Genome-wide analysis of gene transcription in the hypothalamus
Jocelyn M. Bischof and
Rachel Wevrick
Department of Medical Genetics, University of Alberta, Edmonton, Alberta, Canada
 |
ABSTRACT
|
---|
As the genomic regions containing loci predisposing to obesity-related traits are mapped in human population screens and mouse genetic studies, identification of susceptibility genes will increasingly be facilitated by bioinformatic methods. We hypothesized that candidate genes can be prioritized by their expression levels in tissues of central importance in obesity. Our objective was to develop a combined bioinformatics and molecular paradigm to identify novel genes as candidates for murine or human obesity genetic modifiers based on their differential expression patterns in the hypothalamus compared with other murine tissues. We used bioinformatics tools to search publicly available gene expression databases using criteria designed to identify novel genes differentially expressed in the hypothalamus. We used RNA methods to determine their expression sites and levels of expression in the hypothalamus of the murine brain. We identified the chromosomal location of the novel genes in mice and in humans and compared these locations with those of genetic loci predisposing to obesity-related traits. We developed a search strategy that correctly identified a set of genes known to be important in hypothalamic function as well as a candidate gene for Prader-Willi syndrome that was not previously identified as differentially expressed in the hypothalamus. Using this same strategy, we identified and characterized a set of 11 genes not previously known to be differentially expressed in the murine hypothalamus. Our results demonstrate the feasibility of combined bioinformatics and molecular approaches to the identification of genes that are candidates for obesity-related disorders in humans and mice.
hypothalamus; candidate genes; bioinformatics; gene expression microarray
 |
INTRODUCTION
|
---|
OBESITY is a major health issue facing developed nations today. Despite years of population studies, the genetic determinants of healthy body weight are poorly understood (1). Likewise, although there is increased understanding of the role of the hypothalamic-pituitary-adrenal axis in obesity, the proteins implicated in the development and/or function of this organ system are not all identified or characterized. Specific proteins, such as hormones and their receptors, are responsible for the control of appetite and long-term maintenance of body weight (18, 24). An increasing number of studies rely on mapping obesity modifier loci in mice or humans and then choosing candidate genes within the mapped region for further study (11a, 20). Still other studies examine changes in gene expression in mice undergoing dietary challenges or in obesity gene knockout versus wild-type mice. Yet, the reliability of effective choice of candidate genes depends on there being prior knowledge of gene expression or function in searchable databases. Often this knowledge exists but is not readily available to researchers who are unfamiliar with the complexities of bioinformatic databases.
To search for obesity candidate genes, we used a combined bioinformatic and molecular biology approach to identify novel genes important to the function of the hypothalamus. The hypothalamus is a complex region of the brain involved in homeostatic processes including, but not limited to, energy balance. We identified a set of genes already known to be important to hypothalamic function and a set of genes not previously associated with the hypothalamus. We propose that these genes may be important in the function of the hypothalamus, may provide a new source of hypothalamus-specific regulatory elements, and may be important in disorders of hypothalamic function, such as obesity. Furthermore, genes identified in this search are now candidate genes for obesity-related traits genetically mapped to the chromosomal intervals in which they reside.
 |
MATERIALS AND METHODS
|
---|
Bioinformatics.
The Genomics Institute of the Novartis Research Foundation (GNF) murine gene expression atlas (now updated to GNF SymAtlas version 1.0.3) was originally developed as a tool for high-throughput analysis of transcriptional activity in biological samples (25, 26). The expression of specific genes within 45 diverse murine tissues, including the hypothalamus, was quantified by hybridization to Affymetrix mouse (U74A) high-density oligonucleotide arrays and made available as a searchable, on-line database. We filtered the murine dataset to select genes with levels of expression above the median in the hypothalamus but with a moderately high level of expression above the median in no more than five other tissues. According to the GNF SymAtlas database, duplicate samples and/or duplicate hybridizations were performed for each tissue, and these values are integrated into the error bars for each measurement. Because of the configuration of the database, it is not possible to generate estimates of statistical significance; rather, the search criteria were left deliberately broad at the risk of including some genes that do not in fact show as much specificity for hypothalamic expression when analyzed further as might be expected from the database search alone. The median is defined for the specific gene, across all tissues, so that there is no bias against genes that have low to moderate levels of gene expression overall. This first search was performed on the GeneAtlas Version 1 Dataset (25). A second search was performed on the GeneAtlas Version 2 Dataset, Mouse GeneAtlas GNF1M (26). In this data set, a correlation search was performed to identify genes with a high value in the hypothalamus (set at 400), any value in nonbrain tissues, and a low value (set at 1) in other regions of the brain, with a correlation of 0.98. Because the preoptic region is part of the hypothalamus, we allowed any value for expression in the preoptic area as well. Genome localization for selected murine hypothalamic genes was identified from the May 2004 mouse (Mus musculus) draft genome data, build 33 assembly by the National Center for Biotechnology Information (NCBI). The genomic localization of human orthologs was deduced using NCBI database UniGene and Gene.
Expression studies.
RNA in situ hybridization was performed using digoxigenin-labeled antisense RNA probes on 20-µm cryosections of embryonic day 18.5 mouse embryos as previously described (16). Probes for Hcrt, A26100, Dscr1l2, Impact, and Pnck were derived from PCR fragments amplified from reverse-transcribed adult mouse hypothalamus mRNA with recombinant Taq DNA polymerase (Invitrogen) and cloned into a pBluescript vector (Invitrogen). The Hcrt probe includes positions 7573 of hypocretin (Hcrt) mRNA (NM_010410). The A26100 probe includes positions 4321,004 of RIKEN cDNA 2610042L04 gene (2610042L04Rik) mRNA (NM_025940). The Dscr1l2 probe includes positions 177638 of Down syndrome critical region gene 1-like 2 (Dscr1l2) mRNA (NM_022980). The Impact probe includes positions 504955 of imprinted and ancient (Impact) mRNA (NM_008378). The Pnck probe includes positions 6971,252 of pregnancy upregulated nonubiquitously expressed Ca2+/calmodulin kinase (Pnck) mRNA (NM_012040). The Magel2 probe includes positions 9841,754 of Genbank accession number NM_013779 and was previously described (16). All studies involving animals were performed according to the guidelines for the use and care of laboratory animals at the University of Alberta.
 |
RESULTS
|
---|
A major goal of the Human Genome Project was to identify expressed sequences [expressed sequence tags (ESTs)] from RNA isolated from tissues or cell lines. By clustering the ESTs into groups (UniGene clusters) by sequence identity, the resulting UniGene database could be queried with a gene name to discover the tissues in which the gene is expressed. Until recently, there were no public databases that allowed investigators to perform the converse query, which is to discover the expression levels of all the transcripts in a particular tissue with respect to their expression levels in other tissues. However, the results of two large-scale gene array experiments documenting expression levels of most genes in a wide range of mouse tissues, including the hypothalamus, have recently been published. Specifically, labeled cDNA from each mouse tissue was hybridized to Affymetrix Genechip microarrays to profile expression levels of 36,000 genes in those tissues. In this study, multitissue comparisons were performed to normalize expression levels and input into a searchable gene expression database (the GNF database) made available by the Novartis Research Foundation (8, 25). The major advantage over previous studies, which used subtractive methods to derive hypothalamic cDNAs, is that levels of expression can be compared between tissues and between parts of the brain, so that genes expressed more highly or more specifically in the hypothalamus can be easily identified. We queried this database to identify genes differentially expressed in the murine adult hypothalamus.
Strategy for identification of hypothalamic transcripts.
We reasoned that genes important in hypothalamic development or function are likely to be transcribed in hypothalamic precursor tissues or the mature hypothalamus. The laboratory mouse is a model particularly amenable to studies of brain function because of similarities in anatomy to the human hypothalamus and because of its widespread use in obesity studies. We performed three searches using different criteria to query the GNF expression database for genes highly expressed in the postnatal murine hypothalamus. Our first search of the GNF database used low-specificity criteria that required only high expression in the hypothalamus, of at least 8 times the median of the expression level in all 45 mouse tissues. This search identified a set of genes already studied for their role in hypothalamic function, such as pro-opiomelanocortin-
(POMC) and Hcrt, but also identified many genes broadly expressed in the nervous system, such as the microtubule-associated protein-
. Notably, some genes important in obesity and/or hypothalamic function were not identified under these conditions (e.g., the widely expressed Nhlh2). Our second search of the GNF database used high-specificity criteria as that for the first search, high expression in the hypothalamus of at least eight times the median of the expression in other tissues. In addition, the second search required low expression (<3 times the median) in at least 40 other tissues. This second search identified five genes including arginine vasopressin, POMC, Hcrt, and oxytocin, all of which have major roles in endocrine regulation in the hypothalamus. Significantly, this high-specificity search also identified Magel2, which is developmentally regulated and inactivated in the genetic obesity disorder Prader-Willi syndrome. We then formulated a third search with intermediate specificity: expression in the hypothalamus over eight times the median level of all tissues and fewer than five tissues with over three times the median level of expression. The third search of the GNF database identified a set of 15 genes highly expressed in the hypothalamus but moderately to highly expressed in only 5 other murine tissues (Table 1, screen 1). We repeated the third search using an updated database, the 2004 version of the mouse Geneatlas. As with the third search of the GNF database, the search of Geneatlas 2004 identified genes with high expression in the hypothalamus and low expression elsewhere in the brain. We then used the correlation function of the searchable Symatlas database within Geneatlas. The Symatlas search was refined to allow either high or low expression in the preoptic region, which functionally overlaps with the hypothalamus. This last search, still using intermediate specificity criteria, identified 7 genes, 3 of which were identified in the GNF intermediate-specificity search (Table 1, screen 2), for a total of 19 genes.
View this table:
[in this window]
[in a new window]
|
Table 1. Intermediate search of the GNF Gene Expression Atlas to detect genes highly expressed in the hypothalamus and expressed at low levels in most other tissues
|
|
Selection of novel hypothalamic genes.
Of the 19 genes from the two searches of the GNF databases, 3 genes were identified in both searches (Table 1). Because the nature of the search engine provided by GNF changed from the 2002 release to the 2004 release, the two search criteria were not directly comparable. In common between the two searches was the requirement for high expression in the hypothalamus and low expression in other regions of the brain. Because we were interested in identifying genes not known to be important in the hypothalamus, we first set aside genes whose role in hypothalamic function has been well described. These included genes encoding the neuropeptide signaling molecules Hcrt/orexin, pro-opiomelanocortin-
, oxytocin, agouti-related protein, and arginine vasopressin (4). A neuroendocrine role for calcitonin and the calcitonin receptor in growth hormone secretion in the hypothalamus/pituitary had previously been suggested (17). We had previously identified Magel2 as specifically expressed in the developing murine hypothalamus, although its function in this tissue is not yet known (16). Magel2 is of particular interest because its human ortholog, MAGEL2, is one of a set of genes inactivated in the genetic obesity disorder Prader-Willi syndrome (9, 15). Finally, mutations in Patched (Ptch1) can result in holoprosencephaly and developmental delay in humans or, alternatively, cause basal cell nervous syndrome when homozygously inactivated in somatic cells. Ptch1 is an antagonist of the signaling molecule hedgehog, which functions in patterning of the neural tube but is not known to have a specific role in the hypothalamus.
Gene expression profiles.
