpin RNA
   extraction kit (Macherey-Nagel, Düren, Germany). The RNA concentration
   was determined through spectrophotometry using the NanoDrop ND-2000
   (Peqlab, Erlangen, Germany) and RNA quality of final RNA was analyzed
   with the Bioanalyzer 2100 device (Agilent Technologies, Waldbronn,
   Germany). The range of RNA integrity numbers (RIN) was between 6.4 and
   8.9, with an average of 7.8. The absence of genomic DNA contamination
   was confirmed with polymerase chain reaction (PCR) specific for a
   chicken GAPDH fragment [[76]12]. RNA libraries were prepared using the
   Illumina Stranded mRNA preparation kit (Illumina, San Diego, CA, USA)
   and then sequenced using the Illumina NextSeq 2000 platform in a
   paired-end configuration. Quality control and preprocessing of the raw
   sequencing reads were conducted using FastQC (v0.11.7) and Trim Galore
   (v0.5.0; [77]https://www.bioinformatics.babraham.ac.uk/projects/).
   Low-quality reads with mean Q-score less than 20 and reads length
   shorter than 30 base pairs were excluded. The remaining reads were
   aligned to the chicken genome assembly (GRCg6a, Ensembl release 109)
   using Hisat2 (v2.1.0; [78]http://daehwankimlab.github.io/hisat2/).
   HTseq (v0.11.2) was used to summarize read counts to determine the
   expression levels of each gene. Data are available from the EMBL-EBI
   database ([79]www.ebi.ac.uk/arrayexpress) via the accession number
   E-MTAB-14,436. Each of the 80 kidney samples yielded an average of
   22.06 ± 6.80 million paired end reads for downstream analysis.
   Following analyzing the dataset using the arrayQualityMetrics package
   (R, v3.54), 75 samples were used with 5 outliers removed (2 birds from
   LB P + group at week 19, 1 from LSL P + at week 19, and 2 from LSL P-
   group at week 19). In the project consortia, a number of candidate
   genes from an RNAseq run were validated previously on a LightCycler 480
   device (Roche, Basel, Switzerland), demonstrating high consistency
   between RNAseq data and qRT-PCR results [[80]29].

Gene expression analysis

   The count data were passed through strict filtration with 10 or more
   counts in at least 8 samples for subsets representing strain × stage.
   The gene expression analysis was performed with the DESeq2 package in R
   (v1.38.3) per strain and stage to elucidate dietary effects using the
   Wald test. A size factor normalization was applied to account for
   sequencing depth differences between samples. The statistical model
   comprised strain (LB, LSL), week (wk19, wk24), and diet group (P-, P+)
   as fixed effects. Analysis of the applied P diets considered the
   following four contrasts, whose labels were derived from strain and
   week and diet: (i) LBwk19P+ (n = 8) vs. LBwk19P- (n = 10), (ii)
   LSLwk19P+ (n = 9) vs. LSLwk19P- (n = 8), (iii) LBwk24P+ (n = 10) vs.
   LBwk24P- (n = 10), and (iv) LSLwk24P+ (n = 10) vs. LSLwk24P- (n = 10).
   For each of the contrasts, differentially expressed genes (DEGs) at
   P-value < 0.01 were considered for further analyses. Venn diagrams were
   created for each applied contrast to display the DEGs using the online
   tool Venny ([81]https://bioinfogp.cnb.csic.es/tools/venny/).

Functional annotation and pathway enrichment analysis

   The chicken Ensembl-IDs were converted to gene symbols using the online
   tool bioDBnet: db2db
   ([82]https://biodbnet-abcc.ncifcrf.gov/db/db2db.php). For data
   enrichment analysis, DEGs with P < 0.01 were selected and analyzed
   through the online tool g: profiler
   ([83]https://biit.cs.ut.ee/gprofiler/gost). Gene Ontology (GO: BP) and
   Kyoto Encyclopedia of Genes and Genomes (KEGG) were used considering
   maximal term sizes of 500. The KEGG pathway map was visualized using
   the R package ggplot2 (v3.3.6). DEGs were presented by volcano plots
   using VolcaNoseR ([84]https://huygens.science.uva.nl/VolcaNoseR). Genes
   with a -log[10] transformed P-value above 2 were considered as
   functional targets to drive renal adaptation due to reduced dietary P
   intake. In addition, a target genes approach was used to identify
   transcripts involved in P, Ca, myo-inositol, and vitamin D regulation.
   In total, 50 target genes from previous studies (mineral transport
   [[85]15, [86]30–[87]33]; vitamin D metabolism [[88]34]; myo-inositol
   transporters [[89]35]) were selected and compared with DEGs from the
   retrieved expression data.

Results

Plasma hormones and metabolites

   For dietary effects, hens fed P- diets exhibited a lower plasma
   inorganic P concentration in both strains at week 19, whereas at week
   24, only LB hens showed significantly decreased inorganic P plasma
   levels in response to the P- diet (Fig. [90]1). The ALP activity was
   significantly increased in hens fed the P + diet compared to hens fed
   the P- diet at week 24 in LSL strain. No significant differences in ALP
   activity were observed in the LB strain at both sampling stages.
   Dietary P supply did not yield any significant effects on the plasma
   concentrations of total Ca, Mg, calcidiol, calcitriol, albumin and
   triacylglycerides for both LSL and LB strains at weeks 19 and 24.
   However, plasma calcitriol values showed a consistent pattern according
   to the dietary P supply with numerically increased values in P- animals
   of both laying hen strains.

Fig. 1.

   [91]Fig. 1
   [92]Open in a new tab

   Plasma parameters related to mineral homeostasis in LB and LSL hen
   strains before (week 19) and after (week 24) onset of egg laying
   following a high P diet supplemented with 1 g/kg mineral P (P+; dark
   blue) or a low P diet lacking any mineral P supplements (P-, light
   blue). Values are presented as mean ± SE. ^a, b,c, d Different
   superscripts indicate statistical differences between the eight
   experimental groups, i.e., 2 strains × 2 diets × 2 stages (P < 0.05)

   A strain effect was observed on plasma calcidiol levels with
   significantly lower concentrations in the LSL strain compared to the LB
   strain at both weeks 19 and 24, irrespective of dietary P supply. Other
   analyzed metabolites did not consistently differ between the strains.

   Regarding effects of the sampling stage, i.e. sexual maturation,
   inorganic P levels were significantly lower, while total Ca and
   triacylglycerides were higher at week 24 compared to week 19 for both
   strains and diets. Levels of plasma Mg and albumin were higher at week
   19 compared to week 24 in both P + and P- diet groups in LSL but not in
   LB laying hens. Plasma PTH levels were higher in week 19 compared to
   week 24 in LSL hens fed the control P + diet. In contrast, ALP activity
   levels were higher in week 24 compared to week 19 only in LSL hens
   within the control group (P + diet). No consistent effects of sampling
   stage were observed on plasma concentrations of calcidiol, calcitriol,
   and PTH. Even though plasma calcitriol showed no significant
   differences, there were consistent patterns among all hen groups with
   numerically higher calcitriol levels at week 24 compared to week 19.
   The statistical evaluation of the fixed effects considered in the
   linear model showed that the dietary P supply significantly influenced
   plasma concentrations of inorganic P (P < 0.001) and PTH (P = 0.035).
   Strain exhibited a significant effect on plasma calcidiol (P < 0.001)
   and PTH (P = 0.005), whereas sampling stage showed a significant effect
   on plasma inorganic P (P < 0.001), total Ca (P < 0.001), calcitriol
   (P = 0.035) and triacylglycerides (P < 0.001). Interaction effects of
   diet × strain × sampling stage were observed for plasma calcidiol
   (P = 0.033), PTH (P = 0.007), and ALP activity (P = 0.019).

Differential gene expression analysis

   The gene expression analysis in the kidney of LB and LSL laying hen
   strains at week 19 and 24 kept on different dietary P supply (P + vs.
   P-) revealed 1,114 and 556 DEGs specific to the LB strain and the LSL
   strain, respectively, at week 19 (Fig. [93]2A). At week 24, 90 DEGs
   were observed specifically in LB and 146 DEGs specifically in LSL
   (Fig. [94]2B). The overlap of 37 and 1 DEGs was observed being commonly
   regulated due to diet between the two strains at week 19 (Fig. [95]2A)
   and week 24 (Fig. [96]2B), respectively. Respective gene lists of the
   various comparisons of the dietary groups per sampling stage are
   provided in Table [97]S2 (LBwk19P + vs. LBwk19P-), Table [98]S3
   (LSLwk19P + vs. LSLwk19P-), Table [99]S4 (LBwk24P + vs. LBwk24P-), and
   Table [100]S5 (LSLwk24P + vs. LSLwk24P-).

Fig. 2.

   Fig. 2
   [101]Open in a new tab

   Venn diagrams of differentially expressed genes (DEGs) in LB (blue) and
   LSL (yellow) hens under low P diet (P-) compared to a high P diet (P+)
   at week 19 (A) and week 24 (B)

   Volcano plots depicting gene expression data revealed distinct patterns
   in the LB and LSL strains at weeks 19 and 24 (Fig. [102]3). Before the
   onset of lay at week 19, the LB strain exhibited an upregulation
   (P + < P-) of CALB1, CLDN14, and STC1, among others, due to dietary P
   reduction, along with a downregulation (P + > P-) of MAP2K6
   (Fig. [103]3A). In the LSL strain, functionally interesting candidate
   genes such as STC1 were upregulated (P + < P-), while CLDN16 and OCM4
   were notably downregulated (P + > P-) (Fig. [104]3B). Following the
   onset of lay at week 24, the impact of the applied mineral P levels on
   renal gene expression was generally subtle. Regarding relevant
   functional candidate genes, in the LB strain, there was a significant
   upregulation (P + < P-) of SLC5A1 and downregulation (P + > P-) of CA4
   and GREB1 (Fig. [105]3C), whereas the LSL strain demonstrated
   significant downregulation (P + > P-) of AQP7, MYOT, CLDN14 and MAPK12
   (Fig. [106]3D).

Fig. 3.

   [107]Fig. 3
   [108]Open in a new tab

   Volcano plots showing renal gene expression in comparison of a low P
   diet (P-) and a high P diet (P+) in LB (A) and LSL (B) laying hens at
   week 19, and in LB (C) and LSL (D) laying hens at week 24. Significant
   differentially expressed genes (DEGs) are depicted as red (P + < P-)
   and blue (P + > P-) dots. Gray dots remained unchanged due to dietary P
   supply

Gene ontology and pathway enrichment analysis

   KEGG pathway analysis did not provide pathways related to P
   homeostasis, however, an enrichment of pathways related to energy
   metabolism and the cell cycle was observed in both strains at week 19.
   Pathways pertinent to oxidative phosphorylation, phosphatidylinositol
   signaling system, and C-type lectin receptor signaling were highlighted
   in the LB strain. The LSL strain showed functional enrichment of genes
   associated with the metabolism of xenobiotics by cytochrome P450, cell
   cycle, pentose and glucuronate interconversions, and drug metabolism -
   cytochrome P450 (Table [109]1). No enriched pathways were found at week
   24 for both LB and LSL laying hen strains.

Table 1.

   KEGG pathway analysis of differentially expressed genes (DEGs) with
   respect to strain (LB, LSL), dietary P supply (P+, P-), and stage
   (wk19, wk24)
   Contrast of experimental groups Pathway Gene count adjusted P-value
   LBwk19P + vs. LBwk19P- Oxidative phosphorylation 26 < 0.001
   Phosphatidylinositol signaling system 17 0.019
   C-type lectin receptor signaling pathway 16 0.038
   LSLwk19P + vs. LSLwk19P- Metabolism of xenobiotics by cytochrome P450 7
   0.010
   Cell cycle 15 0.037
   Pentose and glucuronate interconversions 5 0.049
   Drug metabolism - cytochrome P450 6 0.047
   LBwk24P + vs. LBwk24P- - - -
   LSLwk24P + vs. LSLwk24P- - - -
   [110]Open in a new tab

Diet-specific expression of target genes involved in P homeostasis

   Based on the RNAseq dataset, a target gene approach was used to
   identify specific transcriptional responses linked with mineral
   utilization. Table [111]2 and Table [112]S6 display the functional
   candidate genes which were upregulated (P + < P-) or downregulated
   (P + > P-) according to the RNAseq expression data in LB and LSL
   strains at weeks 19 and 24. At week 19, the genes STC1, CALB1, CA2 and
   CLDN14 were upregulated in both LB and LSL when comparing the different
   P diets (P + < P-). The gene encoding the P transporter SLC34A1 was
   significantly upregulated, while SLC20A2, SLC34A2 and other solute
   carriers such as SLC2A13 were significantly downregulated in LB strain
   at week 19 (P + > P-). Significant upregulation of SLC4A1, SLC8A1,
   SLC41A2, TRPV6 and TRPM7 and downregulation of CLDN16 was observed in
   LSL at week 19. CLDN10, AQP11 and SLC41A3 was upregulated, while TJP1
   and ATP2B4 were downregulated in LB strain at week 19. The CALCRL gene
   linked to Ca control was significantly downregulated in LB at week 19.

Table 2.

   Target genes with functions and their expression in LB and LSL laying
   hen strains following variable dietary P supply at week 19 and week 24.
   The negative log[2]-FoldChange values indicate lower mRNA abundances
   (P + > P-) and the positive log[2]-FoldChange values indicate higher
   mRNA abundances (P + < P-) in animals kept on low P diets (P-) compared
   to controls (P+)
   LB LSL
   Gene ID Gene Name Expression (baseMean) log[2]-Fold change P-value
   log[2]-Fold change P-value Functional category
   Week 19
   AQP11 Aquaporin 11 871.7 0.40 0.009 0.05 0.701 water channel activity
   ATP2B4 ATPase plasma membrane Ca2 + transporting 4 241.1 -0.38 0.027
   0.03 0.875 calcium transport
   CA2 Carbonic anhydrase 2 6114.9 0.24 0.008 0.38 0.009 hydration of
   carbon dioxide to form bicarbonate
   CALB1 Calbindin 1 7142.8 0.64 < 0.001 1.00 < 0.001 calcium binding
   capability
   CALCRL Calcitonin Receptor Like Receptor 80.5 -0.33 0.031 0.13 0.398
   calcium regulation
   CLDN10 Claudin 10 4976.9 0.25 0.002 -0.21 0.110 passive transport
   CLDN14 Claudin 14 38.9 1.68 < 0.001 1.38 < 0.001 passive transport
   CLDN16 Claudin 16 370.4 0 0.995 -0.44 < 0.001 passive transport
   SLC2A13 Solute carrier family 2 member 13 180.0 -0.51 0.015 -0.09 0.745
   myo-inositol: proton symporter activity
   SLC4A1 Solute carrier family 4 member 1 271.1 -0.14 0.357 0.26 0.029
   chloride/bicarbonate exchanger
   SLC8A1 Solute carrier family 8 Member 1 668.1 0.06 0.782 0.67 0.002
   intracellular Ca transport
   SLC20A2 Solute carrier family 20 member 2 1486.4 -0.28 0.031 -0.15
   0.195 P transport
   SLC34A1 Solute carrier family 34 member 1 41142.8 0.38 0.025 0.28 0.200
   P transport
   SLC34A2 Solute carrier family 34 member 2 11.8 -0.80 0.031 -0.18 0.573
   P transport
   SLC41A2 Solute carrier family 41 Member 2 111.2 -0.01 0.943 0.35 0.025
   Mg transport
   SLC41A3 Solute carrier family 41 Member 3 418.1 0.31 0.031 0.03 0.835
   Mg transport
   STC1 Stanniocalcin-1 57.2 1.65 < 0.001 1.87 < 0.001 Ca and P transport
   TJP1 Tight junction protein 1 556.2 -0.33 0.020 0.26 0.181 passive
   transport
   TRPM7 Transient receptor potential cation channel subfamily M member 7
   280.8 -0.25 0.102 0.46 0.015 Ca and Mg transport
   TRPV6 Transient receptor potential cation channel, subfamily V, member
   6 279.0 -0.26 0.325 0.81 0.009 Ca transport
   Week 24
   AQP7 Aquaporin 7 1184.3 0.40 0.003 0.53 < 0.001 water channel activity
   ATP2B4 ATPase plasma membrane Ca2 + transporting 4 306.9 0.03 0.790
   -0.25 0.030 calcium transport
   CA4 Carbonic anhydrase 4 31.8 -1.56 < 0.001 -0.141 0.596 hydration of
   carbon dioxide to form bicarbonate
   CLDN14 Claudin 14 39.0 0.03 0.932 0.96 0.001 passive transport
   CYP24A1 Cytochrome P450 Family 24 Subfamily A Member 1 3127.6 -1.30
   0.026 0.46 0.365 vitamin D elimination
   FGFR1 Fibroblast Growth Factor receptor 1 1045.1 0.05 0.579 -0.22 0.039
   P reabsorption
   SLC8A1 Solute carrier family 8 Member 1 438.3 -0.42 0.012 -0.23 0.211
   intracellular Ca transport
   SLC20A1 Solute carrier family 20 member 1 678.3 -0.08 0.831 -0.84 0.017
   P transport
   STC1 Stanniocalcin-1 179.7 0.08 0.867 0.70 0.005 Ca and P transport
   [113]Open in a new tab

   After the onset of lay at week 24, the Ca exchanger encoding gene
   SLC8A1 was downregulated in the LB strain while the P transporter gene
   SLC20A1 was downregulated in the LSL strain when comparing the
   different P diets (P + > P-). In both strains, the aquaporin AQP7 was
   significantly upregulated at week 24. In addition, CA4 was
   downregulated in the LB strain. In the LSL strain at week 24, CLDN14
   and STC1 were upregulated, with a downregulation observed in ATP2B4.
   Regarding vitamin D metabolism, FGFR1 in LSL and CYP24A1 in LB were
   significantly downregulated.

Discussion

   Prior to the onset of egg laying, a series of complex biological
   processes take place that are controlled by the endocrine system and
   comprise the gastrointestinal tract as well as tissues involved in
   post-absorptive metabolism such as kidney and liver. These processes
   regulate nutrient and mineral utilization based on developing needs in
   reproductive birds. The present study focused on animal-intrinsic
   effects of P homeostasis in the kidney and associated plasma
   metabolites, as the feed composition remained consistent at the two
   examined sampling stages to avoid confounding effects according to the
   interaction of diet × stage.

Metabolic adaptation via plasma hormones and metabolites

   The resulting plasma P reflected the dietary P intake in both strains
   at week 19, i.e. lowered P levels in P- diets, aligning with reports on
   reduced serum P values in P- fed hens at 25 weeks [[114]36]. As the P-
   diets lacked mineral P and phytase supplements, the plasma P levels
   could be attributed to dietary NPP (1.3 g/kg) and phytate P in feed
   partially available due to mucosal and microbial enzyme activity
   [[115]37]. Stage effects appeared to be significant both in LB and LSL
   with a decrease in plasma P levels from week 19 to 24 in accordance
   with previous studies [[116]38]. The elevated plasma Ca values from
   week 19 to 24 are due to higher Ca intake following increased feed
   intake [[117]27] and mobilization due to intensified Ca demand for
   eggshell synthesis [[118]39]. The increase in ingested Ca led to a drop
   in plasma P levels as high Ca impairs the P absorption and phytate
   degradation by raising digesta pH [[119]13] and forming Ca-phytate
   complexes [[120]14]. In addition, these dynamics in blood Ca and P
   levels in matured laying hens are also a response to the significantly
   increased mineral turnover. Indeed, the medullary bone is involved in
   intermediate deposition of minerals and their rapid mineral
   mobilization during sexual maturation, which is important for the onset
   of eggshell production [[121]40].

   Through ALP activity, Ca mobilization from the bone and its deposition
   onto eggshell can be enhanced [[122]41], thus the observed increase in
   ALP activity in LSL hens after onset of lay (week 24) may follow Ca
   needs. High levels of plasma Ca are known to inhibit the secretion of
   PTH from parathyroid glands [[123]42], which might explain the decrease
   in PTH from week 19 to 24 in P + fed LSL hens. LSL hens fed P + diets
   could also reduce PTH secretion to ensure renal P reabsorption. The
   increase in triacylglycerides at week 24 corresponds to the involvement
   of hepatic metabolism for egg production. Since large amounts of the
   egg yolk proteins and lipids are synthesized in the liver and
   transported to the ovary during the laying period [[124]43],
   triacylglyceride levels in blood increased about 10-fold after the
   onset of lay. The fluctuations in the plasma levels of albumin and Mg
   could be due to their deposition in the egg and thus the individual
   state of sexual maturity of the hens. The more pronounced effects of
   diet and sampling stage on plasma metabolites in LSL compared to LB
   hens might reflect a presumably higher degree of sexual maturity of LSL
   compared to LB.

   Strain-specific differences of the analyzed plasma metabolites were
   pronounced for calcidiol, i.e. the storage form of vitamin D. This is
   consistent with previous reports where higher calcidiol levels were
   found in LB compared to LSL laying hens [[125]11, [126]44]. Although no
   significant effects of dietary P supply and stage were observed for
   calcitriol levels, the numerical differences suggest established
   animal-intrinsic mechanisms acting on renal calcitriol synthesis. These
   endocrine adaptations may help to increase the intestinal mineral
   absorption following low P intake [[127]45] and to meet the mineral
   requirements due to egg yolk deposition as well as eggshell
   calcification [[128]46]. The variation in plasma calcitriol levels
   between hens, particularly at week 24, may reflect the individual
   status in sexual maturity and the associated adaptations of intestinal
   and post-absorptive tissues.

Transcriptional profiling reveals metabolic interlinkages

   In addition to variations observed at the plasma level, the difference
   in renal expression profiles between the two strains also indicates
   that the strains have adapted in a strain-specific manner, as has been
   previously suggested in relation to mineral homeostasis [[129]5]. This
   difference in intrinsic mechanism driving mineral homeostasis
   corresponds to the genetic background of these strains [[130]47].
   Accordingly, the KEGG pathway enrichment analyses revealed pathways
   attributed to energy metabolism in LB hens and cell cycle in LSL hens
   at week 19. Likewise, pronounced strain differences in terms of
   immunity and intestinal capacity were described previously [[131]26,
   [132]48]. The lack of enriched pathways in the kidney following varying
   mineral P supply at week 24 in both strains could be attributed to the
   considerable dynamics in Ca turnover at onset of egg laying which seems
   to outweigh the effects of dietary P reduction.

   The expression pattern of the target genes found in our RNAseq data
   supported the role of the kidney in the regulation of P and Ca
   homeostasis in laying hens, as the adaptive capacity of kidney in
   mineral homeostasis was reported by previous studies [[133]49]. SLC34A1
   encoding NPt2a was identified as the predominantly expressed Na/P
   co-transporter in the avian kidney [[134]12]. The higher mRNA abundance
   of SLC34A1 in LB at week 19 under low P diets is linked with enhanced P
   reabsorption in the kidney, as discussed previously for broiler
   chickens [[135]50]. Similarly, a low P diet (0.15% vs. 0.41% vs. 0.82%
   available P) revealed an upregulated SLC34A1 expression in kidney of
   laying hens as reported previously [[136]32]. However, no effect on
   avian renal SLC34A1 expression following a reduction in dietary NPP
   levels (0.14% vs. 0.32% NPP) was reported previously [[137]51]. The
   annotated Na/P co-transporter SLC34A2 was expressed at very low levels
   in avian kidney. LB hens showed a reduced SLC34A2 mRNA abundance under
   P- diets as observed previously in growing broiler chickens [[138]12].
   Another Na/P co-transporter, SLC20A2, was downregulated (P + > P-) in
   LB at week 19 similar to findings in kidney of growing broiler chickens
   [[139]12], while a higher expression was reported in fattening pigs
   [[140]52] and rats [[141]53] when subjected to a low P diet. Since the
   SLC20A2 gene plays a housekeeping role in cellular P uptake by encoding
   PiT2 [[142]54], its higher expression under P + diets could be linked
   to reduced cellular P efflux and prioritize available P for essential
   metabolic processes. The rate of P reabsorption to adjust plasma P
   levels may involve Na/P co-transport activity through hormonal control
   such as PTH and vitamin D [[143]55], but may also be effectively
   regulated by STC1, which enhances renal P reabsorption and prevents
   hypercalcemia [[144]56]. The considerable upregulation of STC1
   (P + < P-) observed in our study was similar to results in porcine
   kidney studying effects of low P diets [[145]57]. The regulatory role
   of STC1 is considered a conserved mechanism, as also in fish
   stanniocalcin lowers Ca uptake by the gills and intestine and increases
   P reabsorption by the renal proximal tubules through an adenosine
   3′,5′-cyclic monophosphate (cAMP)-dependent pathway, resulting in a
   decrease in Ca and increase in P levels [[146]58].

   In terms of passive mineral transport, claudin-14 (CLDN14) was
   significantly upregulated in P deficient groups in both strains at week
   19 and only in LSL at week 24. The upregulation could be linked with
   maintaining Ca levels in the blood as CLDN14 was reported to regulate
   the paracellular reabsorption of Ca and Mg in the thick ascending limb
   in humans [[147]59]. The observed upregulation of claudin-10 (CLDN10)
   in the LB group at week 19 corresponds to its role in regulating Ca and
   Mg transport in the thick ascending limb of the kidney. This gene
   encodes two alternatively spliced isoforms of claudin-10 proteins in
   humans where claudin-10a functions as an anion channel, and claudin-10b
   acts as a cation channel [[148]60]. Moreover, claudin-16 (CLDN16) was
   downregulated in LSL at week 19, while tight junction protein coding
   gene TJP1 was downregulated in the LB group at week 19. CLDN16 mediates
   Mg and Ca resorption through paracellular permeation in a porcine renal
   cell line [[149]61]. Therefore, the renal expression profile provides
   evidence that both active and passive transport mechanisms were
   affected by the dietary mineral P level, although the specific gene
   functions in birds can so far only be derived from mammalian research.
   Interestingly, the varying mineral P level revealed an effect on
   aquaporin 7 (AQP7) and aquaporin 11 (AQP11), belonging to the
   transmembrane water and small solute channels family involved in water
   and glycerol transport. Although their role in Ca and P transport is
   still unknown, the higher expression of AQP11 under P- diets in LB
   laying hens at week 19 is aligned with previous studies, where AQP11
   was significantly upregulated in the kidney of reproductive pigs due to
   a low P diet [[150]62]. It is therefore conceivable that the dietary P
   intake and subsequent endogenous responses comprise secondary effects
   on the water balance.

   The onset of egg production in laying hens necessitates a higher Ca
   turnover to meet the demands of eggshell formation. Calbindin-D28k,
   encoded by CALB1, plays a crucial role in regulating renal Ca and Mg
   reabsorption in the distal nephrons [[151]63]. In the current study,
   CALB1 was highly expressed in both strains of laying hens at week 19,
   with an upregulation observed in response to the P- diet. However,
   previous studies reported conflicting findings regarding the expression
   levels of CALB1 and its encoded protein in the kidneys of laying hens
   subjected to varying dietary mineral levels at different stages
   [[152]32, [153]33, [154]64]. Interestingly, while the ovulatory cycle
   did not significantly affect CALB1 expression in the kidney, a
   significant response was observed in the shell gland [[155]65]. In our
   study, CALB1 expression levels indicate effectiveness of
   calcitriol-mediated renal mineral reabsorption, although calcitriol was
   not significantly but numerically increased by the applied P- diet.

   The downregulation of CYP24A1 due to P- diets in LB at week 24 suggests
   a reduction in renal 24-hydroxylation of calcitriol and thus reduced
   excretion via urine. This is consistent with the results from pigs fed
   low P diets [[156]34], consequently ensuring adequate circulating
   levels of calcitriol, similar to the results described in laying hens
   [[157]66]. In this context, downregulation of CALCRL in LB hens at week
   19 may indicate calciotrophic effects to diminish the
   calcitonin-mediated inhibition of Ca reabsorption in the kidney,
   supporting a renal response to P- diets maintaining Ca homeostasis.
   However, the contribution of Ca regulation by calcitonin in birds
   remains unclear both in terms of efficacy and signal transduction
   [[158]65]. In addition, systemic effects triggered by the phosphatonin
   FGF23 can be observed based on the renal transcription profile. The
   downregulation of fibroblast growth factor receptor 1 (FGFR1) in LSL at
   week 24 is consistent with a previous study, which revealed a reduced
   FGFR1 expression following low P diets in laying hens [[159]51].
   Reduced FGF23 signaling ultimately decreases P excretion [[160]67], as
   FGF23 negatively affects renal vitamin D biosynthesis by activating the
   catabolic enzyme CYP24A1 and suppresses renal P reabsorption by
   decreasing the expression of SLC34A1 [[161]68]. This might correspond
   to recent findings as LSL laying hens revealed significantly reduced P
   excretion compared to LB laying hens [[162]27]. However, FGF23 also
   depends on its soluble receptor α-Klotho (KL), the renal expression of
   which remained unaltered due to P diets in accordance with earlier
   findings [[163]51].

