Endocrinology Basic and Clinical Principles - part 8 pdf

Chapter 19 / Oncogenes and Tumor Suppressor Genes 303 when injected into suitable hosts, unless specific chromosomes were lost from the hybrids. This phenomenon of tumor suppression suggested that recessive genetic changes were responsible for the tumorigenic phenotype. Although cell hybrid studies of tumor suppression pro- vided very useful information on chromosomal assign- ment, this method could not lead to gene identification, and, to date, no suppressor gene thus has been isolated. The identification of tumor suppressor genes has been greatly facilitated through studies of familial cancers. This is best illustrated in the identification and isolation of the retinoblastoma susceptibility gene (Rb). The essential features of retinoblastoma are that, in the familial forms of the tumor, the affected individual inherits a mutant, “loss-of-function” allele from an affected parent and a second somatic mutation inac- tivates the normal allele derived from the unaffected parent. By contrast, the sporadic forms of the tumor involve two somatic mutational events. According to this recessive mutation model, homozygous deletion of the Rb gene would be expected in some tumors, and Rb gene expression could be altered in retinoblastoma compared to normal tissue. Homozygous deletion of chromosome 13q14 was in fact detected in retinoblas- tomas, and using chromosomal walking to identify sequences conserved in evolution that were expressed in retinoblasts but were absent or altered in retinoblas- tomas, the candidate gene for Rb was isolated. Another powerful strategy for discovering new tumor suppressor genes in human tumors is to identify specific genetic markers that are repeatedly reduced to homozygosity in many tumors of a given type. Loss of heterozygosity (LOH) for this specific marker suggests the presence of a closely linked suppressor gene whose second allele has been eliminated from tumor cells during tumor pathogenesis. For example, losses of allele on the long arm of chromosome 18 are a frequent occurrence in colorectal carcinomas but not in adenomas, suggesting the presence of a suppressor gene on this chromosome whose loss frequently accompanies the conversion of benign adenomas to carcinomas. Based on this evidence, the responsible gene, termed DCC for “deleted in colorectal carcinoma,” was identified and a transmembrane protein with a sequence similarity to the neural cell-adhesion family of molecules isolated. Simi- lar methodology was used to identify other tumor suppressor genes such as p53 and WT-1 (Table 2). 4. PROTOONCOGENES AND TUMOR SUPPRESSOR GENES IN ENDOCRINE TUMORS Like tumors in other tissues, tumors in the endocrine system arise as monoclonal expansions of a signal- mutated cell. For most tumor types, multiple genes are involved and different combinations of gene mutations may result in similar phenotypes. Some genes contrib- ute to tumors of only one cell type whereas other genes are involved in different types of tumors. The role of protooncogenes and tumor suppressor genes in tumorigenesis of the endocrine system is presently being actively studied. Table 3 summarizes the oncogenes and tumor suppressor genes that have been implicated in endocrine tumor formation, and some of these genes are discussed in detail. Table 2 Representative Tumor suppressor Genes in Human Tumors a Chromosomal localization Name of locus Tumors involved Properties of gene product 5p APC Familial adenomatous polyposis, Cytoplasmic protein colotectal carcinoma 10q — MEN 2, astrocytoma — 11p WT-1 Wilms tumor, rhabdomyo- sarcoma, hepato- DNA-binding protein blastoma, bladder and lung carcinoma 11q Menin MEN 1 Transcription factor 13q Rb-1 Retinoblastoma, osteosarcoma, breast and DNA-binding protein bladder carcinoma 17p p53 Small-cell and squamous cell lung carcinoma, DNA-binding protein breast carcinoma, colorectal carcinoma, and other 17q NF-1 Neurofibromatosis type 1 Induces GTP hydrolysis of ras protein 18q DCC Colorectal carcinoma Cell-surface receptor a Adapted from Weinberg (1994). 304 Part IV / Hypothalamic–Pituitary 4.1. Multiple Endocrine Neoplasia Type 1 Multiple endocrine neoplasia type 1 (MEN 1) is an autosomal dominant disorder characterized by tumors of the parathyroid gland, pancreatic islet, and anterior pituitary. Based on the assumption that tumorigenesis involves loss of function for a tumor suppressor gene (see above), and by utilizing restriction fragment length polymorphisms, the MEN 1 locus was mapped to chromosome 11q13. Extensive characterization of the 11q13 region in a panel of MEN 1–associated tumors identified a gene, called menin, that functions as a nuclear protein that interacts specifically with the activator protein-1 transcription factor JunD. Mutations in menin, resulting in failed menin binding to JunD, have been identified in tumors derived from MEN 1 kindreds, sup- porting the hypothesis that MEN 1 encodes a tumor suppressor gene. The clinical phenotype can also be largely recapitulated in transgenic mice in which the menin gene is inactivated. It is now possible to test MEN 1 kindreds for menin mutations at an early age, allowing affected individuals to be monitored for the development of endocrine tumors (Table 4). 