Steroid biomarkers and genetic studies reveal inactivating mutations in hexose-6-phosphate dehydrogenase in patients with cortisone reductase deficiency

CONTEXT

Cortisone reductase deficiency (CRD) is characterized by a failure to regenerate cortisol from cortisone via 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1), resulting in increased cortisol clearance, activation of the hypothalamic-pituitary-axis (HPA) and ACTH-mediated adrenal androgen excess. 11beta-HSD1 oxoreductase activity requires the reduced nicotinamide adenine dinucleotide phosphate-generating enzyme hexose-6-phosphate dehydrogenase (H6PDH) within the endoplasmic reticulum. CRD manifests with hyperandrogenism resulting in hirsutism, oligo-amenorrhea, and infertility in females and premature pseudopuberty in males. Recent association studies have failed to corroborate findings that polymorphisms in the genes encoding H6PDH (R453Q) and 11beta-HSD1 (Intron 3 inserted adenine) interact to cause CRD.

OBJECTIVE

Our objective was to reevaluate the genetics and steroid biochemistry of patients with CRD.

DESIGN

We analyzed 24-h urine collection for steroid biomarkers by gas chromatography/mass spectrometry and sequenced the HSD11B1 and H6PD genes in our CRD cohort.

PATIENTS

Patients included four cases presenting with hyperandrogenism and biochemical features clearly indicative of CRD.

RESULTS

Gas chromatography/mass spectrometry identified steroid biomarkers that correlated with CRD in each case. Three cases were identified as homozygous (R109AfsX3, Y316X, and G359D) and one case identified as compound heterozygous (c.960G-->A and D620fsX3) for mutations in H6PD. No mutations affecting enzyme activity were identified in the HSD11B1 gene. Expression and activity assays demonstrate loss of function for all reported H6PDH mutations.

CONCLUSIONS

CRD is caused by inactivating mutations in the H6PD gene, rendering the 11beta-HSD1 enzyme unable to operate as an oxoreductase, preventing local glucocorticoid regeneration. These data highlight the importance of the redox control of cortisol metabolism and the 11beta-HSD1-H6PDH pathway in regulating hypothalamic-pituitary-adrenal axis activity.

Lack of Association of the 11beta-hydroxysteroid dehydrogenase type 1 gene 83,557insA and hexose-6-phosphate dehydrogenase gene R453Q polymorphisms with body composition, adrenal androgen production, blood pressure, glucose metabolism, and dementia

CONTEXT

Recently, it was proposed that a combination of the 83,557insA polymorphism in the 11beta-hydroxysteroid dehydrogenase type 1 (HSD11B1) gene and the R453Q polymorphism in the hexose-6-phosphate dehydrogenase (H6PD) gene interacts to cause cortisone reductase deficiency (CRD) when at least three alleles are affected.

OBJECTIVE

The aim was to study the separate and combined effects of these polymorphisms on body composition, adrenal androgen production, blood pressure, glucose metabolism, and the incidence of dementia in the healthy elderly population.

DESIGN/SETTING/PARTICIPANTS

The Rotterdam study (n = 6105) and the Frail Old Men study (n = 347) are population-based cohort studies in the elderly.

MAIN OUTCOME MEASURES

Genotype distributions and influences of (combined) genotypes on body mass index, adrenal androgen production, waist to hip ratio, systolic and diastolic blood pressure, fasting glucose levels, glucose tolerance test, and incidence of dementia were measured.

RESULTS

No influence of the HSD11B1 83,557insA (allele frequencies 22.0 and 21.5%) and H6PD R453Q (allele frequencies 22.9 and 20.2%) variants was found for the different outcome measures that were investigated, either separately or when at least three alleles were affected.

CONCLUSIONS

Two population-based studies among Caucasian elderly showed no evidence for (combined) effects of two polymorphisms in the HSD11B1 and H6PD genes on body composition, adrenal androgen production, blood pressure, glucose metabolism, and incidence of dementia. Moreover, the high frequencies observed for these two polymorphisms do not correspond to the low incidence of CRD observed in the general population. Altogether, it is unlikely that these polymorphisms cause CRD.

Variants implicated in cortisone reductase deficiency do not contribute to susceptibility to common forms of polycystic ovary syndrome

OBJECTIVE

There are close phenotypic similarities between cortisone reductase deficiency (CRD), a rare abnormality of cortisone metabolism, and polycystic ovary syndrome (PCOS). As there is evidence that CRD results from digenic mutations involving the genes encoding 11beta-hydroxysteroid dehydrogenase type 1 (HSD11B1) and hexose-6-phosphate dehydrogenase (H6PD), we sought to establish whether CRD-associated variants in these genes, individually or in combination, influence susceptibility to PCOS.

DESIGN

Case-control, family-based association and quantitative-trait analyses.

PATIENTS

A UK case sample comprising 256 nuclear families ascertained from a PCOS offspring and 213 singleton PCOS cases plus 549 control subjects.

MEASUREMENTS

All subjects were genotyped for CRD-related variants in HSD11B1 (rs12086634) and H6PD (rs6688832). Testosterone was measured with an in-house radioimmunoassay using ether extraction and dextran-coated charcoal separation.

RESULTS

Case-control analyses revealed no differences in genotype distribution between PCOS and controls for rs12086634 or rs6688832 (both P = 0.84). Three per cent of cases and 2.4% of controls had genotype combinations (three or more variant alleles at the two sites) considered characteristic of CRD (P = 0.73). There were no departures from expectation in the family-based association studies, and no significant associations between genotypes (individually or in combination) and BMI, WHR or testosterone.

CONCLUSIONS

The variants in HSD11B1 and H6PD typed, though implicated in causation of CRD, do not influence susceptibility to PCOS. It seems likely that additional variants within these genes are required for the development of CRD.

Genotypes at 11beta-hydroxysteroid dehydrogenase type 11B1 and hexose-6-phosphate dehydrogenase loci are not risk factors for apparent cortisone reductase deficiency in a large population-based sample

CONTEXT

Apparent cortisone reductase deficiency (ACRD) is a rarely ascertained condition characterized by signs of androgen excess in women or children and decreased urinary excretion of cortisol metabolites compared with cortisone metabolites. These findings suggest a deficiency of 11beta-hydroxysteroid dehydrogenase type 1 (11-HSD1; encoded by the HSD11B1 gene), which normally converts cortisone to cortisol. Common polymorphisms in both HSD11B1 and the hexose-6-phosphate dehydrogenase (H6PD) gene encoding hexose-6-phosphate dehydrogenase have been found together in ACRD patients, who carry three of a possible four minor alleles at the two loci.

OBJECTIVE

The objective of this study was to confirm the postulated digenic inheritance mechanism for ACRD.

DESIGN

This was a population-based association study (Dallas Heart Study). Subjects were genotyped for the 1971T>G polymorphism in intron 3 of HSD11B1 and the R453Q polymorphism in H6PD.

SUBJECTS

The study comprised 3551 individuals in a population-based sample (50% black, 35% white, and 15% Hispanic).

MAIN OUTCOME MEASURE

The main outcome measure was association between genotypes and risk for polycystic ovarian syndrome.

RESULTS

Both polymorphisms occurred more frequently than previously reported. Thus, ACRD genotypes (at least three of four minor alleles) occurred in 7.0% of subjects. There were no associations between genotype and body mass index; waist/hip ratio; visceral adiposity; measures of insulin sensitivity; levels of testosterone, FSH, or LH (in females); or risk of polycystic ovarian syndrome. There was no genotype effect on urinary free cortisol/cortisone or corticosteroid metabolite ratios, which were measured in 10 subjects, each carrying zero, three, or four minor alleles.

CONCLUSIONS

Previously reported associations of ACRD with HSD11B1 and H6PD alleles represent ascertainment bias. However, rare severe mutations in these genes cannot be ruled out.

A study of the hexose-6-phosphate dehydrogenase gene R453Q and 11beta-hydroxysteroid dehydrogenase type 1 gene 83557insA polymorphisms in the polycystic ovary syndrome

CONTEXT

The R453Q variant in the hexose-6-phosphate dehydrogenase gene (H6PD) and 83557insA mutations in 11beta-hydroxysteroid dehydrogenase (11betaHSD) type 1 gene (HSD11B1) interact, resulting in cortisone reductase deficiency (CRD), a rare disorder characterized by a polycystic ovary syndrome (PCOS)-like phenotype.

OBJECTIVE

The objective was to study these mutations in PCOS.

DESIGN

The design was a case-control study.

SETTING

The study was conducted in an academic hospital.

PARTICIPANTS

A total of 116 PCOS patients and 76 nonhyperandrogenic controls participated.

MAIN OUTCOME MEASURES

Genotype distributions and influence of genotypes on clinical and biochemical variables and, in 28 patients and 12 controls, estimates of 11betaHSD oxoreductase activity were the main outcome measures.

RESULTS

Four controls and five patients presented three of four mutant alleles in H6PD R453Q and HSD11B1 83557insA, which is the genotype observed in some subjects with CRD. Estimates of 11betaHSD oxoreductase activity were measured in six of these nine women, ruling out CRD. Moreover, H6PD R453Q and HSD11B1 83557insA genotypes, either separately or in combination, did not influence 11betaHSD oxoreductase activity. The distribution of H6PD R453Q genotypes (R/R, R/Q, and Q/Q) was different in patients and controls (42% of controls and 63% of PCOS patients were R/R; 53% of controls and 31% of PCOS patients were R/Q; and 5% of controls and 6% of PCOS patients were Q/Q; chi(2) = 9.1; P = 0.011). Patients homozygous for R453 alleles presented with increased cortisol and 17-hydroxyprogesterone levels, compared with carriers of Q453 alleles, but these differences were not observed in controls. On the contrary, HSD11B1 83557insA genotypes were not associated with PCOS and did not influence any phenotypic variable.

CONCLUSIONS

Digenic triallelic genotypes of the H6PD R453Q variant and HSD11B1 83557insA mutation do not always cause CRD. On the contrary, the H6PD R453Q variant is associated with PCOS and might influence its phenotype by influencing adrenal activity.

11beta-hydroxysteroid dehydrogenase type 1: a tissue-specific regulator of glucocorticoid response

11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) interconverts inactive cortisone and active cortisol. Although bidirectional, in vivo it is believed to function as a reductase generating active glucocorticoid at a prereceptor level, enhancing glucocorticoid receptor activation. In this review, we discuss both the genetic and enzymatic characterization of 11beta-HSD1, as well as describing its role in physiology and pathology in a tissue-specific manner. The molecular basis of cortisone reductase deficiency, the putative "11beta-HSD1 knockout state" in humans, has been defined and is caused by intronic mutations in HSD11B1 that decrease gene transcription together with mutations in hexose-6-phosphate dehydrogenase, an endoluminal enzyme that provides reduced nicotinamide-adenine dinucleotide phosphate as cofactor to 11beta-HSD1 to permit reductase activity. We speculate that hexose-6-phosphate dehydrogenase activity and therefore reduced nicotinamide-adenine dinucleotide phosphate supply may be crucial in determining the directionality of 11beta-HSD1 activity. Therapeutic inhibition of 11beta-HSD1 reductase activity in patients with obesity and the metabolic syndrome, as well as in glaucoma and osteoporosis, remains an exciting prospect.

Cortisol metabolism and visceral obesity: role of 11beta-hydroxysteroid dehydrogenase type I enzyme and reduced co-factor NADPH

Several factors including genetic and environmental play a role in the development of obesity and the metabolic syndrome. The transgenic mouse overexpressing 11beta-hydroxysteroid dehydrogenase (11beta-HSD) develops visceral obesity. However, it remains unclear how a ubiquitously expressed 11beta-HSD1 enzyme affects adipose tissue so much that it would lead to obesity. In this commentary we explore the possibility that increased intracellular availability of reduced co-factor, NADPH, could exacerbate the enzymatic activity.

11beta-hydroxysteroid dehydrogenase type 1 deficiency ('apparent cortisone reductase deficiency') in a 6-year-old boy

OBJECTIVE

We present the 1st case of prepubertal hyperandrogenism because of a defect in the conversion of cortisone (E) to cortisol (F) by hepatic 11beta-hydroxysteroid dehydrogenase type 1.

METHODS AND RESULTS

Clinical and anthropometric data were obtained. Serum androgens and gonadotropins with luteinizing hormone releasing hormone stimulation test, dexamethasone suppression test, and corticotropin-releasing hormone stimulation test were evaluated. Adrenal imaging and urinary steroid profiling by gas chromatography/mass spectrometry were employed. A 6.9-year-old boy presented with precocious pubarche, height (+2.6 SD), accelerated bone age (11.5 years), and Tanner stage 2 pubic hair and genitalia. Serum androgen levels were elevated and dexamethasone suppressible. Serum F was normal, but the E concentration was increased. Central precocious puberty and congenital adrenal hyperplasia were excluded. The excretion of androgen metabolites was moderately increased, but a highly increased tetrahydrocortisone (THE) and a diminished tetrahydrocortisol (THF + allo-THF) excretion was found with a [THF + allo-THF/ THE] ratio of 0.032 (normal controls 1.05 +/- 0.17). The corticotropin-releasing hormone stimulation test showed an exaggerated adrenocorticotropic hormone response, suggesting a relative deficiency of F. Two months of hydrocortisone treatment (17.5 mg daily) failed to suppress androgens adequately. Treatment with dexamethasone (0.375 mg/daily) resulted in androgen suppression.

CONCLUSIONS

In the case of precocious pubarche and accelerated growth, the diagnosis of 11beta-hydroxysteroid dehydrogenase type 1 deficiency ('apparent cortisone reductase deficiency') should be considered. The diagnosis is based on determinations of urinary steroid metabolites.

Mutations in the genes encoding 11beta-hydroxysteroid dehydrogenase type 1 and hexose-6-phosphate dehydrogenase interact to cause cortisone reductase deficiency

In cortisone reductase deficiency (CRD), activation of cortisone to cortisol does not occur, resulting in adrenocorticotropin-mediated androgen excess and a phenotype resembling polycystic ovary syndrome (PCOS; refs. 1,2). This suggests a defect in the gene HSD11B1 encoding 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1), a primary regulator of tissue-specific glucocorticoid bioavailability. We identified intronic mutations in HSD11B1 that resulted in reduced gene transcription in three individuals with CRD. In vivo, 11beta-HSD1 catalyzes the reduction of cortisone to cortisol whereas purified enzyme acts as a dehydrogenase converting cortisol to cortisone. Oxo-reductase activity can be regained using a NADPH-regeneration system and the cytosolic enzyme glucose-6-phosphate dehydrogenase. But the catalytic domain of 11beta-HSD1 faces into the lumen of the endoplasmic reticulum (ER; ref. 6). We hypothesized that endolumenal hexose-6-phosphate dehydrogenase (H6PDH) regenerates NADPH in the ER, thereby influencing directionality of 11beta-HSD1 activity. Mutations in exon 5 of H6PD in individuals with CRD attenuated or abolished H6PDH activity. These individuals have mutations in both HSD11B1 and H6PD in a triallelic digenic model of inheritance, resulting in low 11beta-HSD1 expression and ER NADPH generation with loss of 11beta-HSD1 oxo-reductase activity. CRD defines a new ER-specific redox potential and establishes H6PDH as a potential factor in the pathogenesis of PCOS.

The functional consequences of 11beta-hydroxysteroid dehydrogenase expression in adipose tissue

Clinical observations have highlighted the link between glucocorticoids and obesity. While exogenous glucocorticoids in excess predispose to the development of central obesity, we have focused on cortisol metabolism within human adipose tissue. 11beta-hydroxysteroid dehydrogenase (11beta-HSD) inter-converts the active glucocorticoid, cortisol, and inactive cortisone. 11beta-HSD1, the only isoform expressed in adipose tissue, acts predominantly as an oxoreductase to generate cortisol. Expression is higher in omental compared to subcutaneous preadipocytes and activity and expression are potently regulated by growth factors and cytokines. Mice over-expressing 11beta-HSD1 specifically within adipocytes develop central obesity. However, the situation is less clear in humans. Globally, there appears to be inhibition of the enzyme, but expression in human obesity is still not fully characterized; its functional role in adipocyte biology remains to be elucidated. In vitro, 11beta-HSD1 appears to function in promoting adipocyte differentiation and limiting preadipocyte proliferation, but the impact of these effects in vivo upon the regulation of fat mass remains to be defined. Clinical studies utilizing selective 11beta-HSD1 inhibitors may help to answer this question.

Apparent cortisone reductase deficiency: a rare cause of hyperandrogenemia and hypercortisolism

A 55-year-old woman presented with androgenetic alopecia which had started at age 40. Her clinical history revealed that, unlike her younger sister, she was unable to conceive and was diagnosed as being sterile at age 30. At age 45, 21-hydroxylase deficiency (late-onset CAH) was assumed and glucocorticoid treatment suggested, but not initiated. There was slight hirsutism, but no other sign of virilization. Retesting of plasma steroids revealed elevated 17-OH-progesterone and free testosterone. Treatment with prednisone, cyproterone acetate, and spironolactone was started with significant clinical success. Surprisingly, the analyses of urinary steroid metabolites revealed a pattern that did not support the diagnosis of 21-hydroxylase deficiency (pregnanetriolone absent, pregnanediol, 17-OH-pregnanolone and pregnanetriol not increased). Abdominal CT showed bilateral adrenal hyperplasia and masses in both ovaries. Bilateral adnexectomy was performed, and cystic teratomas were diagnosed. Postoperative urinary steroid analyses showed a decreased tetrahydrocortisol/tetrahydrocortisone ratio (values around 0.08 as compared to age- and sex-matched controls with a ratio of about 0.5-0.8). Plasma cortisol appeared to be repeatedly elevated with exogenous sources excluded. Mass spectrometry showed that, while the tetrahydro metabolites were mainly cortisone-derived, the metabolites not reduced in A ring were mostly cortisol derivatives. This constellation clearly indicates cortisone reductase deficiency, a defect of hepatic 11 beta-hydroxysteroid dehydrogenase (11 beta-HSD1). This enzyme catalyzes the oxidation of cortisol to cortisone and the reduction of cortisone to cortisol. In contrast to the corresponding kidney enzyme (11 beta-HSD2), its primary activity is, however, reductive. Although this is only the fifth reported case of that defect, more attention should be paid to this condition in hyperandrogenic women, even if elevated 17-OH-progesterone and testosterone suggest a more frequent cause.

Apparent cortisone reductase deficiency: a functional defect in 11beta-hydroxysteroid dehydrogenase type 1

A 36-yr-old woman was referred to the endocrine clinic for investigation of oligomenorrhea, hirsutism, and acne. She was plethoric and overweight with central fat distribution. Plasma cortisol was normal, but her adrenal glands were enlarged (CT scan). Urinary tetrahydrocortisone excretion rate was consistently high, raising the possibility of 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) deficiency. In addition, 5beta- reduction of cortisol and cortisone was markedly enhanced. The levels of all cortisol metabolites were suppressed normally with dexamethasone, but conversion of oral cortisone acetate to plasma cortisol was delayed and subnormal compared with that of healthy volunteers. This was accompanied by a larger than normal increase in plasma cortisone concentration. Thus, the defect appears to be in 11beta-HSD1 activity and not in 5beta-reductase activity. Three close relatives of the subject showed no comparable abnormalities, and analysis of the coding region and exon/intron boundaries of the 11beta-HSD1 gene of the case revealed no differences from the consensus sequence. The defect may lie outside the coding region. Alternatively, some other inherited or acquired defect may lead to inhibition of this enzyme system.

[Silent adrenal gland tumors in patients with adrenogenital syndrome]

Adrenal tumors accidently detected by CT scan are increasingly seen in patients without clinical signs of adrenal diseases. We studied whether enhanced adrenal stimulation is of importance in the development of adrenal tumors. For this purpose 22 patients with adrenogenital syndrome (AGS) were studied by CT scan. One of these patients suffered from C-11 beta-hydroxylase-, one from C-3 beta-hydroxy steroid dehydrogenase-, and 20 from C-21-hydroxylase deficiency. The average adrenal size of these patients was 506 +/- 79 mm2 as compared to 132 +/- 8 mm2 in the controls (P less than 0.001). Only two patients with the late onset form revealed adrenal glands of normal size. There was a significant correlation between adrenal size and patients' age (P less than 0.01). Females with the simple virilizing form revealed adrenal glands larger than those of the late onset form (640 +/- 169 vs 308 +/- 56 mm2). Eighteen patients with AGS exhibited one (n = 11) or several (n = 7) adrenal tumors, the size of which was 5-9 mm in diameter in 9, 10-20 mm in 7, and more than 50 mm in 2 patients. There was a significant correlation between adrenal hyperplasia and tumor diameter (P less than 0.001). No correlation was found between tumor size and plasma concentrations of testosterone or 17-hydroxyprogesterone, patients' age at the time of diagnosis, or clinical signs of androgenization. Again, tumors were larger in females suffering from the simple virilizing form of AGS than in those with the late onset form (14.8 +/- 5.5 vs 7.7 +/- 0.8 mm).(ABSTRACT TRUNCATED AT 250 WORDS)