PresentationEndocrinology

P450 Oxidoreductase Deficiency: A Disorder of Steroidogenesis with Multiple Clinical Manifestations

+ See all authors and affiliations

Sci. Signal.  23 Oct 2012:
Vol. 5, Issue 247, pp. pt11
DOI: 10.1126/scisignal.2003318
A Presentation from the European Society for Paediatric Endocrinology (ESPE) New Inroads to Child Health (NICHe) Conference on Stress Response and Child Health in Heraklion, Crete, Greece, 18 to 20 May 2012.

Abstract

Cytochrome P450 enzymes catalyze the biosynthesis of steroid hormones and metabolize drugs. There are seven human type I P450 enzymes in mitochondria and 50 type II enzymes in endoplasmic reticulum. Type II enzymes, including both drug-metabolizing and some steroidogenic enzymes, require electron donation from a two-flavin protein, P450 oxidoreductase (POR). Although knockout of the POR gene causes embryonic lethality in mice, we discovered human POR deficiency as a disorder of steroidogenesis associated with the Antley-Bixler skeletal malformation syndrome and found mild POR mutations in phenotypically normal adults with infertility. Assay results of mutant forms of POR using the traditional but nonphysiologic assay (reduction of cytochrome c) did not correlate with patient phenotypes; assays based on the 17,20 lyase activity of P450c17 (CYP17) correlated with clinical phenotypes. The POR sequence in 842 normal individuals revealed many polymorphisms; amino acid sequence variant A503V is encoded by ~28% of human alleles. POR A503V has about 60% of wild-type activity in assays with CYP17, CYP2D6, and CYP3A4, but nearly wild-type activity with P450c21, CYP1A2, and CYP2C19. Activity of a particular POR variant with one P450 enzyme will not predict its activity with another P450 enzyme: Each POR-P450 combination must be studied individually. Human POR transcription, initiated from an untranslated exon, is regulated by Smad3/4, thyroid receptors, and the transcription factor AP-2. A promoter polymorphism reduces transcription to 60% in liver cells and to 35% in adrenal cells. POR deficiency is a newly described disorder of steroidogenesis, and POR variants may account for some genetic variation in drug metabolism.

Presentation Notes

Slide 1: Science Signaling logo

The slideshow and notes for this Presentation are provided by Science Signaling (http://www.sciencesignaling.org).

Slide 2: P450 oxidoreductase deficiency: A new disorder of steroidogenesis with multiple clinical manifestations

This talk concerns P450 oxidoreductase, the electron transport protein required by all microsomal forms of cytochrome P450. We describe P450 oxidoreductase deficiency, focusing on its discovery and varied clinical manifestations, and describe the role of P450 oxidoreductase in drug metabolism.

Slide 3: Steroidogenesis: The old way

Steroid hormones are essential for life; hence, their biosynthesis has been studied in detail. Most medical and endocrinology textbooks portray steroid biosynthesis as shown here, emphasizing the chemical structures of the steroids. But if one wishes to understand how steroidogenesis happens and why people can have disorders of steroid biosynthesis, then one must focus on the arrows in the biosynthetic pathway, because each arrow represents a steroidal conversion, each steroidal conversion requires an enzyme, and each enzyme is encoded by a gene that is subject to transcriptional regulation and mutations that can cause disease.

Slide 4: Steroid hormone biosynthesis

Work in my lab and others over the past 30 years has clarified and simplified our understanding of the pathways of steroidogenesis. As shown by color coding, there are fewer enzymes than there are enzymatic reactions. For example, P450c11AS (aldosterone synthase, encoded by CYP11B2), shown in red, catalyzes the three distinct steps, (11-hydroxylation, 18-hydroxylation, and 18-methyl oxidation) needed to convert deoxycorticosterone (DOC) to aldosterone. Similarly, P450c17 (encoded by CYP17, shown in yellow), catalyzes both the 17-hydroxylation of steroids needed to make cortisol and the scission of the C17-20 carbon-carbon bond needed to make the precursors of all sex steroids. In addition, this pathway shows that most steroidogenic reactions are catalyzed by forms of cytochrome P450.

Slide 5: Cytochrome P450 enzymes

The term “cytochrome P450” refers to a group of enzymes that are each about 500 amino acids long and contain a heme group. When reduced with carbon monoxide, they absorb light at 450 nanometers and hence are called “P450” for “pigment 450.” The human genome project has shown that we have 57 genetically encoded P450 enzymes, which sounds like a lot, but fruit flies have 83, mice have 102, and the plant Arabidopsis thaliana has 246. P450s are categorized into two groups: type I and type II. Type I enzymes are found in mitochondria, catalyze essential biosynthetic functions, and receive electrons from nicotinamide adenine dinucleotide phosphate (NADPH) through ferredoxin and ferredoxin reductase. I will not be discussing any type I enzymes. Type II enzymes are found in the endoplasmic reticulum and receive electrons from NADPH through a single flavoprotein called P450 oxidoreductase (POR), which is the topic of this talk. Type II enzymes may be functionally divided into those that metabolize drugs and xenobiotics; those involved in the biosynthesis of essential molecules, such as steroids; and the orphan P450s whose functions are not yet understood.

Slide 6: Type II microsomal P450 enzymes

This diagram shows how microsomal P450s work (1). POR is a butterfly-shaped enzyme with each “wing” containing a flavin group. Electrons from NADPH are taken up by the flavin adenine dinucleotide (FAD) group in one wing, which elicits a conformational change in the protein that brings the two flavins into close opposition. The electrons then hop from the FAD group to the flavin mononucleotide (FMN) group, which elicits another conformational change that returns the POR to its original, open configuration. Recent studies using mass spectrometry confirm that POR undergoes these conformational changes from a compact to an extended form (2). The FMN-containing domain of POR then docks with the redox-partner binding site of the P450 by electrostatic interactions; this permits the electrons to hop to the ferrous iron atom in the heme ring of the P450, which mediates catalysis. All P450 enzymes have essentially the same fold, with a membrane-anchoring domain and a substrate-access channel that leads to the heme group. The entrance to the channel is positioned at the junction between the lipid membrane and the aqueous cytosol, so that both water-soluble and lipid-soluble substrates may enter. Any compound that fits into the substrate-binding pocket is likely to be a substrate.

Slide 7: Type II P450 enzymes

Several type II P450 enzymes are of particular interest to endocrinologists. P450c17 (yellow) catalyzes steroid 17-hydroxylation and 17,20 lyase activity; P450c21 (green) is the steroid 21-hydroxylase that is disordered in the common form of congenital adrenal hyperplasia; P450aro (pink) is the steroid aromatase that converts C19 androgens to C18 estrogens. This color code is used throughout the presentation to relate either to the enzymes, relevant steroid metabolites, or phenotypic findings associated with their disorders. In addition, P450 2R1 is the principal vitamin D 25-hydroxylase, and two other type II P450s participate in cholesterol biosynthesis.

Slide 8: Role of POR in steroidogenesis

This slide illustrates the pathways of steroidogenesis and highlights the steps that require POR.

Slide 9: POR deficiency—beginnings

The story of POR deficiency begins with a report from Ralph Peterson’s laboratory in 1985 describing a 6-month-old genetic male with genital ambiguity, having the features listed on the slide (3). Of note, the patient had been the product of an uneventful pregnancy and no dysmorphic features were reported. His hormonal evaluation showed elevated concentrations of deoxycorticosterone (DOC), corticosterone, pregnenolone, and progesterone, all of which indicate a defect in steroid 17-hydroxylase activity (yellow). But the patient also had high concentrations of 17-hydroxyprogesterone (17OHP) and 21-deoxycortisol, which indicate a defect in steroid 21-hydroxylase activity (green).

Slide 10: Peterson’s conclusion

Peterson knew that these activities were catalyzed by P450c17 and P450c21, because the isolation of these proteins had recently been described, and he proposed that their combined defect was due to an unidentified defect in the insertion of these proteins into the membrane of the endoplasmic reticulum. When I read this paper, we had just shown that P450c17 was encoded on chromosome 10, and it was already known that P450c21 was encoded on chromosome 6. It seemed most unlikely that two unlinked genes could be affected by a single mutation, so I wrote a letter to the New England Journal of Medicine suggesting that Peterson’s patient instead had a previously undescribed disease, POR deficiency (4).

Slide 11: POR knockout mice

We were unable to obtain DNA from this patient, so the idea of POR deficiency in this case went untested. Then in 2002, two different groups showed that knockout of the mouse POR gene caused early embryonic lethality (5, 6). So, POR deficiency seemed to be a beautiful hypothesis shot down by an ugly fact.

Slide 12: Kenji Fujieda’s index patient

It was at about this time that my friend, the late Kenji Fujieda, then Chair of Pediatrics at Asahikawa Medical College north of Sapporo, contacted me about a genetically female baby who had a mildly elevated 17OHP (which suggested 21-hydroxylase deficiency, green) and genital virilization inappropriate for the observed blood concentration of 17OHP (which suggested a problem with P450c17, yellow). In addition, the mother had experienced virilization during the course of the pregnancy (suggesting partial aromatase deficiency, pink). Furthermore, the baby had several skeletal malformations that the dysmorphologists classified as Antley-Bixler syndrome (ABS).

Slide 13: Antley-Bixler syndrome

ABS is a rare multiple malformation syndrome first described in 1975 (7) and characterized by craniosynostosis (premature fusion of the bones of the skull) radio-­humeral (or radio-ulnar) synostosis (fusion of bones in the elbow), femoral bowing, and other bony malformations. About half of these patients also exhibited genital ambiguity (8).

Slide 14: Candidate genes in ABS

Several autosomal dominant syndromes characterized by craniosynostosis (Crouzon, Pfeiffer, Apert, and Jackson-Weiss syndromes) are caused by gain-of-function mutations in the gene encoding fibroblast growth factor receptor 2 (FGFR2), and some patients with ABS who lacked genital ambiguity were also reported to have FGFR2 mutations (9, 10). Other loci, including the genes encoding lanosterol demethylase, P450c17, and P450c21, had also been examined by others, but were unaffected (11).

Slide 15: Patients 1 to 3

By this time, we had obtained DNA from other patients similar to Kenji Fujieda’s index case and who had steroid hormone concentrations that suggested reduced activity of P450c17 and P450c21.

Slide 16: Patient 4

We also obtained DNA from a 23-year-old woman evaluated by Berenice Mendonça in São Paulo, Brazil, for infertility and primary amenorrhea. She had elevated basal progesterone and DOC with low androstenedione, all of which suggested defective P450c17 activity, and she had elevated 17OHP and 21-deoxycortisol, which suggested defective P450c21 activity. This patient was otherwise well and had no skeletal disorder, so we inferred that her genetic lesion must be less severe than in the children with ABS.

Slide 17: P450 oxidoreductase (POR)

Therefore, we sequenced the 15 then-known exons of the human POR gene.

Slide 18: Discovering POR deficiency

Christa Flück and Amit Pandey in my lab and Toshihiro Tajima, who was working with Kenji Fujieda, then found POR mutations in each of these patients (12). Of these, we were especially interested in A287P, because that mutation was homozygous in a severely affected infant. By contrast, the C569Y and V608F mutants, which were found in the minimally affected Brazilian woman, would be predicted to retain some activity. So, to fulfill the genetic equivalent of Koch’s postulates, we needed to show that the mutations we had found could arguably account for the patients’ phenotypes.

Slide 19: Assay of POR activity: Classic assay

The classic assay for POR activity, which was used in its initial isolation and characterization in 1969, is the reduction of cytochrome c. This assay is easy and reproducible, because pure cytochrome c can be purchased and its reduction is readily quantified by absorption at 550 nanometers. However, this is a laboratory convenience, rather than a biologically relevant assay, because POR is microsomal and cytochrome c is found in mitochondria.

Slide 20: Cytochrome c assays with bacterially expressed POR

Nevertheless, we wanted to see the results with this classic assay. To do this assay, you set up an NADPH regeneration system with glucose-6-phosphate and glucose-6-phosphate dehydrogenase, add cytochrome c and bacterially expressed human wild-type (WT) or mutant POR, and assess the reduction of cytochrome c in vitro at 550 nm. One can manipulate the conditions so as to measure oxidation of NADPH as well as reduction of cytochrome c.

Slide 21: Cytochrome c assays

Using this assay system, we found that WT POR readily oxidizes NADPH and reduces cytochrome c and that our POR mutants had reduced activity. However, the A287P mutant found in a homozygously affected infant with severe ABS appeared to have about half of normal activity, whereas the C569Y and V608F mutants found in the mildly affected Brazilian woman appeared to have much less activity (12). Thus, these data did not explain the severity of the phenotypes we had seen in the patients.

Slide 22: POR-P450c17-cytochrome b5 interactions

Therefore, we either had the wrong gene or the wrong assay, so we designed a biologically more relevant assay based on the ability of POR to support catalysis by P450c17. This slide shows a computational model of the interaction of P450c17 (red) with POR (yellow) (13). The interaction of these two proteins is facilitated by the allosteric action of cytochrome b5 (green).

Slide 23: Assay of POR activity: P450c17 assay

The 17,20 lyase activity of P450c17 is the activity most consistently disrupted in patients with POR deficiency, and prior work in our lab had shown that the 17,20 lyase activity of P450c17 is very sensitive to minor perturbations in the interactions of POR with P450c17. Therefore, we used the 17,20 lyase activity of bacterially expressed human P450c17 to assay POR activity in vitro.

Slide 24: P450c17 activity assays

To do this assay, one again sets up an NADPH regeneration system to supply electrons to POR, which then donates them to P450c17. 17-Hydroxylase activity is measured by the conversion of radiolabeled progesterone to 17OHP, and 17,20 lyase activity is measured by the conversion of radiolabeled 17OH-pregnenolone to de­hydroepiandrosterone (DHEA). Each assay point must be determined by quantifying radiolabeled steroids by thin-layer chromatography, which is very tedious and time-consuming.

Slide 25: P450c17 assays

When the POR mutants were tested using the P450c17 assays, we saw that WT POR had robust activity to support both the 17-hydroxylase and 17,20 lyase activities of P450c17, the A287P POR mutant found in a severely affected infant had very little activity, and the two POR mutants found in the mildly affected Brazilian woman had intermediate activity (12). Thus, these assays of mutant POR activity correlated well with the patient phenotypes.

Slide 26: 32 ABS patients from six continents

We obtained DNA samples from 32 individuals from around the world diagnosed with ABS on the basis of the skeletal phenotype, and we sequenced both POR and FGFR2 in these samples. We found that the dominant, gain-of-function FGFR2 mutations and the recessive, loss-of-function POR mutation showed complete genetic segregation (14). Although many of the mutant POR proteins lacked activity that was detectable with either the cytochrome c or P450c17 assays, we noted that all these patients had a missense mutation on at least one allele; hence, we have not seen a patient with two null alleles, which would be analogous to the (lethal) POR knockout mice. Finally, we found that two missense mutations were common: A287P accounted for most mutations in people of European ancestry, and R457H accounted for most mutations in Japan. Using the 17,20 lyase assays and calculating enzyme activity as the maximum rate of the reaction relative to the affinity for substrate (Michaelis constant) (Vmax/Km) compared with WT control set at 100%, we grouped the mutations into those with no measureable activity, those with poor activity, and those with moderate activity.

Slide 27: Locations of POR mutations

To understand the basis of these variations in activity, we wanted to know where each mutation maps on the POR protein. The structure of rat POR, which is 87% identical to human POR, had been determined by x-ray crystallography in 1997 (15); hence, modeling the human protein on the rat structure was trivial. We then mapped each mutated residue onto the model of human POR. The mutations that resulted in no assayable POR activity (red) were all in positions of contact with the FAD, FMN, or NADPH moieties; the mutants with poor activity (blue) mapped a bit further away from these contact sites, and the mutants that had moderate activity (green) all mapped to the surface of the protein (14).

Slide 28: Clinical findings by genotype

With a fairly large number of patients and mutations to survey, we then looked to see if there was any phenotypic feature that distinguishes patients with ABS due to mutations in POR from those who have mutations in FGFR2. We found no differences in the skeletal and dysmorphological features of the two groups. The only features that distinguished the POR patients were the genital ambiguity and abnormal steroid patterns.

Slide 29: POR deficiency phenotypes

Most patients with POR deficiency are term newborn infants with ABS, genital anomalies, and steroid hormone patterns indicating partial deficiencies of P450c17 and P450c21 activities. Others have shown that deficient P450aro activity is seen with the R457H mutation but not with the A287P mutation (16). In addition, both male and female adults have been described with mild POR mutations who have infertility with no associated skeletal anomalies (12, 14).

Slide 30: Why do POR patients have ambiguous genitalia?

POR deficiency is unusual in that it affects genital differentiation in both sexes. The basis of the genital ambiguity in males is easy to understand: Impaired 17,20 lyase activity decreases fetal testicular production of the androgens needed for development of male external genitalia. The partial virilization of females remains under study. In patients carrying the R457H mutation, diminished aromatase activity would certainly result in overproduction of feto-placental androgens, as illustrated by Kenji Fujieda’s index patient (12). In other patients, especially those carrying the A287P substitution, it appears that overproduction of 17OHP diverts steroids into the so-called “back-door” pathway of androgen synthesis, leading to virilization (17). The back-door pathway is an alternative pathway of androgen synthesis that starts with 17OHP and leads to the most potent androgen, dihydrotestosterone, without the intermediacy of testosterone and its classic precursors.

Slide 31: Why do POR patients have skeletal malformations?

The basis of the ABS skeletal malformations in POR deficiency has been perplexing. We initially suggested that POR deficiency, affecting two microsomal P450 enzymes that participate in cholesterol synthesis, inhibited the production of cholesterol and consequently the cholesterol derivitization of hedgehog proteins, which are secreted ligands involved in multiple developmental pathways (18).

Evidence for this idea comes from two recent studies—one showing that tissue-specific knockout of the POR gene in the mouse limb bud induces the synthesis of cholesterol biosynthetic genes (19) and the other showing that knockdown of POR in rat chondrocytes decreased cellular proliferation and differentiation, as well as the expression of Indian hedgehog (20). Whereas these mechanisms certainly may contribute, a study published in 2011 suggests that the most important factor is loss of activity of CYP26B1, a microsomal enzyme that degrades retinoic acid (RA) (21).

Slide 32: Role of CYP26B1

Two consanguineous families were described where offspring were homozygous for one of two CYP26B1 mutations. R363L caused craniosynostosis, radio-humero-ulnar synostosis, bowed femora, and intrauterine death. S146P caused craniosynostosis and other malformations that were clinically diagnosed as ABS. The authors of this study determined the ability of these mutant CYP26B1 proteins to degrade RA by measuring activation of a reporter gene under control of an RA response element in cultured cells expressing transgenes encoding the mutant enzymes. WT CYP26B1 inactivated all the RA, yielding no transcription, whereas the severe R363L mutant had virtually no activity, and the S146P mutant had intermediate activity. RA stimulates the maturation of osteoblasts into osteocytes, and this process is regulated by CYP26B1-mediated degradation of RA. In the absence of CYP26B1—or presumably when its activity is impaired by POR mutations—there is excess RA, which results in excessive, premature osteocyte formation, which lead to the bony fusions of ABS (21).

Slide 33: POR—A candidate for pharmacogenomics

The metabolism of most clinically used drugs is mediated by only eight P450 enzymes in the liver. Genetic variations in these enzymes and in hepatic drug transporters explains some, but not all, genetic variation in drug metabolism. Because POR is required for the activity of all the drug-metabolizing P450 enzymes, we postulated that POR would contribute to the genetic variation in drug metabolism. Therefore, we first wanted to determine the extent of genetic variation in POR in the human population.

Slide 34: The POR gene is highly polymorphic

We sequenced the POR gene in 842 healthy people who identified themselves as belonging to one of four ethnic groups: African-American, Caucasian-American, Chinese-American, or Mexican-American on the basis that all four of their grandparents claimed membership in that group. We found many polymorphisms, but by far the most common was the amino acid sequence variant A503V, which was present on 28% of all alleles, ranging from 19% of African-American alleles to 37% of Chinese-American alleles (22).

Slide 35: Effect of the A503V variant on two steroidogenic P450s

We had previously found that A503V reduced POR activity in our 17-hydroxylase and 17,20 lyase activity assays with human P450c17 (12, 14) but had found no effect on P450c21 activity (23). Thus, we explored its activity with drug-metabolizing P450 enzymes.

Slide 36: Assay two hepatic P450 enzymes—CYP1A2 and CYP2C19

We first tested CYP1A2 and CYP2C19, largely because these enzymes were then commercially available as purified recombinant proteins. The assay used a substrate called EOMCC, which is a fluorescent dye blocked by two groups that can be removed by either CYP1A2 or CYP2C19. The assay is easily set up in a 96-well format, and activity is measured by excitation at one wavelength and emission at another. Many time points and substrate concentrations are assayed, which permits calculation of Vmax and Km.

Slide 37: CYP1A and CYP2C19

Some of the drugs metabolized by CYP1A2 and CYP2C19 are listed below the enzyme names. Our data show that both enzymes have robust activity with WT POR and essentially no activity with the A287P and R457H POR mutants. The A503V mutant has normal activity with CYP2C19 and somewhat diminished activity with CYP1A2. Notably, the mutant Q153R, which was identified as a loss-of-function POR mutation in an ABS patient, had increased activity with both enzymes (24).

Slide 38: Activities of POR missense mutants with hepatic CYP1A2 and CYP2C19

We used this assay, as well as the assays based on cytochrome c and on P450c17, to characterize the activity of the 35 POR missense mutants that had been reported at that time, and calculated Vmax/Km as a percentage of the WT control (24).

Slide 39: Activities of selected POR mutants

The data with six selected mutants illustrate the difficulty in studying the activities of POR: The ability of a mutant POR to support the activity of one P450 enzyme will not predict its ability to support the activity of a different P450 enzyme. For example, A115L has only modest loss of activity in the cytochrome c and P450c17 assays, but no measurable activity with CYP1A2 or CYP2C19. In contrast, Q153R has barely detectable activity in the cytochrome c assays, modest loss of activity in the P450c17 assays, and increased activity with both CYP1A2 and CYP2C19 (24).

Slide 40: Only eight P450 enzymes catalyze ~95% of hepatic drug metabolism

CYP1A2 and CYP2C19 were assayed for convenience, rather than because of their overall importance in drug metabolism. This pie chart shows that by far the most important drug-metabolizing enzymes are CYP3A4 and CYP2D6, which together metabolize about 70% of drugs metabolized by P450 enzymes (25).

Slide 41: CYP3A4 can alter the size of its substrate-binding pocket to metabolize substrates of various sizes

Consideration of CYP3A4 introduces another variable. Crystallography with various substrates bound to human CYP3A4 show that its substrate-binding pocket can expand from about 500 cubic Å to over 2000 cubic Å when binding large substrates (26). Therefore, we characterized the ability of WT, Q153R, A287P, R457H, and A503V POR to support the ability of human CYP3A4 to metabolize four substrates of differents sizes and chemical class.

Slide 42: POR A503V affects metabolism by CYP3A4

A small sample of those data, depicted here, shows that A503V has a modestly decreased capacity to support the metabolism of midazolam and of testosterone (27).

Slide 43: Activity of A503V with different P450 enzymes

In general, A503V has modestly decreased activity to support most, but not all, activities of CYP3A4, which indicates that there is substrate-specific variation in addition to P450-specific variation in the measured activity of POR variants. A503V similarly reduces activity with CYP2D6. However, A503V has at least 50% of WT activity in all assays (28).

Slide 44: The POR gene has a first, untranslated exon 38 kb upstream of the first coding exon

We noted that studies nearly 20 years ago found that the rat POR gene has an untranslated exon far upstream from the first coding exon. Therefore, even though the GenBank entry for human POR showed no such exon, we sought and found a small untranslated exon 38 kb upstream from the nominal exon 1 that contains the translational start site (29). This then permitted us to study the transcriptional regulation of the human POR gene.

Slide 45: Analysis of the POR promoter

To locate transcriptional regulatory elements, we first compared the sequences upstream from the transcriptional start site in seven species and also used programs to predict transcriptional regulatory regions. We then used the computationally predicted regulatory regions to build eight promoter-reporter constructs containing from 100 to 3200 base pairs (bp) of POR upstream DNA. Activity assays in both adrenal and liver cells indicated that all promoter activity was included within the first 325 bases (30). We had sequenced this region in 701 of our 842 healthy subjects and identified three common promoter polymorphisms (22). Introducing these into the 325-bp construct showed that the change of C to A at the –152 position reduced reporter gene transcription to ~60% in human liver HepG2 cells and to ~35% in human adrenal NCI-H295A cells (30).

Slide 46: Pharmacogenetics of the POR promoter

The –152 polymorphism was found in 2.6% of African-American genes, 7.8% of Chinese-American and Mexican-American genes, and 13% of Caucasian-American genes but was not in linkage disequilibrium with A503V (22). Other studies showed that the principal factors regulating human POR transcription are the Smad3/4 factors, thyroid hormone receptors, and the transcription factor AP-2 (30).

Slide 47: Summary and conclusions

From these studies, we can make several conclusions: (i) POR deficiency causes a genetic disorder of steroid hormone synthesis. (ii) POR variant A503V is found on ~28% of human alleles. (iii) Assays based on cytochrome c are poor predictors of mutant POR activity. (iv) Assays based on P450c17 correlate well with patient phenotypes. (v) Neither of these assays can predict the ability of a POR variant to support the activity of drug-metabolizing P450 enzymes. (vi) POR A503V probably contributes to genetic variations in drug metabolism and sex steroid synthesis. (vii) The –152 promoter polymorphism may also be important. Thus, each POR mutant must be assessed independently with each P450 of interest.

Slide 48: Collaborators

I thank the people in my laboratory who did all this work: Christa Flück, Amit Pandey, Ningwu Huang, Vishal Agrawal, Larissa Gomes, Rachel Scott, Duanpen Sandee, Meng Kian Tee, and Taninee Sahakitrungruang. I also thank my collaborators Kathy Giacomini, who furnished the 842 normal DNAs; Tim Tracy, whose work was not discussed; and the many physicians throughout the world who have provided DNA samples for our research

Editor’s Note: This contribution is not intended to be equivalent to an original research paper. Note, in particular, that the text and associated slides have not been peer-reviewed.

References and Notes

Funding: Work done in Dr. Miller’s laboratory was supported by NIH grant GM073020.
View Abstract

Related Content

Navigate This Article