Research Resource: Whole Transcriptome RNA Sequencing Detects Multiple 1α,25-Dihydroxyvitamin D₃-Sensitive Metabolic Pathways in Developing Zebrafish

Abstract

The biological role of vitamin D receptors (VDR), which are abundantly expressed in developing zebrafish (Danio rerio) as early as 48 h after fertilization, and before the development of a mineralized skeleton and mature intestine and kidney, is unknown. We probed the role of VDR in developing zebrafish biology by examining changes in expression of RNA by whole transcriptome shotgun sequencing (RNA-seq) in fish treated with picomolar concentrations of the VDR ligand and hormonal form of vitamin D₃, 1α,25-dihydroxyvitamin D₃ [1α,25(OH)₂D₃)].We observed significant changes in RNAs of transcription factors, leptin, peptide hormones, and RNAs encoding proteins of fatty acid, amino acid, xenobiotic metabolism, receptor-activator of NFκB ligand (RANKL), and calcitonin-like ligand receptor pathways. Early highly restricted, and subsequent massive changes in more than 10% of expressed cellular RNA were observed. At days post fertilization (dpf) 2 [24 h 1α,25(OH)₂D₃-treatment], only four RNAs were differentially expressed (hormone vs. vehicle). On dpf 4 (72 h treatment), 77 RNAs; on dpf 6 (120 h treatment) 1039 RNAs; and on dpf 7 (144 h treatment), 2407 RNAs were differentially expressed in response to 1α,25(OH)₂D₃. Fewer RNAs (n = 481) were altered in dpf 7 larvae treated for 24 h with 1α,25(OH)₂D₃ vs. those treated with hormone for 144 h. At dpf 7, in 1α,25(OH)₂D₃-treated larvae, pharyngeal cartilage was larger and mineralization was greater. Changes in expression of RNAs for transcription factors, peptide hormones, and RNAs encoding proteins integral to fatty acid, amino acid, leptin, calcitonin-like ligand receptor, RANKL, and xenobiotic metabolism pathways, demonstrate heretofore unrecognized mechanisms by which 1α,25(OH)₂D₃ functions in vivo in developing eukaryotes.

Vitamin D₃ through its active metabolite, 1α,25-dihydroxyvitamin D₃ [1α,25(OH)₂D₃], plays a critical role in calcium and phosphorus homeostasis by increasing intestinal calcium and phosphorus transport, thereby maintaining normal serum calcium and phosphorus concentrations and allowing bone mineralization to proceed (1, 2). To function in target tissues, 1α,25(OH)₂D₃ binds to its receptor, the vitamin D receptor (VDR), and the ligand-bound receptor, either as a heterodimer with the retinoic acid X-receptor, or as a homodimer, activates the expression of various genes, many of which encode proteins that alter trans-epithelial calcium transport (1–8). The VDR is expressed in a variety of tissues including the intestine, kidney, bone, skin, and brain of mammals and fish (9–17). Earlier, we showed that the VDR is expressed in the intestine, kidney, gill-transporting cells, and bone of the adult zebrafish, Danio rerio (17). In addition, the skin, olfactory organ, retina, brain, spinal cord, Sertoli cells of the testis, oocytes, pancreatic acinar cells, hepatocytes, and bile duct epithelial cells express substantial amounts of the VDR. In 48 h post fertilization (hpf) and 96 hpf zebrafish embryos/larvae, the VDR was noted in cartilage, brain, retina, and otic vesicles, suggesting that the VDR and its ligand, 1α,25(OH)₂D₃, play a role in the biology and development of these organs (17). The role of 1α,25(OH)₂D₃ in early development, however, has not been defined in vivo.

The effects of 1α,25(OH)₂D₃ on global gene expression in lymphoblast and macrophage lines have been examined using chromatin immunoprecipitation (ChIP) and massively parallel sequencing (CHIP-seq) (18–20). In lymphoblastoid cells, after 1α,25(OH)₂D₃ stimulation, Ramagopalan et al. (18) found 2776 genomic positions occupied by the VDR. Microarray analysis demonstrated significant changes in the expression of 229 genes in response to 1α,25(OH)₂D₃. Two hundred and twenty six genes were up-regulated, and three genes were down-regulated by 1α,25(OH)₂D₃. In human monocytic leukemia cells, Heikkinen et al. (19), identified 2340 VDR binding locations by ChIP-seq, of which 1171 and 520 occurred uniquely with and without 1α,25(OH)₂D₃ treatment, respectively. Microarray analysis demonstrated that the expression of 638 genes was altered after 1α,25(OH)₂D₃-treatment; 408 genes were up-regulated and 230 genes were down-regulated. These studies indicate that 1α,25(OH)₂D₃ regulates the expression of numerous genes in cultured cells. No information, however, is available on the influence of 1α,25(OH)₂D₃ on gene expression in whole animals in vivo, and correlations between changes in gene expression profiles and morphological change in vivo have not been performed.

The small size and transparency of zebrafish in early development allow the simultaneous assessment of morphological and gene expression changes in vivo. The effects of various agents such as sterols [e.g. 1α,25(OH)₂D₃], steroids and peptide hormones can be assessed by adding such substances to the medium in which the zebrafish is maintained (21). In zebrafish, the skeleton develops in a stereotypical pattern in which cartilage formation begins at about 48 h post fertilization (hpf), and is followed by mineralization of the aforesaid structures at about 144 hpf (22, 23). These changes can be readily visualized because of the transparency of the fish and available staining procedures specific for cartilage and bone. Changes in gene expression patterns can be correlated with changes in cartilage development (before ∼120 hpf) and bone (after 120 hpf). Additionally, the brain, eye, and otic structures develop rapidly before 24 hpf and continue to show significant change even after 24 hpf. The effects of 1α,25(OH)₂D₃ on these structures can be assessed as well.

To determine the influence of 1α,25(OH)₂D₃ on gene expression patterns and metabolic pathways in vivo, we treated zebrafish embryos/larvae with 1α,25(OH)₂D₃ or vehicle and assessed RNA expression at various times after addition of 1α,25(OH)₂D₃ or vehicle with whole transcriptome shotgun sequencing (RNA-seq). We demonstrate that multiple, previously unrecognized transcription factors, peptide hormone and metabolic pathways, are altered by 1α,25(OH)₂D₃ in the developing zebrafish.

Materials and Methods

Treatment of zebrafish with 1α,25(OH)₂D₃

Zebrafish embryos were obtained from mating of Segrest wild-type parents under controlled barrier conditions, in the Mayo Clinic Zebrafish Core Facility, in Instant Ocean media (Spectrum Brands, Madison, WI) (24). Zebrafish embryos (100–120) were placed in 20 ml embryo medium (pH 7.2) containing 1-phenyl-2-thiourea [0.003% (wt/vol)] and were maintained at 28–30 C (Sigma-Aldrich, St. Louis, MO). At 24 hpf (1 d post fertilization, dpf), 10 μl 1α,25(OH)₂D₃ in ethanol (gift of Dr. Milan Uskokovic, Hoffmann LaRoche, Nutley, NJ), was added to embryos maintained in 20 ml fresh embryo medium with 1-phenyl-2-thiourea. The final concentration of 1α,25(OH)₂D₃ was 300 pm. Control zebrafish were treated with 10 μl ethanol alone (vehicle controls). The medium containing either 300 pm 1α,25(OH)₂D₃ or vehicle was changed every 24 h. In experiment 1 (see Fig. 1), at 2, 4, 6, and 7 dpf batches of embryos/larvae were removed and immediately frozen at −80 C for later RNA preparations. Twenty five to 30 embryos/larvae were used for preparation of RNA. Five individual cDNA libraries from 1α,25(OH)₂D₃-treated or vehicle-treated fish were prepared at each time point (e.g. 2, 4, 6, or 7 d). Forty individual libraries (20 1α,25(OH)₂D₃-treated and 20 vehicle-treated) were prepared in experiment 1. At the same times, seven to 12 embryos were fixed in 4% paraformaldehyde in 0.75× Dulbecco's PBS (DPBS). In experiment 2 (see Fig. 1), 6 dpf larvae were treated with 1α,25(OH)₂D₃ (300 pm) or vehicle for 24 h. RNA was prepared from three sets of larvae. Six individual libraries (three 1α,25(OH)₂D₃-treated and three vehicle-treated) were prepared in experiment 2.

Fig. 1.

Experimental design. Experiment 1: zebrafish embryos were treated with 1α,25(OH)₂D₃ (300 pm) or vehicle beginning at 24 hpf. Medium and additives were changed to every 24 h (black arrows). Five batches of 25–30, 1α,25(OH)₂D₃-treated, and five batches of 25–30 vehicle-treated zebrafish embryos/larvae were collected at 2, 4, 6, and 7 dpf (red arrows) for isolation of RNA, construction of cDNA libraries, and sequencing. Experiment 2: zebrafish larvae were treated with 1α,25(OH)₂D₃ (300 pm) or vehicle beginning at 144 hpf; 24 h later, three batches of 25–30 1α,25(OH)₂D₃-treated and three batches of 25–30 vehicle-treated zebrafish were collected (red arrow) for isolation of RNA, construction of cDNA libraries, and sequencing.

Measurement of 25-hydroxyvitamin D and 1α,25-dihydoxyvitamin D in zebrafish larvae

The concentrations of 25-hydroxyvitamin D₂, 25-hydroxyvitamin D₃, 1,25-dihydroxyvitamin D₂, and 1,25-dihydroxyvitamin D₃ in 7 dpf zebrafish larvae were measured using mass spectroscopy and hexadeuterated standards (25, 26). Nine hundred zebrafish larvae were suspended in 1.5 ml buffer (80 mm Tris, 200 mm NaCl, pH 7.4) and homogenized with a Polytron homogenizer (Brinkmann Instruments, Ontario, Canada) using three 5-sec pulses. The lysate was centrifuged at 14,000 rpm for 10 min in an Eppendorf 5415C centrifuge (Brinkmann Instruments, Westbury, NY) to obtain a clear supernatant. Protein concentration was measured in the supernatant using the bicinchoninic assay (Thermo Fisher Scientific, Rockford, IL) using the manufacturer's protocol. The homogenization and centrifugation was carried out at 4 C. Ethyl acetate (2 ml) was added to 0.5-ml aliquots of the clarified lysate in a glass test tube at room temperature, and the sample was vortexed for 1 min. The layers were allowed to separate and the upper organic layer was collected. This step was repeated with 1 ml of ethyl acetate. The combined organic layers were dried under a gentle flow of nitrogen. The dried residue was resuspended in 0.5 ml stripped human serum, and deuterated 25-hydroxyvtamin D₃, 25-hydroxyvtamin D₂, and 1,25-dihydroxyvitamin D₃ and 1,25-dihydroxyvitamin D₂ were added to the sample. 25-Hydroxyvitamin D₂ and 25-hydroxyvitamin D₃ and 1,25-dihydroxyvitamin D₂ and 1,25-dihydroxyvitamin D₃ were measured as described elsewhere (25, 26). Both measurements were performed in triplicate.

Alcian blue and alizarin red staining of embryos/larvae

Zebrafish embryos/larvae were stained with alcian blue 8GX (Sigma-Aldrich no. A3157; Sigma, St. Louis, MO) according to Walker and Kimmel (27). The final staining solution contained 0.003% alcian blue, 130 mm MgCl₂, in 70% ethanol. The solution was filtered through a 0.2-μm filter. Fixed zebrafish embryos/larvae were stained with alcian blue solution overnight at room temperature. Embryos/larvae were destained with 20% glycerol-0.25% KOH followed by 50% glycerol-0.25% KOH for several days at room temperature. Embryos/larvae were bleached to remove endogenous pigmentation, with 1.5% H₂O₂-1% KOH for 50 min at room temperature. Embryos/larvae were finally washed once with water and placed in 20% glycerol-0.25% KOH.

For alizarin red staining, embryos/larvae were fixed for 50 min in 4% paraformaldehyde/0.75× DPBS and were washed once with 1 ml DPBS. A solution of alizarin red S, 0.05% (Sigma-Aldrich; no. A-5533), 10 mm MgCl₂, in 70% ethanol was prepared. Embryos/larvae were stained at room temperature overnight. Excess stain was removed by washing with water. Tissues were clarified with 0.25% KOH-20% glycerol overnight, followed by 0.25% KOH-50% glycerol, with several solution changes until background tissue staining was reduced. Endogenous pigments were bleached with 1.5% H₂O₂-1% KOH as described above for alcian blue staining. The intensity of red staining was determined using the KS 400 Image analysis program (Carl Zeiss, North America, Thornwood, NY).

RNA preparation for RNA-seq and QPCR

RNA was prepared using RNA/protein spin columns (CLONTECH Laboratories, Mountain View, CA). Lysis solution was added to 25–30 frozen embryos/larvae, and embryos/larvae were lysed by passage of tissue successively through 21- and 27-gauge needles. Individual embryo/larvae lysates were applied to RNA spin columns for purification. RNA eluted into nuclease-free water was immediately frozen at −80 C. Before library construction, RNA quality was assessed by capillary electrophoresis against a known size standard in the Mayo Molecular Biology Core facility.

Quantitative PCR (QPCR)

QPCR was carried out using a Roche LightCycler 480 QPCR apparatus in 96-well QPCR plates (Roche Diagnostics Corp., Indianapolis, IN). An I-Script RT-PCR kit (Bio-Rad Laboratories, Inc., Hercules, CA) was used to generate double-stranded DNA from RNA. Reverse transcribed I-script product was used to generate PCR products with each primer pair. Product was quantitated and used to generate standard QPCR curves. QPCR data were subjected to quantitation against 18S RNA or tubulin α1, using software supplied with the instrument. Ten RNA samples (five control and five experimental, 7 dpf) were analyzed in triplicate by QPCR. QPCR primers used were: Danio rerio (in all cases) cytochrome P450, cyp24a1, 5′-oligonucleotide: 5′-gcgtctcgctgagttacaga-3′; 3′-oligonucleotide: 5′-ccaggtctgttgccacaat-3′. IGF 2a, igf2a, 5′-oligonucleotide: 5′-tgaagtcggagcgagatgtt-3′; 3′-oligonucleotide: 5′-ggagtacttcacatttatggtgtcc-3′. IGF-binding protein 1b, igfbp1b, 5′-oligonucleotide: 5-′gcagaagctcatccagcag-3′; 3′-oligonucleotide: 5′-gggcaggtagaaactggtga-3′. Kruppel-like factor 11a, klf11a, 5′-oligonucleotide: 5′-tggtctccctctcatcactgt; 3′-oligonucleotide: 5′-caggccactgaggtactgaga-3′. Leptin receptor, lepr, 5′-oligonucleotide: 5′-ccaaaggaatggacttcagg3′; 3′-oligonucleotide: 5′-tcggagaccagcagcaat-3′. Peroxisome proliferator-activated receptor α a (pparaa), 5′-oligonucleotide: 5′-gcaggaacagcttgtctacctt3′; 3′-oligonucleotide: 5′-gaccacggcaagatttgaa-3′. βactin2, bactin2, 5′-oligonucleotide: 5′-aaggccaacagggaaaagat-3′; 3′-oligonucleotide: 5′-gtggtacgaccagaggcatac-3′. 18s ribosomal RNA, 5′-oligonucleotide: 5′-tcgctagttggcatcgtttatg-3′; 3′-oligonucleotide: 5′-cggaggttcgaagacgatca-3′. Tubulin α1, 5′-oligonucleotide: 5′-ttgtagacctggagcccact-3′; 3′-oligonucleotide: 5′-ttcctttcctgtgatgagctg-3′.

Preparation of libraries

mRNA-seq libraries were prepared using the mRNA v1 sample prep kit following the manufacturer's protocol (Illumina, Inc., San Diego, CA). Poly-A containing mRNA was purified using poly-T oligo attached magnetic beads. The purified mRNA was fragmented using divalent cations at 95 C for 5 min and ethanol precipitated. The RNA fragments were copied into first-strand cDNA using reverse transcriptase and random primers. Second-strand cDNA synthesis was performed using DNA polymerase I and ribonuclease H. The cDNA was purified using Qiaquick PCR columns from QIAGEN (Chatsworth, CA). The ends were repaired and phosphorylated using Klenow, T4 polymerase, and T4 polynucleotide kinase, after which an A base was added to the 3′-ends of double-stranded DNA using Klenow exo- (3′ to 5′ exo minus). Paired end DNA adaptors (Illumina) with a single T base overhang at the 3′-end were ligated and the resulting constructs were separated on a 2% agarose gel. DNA fragments of approximately 250–300 bp were excised from the gel using GeneCatcher tips and purified using Qiagen Gel Extraction Kits. The adapter-modified DNA fragments were enriched by 15 cycles of PCR using primers PE 1.0 and PE 2.0 (Illumina). The concentration and size distribution of the libraries were determined on an Agilent Bioanalyzer DNA 1000 chip (Agilent Technologies, Palo Alto, CA).

Libraries were loaded onto paired end-flow cells at concentrations of 6 pm to generate cluster densities of 400,000/mm² following Illumina's standard protocol using the Illumina cBot and cBot Paired end cluster kit version 2. The flow cells were sequenced as 51 × 2 paired end reads on an Illumina HiSeq 2000 using TruSeq SBS sequencing kit version 1 or version 3 and SCS version 1.1.37.0 or HCS version 1.4.8 data collection software, respectively. Base calling was performed using Illumina's RTA version 1.7.45.0.

mRNA-seq data analysis

Mapping of sequence reads

The generated FASTQ sequence reads from an Illumina HiSEQ instrument were aligned to both the latest available zebrafish genome assembly (zv9) and our in-house exon junction database using Burrows-Wheeler alignment (28). In the exon junction database, unidirectional combinations of exon junction database for the sequencing length of 50 bases were generated using exon boundaries defined by the refFlat file from University of California Santa Clara (UCSC) Table Browser. Burrows-Wheeler alignment is a fast and accurate short reads aligner. A maximum of two mismatches was allowed for first 32 bases in each alignment, and reads that had more than two mismatches or were mapped to multiple genomic locations (alignment score less than 4) were discarded. The aligned sequence tags were counted for each annotated gene/exon using a custom program based on the UCSC genome binning approach (29). A total of 14,267 genes were annotated using RefSeq RNA database, and the raw read counts for genes were generated for further downstream analyses.

Differential gene expression analysis

The raw reads count from each gene were normalized by the total reads of each biological replicate and then standardized to reads per million (gene counts/total counts of each biological replicate * 1,000,000). For differential gene expression analysis between different conditions, we eliminated genes without any reads across all samples. Statistical testing was done using the R package DESeq (30). Because the primary goal of our analysis was to explore the underlying cause of differentially expressed genes between vehicle- and 1α,25(OH)₂D₃-treated conditions at each time point, an adjusted P value/false discovery rate (to adjust for multiple testing using the Benjamin-Hochberg method) cut-off of 0.01 was used to select significant changed genes.

Pathway analysis for differentially expressed genes

We conducted pathway analysis of differentially expressed genes using DAVID (Database for Annotation, Visualization and Integrated Discovery v6.7) (31, 32) and DIRAC (Differential Rank Conservation) (33, 34). An adjusted P value less than 0.01 was used to assign biological significance. The goal was to investigate whether the differentially expressed genes at each time point were enriched with genes in canonical pathways and interconnected biological networks involving genes with functional commonalities. DAVID is a web resource consisting of an integrated biological knowledgebase and analytical tools that aims to extract and understand biological themes in large gene lists. It provides the gene functional classification as well as the identification (ID) conversion of the list of gene ID accessions to any other accession of choice. It also provides pathways from other databases such as Kyoto Encyclopedia of Genes and Genomes (KEGG) (35–37), Panther (38), BioCarta, and Reactome (39), and provides tissue expression and disease-specific information based on represented genes. The results were further confirmed with Ingenuity Pathway Analysis 8.6 (IPA; Ingenuity Systems, Mountain View, CA). In IPA, the differentially expressed zebrafish gene ID were first mapped to human orthologs based on BLAST results between the two organisms (ftp://ftp.ncbi.nih.gov/repository/UniGene). With these orthologs and the adjusted P values generated from statistical analysis at each time point, we queried IPA to map and generate putative networks and pathways based on the manually curated knowledge database of pathway interactions extracted from the literature. The gene network was generated using both direct and indirect relationships/connectivity. These networks were ranked by scores that measured the probability that the genes were included in the network by chance alone.

DIRAC analysis

The DIRAC algorithm (33, 34) relies on the ranks of genes within each pathway to estimate perturbation levels. Because genes must be compared within samples as well as across samples, standard read counts or reads per million would introduce bias for physically longer genes. As a result, all read counts were normalized into reads per kilobase exon per million [RPKM (gene counts/kilobase of exon for that gene/total counts of each biological replicate) * 1,000,000] (40). The length of each gene was calculated using the latest Zebrafish genome assembly (zv9). Version 3 of the Gene Set Enrichment Analysis canonical pathways was obtained (http://www.broadinstitute.org/gsea/), which includes 880 pathways from KEGG (35–37), Biocarta, and Reactome (39). Zebrafish genes were mapped to human orthologs using the ZFIN orthology database (http://zfin.org/zf_info/downloads.html#orthology). The DIRAC algorithm calculates the relative ranks of all mapped genes in each pathway for each sample within a treatment condition [e.g. ethanol or 1α,25(OH)₂D₃]. These relative ranks are used to determine a rank template, or the permutation of genes that best represents all samples within each condition. The difference in rank template for each treatment condition is then calculated, and the significance of this difference is measured using 1000 permutations of the class labels. A P value of 0.01 was used to determine significantly perturbed pathways.

Data sharing

All of the sequence data that were analyzed in this report have been deposited in Gene Expression Omnibus (Accession #GSE38575).

Results

We performed genome-wide transcriptional profiling in five replicate cDNA libraries of 1α,25(OH)₂D₃- or vehicle-treated zebrafish, 48, 96, 144, and 168 hpf [Fig. 1, experiment 1; 24, 72, 120, and 144 h continuous treatment with 1α,25(OH)₂D₃ or vehicle]. RNA samples for cDNA library construction were obtained from 25–30, 1α,25(OH)₂D₃-treated zebrafish and 25–30 vehicle-treated zebrafish. Five independent, 1α,25(OH)₂D₃-treated or vehicle-treated libraries were prepared at each time point. Forty independent libraries were subjected to Illumina DNA sequencing. We used 300 pm concentrations of 1α,25(OH)₂D₃ in the embryo medium because circulating serum 1α,25(OH)₂D₃ concentrations range between 200 and 600 pm (41, 42). The concentrations of 25-hydroxyvitamin D₂, 25-hydroxyvitamin D₃ and 1,25-dihydroxyvitamin D₂ and 1,25-dihydroxyvitamin D₃ in fish larvae were below the limit of detection (defined as the signal-to-noise-ratio, m/z >8) of the method (2 ng/ml and 4 pg/ml for 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D, respectively). In a second experiment (Fig. 1, experiment 2; 24 h treatment between dpf 6 and dpf 7), three replicate cDNA libraries from 1α,25(OH)₂D₃- or vehicle-treated zebrafish were prepared. In experiment 1, the RNA-Seq data obtained from 48, 96, 144, and 168 hpf zebrafish RNA showed approximately 235 million, 223 million, 278 million, and 257 million total reads, respectively (Table 1). Of the total number of reads at each time point, 80.7%, 80.5%, 79.2%, and 79.1% were successfully mapped to known genes. Similar results were obtained for 168 hpf zebrafish larvae (24 h treatment) in experiment 2. On average, 13,351 transcripts with more than 10 reads were mapped and identified at the four times analyzed in experiment 1, representing a substantial fraction of all known protein coding genes (n = 18,744) in the Sanger zebrafish genome assembly Zv9 (http://www.sanger.ac.uk/Projects/D_rerio/). Supplemental Fig. 1 (published on The Endocrine Society's Journals Online web site at http://mend.endojournals.org), shows correlation plots between biological replicates in the samples from 1α,25(OH)₂D₃-treated zebrafish on dpf 2.

Table 1.

Overview of RNA-Seq results

	Day 2		Day 4		Day 6		Day 7
	Vehicle	1α,25(OH)2D3	Vehicle	1α,25(OH)2D3	Vehicle	1α,25(OH)2D3	Vehicle	1α,25(OH)2D3
Total reads	2.45E + 08	2.35E + 08	2.53E + 08	2.23E + 08	2.81E + 08	2.78E + 08	2.33E + 08	2.57E + 08
Mapped reads	1.98E + 08	1.9E + 08	2.04E + 08	1.81E + 08	2.24E + 08	2.21E + 08	1.87E + 08	2.04E + 08
Percentage of mapped reads	80.74	80.68	80.06	80.5	79.22	79.24	79.74	79.16

On dpf 2 (48 hpf), after 24 h 1α,25(OH)₂D₃ treatment, RNA for the cyp24A1 encoding the 24-hydroxlase cytochrome P-450 responsible for degradation of 1α,25(OH)₂D_3, increased 6.8-fold relative to vehicle treatment (adjusted P value 5.1⁻⁴⁷). Significant decreases for the npas4, fosb, and fos RNA for the neuronal PAG containing helix-loop-helix transcription factor and the FBJ osteosarcoma viral oncogene homolog B transcription factors, were noted in 1α,25(OH)₂D₃-treated zebrafish when compared with vehicle-treated zebrafish (P < 0.01 for all; 1.9-, 2.5-, and 1.7-fold decreases). We used adjusted P values of <0.01, corrected for multiple hypotheses testing, as a cutoff to provide robust statistical power to assure biological validity of the observed change. Additionally, the analysis of 10 individual libraries [five for 1α,25(OH)₂D₃ treatment and five for vehicle] enhances the statistical and biological validity of our observations. When the data were analyzed using the DAVID (v6.7) (31, 32) program, specific pathways were not shown to be perturbed. Additional analysis with the DIRAC program (33, 34) showed that genes in the classical antibody-mediated complement activation, graft vs. host disease, glucuronidation, pyruvate dehydrogenase, granulocyte, and allograft rejection pathways were perturbed (P = 0.01).

On dpf 4 (96 hpf), after 72 h 1α,25(OH)₂D₃ treatment, cyp24a1 RNA is significantly increased 7.2-fold (P = 8.9⁻⁵⁷), (Supplemental Table 1). Significant changes in RNA for transcription factors are noted. The RNA for klf11a and klf11b (Kruppel-like 11) transcription factors increased 2.3- and 1.5-fold; the RNA for bcl6 (B-cell CLL/lymphoma 6) increased 1.6-fold. In contrast, there was a significant 2.3-fold decrease in the RNA for fosb (FBJ murine osteosarcoma viral oncogene homolog B) and a 2.3-fold decrease for tfcp2l1 transcription factor (CP2-like L1). Ligand-dependent vitamin D receptors, vdra and vdrb (vitamin D receptor) decreased approximately 1.5-fold; the RNA for pparα also decreased 1.8-fold. The RNA for igfbp1a and b (IGF-binding protein 1) were increased 4.1- and 1.9-fold. RNAs for a number of enzymes such as elastase 2 were decreased. The RNA for the slc34a2 (sodium-phosphate transporter IIa) was diminished. Analysis by DAVID analysis showed no specific pathway change. DIRAC analysis, however, demonstrated perturbation in the graft vs. host disease, pentose and glucuronate interconversions, JAK-STAT, hormone ligand binding receptors, glucuronidation, translocation of ZAP-70 to immunological synapse, and nuclear receptors in lipid metabolism and toxicity pathways (P = 0.01).

On dpf 6 [120-h 1α,25(OH)₂D₃ treatment], 1039 RNA were either increased or decreased (Supplemental Table 2). RNA for cyp24a1 increased 7.9-fold (P = 8.3⁻⁸⁶). RNA for transcription factors klf11a, klf11b, klf3, and klfd were increased 5.1-, 1.86-, 1.67-, and 1.61-fold. Klf2b RNA decreased 1.54-fold. B-cell CLL/lymphoma 6 transcription factor RNA increased 1.8-fold. The forkhead box transcription factors, foxo1 and forkhead box L1-like, increased 2.0-fold. Fos-like antigen 2 and mycn (v-myc myelocytamatosis viral-related oncogene) RNAs were increased approximately 1.6-fold. Ligand-dependent vitamin D receptor RNAs, vdra and vdrb, were decreased, as was the RNA for pparα. Notably, RNAs for several hormones and their binding proteins were altered. The RNA for lepa (leptin) was increased 6.2-fold (P = 8.15⁻¹⁹). Ins (insulin) and gh1 (GH) RNA amounts diminished 3.3- and 2.0-fold (both P < 0.01). The igf2a and igf2b RNA (IGF 2a and 2b) increased significantly by 1.5- and 2.1-fold; RNAs for the igfbp1a, 1b, and 4 increased significantly (2.8-, 7.4-, and 2.8-fold), whereas RNAs for the igf2a, 2b, and 6b diminished 2.2-, 2.0-, and 1.6-fold. RNAs for several G protein-coupled receptors were regulated by 1α,25(OH)₂D₃, the most notable of which was the RNAs for oxgr (2-oxoglutarate receptor 1) which was altered 25-fold. Numerous transporter and channel RNAs were altered by 1α,25(OH)₂D₃, the most notable of which was the aqp8 (aquaporin 8) RNA. Analysis by DAVID showed the following pathways to be altered by 1α,25(OH)₂D₃ (Table 2): the PPAR-modulated pathway; fatty acid metabolism; retinol metabolism; glycine, serine, and threonine metabolism; tryptophan metabolism; arginine and proline metabolism; valine, leucine, and isoleucine degradation metabolic pathways. Figure 2 shows that the PPAR pathway is enriched in differentially expressed genes at dpf 6 when analyzed by DAVID. RNAs indicated with a red up or blue down arrow were either increased or decreased after 1α,25(OH)₂D₃ treatment. The change in expression of genes in the PPAR pathway assessed by the DAVID program was confirmed by IPA analysis (P = 0.0374). The different P values obtained from DAVID and IPA analysis is due to different databases used by the two programs. Examples of changes noted at d 6 in the other pathways analyzed with DAVID are shown in Supplemental Figs. 2–4. Analysis by DIRAC (Table 3) showed that additional pathways of relevance to cartilage and bone physiology were altered. For example, the receptor-activator of NFκB ligand (RANKL) and calcitonin-like ligand receptor pathways were altered after 1α,25(OH)₂D₃ treatment. The TALL1 (TNF and TNF receptor signaling) pathway and the angiotensin II pathway were also noted to be perturbed after 1α,25(OH)₂D₃ treatment. The RNA-seq data on d 7 were validated by assessing the cyp24a1, igf2a, igfbp1b, klf11a, lepr, pparaa, bactin2, 18sRNA, and tuba1 by an RT-PCR method. 18sRNA or tuba1 was used to normalize the data. In all instances, RNAs that were increased or decreased on RNA-seq analysis were also changed in the same direction on QPCR analysis. For bactin (1.4-fold change by QPCR vs. 0.94 change by RNA-seq), igf2a (1.7-fold change by QPCR vs. 1.7 change by RNA-seq), igfbp1b (4.2-fold change by QPCR vs. 3.6 change by RNA-seq), klf11a (6.7-fold change by QPCR vs. 11.1 change by RNA-seq), lepr (1.5-fold change by QPCR vs. 1.6 change by RNA-seq), and pparaa (0.8-fold change by QPCR vs. 0.4 change by RNA-seq). Increases in cyp24a1 RNA amounts after treatment of zebrafish for 6 d (144 h) with 30 pm 1α,25(OH)₂D₃ demonstrated a 11.2-fold change cyp24a1 RNA amounts.

Table 2.

Pathways significantly altered by 1α,25(OH)₂D₃ treatment on dpf 6 [120 h of 1α,25(OH)₂D₃ treatment]

KEGG pathway term	Gene count	P Value
PPAR signaling pathway	22	3.21E-12
Glycine, serine, and threonine metabolism	15	2.75E-10
Fatty acid metabolism	14	1.56E-08
Tryptophan metabolism	14	1.13E-07
Arginine and proline metabolism	17	1.52E-07
Valine, leucine, and isoleucine degradation	12	1.60E-05
Drug metabolism	8	1.84E-04
Metabolism of xenobiotics by cytochrome P450	8	1.84E-04
Starch and sucrose metabolism	9	4.06E-04
Biosynthesis of unsaturated fatty acids	7	4.09E-04
β-Alanine metabolism	7	5.69E-04
Glycolysis/gluconeogenesis	12	7.37E-04
Retinol metabolism	8	7.43E-04
Butanoate metabolism	8	9.42E-04
Histidine metabolism	7	0.00103
Pantothenate and CoA biosynthesis	5	0.004057
Glutathione metabolism	8	0.00519
Steroid hormone biosynthesis	7	0.007196
Linoleic acid metabolism	5	0.00972
d-Arginine and d-ornithine metabolism	3	0.010055
Drug metabolism	7	0.011572
α−Linolenic acid metabolism	4	0.018004
Glycerolipid metabolism	7	0.025381
Primary bile acid biosynthesis	4	0.030237
Ascorbate and aldarate metabolism	4	0.030237

Analysis was performed using the DAVID v6.7 program. CoA, Coenzyme A.

Fig. 2.

PPAR signaling pathway is significantly altered by 1α,25(OH)₂D₃ treatment on dpf 6 [120-h 1α,25(OH)₂D₃ treatment]. RNAs increased after 1α,25(OH)₂D₃ treatment (compared with vehicle) are indicated by red up arrows, and RNAs decreased after 1α,25(OH)₂D₃ treatment are indicated by blue down arrows. RXR, Retinoid X receptor; UCP, uncoupling protein; VLDL, very-low density lipoprotein.

Table 3.

Pathways significantly perturbed by 1α,25(OH)₂D₃ treatment on dpf 6 [120 h of 1α,25(OH)₂D₃ treatment]

Pathway	Database	P Value
RNA polymerase III chain elongation	Reactome	0.001367
Graft vs. host disease	KEGG	0.002477
Classical antibody-mediated complement activation	Reactome	0.002477
Calcitonin-like ligand receptors	Reactome	0.002533
RANKL pathway	Biocarta	0.002747
Antigen- dependent B cell activation pathway	Biocarta	0.004460
Apoptotic cleavage of cell adhesion proteins	Reactome	0.004460
ACE2 pathway	Biocarta	0.004460
FAS Signaling	SigmaAldrich	0.006892
α−Linolenic acid metabolism	KEGG	0.006892
JNK phosphorylation and activation mediated by activated human TAK1	Reactome	0.007631

Analysis was performed using the DIRAC algorithm.

Differential gene expression in response to 1α,25(OH)₂D₃ treatment on dpf 7 (144 h treatment) is shown in Supplemental Table 3. The RNA molecules increased to the greatest extent in response to 1α,25(OH)₂D₃ treatment on dpf 7 (144 h treatment) are those for lepb (leptin b), lepa (leptin a), hspa8 (heat shock protein), klf11, oxgr1, slc1a4, il1b (IL 1β), hbbϵ (hemoglobin ϵ), cpa2, scd5, and cyp24a1. The RNA molecules most decreased in response to 1α,25(OH)₂D₃ treatment are those for aquaporin 8 (aqp8), Rh family B glycoprotein (rhbg), rhodopsin (rho), catechol-O-methyltransferase domain containing protein 1 (comtd1), Epstein-Barr virus-induced protein 3 (ebi3), and insulin (ins). Pathway analysis using DAVID demonstrates changes in the PPAR signaling pathway. Other pathways altered by 1α,25(OH)₂D₃ include drug metabolism, fatty acid metabolism, and tryptophan metabolism signaling pathways. The complete list of pathways most significantly altered by 1α,25(OH)₂D₃ treatment is presented in Table 4. Examples of changes noted in pathways are shown in Supplemental Figs. 5–8. DIRAC analysis (Table 5) demonstrated changes in the calcitonin-like ligand receptor, leptin, trkA, TALL1, and JNK phosphorylation and activation pathways.

Table 4.

Pathways significantly altered by 1α,25(OH)₂D₃ treatment on dpf 7 [144 h of 1α,25(OH)₂D₃ treatment]

KEGG pathway term	Gene count	P Value
PPAR signaling pathway	22	2.61E-05
Tryptophan metabolism	17	2.81E-05
Arginine and proline metabolism	22	3.52E-05
Fatty acid metabolism	15	8.11E-05
Glycine, serine, and threonine metabolism	14	1.12E-04
DNA replication	16	1.22E-04
Aminoacyl-tRNA biosynthesis	15	3.47E-04
β-Alanine metabolism	9	0.003056
Valine, leucine, and isoleucine degradation	14	0.003622
Primary bile acid biosynthesis	7	0.003691
Pyrimidine metabolism	24	0.00391
Butanoate metabolism	11	0.004002
Alanine, aspartate, and glutamate metabolism	12	0.006128
Fructose and mannose metabolism	13	0.00835
Biosynthesis of unsaturated fatty acids	8	0.009276
Glycolysis/gluconeogenesis	17	0.010371
Cell cycle	30	0.010998
Starch and sucrose metabolism	11	0.011499
Terpenoid backbone biosynthesis	6	0.028806
Drug metabolism	8	0.029695
Metabolism of xenobiotics by cytochrome P450	8	0.029695
Retinol metabolism	9	0.031453
Arachidonic acid metabolism	10	0.032062
Pyruvate metabolism	11	0.037949
Drug metabolism	10	0.038672
Glycerolipid metabolism	11	0.044725

Analysis was performed using the DAVID v6.7 program.

Table 5.

Pathways significantly perturbed by 1α,25(OH)₂D₃ treatment on dpf 7 [144 h of 1α,25(OH)₂D₃ treatment]

Pathway	Database	P Value
Calcitonin-like ligand receptors	Reactome	4.98E-4
T cytotoxic pathway	Biocarta	0.001732
Leptin pathway	Biocarta	0.002419
RNA polymerase III transcription termination	Reactome	0.004295
Intrinsic pathway	Reactome	0.004295
Glycosaminoglycan biosynthesis keratin sulfate	KEGG	0.004295
Primary bile acid biosynthesis	KEGG	0.007948
TrkA receptor	SigmaAldrich	0.007948

Analysis was performed using the DIRAC algorithm.

To assess whether the changes in RNA amounts on d 7 were dependent on the length of exposure to 1α,25(OH)₂D₃, we treated 6-d-old zebrafish larvae with 1α,25(OH)₂D₃ or ethanol for 24 h (n = 3). We compared RNA amounts by RNA-seq in vehicle-treated vs. 1α,25(OH)₂D₃-treated zebrafish, and RNA in 7 d zebrafish treated with 1α,25(OH)₂D₃ for 6 d with those present after 1 d of treatment. After 1 d (24 h) exposure to 1α,25(OH)₂D₃ between d 6 and d 7, 427 RNA either increased or decreased. klf11a and klf11b RNA amounts increased 11.8- and 5.4-fold (Supplemental Table 4). These changes are similar to those observed after 6-d exposure to 1α,25(OH)₂D₃. Likewise, cyp24a1 RNA increased comparably by 6.3-fold. Pathway analysis using DAVID also demonstrates changes in the PPAR signaling pathway among others. The complete list of pathways most significantly altered by 1α,25(OH)₂D₃ treatment is presented in Table 6. However, as shown in Supplemental Table 5, 1146 RNAs at dpf 7 were increased or decreased in zebrafish exposed to 1α,25(OH)₂D₃ for 6 d (144 h), that were not changed in 7-d-old zebrafish exposed to 1α,25(OH)₂D₃ for 1 d (24 h). These included RNAs for hsp70 (heat shock protein 70), hsp70l, arrhα (aryl hydrocarbon receptor α), and lepb, which were increased only after prolonged treatment with 1α,25(OH)₂D₃. Analysis by DIRAC shows that the leptin, calcitonin-like ligand receptor, T-cytotoxic, and antigen-dependent B-cell activation pathways are more profoundly perturbed after prolonged (144 h) 1α,25(OH)₂D₃ treatment than short-term (24 h) treatment (P = 0.0004, 0.0011, 0.0026, and 0.0033, respectively).

Table 6.

Pathways significantly altered by 1α,25(OH)₂D₃ treatment on dpf 7 [24 h of 1α,25(OH)₂D₃ treatment]

KEGG pathway term	Gene count	P Value
PPAR signaling pathway	12	7.95E-07
DNA replication	10	1.17E-06
Pyrimidine metabolism	12	7.13E-05
Drug metabolism	6	0.002742
Pantothenate and CoA biosynthesis	4	0.004794
Cell cycle	11	0.005883
Biosynthesis of unsaturated fatty acids	4	0.01559
Base excision repair	5	0.018015
Purine metabolism	11	0.01803
Glutathione metabolism	5	0.02395
Metabolism of xenobiotics by cytochrome P450	4	0.02697
ABC transporters	4	0.04598

Analysis was performed using the DAVID v6.7 program. CoA, Coenzyme A.

Figure 3 shows the developmental distribution of the genes altered by 1α,25(OH)₂D₃ treatment. Genes are grouped based on their altered expression at single or multiple stages. Differential expression analysis was evaluated between ethanol and 1α,25(OH)₂D₃ treatment on dpf 2, dpf 4, dpf 6, and dpf 7. In total, 2553 mRNA transcripts were differentially expressed in at least one of the four studied developmental stages.

Fig. 3.

Venn diagram showing the developmental distribution of the genes altered by 1α,25(OH)₂D₃ treatment. Genes were grouped based on their altered expression at single or multiple stages. Differential expression analyses were evaluated between ethanol and 1α,25(OH)₂D₃ treatments on dpf 2, 4, 6, and 7. In total, 2553 mRNA transcripts were differentially expressed in at least one the four studied developmental stages.

To determine whether treatment with 1α,25(OH)₂D₃ altered the appearance and size of pharyngeal cartilage, we examined the length of palatoquadrate and ceratohyal cartilages in zebrafish stained with Alcian blue. The size of the palatoquadrate (PQ) cartilage was larger in 1α,25(OH)₂D₃-treated zebrafish (P = 0.0005) at dpf 7 after 6 d of continuous 1α,25(OH)₂D₃-treatment (Fig. 4). The ceratohyal (CH) cartilage was increased in length (P = 0.015). Alizarin red staining showed that at this time, calcified skeletal elements were more heavily stained in 1α,25(OH)₂D₃-treated zebrafish indicating greater calcification. Figure 5 shows more prominent red staining of the notochord (labeled N) and other bones of the developing skull of 1α,25(OH)₂D₃-treated zebrafish (P = 0.004). Differences in cartilage appearance were not apparent at dpf 4 and dpf 6.

Fig. 4.

Alcian blue-stained, dpf 7 (168 hpf) zebrafish larvae treated with ethanol (vehicle, EtOH) or 1α,25(OH)₂D₃ (300 pm) continuously for 6 d (144 h). The length of the palatoquadrate (PQ) and ceratohyal (CH) cartilages was assessed in each embryo. The palatoquadrate and ceratohyal cartilages were larger in 1α,25(OH)₂D₃-treated zebrafish compared with vehicle-treated zebrafish (palatoquadrate P = 0.0005; ceratohyal P = 0.015). A1, B1, and C1 represent ethanol-treated zebrafish, lateral view; A2, B2, and C2 represent ethanol-treated zebrafish, ventral view. D1, E1, and F1 represent 1α,25(OH)₂D₃-treated, lateral view; D2, E2, and F2 represent 1α,25(OH)₂D₃-treated, ventral view. VitD, Vitamin D.

Fig. 5.

Alizarin red-stained, dpf 7 (168 hpf) zebrafish larvae treated with ethanol (vehicle, EtOH) or 1α,25(OH)₂D₃ (300 pm) continuously for 6 d (144 h). The mineralized skeleton appears darker in the 1α,25(OH)₂D₃-treated zebrafish. n, Notochord. A1, B1, and C1 represent ethanol-treated zebrafish, lateral view; A2, B2, C2 represent ethanol-treated zebrafish, ventral view. D1, E1, and F1 represent 1α,25(OH)₂D₃-treated, lateral view; D2, E2, and F2 represent 1α,25(OH)₂D₃-treated, ventral view. VitD, Vitamin D.

Discussion

The VDR is widely and abundantly distributed in rodent and fish embryos/larvae (13–15, 17). We have described the presence of the VDR in mesenchymal tissue of the vertebrae and limb buds (13), kidney (14), and nervous system (dorsal root ganglia of the spinal cord) (15) of rat and mouse fetuses as early as 13 d post coitum. Zebrafish express the VDR in substantial amounts at 48 hpf (17). At this time, mineralized bone has not begun to form, although cartilage has just begun to develop. The biological function of the VDR at this stage of development remains unknown. VDR knockout mice do not have overt abnormalities in their morphology and blood chemistry at birth (7, 8, 43) because trans-placental transfer of nutrients, such as calcium, compensate for biological deficits that might occur as a result of VDR deficiency.