The remaining 11 genes fell into 2 categories: 1) novel genes for which a function has been predicted by sequence similarity to known genes and 2) novel genes with no known function and no or limited similarity to known genes (Table 2). We examined the UniGene clusters associated with these novel genes. Because EST sets made from hypothalamus RNA are represented in mouse EST databases, we expected to see a representation of hypothalamic cDNAs within the Unigene clusters corresponding to these genes. Indeed, all 11 genes had corresponding ESTs derived from the whole brain, with some specifically including the hypothalamus (4) or diencephalon (2) (Table 2). The information in the UniGene clusters did not point to these genes being particularly hypothalamus specific, in contrast to the Unigene cluster for arginine vasopressin, for example, in which 12 of the 19 ESTs were derived from the hypothalamus. We then used the EST ProfileViewer function in Unigene to test whether the expression profile derived from ESTs overlapped with that derived from the Geneatlas databases. Seven of the eleven novel genes had a major contribution of ESTs from the brain or pituitary gland; the hypothalamus was included as a "brain" sample in the EST Profile viewer. In one case (Itih3), the majority of UniGene ESTs were derived from the liver, with a small minority from the brain. This is consistent with the Geneatlas distribution, which shows the highest expression of Itih3 in the liver, with most of the remaining expression in preoptic and hypothalamus samples. In one other case (Dscr1l2), the UniGene ESTs did not distribute into any tissue-specific pattern, although an examination of the developmental timing of expression, also using the function in Unigene, demonstrated that the vast majority of ESTs were derived from the egg as opposed to preimplantation embryos or any postimplantation embryonic stage. This is in contrast to the GeneAtlas 2004 expression profile, which shows very little tissue specificity for the Dscr1l2 transcript, and also contrasts with the GeneAtlas 2002 expression profile, which showed more specific expression in the hypothalamus. Similarly, the Pitpnm1 showed conflicting expression profiles between the GeneAtlas 2002, GeneAtlas 2004, and EST ProfileViewer, with the one commonality being expression in the eye.
In some cases, preliminary expression patterns for the novel genes determined by RNA- or protein-based methods had already been published. This information was most readily available from the JAX bioinformatics server, which references the original publications. It had been proposed that Igsf1 (also known as InhBP/p120) may be involved in pituitary function, but both male and female mice mutant for Igsf1 are viable and fertile (3). The insulin receptor substrates, including Irs4, function in insulin signaling. Absence of Irs4 causes mild defects in growth, reproduction, and glucose homeostasis in an Irs4-null mutant mouse line (5). On the basis of a mouse knockout model, Ptprz was shown to have a role in oligodendrocyte survival and in recovery from demyelinating disease (11). Pnck (6), Nsg1 (21), and Ptpro (2) have been suggested to have roles in the developing nervous system; whereas the function of Itih3 has been studied in the liver, no function has been predicted in the brain (2). No function has been predicted for Impact, although the human ortholog IMPACT has been suggested as a candidate gene for bipolar affective disorder (10, 14). There was also no further information on putative function of the gene represented by the Unigene cluster Mm.321089 (gene name A2610042L04).
RNA in situ hybridization.
We first verified that the 11 selected genes were indeed expressed in the murine hypothalamus by RT-PCR on RNA samples from the adult mouse hypothalamus (data not shown). We then localized the expression to specific hypothalamic nuclei by RNA in situ hybridization using a standard mouse atlas to label the sections (13). A probe for Hcrt displayed a strong signal in the lateral anterior hypothalamus, consistent with the known abundance of RNA transcripts in this region of the brain (Fig. 1) (27). Consistent with the earlier embryonic expression of Magel2 in the suprachiasmatic nucleus (16), we detected Magel2 transcripts in the anterior hypothalamus and suprachiasmatic nucleus (Fig. 1). We then selected four genes that emerged from the bioinformatic analysis but about which little else was known. All four genes displayed high levels of expression in the hypothalamus compared with surrounding brain tissue by RNA in situ hybridization (Fig. 1). Probes for A2610042L04 and Pnck detected abundant expression in the anterior hypothalamus, whereas a probe for Dscr1l2 detected the strongest expressions in the dorsal medial hypothalamus, ventromedial hypothalamus, and arcuate nucleus, and a probe for Impact detected the highest expression in the arcuate nucleus.

View larger version (194K):
[in this window]
[in a new window]
|
Fig. 1. Expression of candidate genes in the late embryonic hypothalamus. Cryosections of the embryonic day 18.5 mouse brain were hybridized to digoxigenin-labeled antisense RNA in situ hybridization probes for Hcrt, Magel2, Dscr1l2, Impact, A2610042L04, and Pnck. Hybridization signals were detected in the anterior hypothalamus (AH), arcuate nucleus (Arc), dorsomedial hypothalamus (Dmh), suprachiasmatic nucleus (Scn), or ventromedial hypothalamus (Vmh). 3V, third ventricle, If, infidibular recess, Oc, optic chiasm. All sections are oriented with the ventral aspect at the bottom. Scale bar = 250 µm.
|
|
Gene mapping.
We used NCBI databases to map the 11 selected genes to chromosomal regions in the mouse genome (NCBI build 33) and human genome (NCBI build 35) (Table 3). We then correlated the map positions of selected genes with the genomic positions of published loci influencing obesity-related traits in either human or other mammals (23) (Table 3) updated and published on-line as "The Obesity Gene Map Database" (11a). We identified eight traits, such as body mass index, fat mass, and percent body fat, that had been mapped to the same chromosomal regions as our selected genes in human studies of weight gain and obesity. We then identified cases where obesity traits had been identified as mapping within the mouse, rat, or pig chromosomal region homologous to human chromosome band(s) containing the selected genes. We did not include traits that had been mapped to an interval that contained more than three chromosome bands, which would contain a very large number of candidate genes. In this way, we identified 13 obesity-related traits in the mouse (11), rat (1), or pig (1) that correlated to the map positions of 6 of the 11 selected genes; these traits were all related to obesity and weight gain. Except for A2610042L04, a poorly characterized predicted mouse gene on chromosome 14, all selected genes were in regions of a mammalian genome where genetic evidence for an obesity-related trait exists.
 |
DISCUSSION
|
---|
The use of microarray-based systems has accelerated high-throughput gene expression profiling in tissues of interest in disease processes. Previous studies designed to identify cDNAs enriched in the hypothalamus used subtractive hybridization (7) or EST cataloguing (12). Although these methods did identify genes expressed in the hypothalamus, there are intrinsic biases toward highly expressed genes in both methods; subtractive hybridization methodologies can also have intrinsic sequence-specific biases. In contrast, high-throughput gene expression profiling provides a robust, unbiased, searchable resource that is freely available to the research community. We examined the relative expression levels of an estimated 30,000 genes in the murine hypothalamus as determined by Affymetrix microarray technology and made available by the Novartis Research Foundation (8). From this, we identified a set of genes not previously recognized to be important in hypothalamic function but highly and preferentially expressed in the murine hypothalamus. It is exciting to think that if such a database had been available at the time that we cloned Magel2 in 1999, this analysis could have facilitated the selection of human MAGEL2 as a candidate gene for the extreme obesity seen in Prader-Willi syndrome.
We identified 11 genes whose expression is enriched in the hypothalamus but which were not previously associated with the development of the hypothalamus or its function. The candidate genes could be involved in any number of the interrelated processes partially mediated by the hypothalamus besides energy and fluid balance, such as circadian rhythm, reproduction, stress responses, and temperature regulation (1). The definitive proof of their involvement in obesity, rather than in other functions of the hypothalamus, would come from a demonstration of gene dysregulation in various models of obesity, including genetic and diet-induced obesity. We concentrated on obesity because of the large number of studies providing genetically mapped obesity-related traits, but the genes that we identified could equally well have roles in other disorders of hypothalamic function. Other regions of the brain, including the cortex and hindbrain, have important roles in food intake (22) but have major roles in other neurological functions. The gut, liver, pancreas, and adipose tissues also have major roles in appetite, and parallel studies could reveal novel obesity susceptibility genes that are highly and specifically expressed in those tissues but not expressed in the brain and hypothalamus.
Because susceptibility to obesity is in part genetically determined, variations in these genes may contribute to familial obesity. The information available about the cellular and physiological function of the proteins encoded by these genes is, however, highly variable. Furthermore, the size of the intervals in which obesity traits have been mapped is also highly variable. Further genetic and molecular studies would clarify possible roles of the candidate genes in obesity. For example, the candidate gene located within the region where specific mouse strains or human subjects have shown genetic predisposition to obesity could be sequenced for coding or regulatory sequence variants. These variants could then be tested in larger populations for association with obesity-related traits, including traits genetically mapped to the chromosomal intervals in which the genes reside. Finally, further investigations of the physiological and cellular functions of the proteins identified in this and other bioinformatic screens may elucidate the role of specific genes and gene products in hypothalamic function and genetic predisposition to obesity.
 |
GRANTS
|
---|
This research was performed with the support of a Canadian Institutes of Health Research Pilot Project Grant (Obesity and Healthy Body Weight Request for Applications) (to R.Wevrick). R. Wevrick is a Senior Scholar of the Alberta Heritage Foundation for Medical Research.
 |
ACKNOWLEDGMENTS
|
---|
We acknowledge Sharee Kuny and Megan O'Neill for helpful discussions.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: R. Wevrick, Dept. of Medical Genetics, 8-42 Medical Sciences Bldg., Univ. of Alberta, Edmonton, Alberta, Canada T6G 2H7 (E-mail: rachel.wevrick{at}ualberta.ca).
10.1152/physiolgenomics.00071.2005.
 |
REFERENCES
|
---|
- Barsh GS, Farooqi IS, and O'Rahilly S. Genetics of body-weight regulation. Nature 404: 644651, 2000.[CrossRef][ISI][Medline]
- Beltran PJ, Bixby JL, and Masters BA. Expression of PTPRO during mouse development suggests involvement in axonogenesis and differentiation of NT-3 and NGF-dependent neurons. J Comp Neurol 456: 384395, 2003.[CrossRef][ISI][Medline]
- Bernard DJ, Burns KH, Haupt B, Matzuk MM, and Woodruff TK. Normal reproductive function in InhBP/p120-deficient mice. Mol Cell Biol 23: 48824891, 2003.[Abstract/Free Full Text]
- Ellacott KL and Cone RD. The central melanocortin system and the integration of short- and long-term regulators of energy homeostasis. Recent Prog Horm Res 59: 395408, 2004.[Abstract/Free Full Text]
- Fantin VR, Wang Q, Lienhard GE, and Keller SR. Mice lacking insulin receptor substrate 4 exhibit mild defects in growth, reproduction, and glucose homeostasis. Am J Physiol Endocrinol Metab 278: E127E133, 2000.[Abstract/Free Full Text]
- Gardner HP, Rajan JV, Ha SI, Copeland NG, Gilbert DJ, Jenkins NA, Marquis ST, and Chodosh LA. Cloning, characterization, and chromosomal localization of Pnck, a Ca2+/calmodulin-dependent protein kinase. Genomics 63: 279288, 2000.[CrossRef][ISI][Medline]
- Gautvik KM, de Lecea L, Gautvik VT, Danielson PE, Tranque P, Dopazo A, Bloom FE, and Sutcliffe JG. Overview of the most prevalent hypothalamus-specific mRNAs, as identified by directional tag PCR subtraction. Proc Natl Acad Sci USA 93: 87338738, 1996.[Abstract/Free Full Text]
- Genomics Institute of the Novartis Research Foundation Gene Expression Atlas. GNF SymAtlas. http://symatlas.gnf.org/SymAtlas/.
- Goldstone AP. Prader-Willi syndrome: advances in genetics, pathophysiology and treatment. Trends Endocrinol Metab 15: 1220, 2004.[CrossRef][ISI][Medline]
- Hagiwara Y, Hirai M, Nishiyama K, Kanazawa I, Ueda T, Sakaki Y, and Ito T. Screening for imprinted genes by allelic message display: identification of a paternally expressed gene impact on mouse chromosome 18. Proc Natl Acad Sci USA 94: 92499254, 1997.[Abstract/Free Full Text]
- Harroch S, Furtado GC, Brueck W, Rosenbluth J, Lafaille J, Chao M, Buxbaum JD, and Schlessinger J. A critical role for the protein tyrosine phosphatase receptor type Z in functional recovery from demyelinating lesions. Nat Genet 32: 411414, 2002.[CrossRef][ISI][Medline]
- Human Genetics Laboratory. Obesity Gene Map Database. http://obesitygene.pbrc.edu/.
- Hu RM, Han ZG, Song HD, Peng YD, Huang QH, Ren SX, Gu YJ, Huang CH, Li YB, Jiang CL, Fu G, Zhang QH, Gu BW, Dai M, Mao YF, Gao GF, Rong R, Ye M, Zhou J, Xu SH, Gu J, Shi JX, Jin WR, Zhang CK, Wu TM, Huang GY, Chen Z, Chen MD, and Chen JL. Gene expression profiling in the human hypothalamus-pituitary-adrenal axis and full-length cDNA cloning. Proc Natl Acad Sci USA 97: 95439548, 2000.[Abstract/Free Full Text]
- Jacobowitz DM and Abbott LC. Chemoarchitectonic Atlas of the Developing Mouse Brain. Boca Raton, FL: CRC, 1998.
- Kosaki K, Suzuki T, Kosaki R, Yoshihashi H, Itoh M, Goto Y, and Matsuo N. Human homolog of the mouse imprinted gene Impact resides at the pericentric region of chromosome 18 within the critical region for bipolar affective disorder. Mol Psychiatry 6: 8791, 2001.[CrossRef][ISI][Medline]
- Lee S, Kozlov S, Hernandez L, Chamberlain SJ, Brannan CI, Stewart CL, and Wevrick R. Expression and imprinting of MAGEL2 suggest a role in Prader-Willi syndrome and the homologous murine imprinting phenotype. Hum Mol Genet 9: 18131819, 2000.[Abstract/Free Full Text]
- Lee S, Walker CL, and Wevrick R. Prader-Willi syndrome transcripts are expressed in phenotypically significant regions of the developing mouse brain. Gene Expr Patterns 3: 599609, 2003.[CrossRef][ISI][Medline]
- Lengyel AM, Toledo AL, Czepielewski MA, Vieira JG, and Chacra AR. Calcitonin suppresses growth hormone (GH) response to growth hormone-releasing hormone (GHRH) in man. J Endocrinol Invest 12: 2529, 1989.[ISI]
- Michaud JL. The developmental program of the hypothalamus and its disorders. Clin Genet 60: 255263, 2001.[CrossRef][ISI][Medline]
- Rankinen T, Perusse L, Weisnagel SJ, Snyder EE, Chagnon YC, and Bouchard C. The human obesity gene map: the 2001 update. Obes Res 10: 196243, 2002.[Abstract/Free Full Text]
- Saberan-Djoneidi D, Picart R, Escalier D, Gelman M, Barret A, Tougard C, Glowinski J, and Levi-Strauss M. A 21-kDa polypeptide belonging to a new family of proteins is expressed in the Golgi apparatus of neural and germ cells. J Biol Chem 273: 39093914, 1998.[Abstract/Free Full Text]
- Schwartz GJ. Biology of eating behavior in obesity. Obes Res 12, Suppl 2: 102S106S, 2004.[Abstract/Free Full Text]
- Snyder EE, Walts B, Perusse L, Chagnon YC, Weisnagel SJ, Rankinen T, and Bouchard C. The human obesity gene map: the 2003 update. Obes Res 12: 369439, 2004.[Abstract/Free Full Text]
- Spiegelman BM and Flier JS. Obesity and the regulation of energy balance. Cell 104: 531543, 2001.[CrossRef][ISI][Medline]
- Su AI, Cooke MP, Ching KA, Hakak Y, Walker JR, Wiltshire T, Orth AP, Vega RG, Sapinoso LM, Moqrich A, Patapoutian A, Hampton GM, Schultz PG, and Hogenesch JB. Large-scale analysis of the human and mouse transcriptomes. Proc Natl Acad Sci USA 99: 44654470, 2002.[Abstract/Free Full Text]
- Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, Cooke MP, Walker JR, and Hogenesch JB. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci USA 101: 60626067, 2004.[Abstract/Free Full Text]
- Van Den Pol AN, Patrylo PR, Ghosh PK, and Gao XB. Lateral hypothalamus: early developmental expression and response to hypocretin (orexin). J Comp Neurol 433: 349363, 2001.[CrossRef][ISI][Medline]