Impaired 17 , 20-lyase activity in male mice lacking cytochrome b 5 in Leydig cells

Departments of Pharmacology (V.S., B.M.O., S.A.K., D.J.M.) and Molecular Biology (S.A.K.) and the Howard Hughes Medical Institute (D.J.M.), UT Southwestern Medical Center, Dallas, TX 75390; Departments of Internal Medicine and Pharmacology (J.L., R.J.A.) and the Michigan Metabolomics and Obesity Center (R.C.), University of Michigan, Ann Arbor, MI 48109; and the Institute of Metabolism and Systems Research, University of Birmingham, Birmingham B15 2TT, United Kingdom (B.A.H., W.A.).


Response to Reviewers
We thank the reviewers for their comments and constructive suggestions.We have addressed each point below (marked **) and have highlighted all changes in the manuscript with yellow.

Reviewers' comments:
Reviewer #1: This is a nice manuscript that phenotypes TesKO mice, which are conditional knockout mice in which cytochrome b5 has been deleted from the Leydig cells in the testis.Cytochrome b5 is an allosteric effector of CYP17A1 and determines whether C21 steroids are converted to C19 steroids by the lyase function of this enzyme.The authors show that in the knockout mice expected changes in the steroid profile are observed.However, normal male genital development and fertility are observed.The authors conclude that the TesKO mice recapitulate the biochemical defect of b5 mutation but do not manifest the physiology of the defect.A number of points should be addressed: 1.In the introduction the authors should clarify the statement "patients with b5 defects have the purest form of isolated 17,20-lyase deficiency" **This sentence has been changed to, "Unlike other forms of isolated 17,20-lyase deficiency, patients with b5 defects demonstrate complete preservation of 17hydroxylase activity, confirming the in vivo importance of b5 in stimulating the 17-20lyase reaction." 2. The authors assert that cytochrome b5 mutation results in abnormal male genital development.Is this the phenotype of 17,20-lyase deficiency or has this been observed in the patients described in reference 21. **Both are true.All 46,XY patients with 17,20-lyase deficiency demonstrate impaired genital virilization irrespective of the genetic defect, including those described in references 21 & 39.In fact, a prior report of a patient with b5 deficiency also showed abnormal male genital development, but androgens were not assessed in this child (Giordano, Kaftory, & Steggles Hum Genet 1994;93:568-570).
3. The manuscript would be improved if there were direct measurement of DHEA in the control and TesKO mice with and without stimulation by hCG since DHEA is the immediate downstream product of 17-OH-pregenolone.**Mice synthesize very little DHEA, because P450 17A1 expression is very low in their adrenal glands, and only a trace derives from the gonads.Nevertheless, we have measured DHEA in plasma from these animals and confirmed that DHEA is consistently much lower than androstenedione or testosterone, <0.02 ng/mL in all samples (Results, lines 240-241).4. Why do the cortisol levels in the WT animals (basal) exceed those of the TesKO mice by 30-fold; and why are these levels non-detectable after hCG stimulation?(Table 1 and  Table 2).**In mouse plasma, corticosterone concentrations exceed those of cortisol by up to 3 orders of magnitude.The plasma cortisol concentrations are quite low in all animals with high variability both before and after hCG stimulation.Consequently, none of these cortisol differences are statistically or physiologically significant.To avoid confusion, we have deleted cortisol and cortisone data from Table 1 & 2 and from Figure 2.
Minor Figure 1B, the western blot for the CYB5 knockdown should be repeated since it is not publication quality **We repeated the b5 immunoblots, which are notoriously difficult, using a variety of modifications to arrive at the best conditions (page 6).We now have reproducible data with good quality blots using both liver and testis homogenates from LeyKO and WT animals, which are shown in Figure 1B.We could not demonstrate a difference in b5 immunoreactivity in either liver or testis using whole tissue homogenates.This result is not surprising, as the Leydig cells comprise only ~5% of the testis mass.We also found a second major protein species in testis homogenates from all animals, which might be related to the difference in sample preparation for testis versus liver proteins.This result and analysis are discussed in the text, page 8.The entire b5 blot is shown below: The mice are referred to as either TesKO or LeyKO only one name should be used.**We regret this error in the version of the abstract entered via the portal.LeyKO has been now used consistently and exclusively.
Reviewer #2: In the manuscript entitled 'Impaired 17,20-lyase activity in male mice lacking cytochrome b5 in Leydig cells' by Sondhi and colleagues describe studies aimed at assessing the effect of targeted disruption of cytochrome b5 on steroidogenic capacity.The authors employ a tissue-specific mouse model where the expression of cytochrome b5 is deleted in testis to define the role of this cofactor in Leydig cell androgen synthesis.Comprehensive metabolomics profiling of steroid metabolites in plasma and urine were performed.In contrast to the phenotype of humans recently reported where mutations in CYB55A confer 17,20-lyase deficiency, deletion of the gene in mice has no effect on fertility.The authors provide data to support the conclusion that the 17,20-lyase activity in mouse exhibits cytochrome b5-independent components that are sufficient to confer normal reproductive development and fertility.Overall this is a brief, but well-written study that describes the importance of developing appropriate models to investigate human disease states.There are no major concerns.However, the authors may want to consider the following: 1.
The quality of the cytochrome b5 western blot should be improved **See comment above regarding new immunoblots shown in Figure 1B.

2.
What would be the effect of incubating testicular homogenates to recombinant cytochrome b5 where mutations found by Idowiak et al., are introduced.**The reviewer is inquiring whether inactive forms of b5 can influence P450 17A1 activity.This is an interesting question, and our laboratory has conducted extensive experiments to identify the key residues on b5 that are responsible for stimulation of 17, 20-lyase activity.We have published these results in references 27, 28, and related papers such as Peng & Auchus Biochemistry 2013;52:210-220.

Introduction
Androgen biosynthesis requires the enzyme steroid 17-hydroxylase/17,20-lyase (P450 17A1, CYP17A1) to convert 21-carbon steroids to 19-carbon steroids.P450 17A1 from all species studied catalyzes 17-hydroxylation with similar rates for substrates pregnenolone and progesterone (Prog), yielding 17-hydroxypregnenolone and 17-hydroxyprogesterone (17OHP), respectively.The 17,20-lyase reaction, however, shows species-specific substrate preferences, with human P450 17A1 catalyzing the conversion of 17-hydroxypregnenolone to dehydroepiandrosterone about 50-fold more efficiently than 17OHP to androstenedione (AD), whereas the rodent enzymes utilize both pathways.The 17,20-lyase reaction is more vulnerable than the 17-hydroxylase reaction to disruption from the abundance of its electron transfer protein cytochrome P450-oxidoreductase (POR) (1), to mutations in POR, and to phospholipid composition.Because the 17,20-lyase activity is the sole gateway to all androgens, the potent P450 17A1 inhibitor abiraterone was developed for the treatment of castration-resistant prostate cancer (2,3).The prodrug abiraterone acetate suppresses testosterone (T) synthesis in prostate cancer patients and normalizes androgen production in adult women with uncontrolled classic 21-hydroxylase deficiency (4).
Cytochrome b5 (b5) is a small, highly conserved 15-kDa hemoprotein cofactor for multiple oxidative reactions, including the metabolism of fats and steroids, reduction of methemoglobin to hemoglobin, and the catabolism of xenobiotics and drugs (5)(6)(7)(8)(9).The full-length microsomal form of b5 is found on the cytoplasmic side of the endoplasmic reticulum in many tissues, and a soluble form lacking the C-terminal membrane anchor is abundant in erythrocytes (10).Multiple studies since the 1960s have confirmed the importance of b5 as a modifier of various cytochrome P450 activities, although its mechanism of action remains controversial and might vary with P450 isoforms and substrates.Several mechanisms have been proposed, including direct transfer of electrons to P450s from either NADHcytochrome b5 reductase or POR (11) versus an allosteric effect on the P450s and/or POR (12,13).
Depending on the substrate and enzyme, b5 can serve as an obligate component of the reaction or as a modifier of a reaction for the same P450.For example, P450 2B4-catalyzed metabolism of methoxyflurane shows an absolute requirement for b5 (14); in contrast, b5 inhibits the metabolism of benzphetamine by P450 2B4 (15).In the P450 4A subfamily, which consists of eicosanoid and fatty acid hydroxylases, b5 can increase or decrease the Km of the reaction, depending on the substrate involved (16).P450 17A1 also shows substrate-specific modulation via b5 (12).Similar to the substrate-specific modulation of activity observed with b5 and other P450s, b5 stimulates the 17, 20-lyase reaction rate >10fold when 17-hydroxypregnenolone or 17OHP are substrates, but b5 only increases the conversion of 5pregna-3,17-diol-20-one to androsterone-the most efficient 17,20-lyase reaction for human P450 17A1-by 3-fold (17,18).Evidence from experiments with apo-b5, which lacks the heme and thus cannot transfer electrons, suggests that b5 action on P450 17A1 does not involve direct electron transfer (12), and b5 might act as an allosteric modulator, promoting interaction of POR with P450 17A1 and/or conformational changes that enhance 17,20-lyase activity.
Combined 17-hydroxylase/17,20-lyase deficiency is found in rare patients, and isolated 17,20lyase deficiency is one of the rarest hereditary defects in steroidogenesis (19).Genetic screening allowed the identification of mutations in both P450 17A1 and POR that almost completely ablate 17,20-lyase activity yet only partially impair 17-hydroxylase activity (20).Most recently, two reports have identified consanguineous families with isolated 17,20-lyase deficiency due to mutation in the CYB5A gene encoding b5 (21).Unlike other forms of isolated 17,20-lyase deficiency, patients with b5 defects demonstrate complete preservation of 17-hydroxylase activity, confirming the in vivo importance of b5 in stimulating the 17-20-lyase reaction.
The in vivo functions of b5 have recently been explored with mouse models of b5 deletion, including a hepatic only deletion (22) and a complete b5 null mouse (23).Complete deletion of b5 causes profound changes in the metabolism of various substrates, in accordance with its role as a regulator of numerous P450 activities.Surprisingly, McLaughlin et al observed only a 50% reduction in intratesticular T in b5-null animals, despite a profound defect in conversion of 17OHP to AD.The b5-null mice, unlike human patients with b5 mutations, were fertile and anatomically indistinguishable from wild-type littermates (23).Nevertheless, the contribution of altered hepatic T metabolism, which is also a b5-dependent process, confounds the interpretation of these studies.To further explore the importance of b5 in the P450 17A1-catalzyed 17,20-lyase reaction, we have generated a mouse model with deletion of b5 in testicular Leydig cells but not the liver.Through in vivo analysis of androgen synthesis and in vitro analysis of P450 17A1 activity, we demonstrate the physiologic importance of b5 activating the 17,20lyase reaction and the production of 19-carbon sex steroids from 21-carbon precursors.

Generation of animals, breeding experiments, and histology
The Cyb5 flox/-mice were generously provided by Professor C. Roland Wolf and Dr. Colin J.
Henderson, Cancer Research UK, Ninewells Hospital and Medical School, Dundee, United Kingdom via Dr. Hao Zhu, University of Kansas.Cyb5 flox/flox mice were maintained by random breeding on a 129P2 * C57BL6 genetic background, and Sf1-Cre transgenic mice were maintained on a C57BL6 background as described (24).Cyb5 flox/flox :Sf1-Cre mice were backcrossed to Cyb5 flox/flox mice to generate conditional b5 knockout animals.The presence of the floxed Cyb5 alleles and Sf1-Cre transgenes was determined as previously described (22,25).Cyb5 flox/flox :Sf1-Cre:Srd5a1 -/-mice were generated by crossing Cyb5 flox/flox :Sf1-Cre with Srd5a1 -/-mice.Mice were genotyped for the presence of Srd5a1 as previously described (26).Male mice were studied to provide interpretable data focusing on androgen synthesis without the complications of cyclicity and high conversion to estrogens.The Institutional Animal Care and Research Advisory Committee of the University of Texas Southwestern Medical Center approved all animal protocols.All mice used in experiments were fed a standard irradiated chow diet and housed in a temperature-controlled environment with a 6 AM -6 PM light/dark cycle.Unless specified, mice used in experiments were 6-8 weeks old.
Fertility was assayed by placing fertile wild-type (WT) female mice with individually housed male mice.Breeding cages were surveyed daily for the presence of pups that if found were removed and sacrificed.Adult animals were sacrificed by asphyxiation with isoflurane and, then exsanguination via cardiac puncture; pups were sacrificed by decapitation.For stimulation of steroidogenesis, hCG (Sigma) was dissolved in a 0.9% saline solution for a final concentration of 100 mIU/mL.Mice were administered 100 L of hCG solution or vehicle via an intraperitoneal injection and sacrificed 2 hours later.Tissue fixation, hematoxylin and eosin staining, and microscopy for histological analyses were conducted according to standard protocols.

RT-PCR, immunoblotting, urine and blood collection, and dynamic testing
For RNA extraction, frozen liver or testis samples were homogenized in 500 L of RNA-STAT60 (Isotex Diagnostics), 100 L of chloroform was added, and the phases were separated by centrifugation.The upper phase was removed, and RNA was precipitated with 500 L of ice-cold npropanol.Following centrifugation, the pellet was washed with 70% ethanol, followed by DNase treatment and reverse transcription using random hexamers to generate cDNAs.RT-qPCR analyses used 25 ng of cDNA and 150 nmol of primers mixed with SYBR GreenER PCR Master Mix (Invitrogen).
Blood obtained through cardiac puncture as previously described was collected in EDTA coated microfuge tubes.Plasma was separated and used for steroid profiling.Urine was collected from the wells of 96-well plates placed at the base of cages with individual mice.Mice were placed in the urinecollection cages for 2 hours a day for five days to minimize stress, and urine collected from the same mouse on different days was pooled for analysis.

Enzyme assay
Recombinant tetrahistidine-tagged human b5 was prepared as described (27).Testes (80-130 mg) were homogenized in 0.25 mL of 0.25 M sucrose containing 10 mM TrisHCl pH 7.0 and 1 mM EDTA with 10 strokes of Teflon homogenizer.The debris was pelleted by centrifuging 3 min at 5,000 xg and discarded, and the decanted crude homogenate was stored frozen at -20C until use.Aliquots (10 μL) were incubated with 10 μM [ 3 H]-labeled progesterone or 17-hydroxyprogesterone (100,000 CPM per incubation) and 1 mM NADPH in 50 mM potassium phosphate buffer pH 7.4 with and without 2-30 pmol b5 (estimated 5 pmol b5/pmol P450 17A1 based on activity) at 37C in a total volume of 0.5 mL for 30 min.Steroids were extracted with 1 mL 1:1 ethyl acetate/isooctane or 1 mL dichloromethane.After centrifugation at 8000 rpm for 1 min, the organic phase was transferred into glass tubes and concentrated under nitrogen.Samples were reconstituted with 20 µL methanol, and 5 µL samples were injected into an Agilent 1260 Infinity HPLC system equipped with UV detector and β-RAM4 in-line scintillation counter (LabLogic, FL, USA).Steroid standards and samples were resolved on a Kinetex 50 x 2.1 mm, 2.6 m particle size C8 column (Phenomenex) and methanol-water gradients as described (28), mixed with Bio-SafeII scintillation cocktail (Research Products International), and quantitated by integration of radioactivity peaks using Laura4 software (LabLogic).

Plasma steroid profiling by liquid chromatography-tandem mass spectrometry (LC-MS/MS)
A 50 µL aliquot of plasma was deproteinated with acetonitrile and methanol containing deuterated internal standards, and steroids were extracted and quantitated using an Agilent 1290 binary pump HPLC attached to an Agilent 6490 triple quadrupole tandem mass spectrometer as described (29,30).Assay performance characteristics and precursor/product ion pairs were reported in these references.

Urine steroid metabolite profiling by gas chromatography-mass spectrometry (GC-MS)
Steroids in a 0.5-1 mL sample of urine were extracted, derivatized, and quantitated as previously described (31).In mice, the major androgen metabolite is an incompletely characterized androstanetriolone, which when fully silylated gives a molecular ion of M + = 416 m/z.Excretion of metabolites in untimed samples were normalized to creatinine and expressed as g/mg creatinine.

Generation of Leydig cell b5 knockout mice (LeyKO)
To study the in-vivo role of microsomal b5 in testicular steroid synthesis, we generated Ledyigcell b5 knockout mice by crossing Cyb5 flox/flox and Sf1-Cre transgenic strains.Cyb5 mRNA was reduced 40% in testis from Cyb5 flox/+ :Sf1-Cre mice and 80% in Cyb5 flox/flox :Sf1-Cre mice (subsequently referred to as LeyKO), while expression in the liver was unchanged (Figure 1A).Immunoreactive b5 proteins appeared similar in homogenates of liver and testis from LeyKO mice compared to floxed littermates, with two major b5 species in testis from all animals (Figure 1B).Because Leydig cells compose only ~5% of testis mass, the residual b5 expression might derive from other cell types, which comprise the majority of testis tissue, or from incomplete recombination.LeyKO mice were viable and born at the expected Mendelian ratio.LeyKO male mice were observed to be phenotypically normal, with no differences in internal or external sexual development between knockout mice and floxed littermates.

LeyKO mice have normal body weight, sex organ weight, and fertility
Testicular histology revealed no defects in testicular development in LeyKO mice (Figure 1C).To determine the impact on sexual maturation, we measured the weight of the epididymis, vas deferens, seminal vesicles, preputial glands and testis from LeyKO mice and compared them to floxed littermates (Table 1).No significant differences were observed between the two groups.Additionally, no differences in body weight were observed between LeyKO and floxed littermates (Table 1).Decreased androgen production would be expected to relieve feedback inhibition on the hypothalamus and pituitary, leading to an increase in gonadotropins; however, no increase in FSH or LH was seen in LeyKO mice (Table 1).To determine if LeyKO mice had any defects in their ability to reproduce, we individually housed LeyKO and floxed littermates with wild-type females and measured litter size and frequency.No differences in litter size or frequency were observed between the two cohorts (Table 1), indicating that LeyKO mice had no defect in their ability to copulate.These results are consistent with previously published data on global b5-null mice that were found to be viable, fertile, and anatomically indistinguishable from wild type mice (23).

LeyKO mice show slight changes in basal plasma and urine steroid profiles
Given the strong stimulatory effect of b5 on the P450 17A1-catalyzed 17,20-lyase reaction, we anticipated that deletion of b5 from Leydig cells would decrease circulating 19-carbon steroids AD and T and lead to an accumulation of 21-carbon precursors, particularly 17OHP.Using LC-MS/MS to profile plasma steroids, we found neither a significant decrease in AD and T nor a significant increase in 17OHP and Prog (Table 1).Furthermore, plasma 17OHP concentrations were consistently low or undetectable (<0.5 ng/mL) in both LeyKO mice and floxed littermates, and in agreement with prior studies, plasma T did not exceed 0.5-1 ng/mL (32).These results suggest that residual 17,20-lyase activity in the absence of b5 was sufficient to maintain the low basal level of T synthesis in male mice without accumulation of precursors.An alternative explanation is that b5-dependent extra-testicular androgen synthesis compensates for the lack of testicular AD and T production in LeyKO mice; however, this hypothesis is unlikely given the minor reductions in plasma T found in global b5-null animals (23).To confirm these findings, we performed urinary steroid metabolite profiling by GC/MS, which was used to demonstrate impaired 17,20-lyase activity in patients with b5 deficiency (21).Urine samples from 10-week-old LeyKO mice and floxed littermates (n=10 each group) showed slightly lower amounts of the major androstanetriolone androgen metabolite compared to floxed littermates (AUC/mL urine = 2042000  927000 versus 2786000  865000, respectively, p = 0.09).These data demonstrate minimal changes in basal steroid production in LeyKO mice.
An alternate pathway to dihydrotestosterone (DHT), originally described in the tammar wallaby, is also found in both rodents and human steroidogenic cells (37).The 5-reductase isoform Srd5a1, known to convert T to DHT, also utilizes Prog and 17OHP as substrates to form 5-reduced pregnanes, which are subsequently converted through a series of steps, including the P450 17A1-catalyzed 17,20lyase reaction, into DHT.The 17,20-lyase reaction with key intermediate substrate 5-pregnan-3,17diol-20-one, however, is minimally dependent upon b5 (17).Embryonic activation of the alternate pathway to DHT in LeyKO mice might compensate for low T production, explaining the complete virilization and lack of phenotype observed in the LeyKO mice.To test this hypothesis, LeyKO mice were crossed with Srd5a1-null mice to generate Cyb5 flox/flox :Sf1-Cre:Srd5a1 -/-mice (referred to as LeyKO/Srd5a1 -/-).The LeyKO/Srd5a1 -/-mice were born in normal Mendelian ratios and had no overt differences in internal and external sexual development.Fertility experiments revealed no differences in litter frequency or size between the LeyKO/Srd5a1 -/-mice and Cyb5 flox/flox littermates.These results suggest that the normal sexual development in LeyKO mice is likely due to adequate fetal androgen production by residual 17,20-lyase activity despite the absence of b5 and not due to activation of the alternate pathway to DHT that bypasses the need for b5.

Discussion
Abundant biochemical and limited genetic data have shown that the small hemo-protein b5 provides functions as a cofactor in methemoglobin reduction, fatty acid desaturation, and several cytochrome P450-catalyzed reactions in the metabolism of xenobiotics and steroid hormones (8,10,22,23,38).For the P450 17A1, b5 significantly increases the rate of the 17,20-lyase reaction that converts 21-carbon pregnanes into 19-carbon androgens.This report describes a mouse model of conditional b5 knockout in the testicular Leydig cell (LeyKO).While LeyKO mice were found to be phenotypically identical to floxed littermates with similar basal plasma steroids, acute stimulation of steroidogenesis led to profound differences in plasma steroids using LC-MS/MS assays, with accumulation of 21-carbon pregnanes Prog and 17OHP behind the disrupted 17,20-lyase reaction and impaired androgen synthesis.In vitro studies with testicular homogenates confirmed that the biochemical phenotype resulted from the loss of Leydig cell b5.This study provides the most complete characterization of the physiologic function of b5 in activating the 17,20-lyase activity of P450 17A1.
One limitation to our study is the use of Sf1-Cre transgene for conditional deletion of the Cyb5a gene.Sf1 is also expressed in the adrenal cortex, gonadotropes, and certain regions of the hypothalamus.
In mice, P450 17A1 expression in the adrenal cortex is low, as evidenced by the high plasma corticosterone in all animals (Figure 2).While our data hint that other 21-carbon steroids of adrenal origin also accumulate in LeyKO animals after hCG stimulation, the effect is small compared to the massive rise in 17OHP.Thus, the simultaneous loss of b5 in the adrenal cortex is unlikely to confound our conclusions, whereas the preservation of hepatic b5 expression is a unique advantage of our mouse model (Figure 1).Furthermore, the hypothalamus and pituitary are not important sites of P450 activity or androgen biosynthesis from 21-carbon steroids.The poor 17,20-lyase activity in homogenates of testes from LeyKO animals and its rescue upon addition of b5 is strong evidence that loss of b5 in the Leydig cells is sufficient to explain the consistent and marked accumulation of 21-carbon steroid precursors after hCG stimulation in these mice.The residual Cyb5a mRNA and trace b5 immunoreactive protein in LeyKO testis (Figure 1A,B) might derive from the seminiferous tubules, which compose the majority of the testicular tissue mass.Alternatively, incomplete recombination might account for some of the residual 17,20-lyase activity in the LeyKO mice.Even if the LeyKO Leydig cells contain some b5 protein, the reduced 17,20-lyase activity in testis homogenates and its restoration upon addition of recombinant b5 support our conclusions, although our data might underestimate the true magnitude of the b5 effect.We studied only male animals due to the additional complexities introduced from the estrus cycle and dominant conversion of androgens to estrogens in the ovary of female mice.Nevertheless, subsequent experiments with female "LeyKO" animals, engineered to lack b5 in the theca and granulosa cells, would be a logical topic of future studies.
Previous work on b5 physiology has focused on its in vitro modulation of various P450s or its in vivo effects on metabolism of various xenobiotics (22).A global b5-null mouse has been described with substantial defects in hepatic and extra-hepatic drug metabolism (23).The global b5-null mice were born in normal Mendelian ratios, and were unexpectedly found to be fertile with no overt phenotype.Similar to observations with the global b5-null animals, LeyKO mice had normal fertility without gross or histological changes in the testis, epididymis, vas deferens, seminal vesicles or preputial glands.Our data demonstrate that normal androgen physiology is preserved in this strain of male mice whether b5 is deleted from the Leydig cells alone or simultaneous deletion from the liver as well, which is likely to reduce testosterone catabolism.
In contrast to these findings in mice, human patients with CYB5A mutations and a 46XY karyotype are born with ambiguous genitalia and are universally infertile (21,39).The 17,20-lyase activity of microsomes isolated from LeyKO mice is significantly reduced and comparable to the decrease in 17,20-lyase activity noted in transfected HEK293 cells expressing human P450 17A1 and b5 mutation H44L (21), demonstrating that b5 is required for maximal 17,20-lyase activity in the Leydig cells of both mice and humans.The differences in sexual development and maturation observed between LeyKO mice and human patients potentially stem from differences in the specific androgen concentrations required for fetal development.Decades of research has revealed that human fetal male sexual development is dependent on the production of T and its subsequent conversion to DHT via SRD5A2 (40).Human amniotic T concentrations, measured between 11 and 21 weeks gestation, average 0.17 ng/mL for male fetuses and 0.1 ng/mL for female fetuses (41).Direct measurements of fetal T from fetal vein sampling between 15 and 38 weeks of gestation average 0.61 ng/mL for male fetuses and 0.22 ng/mL for female fetuses (42).T concentrations below a threshold lead to graded degrees of impaired virilization, which are inversely correlated.Male mouse fetal testes predominately produce T plus 5-androstene-3,17-diol at day 16 and 17, when virilization of the mouse urogenital tract is known to occur (43); however, targeted disruption of both Srd5a1 and Srd5a2 does not impair virilization (44).Furthermore, one-third of mice engineered to express a phosphorylation-deficient steroidogenic acute regulatory protein (StAR) are normally virilized, despite undetectable postnatal T synthesis (45).Thus, mouse virilization is exquisitely sensitive to T and does not require DHT as in humans.Consequently, the reduced T production in LeyKO animals is still sufficient to allow normal virilization of male mice.
In summary, we have shown impaired in vivo and in vitro 17,20-lyase activity in testes of LeyKO male mice, with marked accumulation of 21-carbon steroids after hCG stimulation.Despite a 200-fold increase in the stimulated [Prog+17OHP]/[AD+T] (precursor/product) ratio compared to wild-type littermates, the lack of b5 in Leydig cells does not affect sexual maturation, fertility, or basal steroid levels of male mice.In contrast to the impaired virilization and fertility of patients with b5 deficiency, the residual 17,20-lyase activity of the LeyKO mice is sufficient to maintain normal androgen physiology in male animals.We conclude that LeyKO mice represent a good model for the biochemistry but not the physiology of b5 deficiency in human beings, which is one of the rarest disorders of steroidogenesis.Click here to download Figure Fig2Rev1.tif