   In order to maintain a physiological Ca: P ratio in the body
   compartments, the results indicate that both laying hen strains have
   adapted to the lack of additional mineral P sources by regulating Ca
   metabolism, as evidenced by changes in gene expression patterns in the
   kidney. The pronounced effect of the experimental factors on Ca
   metabolism is further illustrated by the variable expression of other
   key genes, including TRPV6, SLC8A1 and ATP2B4. In fact, these genes are
   known to facilitate Ca transport whereas their role in renal Ca
   utilization is sparsely described in poultry [[164]69]. Notably, the
   CaSR was constantly expressed in the kidney across all experimental
   groups, with no evidence of regulation due to dietary P reduction. In
   addition to DEGs involved in Ca metabolism, the divergent P diets also
   tackled genes encoding transcripts involved in Mg metabolism. In LSL
   hens at week 19, TRPM7 was upregulated due to P- diet, which paralleled
   the numerically higher levels of plasma Mg. TRPM7 regulates
   transcellular Mg reabsorption in the distal convoluted tubule of kidney
   [[165]70]. Moreover, the upregulation of SLC41A2 in LB and SLC41A3 in
   LSL at week 19 indicates further interlinkages between P and Mg
   homeostasis as the SLC41 family is involved in Mg transport and
   indirectly regulates renal Mg reabsorption [[166]71, [167]72].

   The transcription profile showed that the SLC4A1 abundance was higher
   in the P- group compared to the P + group in LSL at week 19, whereas
   the opposite was observed in reproductive pigs [[168]62]. SLC4A1 is an
   anion exchanger and mainly involved in intracellular bicarbonate
   (HCO₃⁻) and extracellular chloride (Cl⁻) exchange [[169]73]. Moreover,
   an effect due to dietary P level was observed at the expression levels
   of carbonic anhydrases CA2 and CA4. CA2 was upregulated due to the P-
   diet in both strains at week 19. The Ca^2+ and HCO[3]^− ions are
   continuously supplied to the eggshell gland via transepithelial
   transport [[170]74] to form calcium carbonate (CaCO₃). The upregulation
   of CA2 was in line with previous findings obtaining a 4-fold increased
   expression of CA2 in shell gland of laying hens during oviposition
   [[171]65]. CA2 was also identified in quail renal tubules [[172]75] as
   a cytosolic coenzyme for bicarbonate reabsorption in the avian kidney.
   CA2 is considered the major cytoplasmic isoform of carbonic anhydrase
   in the nephron, while a minor activity is thought to be delivered by
   CA4. CA4, also described for its activity in bicarbonate reabsorption
   and transport mechanisms in rabbit kidney cells [[173]76], was not
   expressed at week 19, however, the transcript was expressed at a low
   rate at week 24 and was significantly downregulated in LB hens due to
   the P- diet. There is evidence that canine carbonic anhydrases are
   involved in the regulation of P excretion through an unknown mechanism,
   likely affecting acid-base balance in the kidneys [[174]77].

Conclusions

   Both LB and LSL strains subjected to a diet without mineral P exhibited
   strain-specific endogenous adaptation to maintain P homeostasis via
   renal gene expression and endocrine profiles. The lack of dietary
   mineral P caused endogenous responses prior to the onset of laying in
   week 19, which, however, declined with increasing maturity of the hens.
   Associated metabolic adaptation showed particular significance for Ca
   homeostasis and triglycerides. Differential expression analysis
   underlines the important role of the renal functions for mineral
   homeostasis, i.e. passive/active mineral transport, Ca transport, and
   vitamin D elimination. Transcriptional profiling revealed new
   P-dependent targets such as carbonic anhydrase 2 (CA2) in both laying
   hen strains, which might offer new insights into metabolic
   interlinkages of mineral P depletion as well as trade-offs for animal
   health and performance.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [175]Supplementary Material 1^ (3.2MB, xlsx)

Acknowledgements