THE JOURNAL or BIOLOGICAL CHEMISTRY
`© 1989 by The American Society for Biochemistry and Molecular Biology, Inc.
`
`Vol. 264, No. 30, Issue of October 25, pp. 18076-18082, 1989
`Printed in U. SA.
`
`Deletion of a Phenylalanine in the N-terminal Region of Human
`Cytochrome P-45017,, Results in Partial Combined
`1 7a-Hydroxylase/ 1 7 ,20-Lyase Deficiency*
`
`Toshihiko Yanasei§, MasaakiKagimoto:l:1l, Shin Suzuki|| , Kunitake Hashiba” , Evan R. Simpsoni,
`and Michael R. Watermanzt
`
`From the iDepartments of Biochemistry and Obstetrics and Gynecology and the Cecil H. and Ida Green Center for Reproductive
`Biology Sciences, University of Texas Southwestern Medical Center, Dallas, Texas 75235 and the || Third Department of Internal
`Medicine, Nagasaki University, Nagasaki 852, Japan
`
`(Received for publication, May 24, 1989)
`
`Steroid 17oz-hydroxylase and 17 ,20-lyase activities
`reside within the same polypeptide chain (cytochrome
`P-45017“), and consequently human 17oz-hydroxylase
`deficiencies are characterized by defects in either or
`both of these activities. Human mutants having these
`deficiencies represent an excellent source of material
`for investigation of P-45017,, structure-function rela-
`tionships. The CYP17 gene from an individual having
`partial combined 17a-hydroxylase/17,20-lyase defi-
`ciency has been characterized structurally and the
`homozygous mutation found to be the deletion of the
`phenylalanine codon (TTC) at either amino acid posi-
`tion 53 or 54 in exon 1. Reconstruction of this mutation
`into a human P-45017,, cDNA followed by expression
`in COS 1 cells led to production of the same amount of
`immunodetectable P-45017,, protein as found with
`expression of the normal human P-45017,, cDNA. How-
`ever, 17a-hydroxylase activity of this mutant protein
`measured in intact cells was less than 37% of that
`observed upon expression of the wild-type enzyme,
`whereas 17,20-lyase activity of the mutant was less
`than 8% of that observed with the normal enzyme.
`When estimated in intact cells, the Km for 17a-hydrox-
`ylation of progesterone was increased by a factor of 2
`in the mutant enzyme, whereas the Vmax was reduced
`by a factor of 3. In order to estimate the kinetic param-
`eters for the 17 ,20-lyase reaction, microsomes were
`isolated from transfected COS 1 cells to enrich for this
`activity. Surprisingly, the specific activity of the mu-
`tant 17a-hydroxylase in microsomes was 3-fold less
`than that observed in intact cells, indicating that the
`structure of mutant P-45017,. was dramatically altered
`upon disruption of COS 1 cells. Apparently the deletion
`of a single phenylalanine in the N-terminal region of
`P-45017,. alters its folding in such a way that both
`enzymatic activities are dramatically decreased, lead-
`ing to the partial combined deficiency observed in this
`individual.
`
`* This research was supported by Grant 1-1084 from the March of
`Dimes Birth Defects Foundation and United States Public Health
`Service Grant GM37942. The costs of publication of this article were
`defrayed in part by the payment of page charges. This article must
`therefore be hereby marked “advertisement” in accordance with 18
`U.S.C. Section 1734 solely to indicate this fact.
`§ Permanent address: Third Department of Medicine, Kyushu
`University, Fukuoka, Japan.
`ll Permanent address: First Department of Medicine, Kyushu Uni-
`versity, Fukuoka, Japan.
`
`The initial step common to the steroidogenic pathways in
`the adrenal cortex and gonads is the the side chain cleavage
`of cholesterol catalyzed by cytochrome (P'45Oscc)1 to form
`pregnenolone, which can be dehydrogenated at the 3B-posi-
`tion to form progesterone. Subsequent 17a-hydroxylation of
`pregnenolone or progesterone is a branch point for the for-
`mation of cortisol and sex hormones in the steroidogenic
`pathways. The side chain of the C21 steroids 17a-hydroxy-
`pregnenolone and 17-hydroxyprogesterone can be further
`cleaved during the 17,20-lyase reaction to yield the C19 steroid
`precursors of testosterone and estrogens, dehydroepiandros-
`terone (DHEA) and androstenedione, respectively. Evidence
`obtained using both purified protein samples and in situ COS
`1 cell expression systems has proven that a single polypeptide,
`cytochrome P-450,7“, can catalyze both the 17a-hydroxylase
`and 17,20-lyase reactions (Nakajin et al., 1984; Zuber et al.,
`19863). Accordingly, in the adrenal cortex, a fraction of the
`17oz-hydroxysteroids is converted to C19 steroids by the 17,20-
`lyase reaction, the remainder of these 17a-hydroxysteroids
`being precursors for cortisol production. In the gonads, all of
`the 17oz-hydroxysteroids are converted to C19 steroids.
`Human 17a-hydroxylase deficiency is an autosomal reces-
`sive disorder and is one of the causes of congenital adrenal
`hyperplasia (Biglieri et al., 1966). In this disease, impaired
`production of cortisol provokes elevation of ACTH and con-
`sequently overproduction of mineralocorticoids, principally
`deoxycorticosterone, resulting in hypertension and hypoka-
`lemia. In addition,
`impaired production of sex hormones
`causes abnormalities of sexual development, leading to failure
`of virilization in the male and sexual infantilism in the female
`in the most severe cases. We have reported previously the
`molecular basis of two examples of complete combined 17a-
`hydroxylase/17,20-lyase deficiency (Kagimoto et al., 1988,
`1989; Yanase et al., 1988). Although many such patients show
`complete blocks of both 17oz-hydroxylase and 17 ,20-lyase ac-
`tivities, some individuals are found to have partial deficiencies
`of both activities leading to ambiguous external genitalia
`(New et al., 1970; Jones et al., 1982). In addition, a few patients
`have been reported to show deficiency of only a single activity,
`either 17a-hydroxylase (Vargas et al., 1981) or 17,20-lyase
`activity (Zachman et al., 1982). Presumably, the elucidation
`of the molecular basis of these variants leading to partial
`
`1 The abbreviations used are: P-450m, Cholesterol side chain cleav-
`age cytochrome P-450; P-45017“, 17a-hydroxylase cytochrome P-450;
`P—450n,«3, 116-hydroxylase cytochrome P-450; P—450c2;, steroid C21-
`hydroxylase cytochrome P-450; P—450c,,,,,, camphor hydroxylase cyto-
`chrome P-450; DHEA, dehydroepiandrosterone; ACTH, adrenocor-
`ticotropin; bp, base pair(s); kb, kilobase pair(s); SDS, sodium dodecyl
`sulfate; Pipes, 1,4-piperazinediethanesulfonic acid.
`
`18076
`
`Amerigen Exhibit 1156
`Amerigen Exhibit 1156
`Amerigen v. Janssen IPR2016-00286
`Amerigen v. Janssen IPR2016-00286
`
`

`
`Phenylalanine Deletion in Human Cytochrome P-45017,,
`
`18077
`
`deficiencies holds the promise of contributing substantially to
`the general understanding of the structure-function relation-
`ships in P-45017“. In particular, reconstruction of mutations
`into the human P-45017.. cDNA, expression of such mutant
`enzymes, and analysis of the resultant enzymatic properties
`should be particularly informative.
`We have now investigated the molecular basis of an exam-
`ple of partial combined deficiency in a female Japanese pa-
`tient. As a result of a phenylalanine deletion at either amino
`acid 53 or 54 in the N-terminal region of P-450170, both
`enzymatic activities are dramatically reduced but not absent
`in this patient. Upon isolation of microsomes from COS 1
`cells transfected with the mutant cDNA, both activities are
`even more dramatically reduced compared with those of the
`normal enzyme when compared with results of in situ assays
`carried out in intact transfected COS 1 cells. We surmise that
`the deletion leads to altered folding of the protein and con-
`sequently producing an. enzyme having reduced 17a-hydrox-
`ylase and 17,20-lyase activities.
`
`MATERIALS AND METHODS
`
`Case Record—The CYP17 (Nebert et al., 1989) gene examined in
`this study was that of a Japanese female having a 46 XX karyotype,
`the subject of a case report by Suzuki et al. (1974). In summary, the
`patient (JG) was 20 years old when reported and currently is 35 years
`old, married, and has no children. Her parents and siblings had no
`apparent clinical abnormalities. When she visited a hospital because
`of an occipital headache, she was found to suffer from hypertension.
`Her blood pressure was 170/110 mmHg, and she was also hypoka-
`lemic. Importantly, she had menstruation, although irregularly, and
`physical examination revealed hypoplastic breasts with no pubic or
`axillary hair. In vivo studies showed decreased levels of 17a-hydrox-
`ylated corticoids such as cortisol, 11-deoxycortisol, and their metab-
`olites. Adrenal androgene and estrogens were also decreased but
`detectable. On the other hand, 17-deoxycorticoids such as corticos-
`terone were increased. These data plus the presence of irregular
`menstruation, suggesting some level of estrogen production, indicate
`that this patient suffers from partial combined 17a-hydroxylase/
`17,20-lyase deficiency.
`DNA Extraction and C YP1 7 Sequence Analysis—Heparinized
`blood was obtained from the patient, and genomic DNA was prepared
`from the leukocytes as described by Blin and Stafford (1976). South-
`ern analysis of the CYP17 gene was performed following EcoRI
`digestion of the DNA using a nick-translated BamHI insert of a
`human P-450,7“ cDNA clone (Bradshaw et al., 1987). Exonic sequence
`analysis of the CYP17 gene was carried out following the protocol
`described previously by this laboratory (Kagimoto et al., 1988; Yanase
`et al., 1988). Briefly, a genomic DNA library was prepared for this
`patient by EcoRl digestion, ligation with )\gtWES EcoRI arms (Be-
`thesda Research Laboratories), in vitro packaging (Packagene, Pro-
`mega-Biotec), and plaque formation on agar plates. Plaque hybridi-
`zation was used to screen approximately 8 X 10" plaques with the
`nick-translated BamHI fragment of human P-45017., cDNA. Six pos-
`itive clones were identified by this procedure, three of which contained
`the 5.7-kb EcoRl fragment associated with the CYP17 gene and three
`of which contained the 6.9-kb EcoRI fragment of this gene. Together,
`these fragments contain all eight exons of the CYP17 gene (Kagimoto
`et al., 1988). Both the 5.7- and 6.9-kb EcoRl fragments were further
`subcloned into pUC19 (Yanase et al., 1988) and sequenced by the
`dideoxynucleotide chain termination method (Sanger et al., 1977)
`using universal,
`reverse. or appropriate custom oligonucleotide
`primers.
`Determination of the Patients Genotype—To determine the geno-
`type of the patient, we utilized the ribonuclease A (RNase A) cleavage
`method originally developed for detection of single base substitutions
`by Meyers et al. (1985). A 1.9-kb Kpnl-Sacl genomic fragment con-
`taining the mutation was subcloned into Bluescript (Stratagene).
`After digestion with Pvull, a 3"’P—labeled RNA probe (247 bp) com-
`plementary to the 234-bp Sacl-Pvull mutant DNA template plus 13
`bp of Bluescript was prepared by transcription using T3 RNA polym-
`erase in the presence of [oz-32P]UTP. The reaction mixture (20 pl)
`contained 40 mM Tris-HCl, pH 7.5; 6 mM MgCl2; 10 mM NaCl; 2 mM
`spermidine (Sigma); 10 mM dithiothreitol (Sigma); 0.5 mM each ATP,
`GTP, and CTP; 12 ,uM UTP; 50 pCi of [ac-32P]UTP (Amersham
`
`Corp.); 1 pg of linearized template DNA; 20 units of RNasin (Pro-
`mega); and 50 units of T3 RNA polymerase (Bethesda Research
`Laboratories). Incubation was carried out at 37 °C for 60 min. The
`sample was then diluted with 100 pl of 10 mM Tris-HCl, pH 7.5, 1
`mM EDTA, 0.1% SDS, and used as a probe. A similar RNA probe
`(250 bp) complementary to the normal DNA fragment was also
`prepared by the same procedure. A mixture of EcoRl-digested genomic
`DNA (5 pg) and either the mutant or wild-type cRNA probe ( 5 X
`105 cpm) was heated at 90 “C for 10 min and annealed in 30 pl of a
`hybridization buffer (80% formamide, 40 mM Pipes, pH 6.4, 0.4 M
`NaCl, and 1 mM EDTA) at 45 °C for 12 h, after which 350 pl of a
`solution containing RNase A (40 pg/ml) in 10 mM Tris-HCl (pH 7.5),
`1 mM EDTA, 200 mM NaCl, and 100 mM LiCl was added and
`incubated at 25 “C for 30 min. The mutant or normal cRNA probe (1
`X 105 cpm) was also hybridized with 50 ng of mutant or normal
`cloned DNA (5.7-kb EcoRl fragment containing CYP17 exon 1) at
`45 °C for 60 min, and these hybrids were also subjected to RNase A
`treatment. The reaction was stopped by the addition of 10 pl of 20%
`SDS and 10 pl of proteinase K (10 mg/ml) (EM Biochemicals)
`followed by incubation at 37 °C for 15 min. After phenol extraction
`and ethanol precipitation, the samples were subjected to electropho-
`resis on a denaturing polyacrylamide gel (5%) (Maniatis et al., 1982),
`and the results were visualized by autoradiography.
`Construction of Mutant P-450,7“ cDNA--The overall procedure for
`constructing the TTC (phenylalanine) deletion in our full length
`human P-45017,. cDNA is summarized in Fig. 1. Basically, the N-
`terminal region of a genomic clone containing this mutation was
`joined with a fragment of the cDNA containing the remainder of the
`human P-45017.. sequence. A 1.6-kb PstI—PstI fragment of the 5.7-kb
`genomic subclone (see Yanase et al., 1988) containing the TTC
`deletion at amino acid 53 or 54 in exon 1 was inserted into pUC19
`(pUCJG17a-1). pUCJG17a-1 was then cut with Nael (5’-flanking
`region) and Smal (pUC19 polylinker) and blunt end ligated into
`pUC19 (pUCJGl7tx-2). A 188-bp Kpnl-Pstl fragment of pUCJG17a-
`2 was further cloned into Bluescript (pBSJG17a-1). The full length
`human P-45017., cDNA in Bluescript (pBSH17a-1) was prepared from
`a BamHl-BamHI fragment of pCD17a-H (Bradshaw et al., 1987) by
`using Bal31 digestion followed by blunt end ligation into the Smal
`site. A 143-bp Pstl-Xbal cDNA fragment was cloned from pBSH17a-
`1 into pBSJG17a-1, giving pBSJG17a-2. A 1410-bp Xbal-Hindlll
`cDNA fragment of pBSH17a-1 was cloned into pUC19, giving
`pUCH17a-1. The mutant cDNA construct was completed by ligation
`of a 325-bp Xbal-Kpnl fragment of pBSJG17a-2 into pUCH17a-1,
`producing pUCJG17a—3. The cDNA construct was sequenced by the
`dideoxy chain termination method to validate its structure. Finally,
`the mutant cDNA was inserted in an expression vector, pCMV,
`developed by David W. Russell at this institution (Andersson et al.,
`1989) using the Kpnl and Hindlll sites, the construct being named
`pCMVJG17oz—H. Wild-type human P-45017,. cDNA in pCMV was
`constructed from pBSH17a-1, a BamHI-Kpnl fragment containing
`the full length 17a-cDNA being cloned from pBSH17a-1 into pCMV
`using the Bglll and Kpnl sites (pCMV1’7a-H).
`Transfection of C08 1 Cells and Enzymatic Assays-Transfection
`of COS 1 cells with plasmids purified by CsCl density gradient
`centrifugation was carried out as described previously (Zuber et al.,
`1986a). 17oz-Hydroxylase and 17,20—lyase activities were assayed by
`conversion of pregnenolone/progesterone to 17a-hydroxypregneno-
`lone/17a-hydroxyprogesterone and of 17a-hydroxypregnenolone to
`DHEA, respectively. The medium (4 ml) contained 3H-labeled sub-
`strate (105 cpm/ml) plus 1.0 or 2.5 pM substrate. At times after
`addition of substrate, aliquots of culture media were removed (6 and
`12 h, 0.5 ml; 24 h, 1.0 ml), steroids extracted and subjected to analysis
`by thin layer chromatography (TLC). For estimation of K,,. and Vm.
`in intact cells, 3 ml of medium was used rather than 4 ml. The steroids
`from 0.5 ml of media were extracted with 5 ml of methylene chloride
`by Vortex mixing and centrifuging for 10 min at 1000 rpm. The
`aqueous layer was removed, and the organic phase was evaporated to
`dryness, dissolved in 0.1 ml of methylene chloride, and chromato-
`graphed by TLC. Pregnenolone, 17a-hydroxypregnenolone, and
`DHEA were separated in chloroformzethyl acetate (9525) and detected
`with a sulfuric acid spray. Progesterone, 17a-hydroxyprogesterone,
`and 16a-hydroxyprogesterone were isolated in chloroformzethyl ace-
`tate (80:20) and detected using short wave UV light. The radioactivity
`in each TLC lane was determined following separation of the lane
`into 11 equal portions, which were placed into scintillation vials and
`counted after the addition of 10 ml of OCS (Amersham) containing
`2% methanol.
`Immunoblots of Cellular Proteins—COS 1 cells were rinsed once
`
`

`
`Phenylalanine Deletion in Human Cytochrome P—45017,,
`
`(pUC19)
`Sma I, Nae I (Blunt)
`
`Pal I
`Kpn I
`SMOKE!
`A G ‘
`pUCJG17a-2
`
`Kpn I
`
`Kpn I, Pal I
`Psi I
`
`Kim I
`
`Ami‘Pstlxbal
`
`aauaanawua
`
`PBSJG l7 u-1
`
`(35)
`
`3-
`
`I
`
`I
`
`Hlnd I
`
`5'
`XbaIPstI Smal
`Smal
`-M11-
`TAA
`ATG
`DBSH 17a-1
`
`(85)
`
`I Xba |,HInd III
`Hlnd III Smal
`Xbal
`
`TM
`
`Kpnlxbmi
`
`Smal Hlnd III
`TAA
`pUCH 17a-1
`
`(pUC19)
`
`Xbal Pstl
`1:1
`
`Kpnl
`
`Pstl
`
`Xbal
`
`A Kpnl Psll
`—-
`pBSJG17a-2
`Kpn, Am-
`Xba I
`
`Xbal
`
`5'
`
`Psi I Xba I Sma I
`Kpn N§flfl!I22—
`ATG ‘
`TAA
`pUCJG17Ix-3
`
`3'
`
`Hlnd III
`
`i
`Kpnl Psll Xbal
`
`Kpn I, Hlnd III
`Hlnd III
`
`Pst I xba I Smal
`HOOONDZT
`ATG ‘
`TAA
`pCMVJG17a-H
`
`3.
`Hlnd III
`
`(PCMV)
`
`FIG. 1. Schematic description of the construction of a eu-
`karyotic expression vector containing the mutant P-45017,,
`cDNA (pCMVJG17a-H). pCD17a—H contains the full
`length
`P450”, CDNA prepared in this laboratory (Bradshaw et al., 1987).
