University of Connecticut Health Center Surgical Research Center Farmington, Connecticut 06030-1110
Address correspondence and requests for reprints to: Carl D. Malchoff, M.D., University of Connecticut Health Center, Surgical Research Center, 263 Farmington Avenue, Farmington, Connecticut 06030-1110.
![]() |
Introduction |
---|
![]() ![]() ![]() |
---|
CNC refers to the familial association of myxomas, spotty pigmentation, and endocrine overactivity. In 1985, J. Aidan Carney and colleagues (1) carefully described many of the clinical characteristics of this disorder based on his analysis of 40 patients. Their findings included cardiac myxomas, skin lesions (both myxomas and pigmented lesions), primary pigmented nodular adrenocortical disease (PPNAD), which causes ACTH-independent Cushings syndrome, myxoid fibroadenomas of the breast, growth hormone-secreting pituitary tumors, and both Sertoli cell and Leydig cell (steroid-type) testicular tumors. Inheritance is autosomal dominant with incomplete penetrance. This rare familial tumor syndrome often presents to the endocrinologist because the clinical features include Cushings syndrome, acromegaly, and male precocious puberty.
Multiple hypotheses were entertained to explain CNC and its diverse
manifestations. These included adrenal stimulating antibodies
(2) and activating mutations of the -subunit of Gs, as
occurs in the McCune-Albright syndrome (3). Interestingly,
these hypotheses were not as disparate as one might think. Because
PPNAD is a relatively frequent component of this disorder, these
hypotheses all proposed that the inherited susceptibility gene would
activate the signal transduction pathway that mediates adrenal cortisol
production. Components of this pathway (Fig. 1
) include ACTH, the type 2 melanocortin
receptor that binds ACTH, the G protein complex that transduces the
ACTH binding message into adenylyl cyclase activation,
phosphodiesterases that degrade intracellular cAMP, cAMP-dependent
protein kinase (PKA) and its regulatory subunits, and further
downstream events. Indeed, clinical syndromes caused by activation of
these signal transduction pathways have been described (4, 5). However, there are multiple components to this pathway,
there could be unsuspected components, and the CNC susceptibility gene
need not be a component of this pathway. Therefore, powerful and
complementary genetic methodologies were chosen to provide additional
information.
|
The careful clinical evaluations continue to clarify the CNC phenotype. In a 1996 publication, Stratakis et al. (6) evaluated 101 persons from 11 families with CNC. About half of those evaluated were determined to have CNC by the presence of two or more clinical features of this disorder. The most common findings were those of skin pigmentation, which occurred in 96% of affected subjects. Other features included cardiac myxomas (36%), skin myxomas (63%), breast myxomas (22% of all affected subjects divided as 37% in women and 4% in men), PPNAD (32%), acromegaly (8%), Sertoli cell testicular tumors (10%), and thyroid tumors (10%). Subsequently, these investigators have aggressively evaluated affected individuals for tumors in specific organs. In this issue of the JCEM, they combine the North American experience with a European experience provided by Dr. W. H. Oelkers and report a prospective and retrospective analysis of CNC ovarian neoplasms (7). They noted that pigmented skin lesions are a clinical overlap of CNC with Peutz-Jeghers syndrome (PJS) and looked for other evidence of clinical overlap. Because both testicular and ovarian tumors are a component of PJS, they postulated that ovarian neoplasms might be a component of CNC. In addition, they argue that a familial tumor syndrome with an increased incidence of testicular neoplasms may have an increased incidence of ovarian neoplasms (7). Although Dr. Carneys 1985 CNC description noted only two ovarian lesions, they reasoned that a more aggressive evaluation would uncover more lesions. Eighteen women with CNC (median age, 33 yr) were evaluated prospectively with ultrasound examination over a mean period of 34 months, and 178 women with CNC from a large international registry were evaluated retrospectively. Interestingly, there were no stromal tumors of the ovaries as seen in PJS. Of the 18 women evaluated prospectively, 12 developed ovarian cysts as compared with 1 of 11 in the control group (P = 0.003). Of the affected women, two had progression of cysts and underwent ovariectomy for serous cystadenomas. The retrospective analysis identified only four women with ovarian neoplasms requiring surgery, and one was felt to be metastatic adenocarcinoma. Two ovarian lesions removed in the prospective study demonstrated a copy gain in the chromosomal region of 2p16, which may represent gene amplification. Other studies of affected subjects have more clearly defined the thyroid lesions and the paradoxical response of cortisol production to dexamethasone administration (8, 9).
Careful clinical evaluation greatly facilitated the genetic linkage
analysis. Genetic heterogeneity was observed, and the protein kinase A
type I- regulatory subunit gene (PPKAR1A) was identified
as a CNC susceptibility gene. Linkage analysis identified two CNC
genetic loci: a 6.4-cM region at 2p16 (6) and
a 17-cM region at 17q23-24 (10). A number of
CNC candidate genes were excluded by these analyses. The
proopiomelanocortin gene was excluded because its chromosomal location
on chromosome 2 is telomeric to the CNC locus. There were two known
tumor suppressor genes on the long arm of chromosome 17:
BRCA1 and NF1, tumor suppressor genes for
familial breast carcinoma and for neurofibromatosis type I,
respectively. However, these are excluded because their chromosomal
locations are centromeric to the CNC locus. In affected subjects
tumor-specific loss of heterozygosity (LOH) was found within the 17q
linkage region in eight different neoplasms including adrenal lesions,
myxomas, GH-secreting pituitary tumors, a follicular thyroid carcinoma,
and a testicular tumor (11). Tumor-specific LOH suggests
that this CNC susceptibility gene is a tumor suppressor gene.
Furthermore, the common region of LOH markedly restricted the
chromosomal location of this tumor suppressor gene. The
PPKAR1A gene is contained within this chromosomal region.
The PKA holoenzyme (Fig. 1
) is an inactive tetramer consisting of two
regulatory and two catalytic subunits. When the regulatory subunits
bind cAMP, they dissociate from the catalytic subunits. The free
catalytic subunits phosphorylate serine and threonine residues of
proteins critical to the activation of downstream processes. In
addition to PPKAR1A, the RI-
regulatory subunit, there are three
other PKA regulatory subunits: RI-ß, RII-
, and RII-ß. Each has a
different pattern of tissue expression. Because the PPKAR1A normally
inhibits downstream signaling by PKA, it was considered a likely
candidate for the CNC tumor suppressor gene. Loss of activity of this
regulatory subunit might activate PKA and subsequent downstream
signaling. The end points could include hormone production (cortisol,
GH, and testosterone) and tumorigenesis. Two groups publishing
in September 2000 provide evidence that PPKAR1A is a CNC
susceptibility gene. Mutations with a high likelihood of inactivating
the PPKAR1A protein were identified in a total of nine CNC families,
and in a single sporadic subject with CNC (11, 12).
Furthermore, the cAMP-dependent PKA activity in CNC steroid-producing
neoplasms increased to a greater degree following cAMP stimulation than
the PKA activity from similar sporadic neoplasms (11).
Interestingly, the basal kinase activity was equal in CNC and control
neoplasms, suggesting that other regulatory proteins can substitute for
PPKAR1A. A diminished amount of apparently normal PPKAR1A was found by
Western analysis in one myxoma, suggesting that deletion is not a
necessary requirement for myxoma formation (12). In
summary, the PPKAR1A gene mutations cause CNC in some
families, and this disorder is now referred to as CNC1. The gene on
chromosome 2p that is mutated in CNC2 remains unknown, for now.
With clarification of the CNC1 susceptibility gene, there is now a molecular gold standard for the diagnosis of CNC1. This will facilitate a clearer clinical description of CNC1 and rapid determination of the role of acquired PPKAR1A mutations in the development of sporadic CNC type neoplasms. The different endocrine abnormalities may have different frequencies in CNC1 as compared with CNC2. We already know that PPNAD and GH-producing pituitary tumors can occur in CNC1, because these tumors demonstrated LOH at the PPKAR1A chromosomal locus and the individuals carrying these tumors had mutations of this gene (11). It is possible, and even likely, that phenotypic diversity exists within a given syndrome as occurs in other familial tumor syndromes, such as the multiple endocrine neoplasia type 2 syndromes. No doubt these same authors will soon let us know if acquired PPKAR1A mutations occur commonly in sporadic cardiac myxomas, spotty skin lesions, GH-producing pituitary adenomas, testicular tumors, and cortisol producing adrenal adenomas.
Other intriguing questions may not be answered as quickly. Will
identification of the CNC1 susceptibility gene facilitate the
identification of the CNC2 susceptibility gene? Will there be more than
one susceptibility gene or phenotype for CNC2? Can identification of
the CNC1 gene help us to predict the clinical phenotype other disorders
caused by abnormalities of the same signal transduction pathways? It is
attractive to speculate that another PKA regulatory subunit causes
CNC2. However, a quick review of available databases suggests that this
is not the case. The CNC2 gene is on chromosome 2p, whereas the genes
for the other known regulatory subunitsRI-ß, RII-, and
RII-ßare on chromosomes 7p, 3p, and 7q, respectively. In addition,
tumor-specific LOH has not been a consistent feature of CNC2
(7), as might be expected if a regulatory subunit
abnormality were the cause of this disorder. Therefore, it seems
unlikely other known PKA regulatory subunits are the CNC2
susceptibility gene. Because tumor specific gene amplification may
occur in CNC2, it is attractive to postulate that a PKA catalytic
subunit is the CNC2 gene. However, the
, ß, and
catalytic PKA
subunits have chromosomal locations at 19p, 1p, and 9q, respectively.
Therefore, they cannot cause CNC2, which has been mapped to chromosome
2p. Identification of the CNC2 susceptibility gene may prove difficult.
However, it is likely to be worth the effort, because new information
concerning tumorigenic mechanisms in man is likely to be uncovered.
In summary, careful clinical evaluation has been combined with genetic techniques to uncover new and critical information concerning tumorigenic mechanisms in CNC. Conversely, the identification PPKAR1A as the CNC1 susceptibility gene will rapidly elucidate the full clinical spectrum of CNC1 by affording a diagnostic gold standard and generating novel hypotheses concerning phenotypic expression. Subsequently, the lessons learned from investigations into CNC1 may be applicable to other familial syndromes or to sporadic neoplasms of the type found in CNC1.
Received September 19, 2000.
Accepted September 19, 2000.
![]() |
References |
---|
![]() ![]() ![]() |
---|
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |