Most cases of CNC are caused by inactivating mutations in the gene encoding one of the subunits of the protein kinase A (PKA) tetrameric enzyme, namely regulatory subunit type 1 α (PRKAR1A), located at 17q22– 24. Although a second locus (2p16) has been implicated, sequencing of the region in the linked families did not reveal alterations in known coding sequences.
PRKAR1A extends to a total genomic length of approximately 21 kb and consists of 11 exons, encoding a total of 381 amino acids, with a dimerization/ docking domain, and two cAMP binding do mains, A and B. Since the identification of PRKAR1A mutations in CNC, more than 140 disease- causing pathogenic sequence changes have been reported (http:// prkar1a.nichd.nih.gov/ hmdb/ intro. html); they are spread over the entire coding sequence of the gene, without a notable preference for a region or exon. Structurally, the majority of the mutations consists of base substitutions, small deletions, and insertions or combined rearrangements, involving up to 15 bp; although rare, large PRKAR1A deletions have been reported.
Mutations in PRKAR1A are seen in more than 70% of the patients with classical CNC and, in most cases, they lead to complete inactivation of one of the PRKAR1A alleles as a result of premature stop codon generation and subsequent nonsense- mediated mRNA decay (NMD). In its inactive form, PKA is a tetramer composed of two regulatory and two catalytic subunits. The decreased cellular concentration of regulatory subunits results in a balance shift between the formation and the disassembly of the PKA tetramer, towards the release of the catalytic subunits. The free catalytic subunits, which are active serine– threonine kinases, further phosphorylate a series of targets that regulate downstream effectors and transcription of specific genes, mediating cell growth and differentiation. Thus, functionally, the mechanism by which PRKAR1A haploinsufficiency causes CNC is through excess cellular cAMP signalling in affected tissues. CNC lesions frequently show loss- of- heterozygosity, suggesting a tumour suppressor function for PRKAR1A.
Although significantly less frequent, mutations that escape NMD and lead to the expression of an abnormal, defective PRKAR1A protein have been reported [24, 28– 30]. These expressed mutations may lead to a characteristic phenotype that reflects the location and the type of the genetic change. Examples include a germline in- frame deletion of exon 3 which results in severe expression of the majority of the CNC manifestations— a phenotype illustrating the importance of exon 3 in linking the dimerization/ docking and the first cAMP binding domain [24]. In contrast, another in- frame variant— a splice- site deletion that eliminates exon 7— is seen associated mostly with lentiginosis and the adrenal component of CNC, PPNAD. Just as lentiginosis is the most common non- endocrine CNC manifestation, PPNAD is the most frequently observed endocrine tumour of the disease. Thus, the presence of only two features of CNC, the most common ones, with this splice- site variant is consistent with the anticipation of a milder phenotype associated with certain splice mutations, due to their incomplete penetrance at the mRNA level (i.e. not all DNA molecules harbouring the splice variant result in mRNA species lacking exon 7).
Apart from the above- mentioned, expressed mutant PRKAR1A protein isoforms, several other expressed isoforms that result from single amino acid substitutions have been reported. Detailed in vitro analysis of their effects on protein function have revealed important PRKAR1A domain features. The six naturally occurring missense substitutions examined by this study (p.Ser9Asn, p.Arg74Cys, p.Arg146Ser, p.Asp183Tyr, p.Ala213Asp, p.Gly289Trp) are spread over all the functional domains of the protein. Although, as mentioned before, the low number of individuals affected by each of these mutations prevented detailed phenotype– genotype analysis, these studies support the previous suggestion that the alteration of PRKAR1A function alone (and not only its complete loss) is sufficient to increase PKA activity, leading to CNC.
Until recently, no genotype– phenotype correlations had been found for the different stop codon mutations, which are expected to uniformly lead to lack of the PRKAR1A mutant allele’s protein product in cells. This was because most of the mutations were identified in single patients only and only two (c.491– 492delTG/ p.Val164fsX4, and c.709(- 7- 2) del6(TTTTTA)) had been seen in more than three kindreds . The first study to explore all PRKAR1A mutations found to date against all CNC phenotypes was recently completed; 353 individuals, 258 of whom (73%) were positive for a PRKAR1A mutation, were studied. Several features that distinguish PRKAR1A mutation carriers from mutation- negative CNC patients were identified; the former presented more frequently and earlier in life with pigmented skin lesions, myxomas, thyroid, and gonadal tumours. In addition, essential correlations between certain genetic defects and the severity and type of CNC manifestations were found. Bertherat et al. outlined subgroups of patients; the first group presented with isolated PPNAD, in some cases accompanied with lentiginosis. In this group the following tendencies were observed: (1) patients diagnosed before 8 years of age were rarely carriers of PRKAR1A mutations; and (2) most of the patients with isolated PPNAD and the presence of PRKAR1A mutation were carriers of either the c.709(- 7- 2) del6(TTTTTA) mutation (P G/ p.Met1Val substitution affecting the initiation codon of the protein. These observations were in line with previously published reports and both mutations are rather rare. Although the molecular mechanism of the p.Met1Val substitution is not completely clear, it is the only mutation that alters the protein initiation site, and may, in theory, result in alternative initiation; c.709(- 7- 2) del6(TTTTTA) is a splice variant that is expected to result in an exon skip, frame shift, and premature stop codon generation. However, since it does not affect the two immediate nucleotides on either site of the splice junction, it is expected to lead to splicing in less than 100% of the molecules that harbour it, and thus, presumably, to lead to a milder phenotype. The fact that a milder phenotype involves only the adrenal and skin is suggestive of their high sensitivity to changes in PKA activity.
The second group of CNC patients that was suggested to have a particular genotype– phenotype correlation comprised individuals with myxomas (affecting all locations— skin, heart, and breast), PMS, thyroid tumours, and large- cell calcifying Sertoli cell tumours (LCCSCT). In these patients, PRKAR1A mutations were seen substantially more often. Related to this is the recognition that certain tumours present at a significantly younger age in PRKAR1A mutation carriers: cardiac myxomas (P = 0.02), thyroid tumours (P = 0.03), and LCCSCTs (P = 0.04) [11]. Another finding in these patients was that mutations that escaped NMD and led to an alternate, usually shorter, protein were associated with an overall higher total number of CNC manifestations (P = 0.04).
In terms of pigmented skin lesions in CNC, two important correlations have been observed: (1) lentigines (as well as PMS, acromegaly, and cardiac myxomas), were seen significantly more often in CNC patients with exonic PRKAR1A mutations, compared to those with intronic ones (P = 0.04); and (2) lentigines (as well as cardiac myxoma and thyroid tumours) were significantly associated with the hot spot c.491– 492delTG mutation compared to all other PRKAR1A defects (P = 0.03). These data add greatly to the understanding of the molecular mechanisms of the involvement of PRKAR1A in endocrine and other tumourigenesis and, thus, for genetic counselling and prognosis in CNC families.
Interestingly, a 2.3- Mb deletion in chromosome band 17q24.2– q24.3, which involved PRKAR1A together with another 13 genes, resulted in a number of clinical features, including posterior laryngeal cleft, growth restriction, microcephaly, and moderate mental retardation. The only CNC manifestation was numerous freckles and lentigines at a young age; the authors called the observed phenotype ‘CNC plus’.
To date, the molecular causes underlying the formation of pigmented skin lesions in CNC are not fully understood. A possible mechanism involves the PKA- mediated activation of pathways downstream of the melanocortin receptors (MCRs), which form a subfamily of the G protein- coupled receptors (GPCRs) and regulate a wide variety of processes, including skin pigmentation. The melanocortin 1 receptor (MC1R) is expressed preferentially in epidermal melanocytes and is known to be the key regulator of mammalian pigmentation. MC1R is stimulated by the proopiomelanocortin- derived melanocyte- stimulating hormone and adrenocorticotrophic hormone (ACTH) and, in turn, activates the rate- limiting enzyme in melanin synthesis, tyrosinase. As a GPCR, MC1R is positively coupled with adenylate cyclase, and its actions are mainly mediated by PKA, in coordination with other signalling molecules involving protein kinase C (PKC) and MAPKs.
