The goal of our work is to understand the genetic and molecular mechanisms leading to disorders that affect the adrenal cortex, with emphasis on disorders that are developmental, hereditary, and associated with adrenal hypoplasia or hyperplasia, multiple tumors, and abnormalities in other endocrine glands (i.e., the pituitary gland). In this context, our laboratory has studied congenital adrenal hypoplasia caused by Allgrove syndrome (or triple A syndrome) and multiple endocrine deficiencies (APECED syndrome), familial hyperaldosteronism, adrenocortical and thyroid cancer, pituitary tumors and multiple endocrine neoplasia syndromes (MEN 1, Cowden disease) affecting the pituitary, thyroid, and adrenal glands, and Carney complex, an autosomal dominant disease that affects the adrenal cortex. Carney complex is the only inherited form of adrenal gland-dependent Cushing’s syndrome; it is also a mutliple endocrine neoplasia affecting the pituitary and thyroid glands and the gonads and is associated with a variety of other tumors, including myxomas (heart and other myxomas) and schwannomas (it is one of three conditions associated with inherited schwannomas; the other two are neurofibromatosis and isolated schwannomatosis), and skin pigmentation defects (lentigines, café-au-lait spots, and nevi). Recently, our laboratory identified the regulatory subunit type 1-_ (Ri_) __of protein kinase A (PKA), which is coded by the PRKAR1A gene, the gene responsible for almost half the cases of Carney complex. Since then, our laboratory has largely focused on PKA-stimulated signaling pathways, PKA effects on tumor suppression and/or development, the cell cycle and chromosomal stability, and PKA-dependent hormonal interactions.

Carney Complex Genetics
Kirschner, Bei, Bourdeau, Matyakhina, Sandrini, Stratakis in collaboration with J.A. Carney (Mayo Clinic)
Families with Carney complex (CNC) and related syndromes from a number of collaborating institutions worldwide have been accessed. Genetic linkage analysis located two loci harboring genes for CNC on chromosomes 2 (2p16) and 17 (17q22-24) while a possible third locus for this genetically heterogeneous condition is currently under investigation. With the application of state-of-the-art molecular cytogenetic techniques, the participation of the two genomic loci on chromosomes 2 and 17 in the expression of the disease is undergoing study. A comprehensive genetic and physical map of the 2p16 chromosomal region was constructed for the cloning of the CNC-associated sequences from this region. Studies in cultured primary tumor cell lines (established from our patients) identified a region of genomic instability in the center of the map. Tumor studies (Figure 1) led to the identification of the PRKAR1A gene on 17q22-24 as the gene responsible for CNC in approximately 40 percent of cases of the disease. PRKAR1A was identified as a novel tumor suppressor gene in CNC-associated tumors; it is also the main regulatory subunit of protein kinase A (PKA), a central signaling pathway for many cellular functions and hormonal responses. With more patients with CNC now participating in genotype-phenotype correlation studies, the studies are expected to shed light on the complex biochemical and molecular pathways regulated by PRKAR1A and PKA.

Figure 3

Mapping of the PRKAR1A gene on chromosome 17 and identification of genetic markers of the PRKAR1A locus that were used for genetic studies in CNC families (left panel). A probe containing the PRKAR1A gene from a chromosome 17 human genome library was then used in an ovarian tumor cell line from a patient with CNC in a fluorescent in situ hybridization experiment (right panel); the cells demonstrate allelic loss of the PRKAR1A-containing probe (one copy per cell instead of two), suggesting that PRKAR1A is a tumor suppressor gene.

PRKAR1A Effects on Protein Kinase A Activity and Endocrine Tumor Development
Kirschner, Matyakhina, Sandrini, Lenherr, Gosselink, Robinson-White, Stratakis in collaboration with Y. Cho-Chung (NCI)
The functional consequences of PRKAR1A mutations are being investigated in cell lines established from CNC patients and their tumors. Both cAMP and PKA activity are measured in the cell lines, along with the expression of the other subunits of the PKA tetramer. We have established stable transfectants of antisense PRKAR1A constructs in mouse endocrine and other commercially available cell lines and have studied the cell lines for the effects of PRKAR1A silencing on cell growth, differentiation, and proliferation. It is hypothesized that the tumorigenicity of PRKAR1A-inactivating mutations relies on the switch from type-I PKA (based almost exclusively on endocrine cells on PRKAR1A) to type-II PKA activity; cell lines with an antisense PRKAR1A construct are believed to be a representative model of the in vivo situation in CNC patients. In addition, we are looking for mutations of the PRKAR1A gene that would further establish its role as a general tumor suppressor in sporadic endocrine and nonendocrine tumors (thyroid adenomas and carcinomas, adrenocortical adenomas and carcinomas, ovarian carcinomas, melanomas and benign and malignant pigmented lesions, and heart myxomas); a variety of investigators within the NIH and around the world provide specimens on a collaborative basis.

PRKAR1A Animal Models
Kirschner, Gosselink, Lenherr, Stratakis in collaboration with H. Westphal (LMGD, NICHD)
Given that the PRKAR1A-knockout (KO) (-/-) mouse (created several years ago by S. McKnight of the University of Washington, Seattle) dies on day nine of embryonic development of heart and central nervous system abnormalities and PRKAR1A’s involvement in CNC, our laboratory became interested in developing alternative animal models that would address the complex in vivo effects of PRKAR1A inactivation. The several animal models under development in our laboratory fall into two categories: conditional PRKAR1A KOs in endocrine tissues (adrenal cortex, anterior lobe of the pituitary and the thyroid gland) and (2) transgenic antisense PRKAR1A expression.