To assess the function of the VDR, we used whole transcriptome shotgun sequencing of zebrafish embryo RNAs after treatment of fish with 1α,25(OH)₂D₃ or vehicle. RNAs and pathways altered by 1α,25(OH)₂D₃ were assessed using bioinformatics tools adapted to assess zebrafish genes and RNA. Several pathways that previously have not been thought to be subject to regulation by 1α,25(OH)₂D₃ are shown to be dramatically regulated in the zebrafish embryo in vivo. For example, regulation of several RNAs in the PPAR signaling pathway demonstrates that in zebrafish embryos/larvae, 1α,25(OH)₂D₃ may significantly alter lipid metabolism and the adipocyte differentiation. The early change (at 4 dpf) in RNA for the Kruppel-like 11 transcription factors could play a role in altering fat metabolism in zebrafish because klf11 transcription factors alter chromatin remodeling, brown adipocyte differentiation and uncoupling protein UCP1 and thermogenic gene expression (44, 45). The decrease of RNAs encoding enzymes important in fatty acid oxidation suggests that 1α,25(OH)₂D₃ decreases fatty acid oxidation. These data are supported by information in the VDR knockout mouse in which energy expenditure is elevated and β-oxidation is increased (46, 47). In VDR knockout mice, leptin concentrations are decreased, whereas in zebrafish treated with 1α,25(OH)₂D₃ there are dramatic elevations in leptin a and b and leptin receptor RNA. Experiments in cultured human adipocytes (48) have shown that 1α,25(OH)₂D₃ inhibits uncoupling protein 2 expression. RNAs for enzymes important in bile acid biosynthesis are also affected by 1α,25(OH)₂D₃.

The regulation of key enzymes involved in the metabolism of various amino acids (e.g. tryptophan, arginine, proline, glycine, serine, and threonine among others) by 1α,25(OH)₂D₃ is also a previously unrecognized effect of the hormone. As expected, the RNA for the cyp24a1, which encodes the cytochrome P450, cyp24a1, a key part of the 1α,25(OH)₂D₃-24-hydroxylase that degrades 1α,25(OH)₂D₃ to 1,24,25-trihydroxyvitamin D₃ (49, 50), is consistently increased even at times when the zebrafish embryo is generally unresponsive to 1α,25(OH)₂D₃. In contrast to the induction of cyp24a1 RNA, the RNAs for a number of other cytochrome P450 involved in xenobiotic metabolism are reduced (cyp1A1, cyp3, cyp3A4, cyp26).

A number of RNAs important in cartilage and bone remodeling are altered by 1α,25(OH)₂D₃ treatment. These include RNAs for IGF and the IGF binding proteins. The effects of 1α,25(OH)₂D₃ on IGF and its binding proteins are consistent with previous findings regarding the effects of the hormone on these substances (51). The DIRAC algorithm finds several relevant pathways associated with osteoclast development to be perturbed in response to 1α,25(OH)₂D_3. Both the RANKL pathway and the calcitonin receptor pathway are highly perturbed on dpf 6 in larvae treated with 1α,25(OH)₂D₃. The RANKL pathway is known to be essential for osteoclastogenesis and bone resorption and to be regulated by 1α,25(OH)₂D₃ (52). Interestingly, the RANKL pathway is no longer perturbed in dpf 7 larvae, whereas the calcitonin receptor pathway becomes more highly perturbed. Also highly perturbed in dpf 7 is the leptin pathway, which has been linked to bone resorption via the sympathetic nervous system (53, 54).

Changes in RNA expression are associated with changes in cartilage morphology in 1α,25(OH)₂D₃-treated zebrafish. Cartilage appears to be larger in hormone-treated fish. Additionally, calcification of bone appears to be enhanced. It is possible that changes in IGF and IGF binding proteins induced by 1α,25(OH)₂D₃ are important in changes observed in cartilage.

In 48 hpf zebrafish, five RNAs are differentially expressed after 1α,25(OH)₂D₃ treatment for a period of 24 h. In contrast, on dpf 7, 24 h of 1α,25(OH)₂D₃ treatment results in a change in the RNA for 482 genes. Changes in the number and types of RNA present after treatment with 1α,25(OH)₂D₃ is thus dependent on the stage of zebrafish development. The difference cannot be attributed to a lack of expression of the VDR or its heterodimeric partner, retinoid X receptor-α, both of which are expressed at 48 hpf. Thus, other factors likely play a role in the development of sensitivity to hormone in older zebrafish larvae. Precedents for stage-specific competence to steroid hormones, such as ecdysone, have been noted in Drosophila and Aedes (55–57). The orphan receptor, β FTZ-F1, is responsible for the development of competence to ecdysone in both Drosophila and the mosquito. The factor (or factors) responsible for the development of enhanced RNA response to 1α,25(OH)₂D₃ in Danio are not known. Transcription factors, such as Kruppel-like factor 11, B-cell CLL/lymphoma 6, forkhead transcription factor, fos and fosB transcription factors, could conceivably play a role because RNA for these factors is consistently increased early after the addition of 1α,25(OH)₂D₃, and more prolonged exposure results in the increase or decrease in 2407 RNA, representing RNA for approximately 10% of all protein coding genes in Danio. This massive change in gene expression by 1α,25(OH)₂D₃ has not been documented in developing whole organisms before. Although it is likely that a majority of the change in RNA amounts is due to changes in gene transcription, we cannot be certain that the changes in RNA amounts occur as a result in changes in gene expression. Other mechanisms such as changes in RNA stability may be playing a role as well.

Conclusion

In developing zebrafish embryos/larvae, 1α,25(OH)₂D₃ dramatically alters RNAs encoding proteins that play a key role in fatty acid, amino acid, and xenobiotic metabolism pathways, and RNAs of transcription factors, leptin, peptide hormones, receptor-activator of NFκB ligand (RANKL), and calcitonin-like ligand receptor pathways. The results demonstrate that 1α,25(OH)₂D₃ regulates multiple, biologically diverse pathways during development that were not previously recognized as vitamin D targets.

NURSA Molecule Pages^†:

• Nuclear Receptors: VDR;

• Ligands: 1α,25-dihydroxyvitamin D₃.