`BS, Bluescript; ATG and TAA, the initiator and terminator codons,
`respectively, of the human F-45017“ A, the location of the phenylala-
`nine deletion. The pCMV expression vector contains the major im-
`mediate early promoter of cytomegalovirus (Thomsen et al., 1984).
`
`with phosphate-buffered saline and were collected by scraping with a
`rubber policeman using a lysis solution containing 0.1% SDS and 1%
`cholate. Solubilization of total cell protein was accomplished by
`repeated cycles of freezing and thawing. Following protein assay (BAC
`kit, Pierce Chemical Co.), 200 pg of cellular protein was placed in a
`loading buffer (0.05 M Tris-HCI, pH 6.8, 2% SDS, 5% [3—mercaptoeth—
`anol, 10% glycerol, 2 mM EDTA, and 0.05% bromphenol blue) and
`boiled for 2 min. The samples were loaded on 10% SDS-polyacryl-
`amide gels and electrophoresed in a running buffer (50 mM Tris, 385
`mM glycine, 0.5 M EDTA, and 10% SDS). Molecular weight markers
`(Bethesda Research Laboratories) were run on each gel. The proteins
`were then transferred electrophoretically to nitrocellulose paper at 35
`V for 12 h using 20% methanol, 20 mM Tris—base, and 150 mM glycine
`and immunoblotted essentially as described by Towbin et al. (1979).
`For blocking, the paper was first incubated for 2 h at room tempera-
`ture with 10 mM Tris—HCl (pH 7.4), 0.15 M NaCl, 0.2% Nonidet P-
`40, and 5% nonfat milk powder, after which an antibody against
`porcine testis P-45017,. (kindly provided by Dr. J. 1. Mason of this
`institution) was added for a 2—h incubation at room temperature. The
`paper was washed six times with washing buffer (10 mM Tris-HCI,
`pH 7.4, 0.15 M NaCl, 0.2% Nonidet P-40, 0.25% sodium deoxycholate,
`and 0.1% SDS) and then incubated with ”5l—labeled protein A (Du
`Pont-New England Nuclear) for 30 min. The paper was washed six
`times with the same washing buffer, and the specific bands corre-
`
`sponding to P-45017,, were visualized by autoradiography.
`Preparation of Microsomes from COS I Cells—COS 1 cells were
`washed once with phosphate—buffered saline, collected by scraping
`with a rubber policeman, suspended in 24 volumes (w/v) of ice-cold
`medium containing 0.5 M sucrose, 1 mM EDTA, 0.04 units/ml apro-
`tinin (Sigma), and 0.1 pg/ml leupeptin (Boehringer Mannheim), and
`homogenized by a motor-driven homogenizer. The homogenization
`was by 10 strokes at 2,500 rpm and 5 strokes at 5,000 rpm. Following
`homogenization, the sample was centrifuged at 760 X g for 15 min to
`remove cell debris and nuclei. The supernatant was then centrifuged
`at 11,000 X g for 15 min to remove mitochondria. To increase the
`recovery, this pellet was resuspended and resedimented (11,000 X g
`for 15 min). The pooled 11,000 X g supernatants were then sedimented
`at 206,000 X g for 60 min using a TL-100 Ultracentrifuge (Beckman).
`The recovery of the microsomes was approximately 300 /1g of protein/
`culture dish (100 mm). The microsomal pellet was finally suspended
`in 0.05 M Tris—HCl buffer (pH 7.5) and frozen at -70 “C. The
`microsomal 17a-hydroxylase activity was assayed by measuring con-
`version of progesterone to 17:2-hydroxyprogesterone in incubations
`carried out at 37 “C in 500 pl of 0.05 M Tris—HCl buffer (pH 7.5)
`containing 1 mM NADPH (Sigma), 10 mM glucose 6-phosphate
`(Sigma), 2.4 units/ml glucose-6-phosphate dehydrogenase (Sigma), 3
`mM MgCl2, "H-labeled (105 cpm/ml) and nonlabeled progesterone,
`and 400 pg/ml COS 1 cell microsomes. The reaction was started by
`the addition of microsomes and stopped by putting samples on ice
`and the subsequent addition of 5 ml of methylene chloride. Steroids
`were then analyzed by TLC as described.
`
`RESULTS
`
`Southern blot analysis of genomic DNA from the patient
`following EcoRI digestion showed 5.7- and 6.9-kb EcoRI frag-
`ments that are identical to those of the normal human CYP17
`
`gene (data not shown). This indicates that no large deletions
`or structural alterations exist within the P-45017,. gene of this
`patient. Sequence analysis of the patient’s P-45017,. gene re-
`vealed a deletion of a phenylalanine codon (TTC) at either
`amino acid position 53 or 54 in exon 1 (Fig. 2). In the normal
`gene, tandem TTC triplets are found at this position, and it
`is unknown which one has been deleted in the patient’s gene.
`The sequence of the region containing the deletion was con-
`firmed in both the 5’ -—> 3’ and the 3’ —> 5’ directions using
`the universal primer and a custom synthesized oligonucleotide
`primer. As a result of this deletion, the mutant P-45017..
`(P-450%"f,“°""“ °' 5")
`is 1 amino acid shorter (507) than the
`normal P-45017,, (508). The mutation was found in all three
`5.7-kb fragments analyzed. However, to be certain that this
`
`Normal
`
`Mutant
`
`GATC GATC
`vlnuuh
`
`FIG. 2. Nucleotide sequences of the region of exon 1 of the
`CYP17 genes showing the deletion of a phenylalanine codon
`(TTC) at the position of amino acid 53 or 54. In the normal
`CYP17 gene, codons 53 and 54 are both TTC (phenylalanine). The
`CYP17 gene of this patient also contained two silent mutations in
`exon 1, as reported previously (Kagimoto et al., 1988). Histidine 46 is
`encoded by CAC rather than CAT, and serine 65 is encoded by TCG
`in place of TCT.
`
`

`
`Phenylalanine Deletion in Human Cytochrome P-45017,,
`
`FIG. 3. RNA~DNA heteroduplex
`analysis by RNase A cleavage dem-
`onstrating the homozygous presence
`of the TTC deletion in the patient’s
`genomic DNA. Probe (JG) and Probe
`(N) indicate mut.ant CRNA (247 bases)
`and normal CRNA (250 bases), respec-
`tively. (‘lone (JG) and clone (N) indicate
`the 5.7-kb EcoRI fragment containing
`CYP17 exon 1 from the mutant gene and
`the normal gene, respectively. Genomic
`(JG) indicates the total genomic DNA
`from the patient
`following complete
`digestion with EcoRI. The faint band
`below the 81-nucleotide band, best visu-
`alized in the probe (N) + clone (JG) lane,
`probably results from cleavage of codon
`53, whereas the 81-nucleotide fragment
`results from cleavage at codon 54. Both
`the J0 and N DNAs contained the silent
`mutations noted in the legend to Fig. 2.
`nt, nucleotides.
`
`FAM C WT JG
`
`FIG. 4. Immunoblot analysis of P-45017,, following expres-
`sion in COS 1 cells. C, WT, and JG indicate COS 1 cells transfected
`with no plasmid, pCMV17a-H (wild type), and pCMVJG17a-H (mu-
`tant), respectively. Each of these lanes contained 200 pg of protein.
`Fifteen pg of human fetal adrenal microsome (FAM) was also run as
`a positive control. Molecular masses of standard proteins electropho-
`resed on the same gel are marked at the left of the panel. The arrow
`indicates the position of P-4501-,,,.
`
`mutation was present on both alleles (homozygous), we used
`a RNase A cleavage analysis based on the principle that a
`mismatch in a RNA-DNA heteroduplex is cleaved by RNase
`A. When a mutant CRNA probe complementary to the 234-
`bp Sacl-Puull DNA from exon 1 of the mutant CYP17 gene
`was hybridized to genomic DNA from the patient or the 5.7-
`kb CYP17 fragment from this individual cloned into pUC,
`the radiolabeled CRNA was completely protected from RNase
`A cleavage (Fig. 3). However, when this probe was hybridized
`to the normal cloned 5.7-kb CYP17 fragment, cleavage led to
`two fragments, 166 and 81 nucleotides, indicating cleavage at
`the site of the TTC deletion. When the experiment was
`reversed, and a normal CRNA probe was used with either
`genomic DNA from the patient or her cloned 5.7-kb CYP17
`fragment, complete cleavage leading to 166- and 81-nucleotide
`fragments was observed (Fig. 3). This result clearly indicates
`that both alleles of the CYP17 gene in this patient contain
`the phenylalanine deletion.
`In order to evaluate the function of the P-450i‘7‘f,"*“"3 °’ ""‘“, a
`
`mutant cDNA was constructed in a eukaryotic expression
`vector (Fig. 1) and transfected into nonsteroidogenic cells
`(COS 1 cells). Transfection with the mutant P-45017,. cDNA
`construct led to the same amount of immunodetectable P-
`
`45017,, protein as found upon expression of the wild-type P-
`45017,, cDNA (Fig. 4). In the presence of the same amount of
`protein, the activities of P-450%‘f,*“"""3 °’ 5“ in COS 1 cells were
`shown to be dramatically reduced (Fig. 5 and Table I). In
`addition to the catalysis of both the 17a-hydroxylase and
`17 ,20-lyase reactions (Bradshaw et al., 1987), normal human
`P-45017,, cDNA is also known to catalyze 16a-hydroxylation
`of progesterone.’ 17a-Hydroxylase and 16a-hydroxylase ac-
`tivities of P-450%‘:,"e"”" "' 5“ were detectable by 6 h and contin-
`ued to increase to at least 24 h (Table 1). However, 17,20-
`lyase activity of P-450;‘7’f.‘‘*‘‘‘” ‘” 54’ was low, being detectable
`only at longer times of incubation. At 24 h, the 17a-hydrox-
`ylase and 16a-hydroxylase activities of the P-450%*:,"°“:’3 °' 54’
`were 37% or less of those of wild-type levels, whereas 17,20-
`lyase activity of the mutant was found to be less than 6% of
`that of wild type (Table II). However, because the 17,20-lyase
`activity is so low, it was not possible to say with certainty
`from these in situ measurements that it
`is altered in the
`
`P-4501‘7’f,"""“ "‘ 54’ disproportionately to the 17a-hydroxylase
`activity, even though every experiment carried out showed
`this trend. The data in Tables I and II also indicate that the
`substrate preference (pregnenolone versus progesterone) does
`not appear to be significantly affected by the phenylalanine
`deletion.
`The Km and Vmm. for the 17oz-hydroxylation of progesterone
`were estimated in transfected COS 1 cells using substrate
`concentrations of 0.5, 1.0, 2.5, 4.0, and 5.0 pM. Assays of
`P-4501*7‘f,“""r"’ "’ 5“ were all carried out at 6 h of incubation,
`whereas those of the wild-type P-45017.. were all carried out
`at 2 h of incubation. In two separate experiments utilizing
`different transfections, the wild-type P-45017,, K,,, was found
`to be 1.67 and 1.47 ;rM, whereas the Km of P-450f{’,'‘°‘53 ‘" 54’
`was found to be 3.35 and 3.09 ;LM. The Vmax for the wild-type
`P-45017,, was estimated to be 0.86 and 0.53 nmol/h/dish,
`whereas the Vmax of P-450%’f.'‘°‘''’‘’ ‘” "““ was estimated to be 0.21
`and 0.16 nmol/h/dish. Thus, both the K,,. and Vmax appear to
`be altered in P-45O;‘7‘f,"e""‘ "’ 54’.
`In order to establish with certainty whether the 17,20-lyase
`
`2 J. 1. Mason, unpublished observations.
`
`

`
`Phenylalanine Deletion in Human Cytochrome P—45017..
`5,
`
`(A)
`
`'°'
`
`TLC 1raction#
`
`TLC fraction#
`
`TLC fractionxt
`
`FIG. 5. Analysis of normal P-45017,, and P-450€3,'.{'°‘53 °’ 5'” activities by thin layer chromatography.
`After chromatography, TLC plates were fractionated into 11 equal sized pieces, and the radioactivity in each
`fraction was counted. The x axes and y axes indicate the fraction number and the radioactivity (cpm), respectively.
`A, conversion of progesterone (P4) to 17a—hydroxyprogesterone (17OHP..) and 16oz-hydroxyprogesterone (16OHP4).
`B, conversion of pregnenolone (P5) to 17a—hydroxypregnenolone (I7OHP5) and DHEA. C, conversion of 17a-
`hydroxypregnenolone to DHEA. Mock, COS 1 cells put through the transfection protocol without the addition of
`DNA; WT, COS 1 cells transfected with the wild-type (normal) P-45017,, cDNA; JG, COS 1 cells transfected with
`P—4501‘7‘§."°“‘3 °' 54* cDNA.
`
`TABLE I
`
`Rates of formation of steroid products in COS 1 cells
`The formation is expressed as percent of total radioactivity. Two substrate concentrations, 1.0 and 2.5 ;4M, were
`tested. Mock, WT, and JG represent no plasmid, pCMV17a-H (wild type), and pCMVJG17oz—H (mutant),
`respectively. P4, progesterone; 17OHP.,, 17a—hydr0Xypr0gester0ne; 16OHP.,, 16a-hydroxyprogesterone; P5, pregnan-
`olone; 17OHP,»-,, 17a-hydroxypregnenolone. The substrate concentration used in the mock experiment was 1 ;LM.
`—>
`—)
`—>
`(or) —>
`16OHP
`
`4
`
`P5
`
`17OHP_=,
`
`DHEA
`
`17OHP5
`
`%
`
`170HP‘
`
`P,
`
`——>
`
`%
`
`92.5
`93.1
`92.6
`
`76.0/76.9
`45.2/61.7
`28.1/38.5
`
`86.4/89.1
`84.8/87.3
`74.5/78.8
`
`0.29
`0.35
`1.42
`
`13.9/11.2
`34.7/20.1
`43.4/37.4
`
`3.6/2.3
`5.0/3.5
`9.6/7.8
`
`0.17
`0.19
`1.0
`
`1.7/1.1
`8.0/5.5
`16.1/11.7
`
`0.64/0.30
`1.5/1.3
`3.2/2.4
`
`87.0
`87.0
`88.3
`
`59.8/74.0
`48.2/47.7
`20.0/33.2
`
`83.2 /87.5
`77.4/87.9
`68.1 /75.5
`
`0.50
`0.20
`2.9
`
`232/152
`34.3/25.6
`48.1/45.4
`
`4.5/1.9
`6.9/3.8
`14.4/10.7
`
`1.6
`2.1
`1.8
`
`66.4
`62.7
`64.4
`
`4.0/1.5
`10.6/4.2
`21.4/8.8
`
`1.6/1.4
`2.1/1.8
`2.8/2.2
`
`69.9/71.7
`50.8/56.2
`44.3/54.0
`
`76.6/71.5
`63.7/57.0
`63.2/62.2
`
`2.4
`2.2
`1.4
`
`6.6/4.3
`12.1/7.2
`23.2/10.6
`
`2.1/2.7
`2.5/3.1
`2.2/2.1
`
`TABLE II
`
`Conversion of 1 14M substrate in 24 h by transfected COS 1 cells
`Mock, WT, and JG represent no plasmid, pCMV17a-H (wild type),
`and pCMVJG17a-H (mutant),
`respectively.
`P4, progesterone;
`17OHP.., 17a-hydroxyprogesterone; 16OHP.,, 16a-hydroxyprogester—
`one; P5, pregnenolone; 17OHP5, 17a-hydroxypregnenolone.
`Cells P. —> 17OHP.
`(or) —> 16OHP. P5 —> 17OHP5 17OHP5 —> DHEA
`%
`
`Mock
`WT
`JG
`
`1.8
`58.1
`22.8
`
`2.0
`19.6
`7.9
`
`0.7
`54.3
`20.4
`
`0.4
`36.8
`2.4
`
`activity was affected more severely than the 17oz-hydroxylase
`by the phenylalanine deletion, microsomes from transfected
`COS 1 cells were isolated and used for enzymatic assays.
`Much to our surprise as shown in Fig. 6, the 17oz-hydroxylase
`of P-450i‘7‘:,"°‘53 °' 5“ in isolated microsomes was much lower
`than expected relative to the activity of the wild-type enzyme.
`Thus, the ratio of 17a-hydroxylase activity of the wild-type
`P-45017., versus P-450%7‘:,“°‘53 °’ 5‘) was several times greater
`than that measured in intact cells (at 6 h using 1 ,uM proges-
`terone, this ratio was 17 in microsomes, whereas it was 6
`when measured in intact cells at the same time using the
`same substrate concentration). The 17,20-lyase activity of the
`P-4501A7‘”c."e“"3 °’ 54’ was also reduced accordingly in microsomes
`making careful examination of the effect of the mutation on
`the 17,20-lyase activity relative to the 17a-hydroxylase activ-
`
`

`
`Phenylalanine Deletion in Human Cytochrome P—45017,,
`
`%ofcorwersion
`
`FIG. 6. Comparison of the time courses for the conversion
`of 1 uM progesterone to 17a-hydroxyprogesterone determined
`in situ and in isolated microsomes. The assays in intact COS 1
`cells (medium) and isolated microsomes (microsome) were carried out
`as described under “Materials and Methods” using 1.0 uM progester-
`one as substrate. The background was subtracted to determine the
`percent of conversion. The ratio of mutant activity (JG) to wild—type
`activity (WT) at each time point is indicated in parentheses above
`the JG line (— — —).
`
`ity impossible. However, the lower activity of the mutant
`enzyme in isolated microsomes raised the question of whether
`the amount of P-4501‘7‘:,*‘°‘53 °’ 5'“ in microsomes relative to that
`of wild-type P-45017,, in microsomes might be reduced. Im-
`munoblots (Fig. 7) showed this not to be the case, as the
`amounts of both types of P-45017,. in microsomes were ap-
`proximately the same. Therefore, we must conclude that the
`specific activity of the P-4501‘7‘f,“°‘53 °' 5“ is decreased signifi-
`cantly upon isolation of microsomes. Homogenization of COS
`1 cells also led to reduced activity of the P-450?71Z"°‘53 °’ 54’
`relative to the wild-type enzyme but only to approximately
`half the reduction found in microsomes.
`
`DISCUSSION
`
`Site—directed mutagenesis has proven to be useful in inves-
`tigating the role of specific amino acids in P-450 function
`(Atkins and Sligar, 1988; Shimizu et al., 1988). We have
`reasoned that for investigation of structure-function relation-
`ships in forms of cytochrome P-450 involved in endogenous
`substrate metabolism, natural mutations in the human pop-
`ulation will also provide a rich supply of information. Con-
`genital adrenal hyperplasia results from a heterogeneous
`group of diseases (New et al., 1983) including deficiencies in
`
`FIG. 7. Immunoblot analysis of P-45017,. in microsomes fol-
`lowing expression in COS 1 cells. WT and JG represent wild-
`type and mutant expressions, respectively. Cell, cell lysates (250 pg);
`Ms, microsomes (100 pg); Ms, 37C, 60 min, microsomes after incu-
`bation conditions used for steroid assays (100 pg). HFA Ms, human
`fetal adrenal microsomes (15