4.2. Multiple Endocrine Neoplasia Type 2 Multiple endocrine neoplasia type 2 (MEN 2) consists of three clinically distinct, dominantly inherited cancer syndromes. MEN 2A is the most common, and patients develop familial medullary thyroid carcinoma (FMTC), pheochromocytoma, and primary hyperpar- athyroidism. Those with MEN 2B have FMTC, pheochromocytoma, mucosal neuromas of the tongue, hyperplasia of neuronal Schwann cells in the cornea, ganglioneuromatosis of the gastrointestinal tract, and an asthenic marfanoid habitus. In familial MTC, only the thyroid is affected. All three syndromes result from one of several different germline mutations of the receptor tyrosine kinase RET protooncogene, located on chromosome 10q11. The RET gene was identified and cloned through rearrangements that occur in papillary thyroid carcinomas and in vitro during transfection studies. It consists of 21 exons and encodes a receptor kinase. RET mRNA is expressed in developing central and peripheral nervous system and during renogenesis in the mouse embryo. Mice homozygous for mutant RET fail to develop kidneys and enteric neurons and die within 16– 24 h after birth. Table 3 Oncogenes and Tumor Suppressor Genes in Endocrine Tumors Gene Defect Tumor phenotype H-ras Point mutation Thyroid adenomas and carcinomas, pituitary carcinoma metastases K-ras Point mutation Thyroid tumors N-ras Point mutation Thyroid tumors Ret Chromosomal rearrangement, Papillary thyroid carcinoma, MEN 2 point mutations trk chromosomal rearrangement Papillary thyroid carcinoma G s α Point mutation Pituitary tumors, thyroid adenomas PTTG Unclear Various cancers including pituitary and thyroid tumors Menin Mutation MEN 1, parathyroid tumors 11q LOH Pituitary tumors 13q LOH Parathyroid and pituitary carcinomas p53 Point mutation anaplastic thyroid carcinoma PRAD1 Chromosomal rearrangement Parathyroid adenoma c-myc Overexpression Thyroid carcinoma c-fos Over expression Thyroid carcinoma GADD45 Unclear Melanoma, pituitary tumors PPAR-γ /PAX-8 Chromosomal translocation Follicular thyroid carcinoma Table 4 Chromosome 11 Deletions in Sporadic Endocrine Tumors Adenoma a Deletion present (%) Nonfunctioning 20 GH cell 16 PRL cell 12 ACTH cell 28 Parathyroid 25 a PRL, prolactin; ACTH, adrenocortico- trophic hormone Chapter 19 / Oncogenes and Tumor Suppressor Genes 305 RET protooncogene missense mutations have been detected in 95% of patients with MEN 2A and MEN 2B, and in about 88% of families with FMTC. Families with MEN 2A have a RET mutation in exon 10 or 11, involving five conserved cysteine residues (609, 611, 618, 620, and 634) in the cysteine-rich region of the cadherin- like ligand-binding domain. These mutations probably interfere with ligand binding. Missense mutations of RET in the same cysteine residues were also identified in families with FMTC, although additional mutations in exons 13, 14, and 15 have been identified in a small number of families. Germ-line mutations in the RET protooncogene, at exon 16, were found in >95% of MEN 2B cases. Almost all cases have the same missense change (918T), resulting in a methionine-to-threonine change in the substrate-recognition pocket of the tyrosine kinase domain, although recently a methionine- to-valine substitution, involving exon 15, has been identified. The same (918T) mutation was also detected in about 30–40% of sporadically occurring MTCs and pheochromocytomas. It has been suggested that this mutation is a dominant tyrosine kinase–activating mutation, perhaps altering target specificity, which would explain the tissue hyperplasia leading to both tumors and ganglioneuromas. In papillary thyroid cancer, the genetic defect in the RET gene results from rearrangements, rather than a point mutation. The exons that encode the tyrosine kinase domain of RET are fused to 5´-regulatory sequences of other genes, leading to constitutive activation in this malignancy. 5. PITUITARY TUMORS Tumors of the pituitary gland are mostly benign adenomas that are either hormonally functional or non- functional. Functional tumors are characterized by auto- nomous hormone secretion, leading to clinical hormone excess syndromes such as acromegaly or Cushing disease, whereas nonfunctioning pituitary adenomas often secrete clinically inactive glycoprotein hormones or their free subunits. Pituitary carcinomas are extremely rare, and, to date, only about 40 cases have been documented. 5.1 Oncogenes The first point mutations detected in pituitary tumors were localized in the G protein α-subunit. Signal transduction of many peptide hormones and their cell-surface receptors are coupled with G proteins, which consist of three polypeptides: an α-chain that binds to guanine nucleotide, a β-chain, and a γ-chain. Activation of receptor accelerates the binding of GTP, which induces a con- formational change in G protein, releases α-subunit from β and γ, and allows it to interact with target proteins. Hydrolysis of bound GTP to GDP by the intrinsic GTPase acivity of the α-subunit terminates signal transduction (Fig. 1). Growth hormone–releasing hormone (GHRH) utilizes cyclic adenosine monophosphate (cAMP) as a second messenger to stimulate growth hormone (GH) secretion and somatotrope proliferation. The GHRH receptor contains seven transmembrane domains and is coupled to a G protein. Characterization of GH- secreting pituitary adenomas revealed that a subgroup of these tumors has elevated basal cAMP and GH secretion and is no longer responsive to GHRH stimulation. Sub- sequently, point mutations of G s α were identified in approx 40% of GH-secreting and 10% of nonfunctioning pituitary tumors derived from Caucasian and Korean patients, although G s α mutations appear less frequently (approx 10%) in GH-secreting pituitary tumors derived from Japanese patients. These missense mutations, Arg201Cys or His and Gln227 Arg or Leu, occur at two sites critical for GTPase activity and result in a perma- nently activated adenylate cyclase system, by prevent- ing GTP hydrolysis. Subsequent studies have not identified significant differences in the biologic and clinical phenotypes between GH-secreting tumors har- boring the gsp mutant or nonmutant proteins. Moreover, the increased cAMP levels exert a positive feedback on the cAMP phosphodiesterase (PDE4), leading to enhanced degradation of cAMP, potentially limiting the activating effects of gsp mutation. Accordingly, the true oncogenic potential of these mutations is as yet unclear. G s α point mutations have also been found in other types of endocrine tumor (see Section 6). Fig. 1. Activation of G protein on hormone stimulation. Hor- mone binding to its cell-surface receptor results in GTP binding to the G s α subunit, which activates adenylyl cyclase to increase intracellular cAMP levels. Hydrolysis of bound GTP to GDP by intrinsic GTPase activity of G s α results in inactivation of G s α and termination of signal transduction. Mutations in G s α result in inhibition of GTP hydrolysis leading to constitutive activation of G s α. H = hormones, R = receptor, C = catalytic subunit of the G protein; ATP = adenosine triphosphate. 306 Part IV / Hypothalamic–Pituitary Because the pituitary gland is under the control of multiple hormones and growth factors, signal transduction of these factors utilizes pathways other than those coupled to G protein. For example, mutations of the tyrosine kinase receptor, RET, are implicated in the pathogenesis of a subset of pituitary tumors, and a trun- cated fibroblast growth factor (FGF) receptor-4 isoform (ptd-FGFR4) has been reported in approx 40 % of pituitary adenomas. However, no single, causatory factor for sporadic pituitary tumorigenesis has been isolated. The product of protooncogene ras plays an important role in growth factor signal transduction. There are three functional ras genes—H-, N-, and K-ras—that encode for a 21-kDa protein, P 21ras . P 21ras is a guanine nucleotide–binding protein that possesses intrinsic GTPase activity and is associated with the plasma membrane. The similarities between P 21ras and G proteins suggest that ras protein is involved in signal transduction pathways regulating growth and differentiation. Recent evidence indicates that ras is brought into close contact with tyrosine kinase receptors through interaction with other proteins. Signaling downstream of ras involves a cascade of protein kinases that transmit the incoming signal to the nucleus. Point mutations of the ras gene can convert ras into a constitutively active oncogene (see Section 2). ras oncogenes have been implicated in the development of a variety of tumors and represent one of the most common mutations detected in human neoplasia. However, ras gene mutations appear to be a rare event in pituitary tumors. To date, only one invasive prolactinoma was found to harbor a missense mutation of ras. Mutations of ras were also detected in metastastic deposits of pituitary carcinomas, but not in the primary pituitary tumors. These findings suggest that activation of ras oncogene is not the initial event in pituitary tumorigenesis, however, point mutations of ras may be important in the formation or growth of metastases origi- nating from the rarely occurring pituitary carcinomas. Since its initial isolation, from rat pituitary tumor cells, pituitary tumor–derived transforming gene (PTTG) has been identified as the index mammalian securin protein, which functions to ensure faithful chromosomal separation during mitosis. The nucleotide sequence of the human homolog of this transforming gene shares 89% identity with rat PTTG. PTTG exhibits potent in vitro and in vivo transforming actions, and higher basic FGF (bFGF) levels were detected in condi- tioned medium derived from stable PTTG transfectants. PTTG expression is low in most normal adult tissues, and abundant in the testis. By contrast, abundant PTTG expression is observed in a variety of solid and hemopoietic neoplasms including pituitary, thyroid, endometrial, and colorectal tumors. Inappropriately high cellular PTTG expression promotes uneven sister chromatid separation, leading to aneuploidy, a frequent occurrence in endocrine tumors, and this chromosomal gain or loss may render the cell more prone to protooncogene activation or loss of heterozygosity of tumor suppressors. PTTG abundance may therefore be a key early determi- nant of tumorigenesis, and PTTG is the first human transforming gene found to be expressed at increased levels in the majority of pituitary tumors tested. 5.2. Tumor Suppressor Genes Because pituitary tumors comprise part of the MEN 1 syndrome, the menin gene region has been the focus of intense research in sporadic pituitary tumors. LOH of chromosome 11q13 (where menin is located) was found in up to 20% of sporadic pituitary tumors; no inactivat- ing menin mutation has been demonstrated, indicating the existence of a still unidentified tumor suppressor gene in this region. The retinoblastoma susceptibility gene (Rb), a well- characterized tumor suppressor gene, is inactivated on both alleles in a variety of human tumors. Individuals with germ-line mutations on one Rb allele have a >90% chance of developing retinoblastoma during childhood. The Rb gene maps to chromosome 13q14, and loss of LOH at this locus was demonstrated in retinoblastoma cells. The Rb gene product, pRB, is a major determi- nant of cell-cycle control and acts as a signal trans- ducer interfacing the cell-cycle apparatus with the transcriptional machinery. pRb is phosphorylated in a cell-cycle-dependent manner, being maximal at the start of the S phase and low after mitosis and entry into G1, and its state of phosphorylation regulates its activity. pRb interacts with a variety of viral and cellular proteins, and through this interaction, pRb allows the cell- cycle “clock” to control genes that mediate the advance through a critical phase of the cell growth cycle. Thus, loss of pRb function deprives the cell of an important mechanism for controlling cell proliferation through regulation of gene expression. The role of the Rb gene in pituitary tumor formation was initially suggested by studies in transgenic mice, in which one allele of the Rb gene was disrupted. Embryos homozygous for the Rb mutation die between 14 and 15 d of gestation and exhibit neuronal cell death and defective erythropoiesis. Mice heterozygous for the Rb gene mutation are not predisposed to retinoblastoma but, interestingly, develop pituitary tumors at a high frequency. These tumors originate from the inter- mediate lobe of the pituitary and are classified histo- pathologically as proopicmelanocortin immunoreac- tive adenocarcinomas. DNA derived from these tumors shows the absence of the wild-type Rb allele, and reten- Chapter 19 / Oncogenes and Tumor Suppressor Genes 307 tion of the mutant allele. Pituitary tumor tissue in the mouse also displays expression of a dysfunctional Rb protein. In 13 invasive pituitary adenomas and in pituitary carcinomas, allelic deletion of Rb was observed, whereas no LOH at the Rb locus was detected in four noninvasive pituitary tumors, although all these tumors showed normal expression of Rb protein. It is therefore likely that another tumor suppressor gene on chromosome 13 located in close proximity to the Rb locus might be involved in the formation of pituitary tumor. Interestingly, although p53 tumor suppressor gene mutations have been detected in a wide variety of human tumors, no such mutation has been found in pituitary tumors that were comprehensively screened. A recent study examining GADD45γ expression, a member of a growth arrest and DNA damage-inducible gene family, demonstrated that although GADD45γ was abundant in normal human pituitary tissues, it was detectable in only 1 of 18 clinically nonfunctioning pituitary tumors and was not expressed in most GH- or prolactin-secreting pituitary tumors. Transfection of GADD45γ cDNA into pituitary tumor cells inhibited tumor cell colony formation, indicating growth-sup- pressive actions of GADD45γ. 6. THYROID TUMORS Thyroid neoplasia comprises benign follicular adenoma s, differentiated carcinomas (follicular carcinoma and papillary carcinoma), and anaplastic or undifferen- tiated carcinomas, and many distinct molecular events occur in thyroid neoplasia. The most common mutations found in thyroid tumors are point mutations of the ras protooncogene. Mutations of all three ras genes have been identified in follicular adenomas and in thyroid carcinomas. Besides activating point mutations of ras, some thyroid tumors exhibit ras gene amplifications. Thyrotropin (thyroid-stimulating hormone [TSH]) not only regulates differentiated function of thyroid cells, but also acts as a growth factor for thyrocytes. The growth-inducing function of TSH on thyrocyte growth is mediated by cAMP, which is induced after adenylyl cyclase activation by G s α protein. As in GH-secreting pituitary tumors, point mutations that inhibit the intrinsic GTPase activities of G s α have been detected in 25% of hyperfunctioning thyroid adenomas. Some hyperfunctioning adenomas have mutations in the third intracellular loop of the TSH receptor, resulting in a constitutively active receptor and inappropriate activation of adenylyl cyclase. A novel oncogene, PTC-RET, is unique to papillary thyroid carcinomas. PTC-RET arises as the result of an intrachromosome inversion that juxtaposes unrelated gene sequences to the tyrosine kinase domain of the RET protooncogene (see Section 4.2). Other activating gene rearrangements found in this group of thyroid carcinomas include the nerve growth factor receptor (trk). Recently, a translocation t(2;3)(q13;p25) involving the fusion of the thyroid transcription factor, PAX8, and PPAR-γ was suggested to arise in follicular thyroid carcinomas, and this rearrangement may be a useful diagnostic marker to differentiate follicular thyroid carcinoma and adenoma. Anaplastic carcinoma is the most aggressive form of thyroid cancer and displays complete loss of thyroid differentiation. These tumors exhibit a high prevalence of p53 tumor suppressor gene missense mutations that are not present in differentiated thyroid carcinomas. p53 is a sequence-specific DNA- binding protein that binds to DNA as a tetramer and regulates transcription of genes that negatively control cell growth and invasion. p53 induces differentiation and acts as a checkpoint protein that arrests the cell cycle in response to DNA damage, allowing DNA repair to take place, or to activate pathways for apoptosis. The p53 gene is the most frequently mutated locus in human neoplasia, involved in 50% of human cancers. Typically, one allele of the p53 gene is lost and point mutations occur in the remaining allele, resulting in production of a mutant protein. The majority of missense mutations occur within the evolutionarily conserved regions of the gene. Some mutants lose the ability either to bind to DNA or to transactivate target genes. Other mutations seem to affect p53 function by changing the global con- formation of the protein. All mutant p53 proteins have lost the ability to suppress transformation, and some mutants can also act as dominant oncogenes in coopera- tion with ras in transformation of primary cells. The presence of p53 point mutations in anaplastic carcinomas but not in differentiated thyroid tumors suggests that inactivation of the p53 tumor suppressor gene may play a role in transition to the more malignant phenotype of thyroid carcinomas. 7. PARATHYROID NEOPLASIA Parathyroid adenomas arise either sporadically or in association with MEN 1. Multiple chromosomal regions, 1p-pter (40% of adenomas), 6q (32%), 15q (30%), and 11q (25–30%), are missing in individual parathyroid adenomas, probably reflecting the deletion of tumor suppressor genes, some of which have been characterized. For example, the majority of MEN 1– associated and about 25% of sporadic parathyroid adenomas exhibit 11q deletions that are associated with mutations in the menin gene (located on the undeleted 11q). A subgroup of parathyroid adenomas contains a chromosome 11 inversion in which the 5´-regulatory region of the parathyroid hormone is fused to the coding 308 Part IV / Hypothalamic–Pituitary region of the PRAD1 gene. This gene rearrangement leads to overexpression of the PRAD1 gene in tumor cells. Cloning of PRAD1 cDNA revealed that it is struc- turally related to the cyclins, and it is now termed cyclin D1. Cyclins are a group of proteins that play important roles in controlling cell-cycle progression. Cell-cycle progression is regulated at two critical checkpoints: the G2/M border and the G1/S transition. Cyclin D1 is a “G1 cyclin” and functions to propel cells through the G1/S transition checkpoint. Cyclin D1 forms com- plexes with cyclin-dependent kinases, and also interacts directly with the product of Rb gene to participate in cell-cycle regulation. Overexpression of cyclin D1 could lead to excessive cell proliferation. The implica- tion of cyclin D1 as a parathyroid oncogene indicates the important role that the cell-cycle machinery can play in inducing parathyroid tumors, by regulating the biochemical pathways that control parathyroid cell proliferation. Most parathyroid carcinomas show LOH at the Rb allele. As in pituitary tumors, inactivation of Rb is only found in more aggressive tumors, not in benign parathyroid adenomas, suggesting that Rb gene assessment could be used as a diagnostic or prognostic molecular marker for parathyroid carcinoma. A subset of parathyroid tumors harbors amplification of regions on chromosomes 16p and 19p that contain as yet uncharacterized protooncogenes. 8. CLINICAL IMPLICATIONS Genetic changes in protooncogenes and tumor suppressor genes play important roles in the development and progression of endocrine tumors. The identification and characterization of new tumor suppressor genes and oncogenes will not only provide important insights into normal growth regulation of endocrine cells and eluci- date the genetic alterations that lead to tumor formation, but may also provide diagnostic or prognostic tools for these lesions (Table 5). For example, ras protooncogene point mutations have been found in all types of thyroid tumors but rarely in pituitary tumors. The advent of rapid throughput and powerful molecular biologic techniques, such as DNA microarray (Fig. 2) and proteomic analyses, that will ultimately enable characterization of individual tumors Table 5 Clinical Impact of Genetic Screening in Endocrine Tumors 1. Allows early prediction of tumor behavior 2. Portends a response to therapeutic interventions 3. Provides genetic screening for tumor prediction 4. Allows design of novel subcellular therapies Fig. 2. Schematic representation of microarray. Any known DNA sequence, generated by chemical synthesis or polymerase chain reaction, is positioned on specifically treated glass slides with the aid of a robotic device capable of depositing very small drops (nanoliters) in precise patterns. Ultraviolet light is then used to crosslink the DNA to the glass slides, and the microarry can then be probed with fluorescently labeled nucleic acids. For example, mRNA samples are collected from normal and cancer cells (rep- resenting all the genes being expressed in these cells), cDNA probes for each sample are made with nucleotides that fluoresce in different colors, and a mixture of the cDNAs is used to probe the microarray. Spots that fluoresce green represent mRNAs more abundant in the cancer, whereas spots that fluoresce red represent sequences more abundant in normal tissue. The molecular phenotype of the tumor is thereby interrogated and characterized by computer modeling to identify potential drug targets and “tailor design” appropriate treatment. at the mechanistic level, as well as the parallel development of specific receptor and signal transduction cascade targeted therapies, potentially will identify tumors that exhibit an aggressive phenotype, and guide clini- cians in determining the need for more aggressive post- operative management and follow-up. As these tools increase the understanding of key underlying molecular mechanisms of endocrine tumor development, in tan- dem they will pave the way to designing subcellular therapeutic modalities for managing endocrine neoplasia in the future (Table 6). Chapter 19 / Oncogenes and Tumor Suppressor Genes 309 SELECTED READING Agarwal SK, Guru SC, Heppner C, et al. Menin interacts with the AP1 transcription factor JunD and represses JunD activated transcription. Cell 1999;96:143–152. Bos J L. Ras genes in human cancer: a review. Cancer Res 1989;49: 4682. Chandrasekharappak SC, Guru SC, Manickam P, et al. Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 1997;276: 404–406. Evan GI, Littlewood TD. The role of myc in cell growth. Curr Opin Genet Dev 1993;3:44. Ezzat S, Zheng L, Zhu XF, Wu GE, Asa SL. Targeted expression of a human pituitary tumor-derived isoform of FGF receptor-4 reca- pitulates pituitary tumorigenesis. J Clin Invest 2002;109: 69–78. Fagin JA. Minireveiw: branded from the start-distinct oncogenic initiating events may determine tumor fate in the thyroid. Mol Endocrinol 2002:16:903–911. Hahn WC, Weinberg RA. Rules for making human tumor cells. N Engl Med 2002;347:1593–1603. Heaney AP, Horwitz GA, Wang Z, Singson R, Melmed S. Early involvement of estrogen-induced pituitary tumor transforming gene and fibroblast growth factor expression in prolactinoma pathogenesis. Nat Med 1999;5:1317–1321. Hoff AO, Cote GJ, Gagel RF. Genetic screening of endocrine disease. In: Baxter JD, Melmed S, New MI, eds. Genetics in Endo- crinology. Philadelphia, PA: Lippincott Williams & Wilkins 2002:189–221. Knudson AD. All in the (cancer) family. Nature Genet 1993;5:103. Kroll TG, Saraff P, Pecciarini L, Chen CJ, Mueller E, Spiegelman BM, Fletcher JA. PAX8-PPARgamma1 fusion oncogene in human thyroid carcinoma [corrected]. Science 2000;289:1357– 1360. Motokura T, Arnold A. Cyclins and oncogenesis. Biochem Biophys Acta 1993;1155:63. Santoro M, et al. Activation of RET as a dominant transforming gene by germline mutations in MEN2A and MEN2B. Science 1995; 267:381. Weinberg RA. Molecular mechanisms of carcinogenesis. In: Philip Leder, David A. Clayton, Edward Rubenstein, eds. Scientific American Introdtuction to Molecular Medicine, New York, NY: Scientific American, 1994. Table 6 Potential Endocrine Tumor–Targeted Gene Therapy Replacing functions of tumor suppressor gene 13q, 11q, 17p Interrupting self-stimulatory autocrine loops gsp, hypothalamic hormone receptor antagonists Interrupting aberrant signal responsiveness RET Using specific immunotherapy Mutant receptors, growth factors Chapter 20 / Insulin Secretion and Action 311 311 From: Endocrinology: Basic and Clinical Principles, Second Edition (S. Melmed and P. M. Conn, eds.) © Humana Press Inc., Totowa, NJ 20 Insulin Secretion and Action Run Yu, MD, PhD, Hongxiang Hui, MD, PhD, and Shlomo Melmed, MD CONTENTS INTRODUCTION DEVELOPMENT OF ENDOCRINE PANCREAS AND REGULATION OF ISLET Β-CELL MASS INSULIN SYNTHESIS INSULIN SECRETION INSULIN SIGNALING required for normal development. Intact insulin function requires four components: islet β-cell mass; insulin synthesis; glucose-dependent insulin secretion; and, ultimately, insulin signaling at the target cells. Although theoretically any of these components can go awry and cause disease, abnormal insulin signaling is the most common problem, followed by decreased islet cell mass. Abnormal insulin synthesis or secretion rarely causes diseases. Understanding the physiology of all four components, however, is important to prevent and treat diabetes and related diseases. 2. DEVELOPMENT OF ENDOCRINE PANCREAS AND REGULATION OF ISLET β-CELL MASS The islet β-cell mass is dynamic and regulated. Maintaining an appropriate β-cell mass in response to metabolic demand is critical for maintaining glucose homeostasis. Decreased β-cell mass underlies several types of diabetes. In type 1 diabetes, β-cells are destroyed by an autoimmune mechanism. In late-stage type 2 diabetes, β-cells undergo excessive apoptosis owing to glucose toxicity. Intrinsic genetic defects play a central role in causing decreased β-cell mass in maturity-onset diabetes of the young (MODY), diabe- 1. INTRODUCTION Because glucose is the primary energy source of most cells in the body, control of constant circulating glucose levels is of utmost importance. Too little serum glucose (hypoglycemia) suppresses central nervous system functions and prolonged hypoglycemia leads to death. Too much serum glucose (hyperglycemia) as seen in diabetes mellitus, results in grave consequences such as kidney, nerve, eye, muscle, and immune system damage. The body has an elaborate system to control circulating glucose levels in a narrow range (72–126 mg/dL) to prevent untoward fluctuations. For populations in Western societies, the predominant problem in glucose metabolism is diabetes mellitus although other derange- ments are significant but not often encountered. Of all the humoral and neuronal regulatory mechanisms for glucose metabolism, insulin is the hormone that lowers serum glucose whereas most other mechanisms function to increase serum glucose. Insulin is a peptide hormone secreted by β-cells in the pancreatic islets of Langerhans. The main function of insulin is to lower serum glucose. Insulin is a major anabolic hormone that is critical in lipid and protein synthesis, and insulin is also an essential growth factor 312 Part IV / Hypothalamic–Pituitary tes associated with genetic syndromes, and diabetes encountered in genetically modified animal models. Unlimited growth of transformed β-cells comprising insulinoma tumor causes hypoglycemic coma. The net β-cell mass results from the difference between β-cell proliferation and cell death. β-Cell proliferation is achieved in two ways: (1) neogenesis (formation of β− cells from precursor pancreatic ductal cells) and (2) replication of differentiated β-cells. β-Cell death can be due to either apoptosis (programmed and controlled) or necrosis (associated with inflammation). It has recently been realized that β-cells continue to proliferate throughout adult life. The average islet size is more than 10-fold larger in adult mice than in younger ani- mals. The most robust islet growth occurs in the intrau- terine and neonatal period. β-Cell proliferation is also prominent during pregnancy. An islet of Langerhans comprises three main cell types: (1) glucagon-secreting α-cells at the periphery (15–20% of islet cells), (2) insulin-secreting β-cells at the inside (60–80%), and (3) peripheral somatostatin-secreting δ-cells (5–10%). Glucagon functions to increase serum glucose levels and somatostatin suppresses insulin secretion. During embryonic pancreatic development, primordial islet cell clumps are derived from nascent pancreatic ducts and detach from the ducts to expand and coalesce with other clumps (Fig. 1). β-Cells are first detected at the wk 13 of gestation and begin to secrete insulin at the wk 17. β-Cell proliferation and differentiation are under tight control of a number of transcription factors. Pancreatic duode- nal homeobox gene-1 (PDX-1), hepatocyte nuclear factor-1α (HNF-1α), HNF-1β, HNF-4α, insulin pro- moter factor-1 (IPF-1), and β-cell E-box transactivator 2 (BETA2) are some that have clinical implications since their respective mutations result in neonatal diabetes or MODY. PDX-1 is required for normal pancreatic development. A patient with deficient PDX-1 expression has failed pancreatic development. In adults, PDX-1 is also important for normal islet function because it regulates insulin gene expression. HNF- 1α gene mutations are found in MODY3 and HNF-4α gene mutations in MODY1; mutations in HNF-1β, IPF-1, and BETA2 cause MODY5, MODY4, and MODY6, respectively. All those MODY subtypes (1, 3–6) are characterized by insufficient β-cell mass. Genetic defects are also responsible for several clinical diabetes syndromes and diabetes in some animal models. Defective expression of β-cell mitochondrial protein frataxin, a gene that is deficient in Friedreich ataxia, results in decreased β-cell proliferation and increased apoptosis in experimental mice, suggesting that the diabetes associated with Friedreich ataxia may be owing to decreased β-cell mass. Wolframin, an endoplasmic protein that is defective in Wolfram syndrome (diabetes, blindness, and deafness at early childhood), appears to protect β-cells from apoptosis, thus explaining the decreased β_cell mass observed in this syndrome. Dia- betes is found in a number of knockout mice deficient in various genes. In many cases, the experimentally disrupted gene is expressed in many cell types, but the β-cells seem to be particularly vulnerable to the disruption. Securin (also called PTTG), a regulatory protein critical for progression of mitosis, is expressed in all proliferative cells. Disruption of murine securin results in defective proliferation of β-cells, a cell type not known for rapid division, while sparing more proliferative cells such as hemopoietic and spermatogenic cells. Another intriguing feature of diabetes owing to genetic defects is that in most cases, they do not immediately manifest themselves but occur after a considerable latent period in childhood or young adulthood, suggesting that other insults during postnatal growth must cooperate with the genetic defects to result in the clinical phenotype. Besides genetic determinants, β-cell mass is regulated by nutrients, growth factors, and hormones. Glu- cose is a major stimulator of β-cell proliferation, and in rodents, infusion of glucose for 24 h results in a rapid increase in β-cell mass, mostly owing to neogenesis. Ironically, glucose also induces β-cell apoptosis, as seen in prolonged type 2 diabetes. Amino acids and free fatty acids are also potent stimulators of β -cell proliferation. Several growth factors such as epidermal growth factor, fibroblast growth factor, and vascular endothelial growth factor stimulate β-cell proliferation. Many hormones including insulin, insulin-like growth factor-1 (IGF-1), IGF-2, glucagon, gastroinhibitory peptide, gastrin, cholecystokinin, growth hormone (GH), prolactin (PRL), placental lactogen, leptin, and glucagon- like peptide-1 (GLP-1) stimulate β-cell proliferation. Most of the hormones have significant systemic effects and therefore are not good candidates for potential therapeutic agents, whereas GLP-1 is rather specific in increasing β-cell mass. GLP-1, a gastrointestinal peptide hormone secreted by enteroendocrine L-cells, stimulates β-cell mass growth in animal models and in Fig. 1. Schematic illustration of pancreatic β-cell neogenesis from ductal cells. [...]... neuronal production and release of the peptide Endothelial cell production of CNP has been clearly established and the control of peptide secretion partially 324 Part IV / Hypothalamic–Pituitary characterized A variety of cytokines and growth factors (including interleukin-1α [IL-α] and IL-β, tumor necrosis factor-α [TNF-α], and transforming growth factor [TGF-β], as well as ANP and BNP) can stimulate... regions in insulin—amino terminal A-chain (GlyA1-IleA2ValA3-GluA4 or AspA4), carboxyl terminal A-chain (TyrA19-CysA20-AsnA21), and carboxyl terminal Bchain (GlyB23-PheB24-PheB25-TyrB26)—are located at or near the surface of insulin and therefore may interact with the insulin receptor Understanding insulin structure has stimulated development of a number of insulin analogs, and recombinant DNA technology... normal chromaffin cells of the adrenal gland, as well as a variety of other tissues, including brain, kidney, endothelial cells, and VSMCs Posttranslational processing of the 185 -amino-acid prohormone results in the production and secretion of the mature 52-amino-acid form, designated AM, and a 20-amino-acid fragment from the N-terminus designated proadrenomedullin N-terminal 20 peptide (PAMP), a peptide... ischemia, and shear stress 3.3 Site and Mechanisms of Action Two mammalian ET receptor subtypes have been cloned, and they are members of the G protein–linked, seven-transmembrane-spanning domain superfamily of biologic receptors (Fig 3) The ET-A receptor displays a rank order of binding affinity with ET-1 being the preferred ligand (ET-1 Ն ET-2 >> ET-3) This receptor predominates in VSMCs and cardiac... the ET-A receptor is possible and isoform-specific activation of the ET-B receptor can now be accomplished Also available are antagonists that affect both the ET-A and ETB receptor, and a new generation of relatively specific ET-B antagonists Much interest continues regarding the possible existence of a unique, ET-3-selective ET-C receptor in mammals similar to that found in frog melanophores, and it... human ET-1 gene has five exons and four introns, with the peptide coded in the second exon The gene is transcriptionally regulated via cis elements, including a GATA-2 protein-binding site and an AP-1 site that is activated by thrombin, angiotensin II, epidermal growth factor (EGF), basic fibroblast growth factor (bFGF), 3 28 Part IV / Hypothalamic–Pituitary insulin-like growth factor, and TGF-β Other... protein (SHC) and multifunctional docking proteins IRS-1 and IRS-2 (Fig 4) Tyrosine-phosphorylated SHC recruits the small adapter protein growth factor receptor–binding protein 2 (Grb2), which, in turn, recruits and activates the ras-guanosine 5´-diphosphate exchange 316 Part IV / Hypothalamic–Pituitary Fig 4 (A) Insulin receptor structure; (B) signal transduction for IR factor mammalian son-of-sevenless... the 3-position to form phosphatidylinositol-3-phosphates Signaling proteins containing pleckstrin homology domains bind to the membrane-bound phosphaotidylinositol-3-phosphate, and their activity or localization is altered by the binding p110 also has some serine kinase activity that appears to be largely directed toward the p85 subunit and IRS-1 PI3K recruitment increases PI3K activity because p85 interaction... the α-subunit Both isoforms have similar affinities for insulin IR as a whole is an allosteric enzyme The β-subunit has tyrosine kinase activity, and the α-subunit regulates the β-subunit in an insulin-binding-dependent manner The extracellular IR α-subunit is responsible for ligand binding It has two insulin-binding pockets that interact with corresponding regions of insulin molecule The carboxyl-terminal... GC-B receptors and are alternatively called natri- uretic peptide receptor-A (NPR-A) and NPR-B A third receptor subtype, called the clearance receptor or NPRC, shares approx 30% homology with NPR-A and NPRB in the extracellular ligand-binding domain; however, this receptor lacks the intracellular C-terminal extension (i.e., it is missing the kinase and GC domains) This receptor was originally thought . homeobox gene-1 (PDX-1), hepatocyte nuclear factor-1α (HNF-1α), HNF-1β, HNF-4α, insulin pro- moter factor-1 (IPF-1), and β-cell E-box transactivator 2 (BETA2) are some that have clinical implications since. insulin—amino terminal A-chain (GlyA1-IleA 2- ValA3-GluA4 or AspA4), carboxyl terminal A-chain (TyrA19-CysA20-AsnA21), and carboxyl terminal B- chain (GlyB23-PheB24-PheB25-TyrB26)—are located at. insulin-secreting β-cells at the inside (60 80 %), and (3) peripheral somatostatin-secreting δ-cells (5–10%). Glucagon functions to increase serum glucose levels and somatostatin suppresses