Steroid 17α-Hydroxylase Deficiency—Not Rare Everywhere

The congenital adrenal hyperplasias (CAH) have fascinated endocrinologists for generations. The use of classical chemistry in the form of urinary steroid assays delineated “blocks” in single enzymatic steps, leading to the successful treatment of 21-hydroxylase deficiency with cortisol in 1950 (1, 2). The union of chemistry, physiology, and clinical insight soon explained most forms of CAH. A “blocked” enzyme led to decreased secretion of downstream steroids (cortisol and aldosterone in 21-hydroxylase deficiency) and accumulation of upstream steroids (such as precursors of androgens) that accounted for the complex and seemingly unrelated clinical findings of salt loss and genital virilization. By the early 1960s, Bongiovanni and Kellenbenz (3) used the same conceptual and experimental framework to characterize 3β-hydroxysteroid dehydrogenase deficiency, and in 1966 Biglieri et al. (4) similarly described the first patients with 17α-hydroxylase deficiency (17OHD), in which patients had sexual infantilism (from lack of sex steroid precursors) and apparent mineralcorticoid excess with salt retention and hypertension. Through patient, incisive investigation, Biglieri deduced that the responsible mineralcorticoid was deoxycorticosterone produced in the zona fasciculata under stimulation by ACTH and that this deoxycorticosterone suppressed the renin-angiotensin system and the production of aldosterone (4). The approach of looking for accumulated precursors and deficient products was successful in many other disorders of steroidogenesis, although it was a failure in inferring that congenital lipoid adrenal hyperplasia was due to a disorder in a hypothetical enzyme complex termed 20, 22 desmolase [it is due to mutations in a cholesterol transport protein termed StAR (5, 6), and the hypothetical 20, 22 desmolase complex is really mitochondrial P450scc (7)]. Because the CAHs were genetic disorders of steroidogenic enzymes, the lesions causing these disorders could not be determined until the gene(s) for each enzyme were cloned. The cDNAs and genes for most of the human steroidogenic enzymes were cloned in the mid 1980s (7). The cDNA and gene for human P450c17 were reported in 1987 (8, 9), and the first mutations in this gene in patients with classic 17OHD were reported the following year (10). Genetic analysis, primarily sequencing of the affected genes, showed that the various forms of 21-hydroxylase deficiency were ordinary allelic variants showing remarkably good phenotype/genotype correlations. Similarly, it became clear that the high incidence of certain forms of CAH in certain areas were due to genetic founder effects and not due to any selective advantage of heterozygous carriers, as is the case with the antimalarial protection afforded to heterozygous carriers of hemoglobinopathies such as sickle cell anemia and hemoglobin C disease. A founder effect occurs when one or two founder individuals, usually unaffected heterozygotes, become relatively isolated genetically and breed with a very limited population. As a result, the carrier rate (the percentage of the population that has the mutation in a benign, heterozygous state) increases. The incidence of affected homozygotes will also increase, but these will carry the same mutations introduced to the population by the founders. Founder effects are well-known in genetics. For example, vitamin D 1α-hydroxylase deficiency (formerly called vitamin D-dependent rickets type 1 or pseudo vitamin D-deficiency rickets) is a very rare disorder everywhere in the world except in the Canadian province of Quebec. Population, genetic linkage, and ethnographic studies traced the founder to 17th century French immigrants who settled in the Charlevoix lac St. Jean region, among whom the carrier rate is now about 1 in 26 (11). Not surprisingly, genetic analysis showed that all of these individuals carried the same ΔG958 frameshift mutation (12). Among the CAHs, founder effects are well-established for such diverse disorders as 21-hydroxylase deficiency among the Yupic-speaking Eskimos of Alaska (13), 11-hydroxylase deficiency among Sephartic Jews from Morocco (14), congenital lipoid adrenal hyperplasia in Japan and Korea (6), and corticosterone methyl oxidase deficiency among Jews of Iranian ancestry, mostly from the city of Isfahan (15). Thus, although 21-hydroxylase deficiency is common because of the unusual genetics and high rate of genetic recombination in the human leukocyte antigen locus in which its gene(s) resides (16), it may not be possible to state what the second most common form of CAH is without stating the population being considered. Studies in the United States and Western Europe indicate that 11-hydroxylase deficiency is the second most common form, probably because of its high incidence in populations of European ancestry (17), but if one asks what is the second most common form of CAH in Japan or Korea, the answer is clearly congenital lipoid adrenal hyperplasia (18). In this issue of the JCEM, Costa-Santos et al. (19) provide remarkable new findings about 17OHD, which is quite rare among most populations but is common among the Mennonite descendants of Dutch Frieslanders (20), a distinctly limited population, and now, among the whole of the Brazilian population. Brazil is a vast and ethnically diverse nation that includes a mix of indigenous peoples, Afro-Brazilians and Euro-Brazilians, largely of Portuguese and Spanish ancestry. Working at the Federal University of Sao Paulo, Claudio Kater first noted the surprisingly large number of Brazilians with 17OHD. Having trained in endocrinology in San Francisco with Dr. Ed Biglieri, who first described this disease (4), Kater was uniquely qualified to recognize the disorder and rigorously characterize it hormonally. Over a period of years, he and his colleagues amassed 30 patients with 17OHD. Unlike metaanalyses of multiple individual reports, Kater’s group was able to investigate a large number of individuals with consistent technologies and assays. The ethnic diversity of Brazil might have suggested that these individuals would carry a wealth of new mutations in their CYP17 genes, but when Marivania Costa-Santos from Kater’s group sequenced their genes in Rich Auchus’ laboratory, they found only seven mutations among 28 apparently unrelated alleles in 19 families. When the patients were categorized ethnographically, it was apparent that the W406R mutation was found among Southern Brazilians of Spanish ancestry, whereas the R362C mutation was found among Northern Brazilians of Portuguese ancestry. Although far more Brazilians are of Portuguese than of Spanish ancestry, the Spanish W406R mutation was more common, suggesting that Spanish Brazilians tended to marry within their national group, whereas Portuguese Brazilians were more likely to intermingle. Although both mutations were found in many patients, it is of note that neither of these mutations has been described in Spain or Portugal. This would be consistent with single founders from each country arriving in Brazil many generations ago. Mutation analysis is getting easier due to advances in PCR technology. A second paper from the same group (21) describes a splice-site error in another Brazilian patient with 17OHD. When one finds a mutation, one is obligated to show that that mutation is sufficient to cause disease (the genetic equivalent of Koch’s postulates). For missense (amino acid replacement) mutations, this involves site-directed mutagenesis, expression in a suitable cell system, and assay of the function of the mutant protein. Costa-Santos et al. (19) provide a nice example of how this is done in their first paper. However, mutations that alter the splicing of nuclear precursors to mRNA can be more difficult to characterize. The standard approach for the functional characterization of intronic mutations consists of three steps. First, one builds a minigene for the wild-type and the mutant. The minigene generally has only one intron (the one in which the mutation is found); the rest of the minigene is a cDNA construct. Second, the minigene is expressed in transfected cells. Because the transcription machinery is in the nucleus, only those minigenes that reach the nucleus are transcribed into RNA. The resulting RNA, with its one intron, will be spliced incorrectly if the mutation in the intron is functionally significant. Third, the results of this transcription and splicing are analyzed by RT-PCR and gel electrophoresis. The disadvantage of this approach is that it will not detect splicing forms present in low abundance, including wild-type sequences correctly spliced from the mutant construct that may result in clinically significant partial activity. The only way around this problem has been to perform tedious RNase protection experiments from both the minigene and affected tissue. An example of this concerns the mutation of a single thymidine, 11 bases from the splice acceptor site in intron 4 of the StAR gene (22). Costa-Santos et al. (21) circumvented this problem by expressing the whole gene in an exon-trapping vector, permitting splicing of all the intron/exon boundaries. Instead of assaying the resulting RNA by RT-PCR, they directly assayed the enzymatic activity of the encoded protein product, thus achieving the sensitivity of an RNase protection assay with little more work than a standard minigene experiment. This approach will be limited to smaller genes amenable to amplification in only a few pieces, and to genes encoding readily assayed products. Nevertheless, it is likely that others will adapt this procedure to many systems in the future. Rapid, efficient identification of novel mutations in P450c17, and especially the detailed elucidation of their structural, enzymatic, and clinical consequences, is becoming increasingly important. Mutations that have a major impact on the 17α-hydroxylase activity cause classic 17OHD, the first hypertensive disorder of steroidogenesis identified. The roles of various C-21 17-deoxysteroids in idiopathic hypertension remain unresolved; further studies of P450c17 may illuminate this important area. Mutations that selectively ablate the 17,20 lyase activity, causing isolated 17, 20 lyase deficiency are exceedingly rare (23, 24) and may be the rarest of all disorders of steroidogenesis. Yet analysis of these rare mutations helped to prove that the regulation of 17, 20 lyase activity, and hence of all androgen production, is regulated posttranslationally at the level of electron transfer to P450c17 (25, 26). The posttranslational regulation of the 17,20 lyase activity of P450c17 appears to be the key event in the regulation of the onset of adrenarche and the hyperandrogenism of polycystic ovary syndrome (27). Thus, major advances in all areas of knowledge concerning P450c17 are most welcome and are likely to have impacts far beyond the confines of classic 17OHD. Congenital adrenal hyperplasia; 17α-hydroxylase deficiency.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.