PRKAR1A, the Cell Cycle, Chromosomal Stability, and Other Signaling Pathways
Matyakhina, Robinson-White, Stratakis in collaboration with V. Papadopoulos (Georgetown University, Washington DC)
Genes implicated in cyclic nucleotide-dependent signaling have long been considered likely candidates for endocrine tumorigenesis. Somatic activating mutations in a number of G-protein coupled receptors (GPCRs) and the gene coding for a subunit of the stimulatory G protein (GNAS1) lead to increased cAMP production and are responsible for a number of endocrine tumors of various types. To date, however, there is no convincing evidence that GNAS1 or GPCR activation in the absence of additional genetic abnormalities is involved in cancer. Individuals with McCune-Albright syndrome (MAS, a disease similar to CNC) who bear somatic GNAS1 mutations in their endocrine glands may be predisposed to some cancers. However, activation of additional pathways and/or other changes appear to be required for the in vitro transformation of 3T3 or FRTL5 cells by constitutively active GPCR transgenes or in other settings of increased cAMP signaling that lead to malignant transformation. Thus, other genes that regulate PKA function and increase cAMP-dependent proliferation and related signals may be altered in the process of endocrine tumorigenesis initiated by a mutant PRKAR1A, a gene with important functions in the cell cycle as well as in chromosomal stability. Our work aims to identify the interactions of PRKAR1A by studying mitogenic and other growth signaling pathways in cell lines that express the antisense PRKAR1A contructs referred to above. In addition, chromosomal stability in both human and mouse cell lines in which PRKAR1A has been inactivated is under investigation with the techniques of classic and molecular cytogenetics, including fluorescent in situ hybridization (FISH), spectral karyotyping (SKY), and comparative genomic hybridization (CGH). Finally, more recently, we are investigating proteins that are directly bound to PRKAR1A (RI_) and regulate its function, including a novel protein, PAP7 that a collaborating laboratory identified. Compartmentalization of PKA function is supposed to be mediated by anchoring proteins; almost all the PKA anchoring proteins known to date, however, are bound to type II-PKA. PAP7 may be the first PKA type-I-specific anchoring protein.

Genetic Investigations on Other Adrenocortical Diseases and Tumors
Bourdeau, Sandrini, Farmakidis, Stratakis in collaboration with S. Libutti (NCI), W.Y. Chan (NICHD), J.A. Carney (Mayo Clinic), M Stowasser and D Torpy (University of Queensland, Australia), A. Lacroix (University of Montreal, Canada), and J. Bertherat (Hospital Cochin, France)
Our work aims at using general and pathway-specific microarrays on a variety of adrenocortical tumors to identify genes with important functions in adrenal oncogenetics; examining specific candidate genes (such as TP53 and other tumor suppressors and oncogenes) for their roles in adrenocortical tumors and development; and identifying by positional cloning additional genes with a role in inherited adrenocortical diseases. Our laboratory engages in collaborations for a large part of its work, as is evident from the number of collaborating investigators and institutions. As part of this work, we and our collaborators have recently completed the following: identification of a novel TP53 mutation in a cohort of adrenocortical tumors from southern Brazil with important implications for p53’s function in adrenocortical tumor suppression; development of a genomewide screen for the identification of gene(s) responsible for inherited adrenocortical aldosteronomas (familial hyperaldosteronism type II - FH-II) and that identified a locus for FH-II on chromosome 7 (7p22); ongoing development of a genomewide screen for the identification of a syndrome composed of familial paragangliomas and adrenal, gastric stromal. and pulmonary tumors; and the identification of new mutations in the APECED and Allgrove syndrome genes leading to congenital adrenal hypoplasia.

Genetic Investigations on Other Endocrine Neoplasias and Related Syndromes
Bourdeau, Sandrini, Stratakis
Additional work in our laboratory is exploring the genetics of CNC- and adrenal-related endocrine tumors, including childhood adrenocortical cancer and thyroid and pituitary tumors. As part of this work, we described a novel form of acromegaloid syndrome in which the responsible genetic defect was associated with a chromosome 11 abnormality (see figure). Finally, we are identifying the genetic defects in patients with CNC-related syndromes (the lentigenoses, i.e., Peutz-Jeghers syndrome and others) largely in collaboration with several other investigators at the NIH and elsewhere.

Figure 4

A) G-banding high resolution karyotype of a patient with pseudoacromegaly; the arrows point to a newly described chromosome 11 abnormality [inv(11)(p15.3;q23.3)], which is undetectable without proper analysis of band density. B) Fluorescent in situ hybridization on cultured lymphocytes from the proband with two probes that hybridize to the 11p telomere (11pter) and to the MLL gene on 11q23, respectively, showing that, on one chromosome 11, the two probes were positioned proximal to each other (arrow), whereas they hybridized to their expected positions on the other chromosome.

Clinical Investigations in the Diagnosis and Treatment of Adrenal and Pituitary Tumors
Kirschner, Stratakis, Keil
Patients with adrenal tumors and other types of Cushing’s syndrome (and occasionally other pituitary tumors) come to the NIH Clinical Center for diagnosis and treatment. Ongoing investigations include the prevalence of ectopic hormone receptor expression in adrenal adenomas and massive macronodular adrenocortical disease; the diagnostic use of high- sensitivity magnetic resonance imaging for the earlier detection of pituitary tumors; the diagnosis, management, and postoperative care of children with Cushing’s disease and other pituitary tumors.