Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2013 Aug 22;154(11):4396–4407. doi: 10.1210/en.2013-1432

Expression Profiling of Intestinal Tissues Implicates Tissue-Specific Genes and Pathways Essential for Thyroid Hormone-Induced Adult Stem Cell Development

Guihong Sun 1,*, Rachel A Heimeier 1,*,✉, Liezhen Fu 1,*, Takashi Hasebe 1, Biswajit Das 1, Atsuko Ishizuya-Oka 1,✉, Yun-Bo Shi 1,✉
PMCID: PMC3800751  PMID: 23970787

Abstract

The study of the epithelium during development in the vertebrate intestine touches upon many contemporary aspects of biology: to name a few, the formation of the adult stem cells (ASCs) essential for the life-long self-renewal and the balance of stem cell activity for renewal vs cancer development. Although extensive analyses have been carried out on the property and functions of the adult intestinal stem cells in mammals, little is known about their formation during development due to the difficulty of manipulating late-stage, uterus-enclosed embryos. The gastrointestinal tract of the amphibian Xenopus laevis is an excellent model system for the study of mammalian ASC formation, cell proliferation, and differentiation. During T3-dependent amphibian metamorphosis, the digestive tract is extensively remodeled from the larval to the adult form for the adaptation of the amphibian from its aquatic herbivorous lifestyle to that of a terrestrial carnivorous frog. This involves de novo formation of ASCs that requires T3 signaling in both the larval epithelium and nonepithelial tissues. To understand the underlying molecular mechanisms, we have characterized the gene expression profiles in the epithelium and nonepithelial tissues by using cDNA microarrays. Our results revealed that T3 induces distinct tissue-specific gene regulation programs associated with the remodeling of the intestine, particularly the formation of the ASCs, and further suggested the existence of potentially many novel stem cell-associated genes, at least in the intestine during development.

When a multicellular organism suffers damages to tissues/organs it heals itself by either substituting the lost cellular matrix by scar formation or regenerating the lost tissues. Regeneration and repair are suggested to occur by recapitulation of the process that formed the tissue/organ during development. Many processes regulating organ development are based on epithelial interactions and communications with the underlying mesenchyme to strictly control stem cell activity. As such, adult organ-specific stem cells are critical for organ regeneration and repair. Perturbed regulation of stem cell activity and the signaling cascades through mutational activation or inactivation can disrupt epithelial homeostasis and may result in cancer on the tissue/organ. Elucidation of the molecular basis of these processes and their signaling pathways is therefore vital for clinical research in order to develop novel therapies in regenerative medicine.

The intestine is a model organ for studying proliferation and differentiation and stem cell function because the intestinal epithelium is rapidly and continuously renewed throughout the entire lifespan (1–7). The mammalian intestine comprises crypts and villi. The continuous renewal of the intestinal epithelium is achieved by a small number of stem cells residing near the base of each intestinal crypt in a protective niche that proliferate and differentiate along the crypt-villus axis of the intestine. After a finite period of time, the differentiated epithelial cells undergo cell death at the tip of the villi, where they are shed into the lumen. This process is unique among different organs and conserved across vertebrates. In adult mice, the epithelium is replenished every 5 days (1–3), whereas in adult frogs the epithelium is renewed in 2 weeks (8). A handful of signaling pathway cascades required for intestinal development and cell renewal have been identified and contribute to our understanding of intestinal homeostasis and neoplasia development (3, 9). However, the molecular mechanisms regulating the formation of the adult stem cells during development are poorly understood. Currently, there is an urgent need to identify unique regulator/marker genes for stem cells and the crypt niche to study how adult stem cells (ASCs) are formed and to understand their role in self-renewal.

Among vertebrates, amphibian metamorphosis shares strong similarities with postembryonic development in mammals, including intestinal maturation, and therefore offers a unique opportunity to study the complexities involved during organogenesis and cell regeneration in vertebrate development. Amphibian metamorphosis is controlled by T3 and resembles mammalian postembryonic development, a period around birth when plasma T3 levels are also high (10, 11). During metamorphosis, the intestine is remodeled into a form similar to that in adult mammals. More importantly, a number of recent molecular and genetic studies suggest that the formation of mammalian adult intestine is dependent on T3 and involves similar mechanisms (4, 5, 7, 12–18). Thus, intestinal remodeling during metamorphosis serves as a valuable model to study ASC development in vertebrates.

The larval intestine of Xenopus laevis is a simple tubular structure consisting of a single layer of primary epithelium and thin layers of immature connective tissue and muscles. During metamorphosis, the larval epithelial cells undergo degeneration through programmed cell death or apoptosis. Concurrently, adult intestinal stem cells are formed de novo through dedifferentiation of cells within the larval epithelium (4, 19). The stem cells, as well as the cells of the connective tissue and muscles, proliferate and eventually, the adult cells differentiate to form a multiply folded epithelium surrounded by well-developed connective tissue and muscles. After metamorphosis, the adult epithelium is renewed along the trough-crest axis of the intestinal folds, similar to the mammalian crypt-villus axis.

The intestinal epithelium (Ep) and the rest of the intestine, or the nonepithelium (non-Ep), not only change dramatically during metamorphosis but also influence each other's remodeling (4, 20–22). Previously we have shown that the non-Ep is essential for adult epithelial development likely by providing the microenvironment (niche) around the developing stem cells (20, 23). Our earlier studies suggest that T3 can induce some cells within the larval epithelium to undergo autonomous transformation into the precursor form of the adult intestinal stem cells. However such cells will not dedifferentiate into adult stem cells unless T3 signaling is also active in the non-Ep (23). These findings suggest that both cell-autonomous and cell-cell interaction-dependent signaling pathways are important for the formation of the adult stem cells, with the latter likely functioning via the formation of the ASC niche. Therefore, amphibian metamorphosis also represents an excellent model for examining T3-dependent regulation of the formation of the stem cells and its associated niche.

To understand ASC development, it is critical to determine the genes and signaling pathways involved in this process. In this study our objective was to profile the gene expression programs in the epithelial and nonepithelial tissues in X. laevis intestine to systematically determine changes underlying the cell-autonomous and cell-cell interaction-dependent processes that are required for stem cell formation. We isolated the intestines of tadpoles from premetamorphosis (stage 56), metamorphic climax (stage 61/62, when larval epithelial cell death is near completion, cell proliferation is rapid and most of the cells are adult progenitor/stem cells), and end of metamorphosis (stage 66, when adult intestine is formed) and separated the Ep from the non-Ep of each intestine. High throughput gene expression profiling on the isolated tissues revealed that most of the genes regulated during intestinal metamorphosis were tissue specific and that distinct signal transduction pathways were affected by T3 in the Ep and non-Ep, suggesting that their differential role in cell autonomous functions in stem cell development vs in the stem cell niche formation.

Materials and Methods

Experimental animals

X. laevis tadpoles from pre- (stage 56), climax (stage 61–62), and post- (stage 66) metamorphosis were purchased from NASCO or produced in the laboratory. Developmental stages were determined as described elsewhere (24). All animals were maintained and used as approved by Animal Use and Care Committee of National Institutes of Child Health and Human Development, National Institutes of Health or Nippon Medical School Animal Use and Care Committee.

Tissue collection and microarray analyses

The anterior region of the intestine from tadpoles at indicated stages was dissected, flushed of contents, treated with 1000 U/mL dispase (Godo) to separate the Ep from non-Ep, with the latter consisting predominantly of connective tissue and muscles (23). RNA was isolated and 2 biological replica RNA samples for stages 56, 61, and 66 were subjected to microarray (Agilent slides AMADID 013665) analysis by using a 2-color reference design system (25, 26). To obtain sufficient RNA for analysis, 60–100 tadpoles were pooled for each biological replicate per developmental stage. To identify statistically significant gene expression changes, ANOVA was performed across all developmental time points with a False Discovery Rate of ≤ 5% for multivariate correction. Bioinformatics analysis was performed with the software from NIA microarray analysis tool (http://lgsun.grc.nia.nih.gov/ANOVA) (27, 28), GenMAPP and MAPPFinder with GO terms (29–31), and GOMiner (32, 33).

Quantitative RT-PCR (qRT-PCR)

This was done as described (18, 34) by using primers and either FAM dye-labeled TaqMan probes or SYBR Green I dye (Applied Biosystems) with an ABI 7000 system (Applied Biosystems). The genes examined with the FAM dye-labeled Taqman probes were stromelysin-3 (ST3), gelatinase A (GelA) GelA, intestinal fatty acid binding protein (IFABP), sonic hedgehog (XHH), and T3-responsive basic leucine zipper transcription factor (TH/bZIP). The expression level of each gene was normalized to that of the control rpl8 gene (35). Additional genes were analyzed with SYBR Green I dye, with the expression level of each gene being normalized to that of the control gene, the somatic EF-1α (36). All gene-specific primers are summarized in Supplemental Table 1 published on The Endocrine Society's Journals Online web site at http://endo.endojournals.org.

In situ hybridization

To generate hybridization probes, cDNA fragments were amplified by PCR and cloned into pCR TOP vectors for sequence confirmation (Invitrogen). The primers for PCR amplifications were: 5′-TGAATGCCGCTTCCTTTGG-3′ and 5′-CGTCATCATCCATCTCTTTCCTGG-3′ for apolipoprotein A1 (APOA1) 5′-GGCATCCGAAGAGTATGGTGAAC-3′ and 5′-GAACAAGGAACACAGGACACAGC-3′ for solute carrier family 44, member 4 (SLC44A4); 5′-TTTTCCTGAACAGCAATGAGGC-3′ and 5′-CCACACACAGTTATGACGCACTCC-3′ for lipoprotein A (LPA). The plasmids were linearized to synthesize either sense or antisense probes with T7 or SP6 RNA polymerase, respectively, by using digoxigenin RNA labeling mix (Roche Applied Science). In situ hybridization was performed on 4% paraformaldehyde-fixed cryosectioned tissues as previously described (37).

Results

Microarray analysis identifies the genes regulated in the Ep and non-Ep during intestinal metamorphosis

Our earlier studies suggest that T3 acts in a tissue-autonomous manner to initiate the dedifferentiation of some larval epithelial cells to form adult intestinal stem cells (23). However, such cells will not develop into ASCs unless T3 signaling is also active in the rest of the intestine or the non-Ep (22, 23). Thus, ASC development involves T3-induced gene regulation program specific to the Ep and non-Ep. To determine these T3-induced gene regulation programs in the Ep and non-Ep, we separated the Ep from the non-Ep of the small intestine of tadpoles at different stages, including stage 56, the onset of metamorphosis when T3 is first detectable (referred to as a “premetamorphic stage” for simplicity); stage 61, the climax of metamorphosis when T3 is near the peak levels, a large fraction of the cells in the epithelium are adult progenitor/stem cells, and the larval epithelium undergoes apoptosis (19); and stage 66, the end of metamorphosis when T3 levels are lower and the adult epithelium has developed. Total RNA was then purified from the isolated tissues.

To identify tissue-specific gene regulation programs underlying metamorphosis, it is critical to have well-separated tissues for microarray analyses. To ensure that the RNA samples represented signature gene expression patterns of both stage- and tissue-specific program during development, we assessed the expression of established tissue-specific genes regulated by T3 during metamorphosis in each RNA sample prior to microarray analysis. These include one up- and one down-regulated Ep-specific gene (sonic hedgehog and IFABP, respectively) (38, 39), two up-regulated non-Ep-specific genes (ST3 and GelA, expressed in the connective tissue) (37, 40), and one up-regulated ubiquitously expressed gene (TH/bZIP) (41, 42). The tissue specificity and developmental expression profiles of these genes in the isolated Ep and non-Ep samples agreed completely with earlier findings based on expression studies on total intestinal RNA samples and in situ hybridization analyses (Figure 1), confirming that the RNA samples collected represented Ep and non-Ep tissues from premetamorphosis, metamorphic climax, and the end of metamorphosis. Furthermore, quantitative comparison of the expression levels of different tissue-specific genes in Ep and non-Ep indicated that tissue contamination between Ep and non-Ep during tissue isolation, if any, was less than 1%.

Figure 1.

Figure 1.

Open in a new tab

qRT-PCR Characterization of the isolated tissues. The total RNA from the Ep and non-Ep was used to analyze the expression of genes regulated by T3 specifically in the connective tissue (ST3, GelA), epithelium (XHH, IFABP), or ubiquitously (TH/bZIP), during natural development. The results are expressed in arbitrary units after normalization against the control gene, rpL8. Note that all tissue-specific genes were detected in only the expected tissues, with less than 1%, if any, contaminations in the nonexpressing tissue, confirming the purity of the isolated tissue samples.

To identify tissue-specific gene expression changes during intestinal development, we subjected the tissue-specific RNA at different stage to genome-wide microarray analyses. This led to the identification of 1985 genes significantly regulated in the Ep (Supplemental Table 2) and 2826 genes in the non-Ep (Supplemental Table 3) during T3-dependent intestinal remodeling (Supplemental Table 4 lists all genes the expression of which was different among different stages and/or between Ep and non-Ep, together with their expression levels at different stages in the Ep and non-Ep). To validate microarray data, we first analyzed the microarray signals for all genes across all 12 arrays and found that the 2 independent samples for each time point of each tissue type clustered together but away from other time points (data not shown), demonstrating the consistency of tissue isolation and microarray reproducibility. Next we randomly selected a number of genes the expression of which changed by 3-fold or more during development according to microarray data and performed qRT-PCR analyses on independently isolated intestinal Ep and non-Ep RNA and observed that the qRT-PCR results were all consistent with the microarray data (Figure 2A and data not shown). Finally, to confirm tissue specificity of the gene expression, we performed in situ hybridization on intestinal sections for several genes specific for Ep and non-Ep (ie, with at least 10-fold difference between Ep and non-Ep based on the microarray data) and found that the in situ patterns during development agreed with the microarray findings (Figure 2B and data not shown). Thus, the microarray data faithfully reflected spatiotemporal expression profiles of the genes during intestinal metamorphosis.

Figure 2.

Figure 2.

Open in a new tab

Validation of microarray results. A, Consistent gene expression patterns as revealed by microarray and qRT-PCR. A number of regulated genes identified from the microarrays were selected for qRT-PCR analysis. The results were all consistent with the microarray findings, and the expression patterns for 4 representative genes are shown here. All expression data were shown in arbitrary units. HAS2, hyaluronan synthase gene; CV079029, an expression sequence tag (GenBank # CV079029). B, In situ hybridization analysis confirms the tissue specificity of the genes as identified from the microarrays. In situ hybridization on intestinal cross-sections at stages 56, 62, and 66 was carried out with sense or antisense probe for APOA1, SLC44A4, or LPA. Note all 3 genes had little expression at stages 56 or 62 but were highly expressed in the epithelium (APOA1 or SLC44A4) or the connective tissue (LPA) at stage 66, in agreement with the microarray or qRT-PCR data (Figure 2 and data not shown). AE, adult epithelium; CT, connective tissue; L, lumen; M, muscle; Ty, typhlosole.

Most of the regulated genes are developmental stage- and tissue specific

To compare the gene regulation programs in the Ep and non-Ep, the Venn diagram analysis was carried out on the regulated genes. The vast majority of the genes were found to be developmentally regulated only in either the Ep or the non-Ep with only 732 genes commonly regulated in both Ep and non-Ep, representing 37% and 26% of the total regulated genes in the Ep and non-Ep, respectively (Figure 3A). To rule out the possibility that some of the regulated genes might be due to tissue contamination during isolation, we identified all genes the highest expression level of which at any of the three stages in one tissue (eg, Ep) was at least 100 times higher than their highest expression level at any stages in the other tissue (eg, non-Ep) based on microarray data (we chose 100-fold because the data in Figure 1 suggest that contamination, if any, would be less than 1%) and found 590 such tissue-specific genes, ie, with at least 100-fold difference between the expression levels in Ep and non-Ep. Importantly, out of these 590 tissue-specific genes, 239 were developmentally regulated in either Ep or non-Ep but none were among the 732 genes commonly regulated in both Ep and non-Ep (data not shown). These findings indicate that there were no genes falsely identified as developmentally regulated genes due to tissue contamination in either Ep or non-Ep. These microarray data showed that most of the genes regulated during intestinal remodeling are tissue specific, arguing that distinct gene regulation programs underlie the distinct metamorphic changes in different tissues.

Figure 3.

Figure 3.

Open in a new tab

Venn diagram analyses reveal tissue- and stage specificity of developmentally regulated genes. A, Venn diagram shows that most of the developmental regulated genes are tissue-specific. Note that of about 1200–2100 or so genes are uniquely regulated in the Ep and non-Ep, respectively, whereas only 732 genes were regulated in both Ep and non-Ep. The total number of transcripts regulated in Ep and non-Ep were 9.2% and 13.1% of all (21 495) transcripts analyzed. B, Venn diagrams showing the number of genes regulated between different pairs of two different stages in the Ep and nonepithelial tissues in the intestine during natural metamorphosis. Note that distinct genes were up-regulated or down-regulated from 56 to 61 compared with those from 61 to 66.

Next, we compared the genes regulated between stage 56 and 61 to those regulated between stage 61 and 66. In the Ep, there were 830 genes that were up-regulated between 56 and 61 and 322 between stage 61 and 66 (Figure 3B). However, only 5 genes were common between the two up-regulated gene lists, ie, these 5 genes were up-regulated from stage 56 to 61 and were further up-regulated by stage 66. Similarly, of the 350 genes down-regulated in the Ep between stage 56 and 61 and 695 genes down-regulated in the Ep between stage 61 and 66, only 10 were common. In non-Ep, there were only 27 genes commonly up-regulated and 3 genes commonly down-regulated both between stage 56 and 61 and between stage 61 and 66 (Figure 3B). Thus, in both the Ep and non-Ep, the gene regulation programs underlying the changes between stage 56 and 61 and those underlying the changes between stage 61 and 66 are distinct, consistent with the underlying developmental changes. In particular, between stage 56 and 61, the larval epithelial cells undergo apoptosis and concurrently, adult epithelial progenitor/stem cells are formed de novo and rapidly proliferate. In contrast, between stage 61 and 66, the major events in the epithelium are the proliferation and subsequent differentiation of the adult epithelial cells, contrasting sharply with the events between stage 56 and 61. In the non-Ep, the metamorphic transformations are less dramatic. Between stage 56 and 61, the connective tissue increases in mitotic activity, cell types, cell number, and thickness. In particular, the fibroblasts in connective tissue are activated and become rich in rough endoplasmic reticulum, and the basal lamina or basement membrane separating the connective tissue and epithelium becomes thick and amorphous (19). After stage 61/62, the basal lamina becomes thin beneath the newly developed adult epithelium, and the connective tissue cells become ordinary fibroblasts. Such distinct changes are also consistent with the gene regulation data during the two developmental periods.

Temporal tissue-specific gene regulation programs underlying stem cell development

To determine the gene regulation pathways that control Ep and non-Ep tissue transformations during intestinal remodeling, we clustered the Ep- and non-Ep-regulated genes into 8 possible expression patterns by using 1.5-fold cutoff for gene expression change from one stage to the next. In both Ep and non-Ep, about half or more of the regulated genes had either maximal or minimal expression level at the climax of metamorphosis (stage 61) (patterns 1 and 2; Figure 4), indicating that most of the changes in gene expression are only transient and involved in the remodeling process whereas the rest of the changes likely reflect the different nature of the larval and frog intestine.

Figure 4.

Figure 4.

Open in a new tab

Most of the developmental regulated genes have their maximal or minimal expression levels at the metamorphic climax. The regulated genes were clustered into 8 possible development groups based on their development expression profile. The top 2 patterns together contained about half of the regulated genes.

At the climax of metamorphosis, adult progenitor/stem cells represent a large fraction of the cells in the Ep, and in addition, the non-Ep plays an essential role for the stem cell development (23). The genes with peak or minimal expression specifically at the climax of metamorphosis are therefore likely to play important roles in stem cell development and/or proliferation. To identify the gene regulation programs important for the ASCs, we thus focused our analysis on such genes (patterns 1 and 2 in Figure 4). Gene Ontology (GO) analysis was carried out to identify biological categories enriched among these genes based on the human RefSeq homologs (Supplemental Tables 5–8). The GO analysis revealed that in the Ep, GO categories such as cell differentiation (GO:30154), cell proliferation (GO:8283), epithelial development (GO:60429), and cell death (GO:8219), etc., were enriched among the up-regulated genes with peak expression at the climax of metamorphosis (Supplemental Table 5), consistent with the apoptosis of larval epithelium and development of the adult epithelial stem cells. In contrast, in the non-Ep, GO categories such as proteolysis (GO:6508), peptidase activity (GO:8233), metalloproteinase activity (GO:8237), and cellular component biogenesis at cellular level (GO:71843), etc., were significantly enriched among the up-regulated genes with peak expression at the climax of metamorphosis (Supplemental Table 7). When the enriched GO categories were ranked according to the number of genes regulated in each category, the GO categories with the highest numbers of regulated genes in the 4 groups of genes (patterns 1 and 2 in Figure 4 for Ep and non-Ep, respectively) were, not surprisingly, large GO categories such as biological regulations (GO:65007, 2693 genes), regulation of biological process (GO:50789, 2503 genes), cellular component (GO:5575, 5075 genes), etc., (Figure 5 and Supplemental Tables 5–8). Interestingly, many of the GO categories enriched in the group of genes with peak expression at the climax of metamorphosis in Ep were enriched in the group of genes with minimal expression at the climax of metamorphosis in non-Ep. Conversely, a number of GO categories enriched in the group of genes the expression of which were the lowest at the climax of metamorphosis in Ep were also enriched in the group of genes with peak expression at the climax of metamorphosis in non-Ep (see Figure 7and Supplemental Tables 5–8). Thus, these enriched GO categories also support the contrasting changes in the Ep and the non-Ep associated with stem cell development.

Figure 5.

Figure 5.

Open in a new tab

GO analysis reveals tissue-specific enrichment of genes in different functional categories during T3-regulated metamorphosis. The genes the expression of which peaks or troughs at the climax of intestinal metamorphosis were identified and subjected to GO analysis. The significantly enriched functional groups (P < .05) were identified (Supplemental Tables 5–8), and the top 10 categories with the highest number of genes regulated were listed. Heat maps for each group of genes were shown on the left of each panel. Note that a number of GO categories (in the colored boxes) were enriched in both the Ep and non-Ep but with opposite regulation patterns.

Figure 7.

Figure 7.

Open in a new tab

Tissue-specific regulation of genes associated with stem cell development and maintenance. The genes associated with stem cell development and maintenance were identified from the developmentally regulated genes in the Ep, non-Ep, or both, and their expression profiles were shown as clusters of similar expression patterns. Note that 19 of 52 stem cell-associated genes from Ep had peak levels of expression at the climax (stage 61) whereas only 11 of 115 of the non-Ep genes belonged to this pattern (pattern 1, Figure 4). Of the genes regulated in both Ep and non-Ep, the patterns varied among genes.

When we analyzed the GO categories enriched in the 732 commonly regulated genes and the 1253 and 2094 genes specifically regulated in the Ep and non-Ep, respectively (Figure 3A), we found that among the top 20 enriched GO categories (based on the number of genes regulated in each GO category), 5 categories (GO:5737, 44444, 16020, 71944, and 5886) were among the top 20 GO categories that were enriched in the 1253 Ep-regulated genes and 6 of them (GO:32501, 7275, 48856, 48731, 71944, and 5886) were among the top 20 GO categories that were enriched in the 2094 non-Ep-regulated genes (Supplemental Table 9). For example, the GO category cell periphery (GO:71944) was enriched among the commonly regulated genes as well as in the Ep- and non-Ep-specifically regulated genes, involving 86 coregulated genes, 134 Ep-specifically regulated genes, and 209 non-Ep-specifically regulated genes, respectively. Similarly, the GO category plasma membrane (GO:5886) had 84 coregulated genes, 132 Ep-specifically regulated genes, and 204 non-Ep-specifically regulated genes. Thus, even in GO categories commonly enriched in the Ep and non-Ep, distinct genes were regulated, suggesting different regulation and/or mechanisms governing the tissue-specific changes during metamorphosis. For example, 2 well-studied signaling pathways, the MAPK signaling pathway and the Wnt signaling pathway/pluripotency, were among the significantly regulated pathways in both the Ep and non-Ep. However, in both cases, most of the regulated genes were regulated only in the Ep or non-Ep (Figure 6). The involvement of different tissue-specifically regulated genes in these pathways suggests that these pathways play different roles in the different tissues although how such involvement of different genes in the same pathways affects tissue-specific transformations remains to be investigated.

Figure 6.

Figure 6.

Open in a new tab

Representative gene regulation pathways potentially involved in the development of intestinal stem cells as revealed by microarray analysis. Note that both MAPK and Wnt pathways are known to be associated with cell proliferation and stem cells. Interestingly, whereas both pathways are regulated during metamorphosis in both the Ep and non-Ep, distinct genes within the pathways are regulated in the Ep and non-Ep, supporting different roles of the Ep and non-Ep in stem cell development and proliferation.

Regulation of stem cell-associated genes during intestinal metamorphosis

Many genes are known to be associated with stem cells. Given the fact that all larval epithelial cells will be replaced by the newly formed adult progenitor/stem cells at the climax of metamorphosis, we were interested to determine how such stem cell-associated genes are regulated during intestinal metamorphosis. We first identified all known stem cell-associated genes by searching the AmiGO Gene Ontology terms for “Stem cell” against all GO categories. Comparison of the 1397 genes thus identified against the 21 495 probes on the Xenopus microarray slide revealed that 618 probes were derived from these stem cell-associated genes. Of the transcripts hybridized to the 618 probes, 212 transcripts were regulated during development (Supplemental Table 10). Among them, 45 transcripts were commonly regulated in both Ep and non-Ep (Figure 7), and nearly half had minimal expression in the Ep at the climax of metamorphosis, suggesting that many of these commonly regulated genes function only in differentiated cells. Among the 52 transcripts that were uniquely regulated in the Ep, about 40% had maximal expression at the climax of metamorphosis in the Ep when stem cell development and proliferation took place, suggesting that they are likely to function in stem cells. For example, sonic hedgehog has long been shown to be up-regulated in the developing stem cells, and both the sonic hedgehog gene and the related Indian hedgehog gene were among this group of genes. The non-Ep had the most number of uniquely regulated stem cell-associated genes, with 115 transcripts belonging to this group. Unlike the Ep-specific genes, only a very small fraction, 11/115, had peak expression at the climax whereas nearly half had minimal expression at the climax. Thus, the down-regulation of many of the genes specifically regulated in the non-Ep appears to be critical for intestinal metamorphosis.

Discussion

Adult organ-specific stem cells are essential for adult organ homeostasis and tissue repair and regeneration. To date, little is known about the underlying molecular basis responsible for the formation of such ASCs during development. Here, using intestinal remodeling during X. laevis metamorphosis as a model, we investigated gene expression profiles underlying cell-autonomous determination of the adult epithelial progenitor cells and the establishment of the stem cell niche in the non-Ep. Our findings indicate that the vast majority of the regulated genes during intestinal remodeling are tissue- and developmental stage specific. We further show that distinct genes belonging to specific functional groups (GO) and signaling pathways in different tissues are temporally correlated with stem cell development and that extensive interactions among developmentally regulated genes in the Ep and non-Ep are likely critical for this process.

Intestinal remodeling during metamorphosis offers a unique opportunity to study gene regulation programs for ASC development. This is due, in part, to essentially total replacement of the epithelium with the larval epithelial cells undergoing apoptosis and concurrent de novo development of the adult epithelial progenitor/stem cells. Thus, at the late climax of metamorphosis (around stage 61–62), the intestinal epithelium is made of mostly proliferating adult epithelial progenitor/stem cells with some residual dying larval epithelial cells. This makes it possible to easily isolate samples with highly enriched adult epithelial progenitor/stem cells, ie, the Ep at the climax of metamorphosis, as well as the surrounding tissues (non-Ep) likely critical for the formation of the stem cell niche, for global gene expression analysis.

For the microarray analysis, we chose three stages, the premetamorphic stage 56 when T3 level is low before significant metamorphic changes occur in the intestine, the climax stage 61 when a significant fraction of the epithelial cells are adult progenitor/stem cells, and the end of the metamorphosis (stage 66) when basic adult intestine is formed. Our global gene expression analysis revealed that the genes that are regulated between stages 56 and 61 and those between stages 61 and 66 are distinct with little overlap in either the Ep or non-Ep. This correlated well with the developmental processes during intestinal metamorphosis. For example, the Ep first undergoes extensive apoptosis to remove the larval cells (19, 43). Concurrently, adult epithelial progenitor/stem cells are formed de novo and rapidly proliferate to become the major fraction of the cells in the Ep by stage 61/62. The developmentally regulated genes during this first phase would be expected to be important for larval apoptosis and adult cell development and proliferation. After stage 61/62, as larval cell death completes, adult epithelial progenitor/stem cells begin to differentiate to form the adult epithelium. This will involve the down-regulation of stem cell-specific genes or genes involved in proliferation and up-regulation of genes important for the adult epithelium. Consequently, the genes regulated in these 2 phases are expected to be distinct.

In the non-Ep, the metamorphic transformations are less dramatic, but it involves extensive cellular changes. The tadpole intestine has only a very thin layer of connective tissue, except in the typhlosole (the only epithelial fold in the tadpole). Between stage 56 and 61, the connective tissue increases in mitotic activity, cell types, cell number, and thickness. The connective tissue around stage 61 consists of various types of cells such as immature proliferating mesenchymal cells, fibroblasts that are activated and rich in rough endoplasmic reticulum, macrophages, and mast cells (44–46). Furthermore, remarkable changes in the connective tissue occur close to the epithelium. When the primary epithelium begins to degenerate, the basal lamina or the basement membrane, which is thin throughout the larval period, becomes thick and amorphous. Through gaps in such basal lamina, the fibroblasts that possess well-developed rough endoplasmic reticulum often contact the developing adult epithelial progenitor/stem cells. From stage 61/62 to the end of metamorphosis, the basal lamina becomes thin beneath the newly developed adult epithelium, and cell-cell contacts and all cell types of the connective tissue, except fibroblasts, decrease in number. In addition, the muscles rapidly proliferate and increase in thickness in this second phase. By the end of metamorphosis, almost all of the connective tissue cells are ordinary fibroblasts, and the connective tissue and the muscles are much more abundant to support a multiply folded adult epithelium. Such changes are consistent with the distinct genes regulated in these two phases.

Our analysis also showed that the developmentally regulated genes are mostly tissue-specific, ie, most genes are regulated in either Ep or non-Ep but not in both, indicating that distinct gene regulation programs are required in different tissues for the formation of the adult intestine. Our earlier studies have shown consistently that the Ep and non-Ep have distinct roles in the formation of the adult intestine (22, 23, 47). In particular, using recombinant organ cultures with wild-type and transgenic tissues expressing a dominant positive TR under the control of a heat shock-inducible promoter, we have recently shown that T3 action in the larval epithelium can induce tissue-autonomous dedifferentiation of larval epithelial cells, leading to the formation of adult progenitor cells lacking the expression of differentiation marker gene IFAPB (intestinal fatty acid binding protein) while expressing sonic hedgehog, a marker for adult progenitor/stem cells. On the other hand, in the absence of the dominant-positive TR in the non-Ep, such progenitor cells cannot proceed to develop into stem cells expressing markers of adult intestinal stem cells, as observed in mammalian intestine, or form the adult intestine expressing IFABP after prolonged culturing of the organ culture. Thus, T3 action in the non-Ep plays an essential role in the formation of the stem cells from the adult progenitor cells, likely by participating in the formation of the ASC niche during intestinal remodeling. The unique, tissue-specific gene regulation profiles that we identified from our microarray analysis of the Ep and non-Ep are likely responsible for the cell autonomous formation of the epithelial progenitor cells and the stem cell niche, respectively.

At the climax of metamorphosis (stage 61/62), most of the epithelial cells are adult progenitor/stem cells. Thus, genes the expression of which peaks at the climax of metamorphosis in either the Ep and Non-Ep (Figure 4, pattern 1) are involved in adult intestinal stem cell formation and/or proliferation. Interestingly, most of these genes were not previously known to be associated with stem cells (only about 200 known stem cell-associated genes were regulated in either the Ep and/or Non-Ep; Figure 7). Thus, our study here also revealed potentially many novel stem cell-associated genes, at least in the intestine during development.

Growing evidence suggests that the formation of the adult intestinal stem cells is conserved among vertebrates. During mammalian development, a primitive but functional intestine is formed by the end of embryogenesis (6, 19). The intestine subsequently matures into the adult form during the so-called postembryonic development when T3 levels are high (7, 10, 16, 17, 48, 49), resembling amphibian metamorphosis. For example, during mouse development, the morphogenesis of intestinal villi starts around embryonic day 14.5 (E14.5) (16, 17). The mature villus-crypt axis, with the stem cells located in the crypt, is formed during the first few weeks of postnatal life and T3 and T3 receptor (TR) are critical for the normal formation of the adult crypt-villus axis and the proliferation of the crypt cells (12–15, 50). More importantly, several recent studies indicate that the ASCs in the mouse and zebrafish intestine are distinct from the epithelial stem/proliferating cells in the embryonic or neonatal intestine (16, 17, 51, 52) and that this transition occurs when plasma T3 levels were high (53, 54), resembling amphibian metamorphosis. Thus, it is tempting to speculate that many of the stem cell-associated genes and signaling pathways that we identified here are likely applicable to the mammalian development. In support of this, we observed that a number of well-studied signaling processes known to important for adult mammalian intestinal stem cells, such as the Wnt signaling pathways (3), were enriched among the genes regulated during the intestinal metamorphosis. Interestingly, the regulation of these pathways often involved distinct genes in the Ep and non-Ep (Figure 6), suggesting that the same pathways may have different functions in different tissues to enable the formation and/or the proliferation of ASCs during metamorphosis. Such involvement of different genes in the Ep and Non-Ep may be a underlying molecular basis for the cell-autonomous formation of the epithelial progenitor cells by the Ep and contribution to the stem cell niche formation by the non-Ep during ASC formation. It will be of interest to determine whether such genes and signaling pathways are regulated and function during intestinal development in other vertebrates, especially mammals.

Supplementary Material

Supplemental Data
supp_154_11_4396__index.html (979B, html)

Acknowledgments

We thank Ms Mitsuko Kajita for her technical help.

This work was supported by the intramural Research Program of National Institute of Child Health and Human Development, National Institutes of Health, and in part by Narishige Zoological Science Award (to T.H.).

Current address for B.D.: Laboratory of Immunopathogenesis and Bioinformatics, Clinical Services Program, SAIC-Frederick, Inc, Frederick National Laboratory for Cancer Research, Frederick, Maryland 21702.

Disclosure Summary: The authors have nothing to disclose.

Footnotes

Abbreviations:
APOA1
apolipoprotein A1
ASC
adult stem cell
Ep
epithelium
GelA
gelatinase A
GO
Gene Ontology
IFABP
intestinal fatty acid binding protein
LPA
lipoprotein A
non-Ep
nonepithelium
qRT-PCR
quantitative RT-PCR
TR
T3 receptor
SLC44A4
solute carrier family 44, member 4
ST3
stromelysin-3
TH/bZIP
basic leucine zipper transcription factor.

References

  • 1. MacDonald WC, Trier JS, Everett NB. Cell proliferation and migration in the stomach, duodenum, and rectum of man: radioautographic studies. Gastroenterology. 1964;46:405–417 [PubMed] [Google Scholar]
  • 2. Toner PG, Carr KE, Wyburn GM. The Digestive System: An Ultrastructural Atlas and Review. London: Butterworth; 1971 [Google Scholar]
  • 3. van der Flier LG, Clevers H. Stem cells, self-renewal, and differentiation in the intestinal epithelium. Annu Rev Physiol. 2009;71:241–260 [DOI] [PubMed] [Google Scholar]
  • 4. Shi YB, Hasebe T, Fu L, Fujimoto K, Ishizuya-Oka A.The development of the adult intestinal stem cells: insights from studies on thyroid hormone-dependent amphibian metamorphosis. Cell Biosci. 2011;1:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Ishizuya-Oka A, Shi YB. Evolutionary insights into postembryonic development of adult intestinal stem cells. Cell Biosci. 2011;1:37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Cheng H, Leblond CP. Origin, differentiation and renewal of the four main epithelial cell types in the mouse small intestine. III. Entero-endocrine cells. Am J Anat. 1974;141:503–519 [DOI] [PubMed] [Google Scholar]
  • 7. Sun G, Shi YB. Thyroid hormone regulation of adult intestinal stem cell development: Mechanisms and evolutionary conservations. Int J Biol Sci. 2012;8:1217–1224 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. McAvoy JW, Dixon KE. Cell proliferation and renewal in the small intestinal epithelium of metamorphosing and adult Xenopus laevis. J Exp Zool. 1977;202:129–138 [Google Scholar]
  • 9. Sancho E, Eduard Batlle E, Clevers H. Signaling pathways in intestinal development and cancer. Annu Rev Cell Dev Biol. 2004;20:695–723 [DOI] [PubMed] [Google Scholar]
  • 10. Tata JR. Gene expression during metamorphosis: an ideal model for post-embryonic development. Bioessays. 1993;15:239–248 [DOI] [PubMed] [Google Scholar]
  • 11. Shi Y-B. Amphibian Metamorphosis: From Morphology to Molecular Biology. New York: John Wiley & Sons, Inc.; 1999 [Google Scholar]
  • 12. Plateroti M, Gauthier K, Domon-Dell C, Freund JN, Samarut J, Chassande O. Functional interference between thyroid hormone receptor α (TRα) and natural truncated TRδα isoforms in the control of intestine development. Mol Cell Biol. 2001;21:4761–4772 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Flamant F, Poguet AL, Plateroti M, et al. Congenital hypothyroid Pax8(−/−) mutant mice can be rescued by inactivating the TRα gene. Mol Endocrinol. 2002;16:24–32 [DOI] [PubMed] [Google Scholar]
  • 14. Kress E, Rezza A, Nadjar J, Samarut J, Plateroti M. The frizzled-related sFRP2 gene is a target of thyroid hormone receptor α1 and activates β-catenin signaling in mouse intestine. J Biol Chem. 2009;284:1234–1241 [DOI] [PubMed] [Google Scholar]
  • 15. Plateroti M, Chassande O, Fraichard A, et al. Involvement of T3Rα- and β-receptor subtypes in mediation of T3 functions during postnatal murine intestinal development. Gastroenterology. 1999;116:1367–1378 [DOI] [PubMed] [Google Scholar]
  • 16. Muncan V, Heijmans J, Krasinski SD, et al. Blimp1 regulates the transition of neonatal to adult intestinal epithelium. Nat Commun. 2011;2:452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Harper J, Mould A, Andrews RM, Bikoff EK, Robertson EJ. The transcriptional repressor Blimp1/Prdm1 regulates postnatal reprogramming of intestinal enterocytes. Proc Natl Acad Sci USA. 2011;108:10585–10590 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Heimeier RA, Das B, Buchholz DR, Fiorentino M, Shi YB. Studies on Xenopus laevis intestine reveal biological pathways underlying vertebrate gut adaptation from embryo to adult. Genome Biol. 2010;11:R55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Shi Y-B, Ishizuya-Oka A. Biphasic intestinal development in amphibians: embryogensis and remodeling during metamorphosis. Curr Top Dev Biol. 1996;32:205–235 [DOI] [PubMed] [Google Scholar]
  • 20. Ishizuya-Oka A, Shimozawa A. Connective tissue is involved in adult epithelial development of the small intestine during anuran metamorphosis in vitro. Roux's Arch Dev Biol. 1992;201:322–329 [DOI] [PubMed] [Google Scholar]
  • 21. Ishizuya-Oka A, Shimozawa A. Inductive action of epithelium on differentiation of intestinal connective tissue of Xenopus laevis tadpoles during metamorphosis in vitro. Cell Tissue Res. 1994;277:427–436 [DOI] [PubMed] [Google Scholar]
  • 22. Hasebe T, Fu L, Miller TC, Zhang Y, Shi YB, Ishizuya-Oka A Thyroid hormone-induced cell-cell interactions are required for the development of adult intestinal stem cells. Cell Biosci. 2013;3:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Hasebe T, Buchholz DR, Shi YB, Ishizuya-Oka A. Epithelial-connective tissue interactions induced by thyroid hormone receptor are essential for adult stem cell development in the Xenopus laevis intestine. Stem Cells. 2011;29:154–161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Nieuwkoop PD, Faber J. Normal Table of Xenopus laevis. 1st ed Amsterdam: North Holland Publishing; 1956 [Google Scholar]
  • 25. Das B, Cai L, Carter MG, Piao YL, Sharov AA, Ko MS, Brown DD. Gene expression changes at metamorphosis induced by thyroid hormone in Xenopus laevis tadpoles. Dev Biol. 2006;291:342–355 [DOI] [PubMed] [Google Scholar]
  • 26. Buchholz DR, Heimeier RA, Das B, Washington T, Shi YB. Pairing morphology with gene expression in thyroid hormone-induced intestinal remodeling and identification of a core set of TH-induced genes across tadpole tissues. Dev Biol. 2007;303:576–590 [DOI] [PubMed] [Google Scholar]
  • 27. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing/. J Roy Statist Soc Ser B. 1995;57:289–300 [Google Scholar]
  • 28. Sharov AA, Dudekula DB, Ko MS. A web-based tool for principal component and significance analysis of microarray data.. Bioinformatics. 2005;21:2548–2549 [DOI] [PubMed] [Google Scholar]
  • 29. Dahlquist KD, Salomonis N, Vranizan K, Lawlor SC, Conklin BR. GenMAPP, a new tool for viewing and analyzing microarray data on biological pathways. Nat Genet. 2002;31:19–20 [DOI] [PubMed] [Google Scholar]
  • 30. Doniger SW, Salomonis N, Dahlquist KD, Vranizan K, Lawlor SC, Conklin BR. MAPPFinder: using Gene Ontology and GenMAPP to create a global gene-expression profile from microarray data. Genome Biol. 2003;4:R7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Ashburner M, Ball CA, Blake JA, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000;25:25–29 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Zeeberg BR, Feng W, Wang G, et al. GoMiner: a resource for biological interpretation of genomic and proteomic data. Genome Biol. 2003;4:R28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Bussey KJ, Kane D, Sunshine M, et al. MatchMiner: a tool for batch navigation among gene and gene product identifiers. Genome Biol. 2003;4:R27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Heimeier RA, Das B, Buchholz DR, Shi YB. The xenoestrogen bisphenol A inhibits postembryonic vertebrate development by antagonizing gene regulation by thyroid hormone. Endocrinology. 2009;150:2964–2973 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Shi YB, Liang VC. Cloning and characterization of the ribosomal protein L8 gene from Xenopus laevis. Biochim Biophys Acta. 1994;1217:227–228 [DOI] [PubMed] [Google Scholar]
  • 36. Das B, Heimeier RA, Buchholz DR, Shi YB. Identification of direct thyroid hormone response genes reveals the earliest gene regulation programs during frog metamorphosis. J Biol Chem. 2009;284:34167–34178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Hasebe T, Hartman R, Matsuda H, Shi YB. Spatial and temporal expression profiles suggest the involvement of gelatinase A and membrane type 1 matrix metalloproteinase in amphibian metamorphosis. Cell Tissue Res. 2006;324:105–116 [DOI] [PubMed] [Google Scholar]
  • 38. Shi YB, Hayes WP. Thyroid hormone-dependent regulation of the intestinal fatty acid-binding protein gene during amphibian metamorphosis. Dev Biol. 1994;161:48–58 [DOI] [PubMed] [Google Scholar]
  • 39. Stolow MA, Shi YB. Xenopus sonic hedgehog as a potential morphogen during embryogenesis and thyroid hormone-dependent metamorphosis. Nucleic Acids Res. 1995;23:2555–2562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Patterton D, Hayes WP, Shi YB. Transcriptional activation of the matrix metalloproteinase gene stromelysin-3 coincides with thyroid hormone-induced cell death during frog metamorphosis. Dev Biol. 1995;167:252–262 [DOI] [PubMed] [Google Scholar]
  • 41. Shi YB, Ishizuya-Oka A. Autoactivation of Xenopus thyroid hormone receptor β genes correlates with larval epithelial apoptosis and adult cell proliferation. J Biomed Sci. 1997;4:9–18 [DOI] [PubMed] [Google Scholar]
  • 42. Ishizuya-Oka A, Ueda S, Shi YB. Temporal and spatial regulation of a putative transcriptional repressor implicates it as playing a role in thyroid hormone-dependent organ transformation. Dev Genet. 1997;20:329–337 [DOI] [PubMed] [Google Scholar]
  • 43. Shi YB, Ishizuya-Oka A. Thyroid hormone regulation of apoptotic tissue remodeling: Implications from molecular analysis of amphibian metamorphosis. Prog Nucleic Acid Res Mol Biol. 2001;65:53–100 [DOI] [PubMed] [Google Scholar]
  • 44. Ishizuya-Oka A, Shimozawa A. Development of the connective tissue in the digestive tract of the larval and metamorphosing Xenopus laevis. Anat Anz. 1987;164:81–93 [PubMed] [Google Scholar]
  • 45. Ishizuya-Oka A, Shimozawa A. Ultrastructural changes in the intestinal connective tissue of Xenopus laevis during metamorphosis. J Morphol. 1987;193:13–22 [DOI] [PubMed] [Google Scholar]
  • 46. Yoshizato K. Biochemistry and cell biology of amphibian metamorphosis with a special emphasis on the mechanism of removal of larval organs. Int Rev Cytol. 1989;119:97–149 [DOI] [PubMed] [Google Scholar]
  • 47. Ishizuya-Oka A, Hasebe T, Buchholz DR, Kajita M, Fu L, Shi YB. Origin of the adult intestinal stem cells induced by thyroid hormone in Xenopus laevis. Faseb J. 2009;23:2568–2575 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Kress E, Samarut J, Plateroti M. Thyroid hormones and the control of cell proliferation or cell differentiation: paradox or duality? Mol Cell Endocrinol. 2009;313:36–49 [DOI] [PubMed] [Google Scholar]
  • 49. Sirakov M, Plateroti M. The thyroid hormones and their nuclear receptors in the gut: from developmental biology to cancer. Biochim Biophys Acta. 2011;1812:938–946 [DOI] [PubMed] [Google Scholar]
  • 50. Plateroti M, Kress E, Mori JI, Samarut J. Thyroid hormone receptor α1 directly controls transcription of the β-catenin gene in intestinal epithelial cells. Mol Cell Biol. 2006;26:3204–3214 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Matsuda H, Paul BD, Choi CY, Hasebe T, Shi YB. Novel functions of protein arginine methyltransferase 1 in thyroid hormone receptor-mediated transcription and in the regulation of metamorphic rate in Xenopus laevis. Mol Cell Biol. 2009;29:745–757 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Matsuda H, Shi YB. An essential and evolutionarily conserved role of protein arginine methyltransferase 1 for adult intestinal stem cells during postembryonic development. Stem Cells. 2010;28:2073–2083 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. Brown DD. The role of thyroid hormone in zebrafish and axolotl development. Proc Natl Acad Sci USA. 1997;94:13011–13016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Friedrichsen S, Christ S, Heuer H, et al. Regulation of iodothyronine deiodinases in the Pax8−/− mouse model of congenital hypothyroidism. Endocrinology. 2003;144:777–784 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data
supp_154_11_4396__index.html (979B, html)
supp_en.2013-1432v1_EN-13-1432supptabs10.pdf (2.9MB, pdf)

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES