The New England Journal of Medicine
THE PATHOPHYSIOLOGY AND GENETICS OF CONGENITAL LIPOID ADRENAL
HYPERPLASIA
HIMANGSHU S. BOSE, PH.D., TERUO SUGAWARA, M.D., PH.D., JEROME F. STRAUSS III, M.D., PH.D.,
AND WALTER L. MILLER, M.D., FOR THE INTERNATIONAL CONGENITAL LIPOID ADRENAL HYPERPLASIA CONSORTIUM*
ABSTRACT
Background Congenital lipoid adrenal hyperpla-
sia results in severe impairment of steroid biosyn-
thesis in the adrenal glands and gonads that is man-
ifested both in utero and postnatally. We recently
found mutations in the gene for the steroidogenic
acute regulatory protein in four patients with this
syndrome, but it was not clear whether all patients
have such mutations or why there is substantial clin-
ical variation in these patients.
Methods We directly sequenced the gene for ste-
roidogenic acute regulatory protein in 15 patients
with congenital lipoid adrenal hyperplasia from 10
countries. Identified mutations were confirmed and
recreated in expression vectors, transfected into cul-
tured cells, and assayed for the presence and activity
of steroidogenic acute regulatory protein.
Results Fifteen different mutations in the gene for
steroidogenic acute regulatory protein were found in
14 patients; the mutation Gln258Stop was found in
80 percent of affected alleles from Japanese and
Korean patients, and the mutation Arg182Leu was
found in 78 percent of affected alleles from Palestin-
ian patients. We developed diagnostic tests for these
and eight other mutations. Thirteen of the 15 muta-
tions were in exons 5, 6, or 7, and all rendered the
steroidogenic acute regulatory protein inactive in
functional assays. Some mutants with amino acid
replacements were capable of normal mitochondrial
processing, indicating that the activity of steroido-
genic acute regulatory protein is not associated with
its translocation into mitochondria. Steroidogenic
cells lacking the protein retained low levels of steroi-
dogenesis. This explains the secretion of some ster-
oid hormones by the ovaries after puberty before af-
fected cells accumulate large amounts of cholesterol
esters.
Conclusions The congenital lipoid adrenal hyper-
plasia phenotype is the result of two separate events,
an initial genetic loss of steroidogenesis that is de-
pendent on steroidogenic acute regulatory protein
and a subsequent loss of steroidogenesis that is in-
dependent of the protein due to cellular damage
from accumulated cholesterol esters. (N Engl J Med
1996;335:1870-8.)
1996, Massachusetts Medical Society.
1870  De c e m b e r 1 9 , 1 9 9 6
The New England Journal of Medicine
Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
P
ATIENTS with congenital lipoid adrenal
hyperplasia, the most severe genetic disorder
of steroid hormone biosynthesis, have a se-
vere defect in the conversion of cholesterol
to pregnenolone, the first step in adrenal and gonad-
al steroidogenesis. Deficient fetal testicular steroido-
genesis in patients with a 46,XY karyotype results in
phenotypically normal female genitalia. The adrenal
cortex becomes engorged with cholesterol and cho-
lesterol esters; deficient adrenal steroidogenesis leads
to salt wasting, hyponatremia, hypovolemia, hyper-
kalemia, acidosis, and death in infancy,1,2 although
patients can survive to adulthood with appropriate
mineralocorticoid- and glucocorticoid-replacement
therapy.3,4 Some affected infants have immediate
signs of mineralocorticoid deficiency, but others re-
main asymptomatic for months; furthermore, affect-
ed 46,XX females may undergo feminization and
have vaginal bleeding at puberty.5 Thus, it was not
known whether the congenital lipoid adrenal hyper-
plasia syndrome was a single disease, or how a single
genetic defect could account for these clinical varia-
tions.
Affected adrenal or testicular tissues cannot con-
vert cholesterol to pregnenolone in vitro, suggesting
a defect in the cholesterol-side-chain cleavage sys-
tem,4,6-8 which consists of cytochrome P450scc and
its electron-transfer proteins adrenodoxin reductase
and adrenodoxin.9 Adrenodoxin reductase, adreno-
doxin, and several factors thought to participate in
the transport of cholesterol to mitochondria are nor-
mal in patients with congenital lipoid adrenal hyper-
plasia,10,11 so attention focused on P450scc. Howev-
er, the P450scc gene is normal in these patients10,12-14
and the synthesis of pregnenolone in the placenta (a
fetal tissue) is unaffected,15 indicating that the entire
cholesterol-side-chain cleavage system can function
normally in affected patients.
Recently, a 30-kd mouse mitochondrial protein
that appears to be a rapidly inducible, cyclohexi-
From the Department of Pediatrics, University of California at San Fran-
cisco, San Francisco (H.S.B., W.L.M.); and the Department of Obstetrics
and Gynecology, University of Pennsylvania, Philadelphia (T.S., J.F.S.). Ad-
dress reprint requests to Dr. Miller at the Department of Pediatrics, Uni-
versity of California, San Francisco, Bldg. MR-IV, Rm. 209, San Francisco,
CA 94143-0978.
*Other members of the International Congenital Lipoid Adrenal Hyper-
plasia Consortium are listed in the Appendix.
 PAT H O P H YS I O LO GY A N D G E N ET I C S O F C O N G E N I TA L L I P O I D A D R E N A L H Y P E R P L AS I A
mide-sensitive mediator of the acute steroidogenic
response16,17 was cloned and named the steroidogen-
ic acute regulatory protein.18 We cloned the human
complementary DNA (cDNA)19 and gene20 and
found that messenger RNA (mRNA) for this protein
was expressed in the adrenal glands and gonads but
not in the placenta or brain, as expected for a factor
that might cause congenital lipoid adrenal hyperpla-
sia but spare placental steroidogenesis. We found
mutations in the gene for steroidogenic acute regu-
latory protein in four affected families21,22; however,
it was not clear whether all patients with this pheno-
type have such mutations. Furthermore, correlations
between the severity of the mutation and the phe-
notype have not been possible, and it was not clear
how the mutations cause the clinical findings. To
elucidate these issues, we examined the gene for ste-
roidogenic acute regulatory protein in 15 previously
unstudied patients with congenital lipoid adrenal
hyperplasia, from various ethnic groups.
METHODS
Leukocyte genomic DNA was amplified by the polymerase
chain reaction (PCR) and sequenced directly (without cloning)
on both strands with an automated sequencer (exons 1 through
4) or manually (exons 5 through 7). The following oligonucleo-
tide primers were used: Ex1S 5TAACACAGGTTTCTGAGC-
CTCAAT3 and Ex1AS 5ATCAGAATTGGGTGGCCTGAGCC-
TC3 for exon 1, Ex2S 5GTCCCTGCTAGAATACTGTGTT3
and Ex2AS 5AAAGCCACATGCACCACATCA3 for exon 2,
Ex3S 5CAATGAGCAGACCCAGAGCT3 and Ex3AS 5GACT-
GCTGCATGAGACAGGA3 for exon 3, Ex4S 5TGCTGGGAT-
TATAGGCGTGAAC3 and Ex4AS 5GCTAGGGGTCCTCTC-
TTTGATACAG3 for exon 4, Ex5S 5TGCTGTATCAAAGAG-
AGGAC3 and Ex5AS 5AGCCTGCTGCCCGTATTTAC3 for
exon 5, and oligonucleotide B2 is 5GACCACAAGATGAGCAC-
ATTC3. Primers S3 and AS1 for exons 5 through 7 and primers
S2, S3, S4, AS1, and AS5 have been described previously.21,22
Identified mutations were reconfirmed by manual sequencing
from an independent PCR amplification and re-created in a vec-
tor expressing steroidogenic acute regulatory protein cDNA.
Primer sequences for mutagenesis and conditions for PCR have
been deposited with the National Auxiliary Publications Service
(NAPS).* The activity of the protein was determined by measur-
ing the pregnenolone produced from endogenous cholesterol in
COS-1 cells transfected with plasmids expressing the cholesterol-
side-chain cleavage system and steroidogenic acute regulatory pro-
tein, as described previously.19-22 Western immunoblotting with
rabbit antimouse antiserum against the protein was done as de-
scribed previously.19,20
RESULTS
Patients
Fifteen patients from 10 countries were studied
(Table 1). All patients had normal birth weights and
gestational ages, and all had phenotypically normal
*See NAPS document no. 05350 for 3 pages of supplementary material.
Order from NAPS, c/o Microfiche Publications, P.O. Box 3513, Grand
Central Station, New York, NY 10163-3513. Remit in advance (in U.S.
funds only) $7.75 for photocopies or $5 for microfiche. Outside the U.S.
and Canada, add postage of $4.50 for photocopies or $1.75 for microfiche.
There is a $15 invoicing charge on all orders filled before payment.
The New England Journal of Medicine
Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
female genitalia. Their plasma corticotropin and re-
nin values were high; serum cortisol and testoster-
one values varied substantially but responded poorly
to corticotropin and chorionic gonadotropin. There
were substantial variations in the degree of hypo-
natremia and hyperkalemia and in the age at onset
of symptoms, with one child surviving for six months
without hormonal replacement. At least five neo-
nates had hypoglycemia, and at least five had respi-
ratory disorders. To our knowledge, neither of these
features has been described previously in congenital
lipoid adrenal hyperplasia, but both could be caused
by glucocorticoid deficiency. At least 12 patients
had hyperpigmentation at birth, indicating intra-
uterine glucocorticoid deficiency, which caused ex-
cessive corticotropin secretion. We found 15 differ-
ent mutations in the gene for steroidogenic acute
regulatory protein, 13 of which were in exons 5
through 7, in 14 of these 15 patients. Identified
mutations were confirmed by direct sequencing of
DNA in all patients and in their parents and siblings
whenever possible.
Gln258Stop Mutations
In three Japanese patients, four of the six unrelat-
ed alleles for steroidogenic acute regulatory protein
had the mutation Gln258Stop (Table 2). When the
previously described21 homozygous Patients 16 and
19 were included, 8 of 10 affected alleles in the pa-
tients from Japan and Korea, where congenital lip-
oid adrenal hyperplasia is not so rare as it is in the
United States,4,5,14 had this same mutation, suggest-
ing a founder effect. The two other mutations in
the Japanese patients have not been found in other
patients and may represent new mutations. The
Gln258Stop mutation is easily identified by amplify-
ing genomic DNA with primers S4 and AS121 fol-
lowed by digestion with EcoRII, BstN1, or Sex A1,
since the responsible CT mutation (indicated by
the underlined letter) destroys the recognition site
ACCAGGT (Table 2); we estimate that the carrier
rate for this mutation in Japan is about 1 in 200.
Arg182Leu Mutations
Six of our patients were of Palestinian ancestry (Ta-
ble 1). Patients 5 and 6 were siblings, and Patient 4
was from a consanguineous marriage, so these six pa-
tients represented nine unique alleles, seven of which
had the Arg182Leu mutation. These apparently un-
related patients were from Jordan, Israel, Kuwait, and
Denmark. Identification of intronic polymorphisms
and other mutations within the gene for steroi-
dogenic acute regulatory protein showed that the
Arg182Leu mutation was found in various sequence
contexts, confirming that the patients were unre-
lated. The Arg182Leu mutation is easily identified
by amplifying genomic DNA with primers S3 and
Ex5AS followed by digestion with Tsp 45I (Table 2).
Vo l u m e 3 3 5 Nu m b e r 2 5  1871
 The New England Journal of Medicine

TABLE 1. CHARACTERISTICS OF PATIENTS WITH CONGENITAL LIPOID ADRENAL HYPERPLASIA AND MUTATIONS IN THE GENE
FOR STEROIDOGENIC ACUTE REGULATORY PROTEIN.

ETHNIC
GROUP, NO. OF
PATIENT RACE, OR CLINICAL SERUM SERUM SERUM PULMONARY MUTATION AFFECTED
NO.* NATIONALITY SYMPTOMS HRT SODIUM POTASSIUM GLUCOSE PROBLEMS KARYOTYPE NO. EXON ALLELES REFERENCE

age at onset mmol/liter mg/dl

1 Japanese 19 days 19 days XY 1 7 2 Kondo

et al.23
2 Japanese 1 mo 2 mo 137 5.8 92 XY 14 2 1 Kondo and
1 7 1 Saito24
3 Japanese 4 days 3 mo 130 6.7 XY 13 5 1 Kondo
1 7 1 et al.25
4 Palestinian 1 day 1 day 124 7.8 56 Respiratory XY 3 5 1
distress
5 Palestinian 1 day 5 mo 117 7.1 Low Respiratory XX 2 5 Mller
distress et al.26
2
6 Palestinian 5 days 5 days 133 6.6 XY 2 5 Mller
et al.26
7 Palestinian 7 days 12 days 120 8.0 13 XY 2 5 2
8 Palestinian 10 days 22 days 123 8.1 70 Asphyxia XY 7 5 2
2, 4 5 1
9 Palestinian 2 wk 3 wk 126 6.8 36 Asphyxia XY 4 5 2
2 5 2
10 Mexican 8 days 6 wk 126 5.6 11 XY 5 7 2
11 Greek 10 days 35 days 132 7.2 XX 6 7 2
12 British, 6 mo 118 5.9 88 XY 15 5 1
white
13 Canadian, 2 mo 4 mo 123 6.9 63 XY 11 6 1
white 12 7 1
14 Danish, 24 wk 2 mo 107 7.3 XY ? 1 Mller
white et al.26
15 Egyptian 1 wk 16 days 124 8.9 Fluid XY 8 4 1
retention
16 Korean 10 days 113 9.0 Normal XX 1 7 Lin et
al.,10,11,21
Saenger
et al.27

17 Korean 4 wk XY 1 7 2 Lin et al.,10
Saenger
et al.27
18 Korean Abortus XY 1 7 Saenger
et al.15
19 Japanese 4 wk 4 wk 128 6.9 XY 1 7 2 Hauffa
et al.,4
Lin et
al.10,21
20 U.S., white 1 day 9 days 128 7.1 13 Apnea XY 9 5 2 Lin et al.21
21 Vietnamese 4 wk 10 wk 118 8.9 92 XY 10 5 2 Tee et al.22

*Patients 1 through 15 are the subjects of this report; Patients 16 through 21 have been described previously.
HRT denotes hormone-replacement therapy.
To convert values for serum glucose to millimoles per liter, multiply by 0.05551.
The nucleotide and protein changes in mutations 1 to 10 are given in Table 2. Nucleotide and protein changes in mutations 11 to 15 are as follows:
11, C779T (Ala218Val); 12, T950C (Leu275Pro); 13, G631A (Glu169Lys); 14, 247/InsG/248 (frame shift); and 15, DNA insertion (truncated protein).
Compound heterozygotes have two entries per patient (one for each allele), homozygotes have a single entry (two alleles) except for Patients 4, 14,
and 15, whose parents were consanguineous (one allele). Patients 5 and 6 are siblings, as are Patients 16, 17, and 18.

1872  De c e m b e r 1 9 , 1 9 9 6

The New England Journal of Medicine

Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
 PAT H O P H YS I O LO GY A N D G E N ET I C S O F C O N G E N I TA L L I P O I D A D R E N A L H Y P E R P L AS I A

TABLE 2. MOLECULAR DIAGNOSIS OF CONGENITAL LIPOID ADRENAL HYPERPLASIA.

MUTATION NUCLEOTIDE PROTEIN OLIGONUCLEOTIDE RESTRICTION NORMAL MUTANT

NO. MUTATION* MUTATION* PAIR ENZYME FRAGMENTS FRAGMENTS

bp

1 C898T Gln258Stop S4/AS1 EcoRII 115, 62, 33, 133, 62,

26, 18 33, 26
2 G671T Arg182Leu S3/Ex5AS Tsp45I 223, 122 345
3 A632G Glu169Gly Ex5S/Ex5AS AluI 162, 103, 30 265, 30
4 2T593 Frame shift Ex5S/Ex5AS Sau96I 204, 91 295
or AvaII
5 940942 Deletion of Arg272 B2/AS1 FspI 107, 98 205
6 947/InsA/948 Frame shift B2/AS1 HhaI 108, 91, 6 108, 97
7 C650 Frame shift S3/Ex5AS HaeIII 99, 75, 68, 62 99, 89, 75,
68, 62
8 548/InsTT/549 Frame shift Ex4S/Ex4AS NlaIII 144, 123, 33 156, 144
9 C703T Arg193Stop Ex5S/Ex5AS HaeII 295 197, 98
10 TA @11I4 E5 Frame shift S2/Ex5S NcoI 207, 151, 74 358, 74
AS5/EX5AS

*Nucleotide and amino acid numbers are given according to the cDNA sequence19 (Genbank U17280). Codon 203
should be GCC (Ala) rather than GAC (Asp), as in the Genbank entry.
Use of the isoschizomer BstN1gives identical results; SexA1, which recognizes 7 bp, does not cut the nearby
EcoRII/BstN1 sites; hence, normal DNA yields 139- and 115-bp fragments and mutant DNA yields a 254-bp fragment.
TA @11I4E5 denotes a change from thymidine to adenine 11 bp upstream from the junction of intron 4 and exon 5.

Patient 15, an Egyptian of Coptic Christian heritage
and hence probably a member of a different gene
pool, had an unrelated frame-shift mutation.
Clustering of Mutations in the Gene for Steroidogenic
Acute Regulatory Protein
Five additional patients from assorted ethnic
groups had various mutations (Table 1). Patient 10,
a Mexican of Native American ancestry, was ho-
mozygous for a deletion of the Arg272 codon, and
Patient 11, from Greece, was homozygous for a
frame-shift mutation, but no history of consanguin-
ity was found in the families of these patients. Pa-
tient 12, a white patient from Britain, was heterozy-
gous for the insertion of a foreign DNA segment
beginning in exon 5; the mutation on the maternal
allele has not been found. Patient 13, a white patient
from Canada, was a compound heterozygote for the
Leu275Pro and Ala218Val mutations. In Patient 14,
the product of a consanguineous union, no muta-
tion was found; this patients mutation could be the
result of a promoter mutation, an uninvestigated up-
stream splicing mutation, or a mutation in some
other gene. Among 33 affected alleles we found 15
mutations, all but 2 of which affected exon 5, 6, or
7 (Table 1).
Genetic Diagnosis of Congenital Lipoid Adrenal
Hyperplasia
The Gln258Stop and Arg182Leu mutations ac-
counted for 70 to 80 percent of the mutations in the
Japanese and Palestinian patients, providing the op-
portunity for genetic screening. We developed diag-
nostic tests for these and eight other mutations (Ta-
ble 2). Genomic DNA was amplified by PCR with
primers that encompass the suspected mutation, the
DNA was then cut with a restriction endonuclease
whose recognition sequence was created or de-
stroyed by the mutation, and the products were ex-
amined by gel electrophoresis. Examples of genetic
diagnosis with Gln258Stop and Arg182Leu are
shown in Figure 1.
Activity of the Mutants
To determine whether the identified mutations
caused the patients disease and to study the struc-
tural and functional requirements of the steroi-
dogenic acute regulatory protein, we tested each
mutant in vitro. The mutations were recreated in
vectors expressing the protein and transfected into
nonsteroidogenic COS-1 cells, cotransfected with a
vector expressing the three components of the cho-
lesterol-side-chain cleavage system as a single mono-
molecular fusion protein (H2NP450sccadrenodox-
in reductaseadrenodoxinCOOH) termed F2.28
This construct optimizes the activity of cytochrome
P450scc and eliminates variations in P450scc activi-
ty due to variation in the molar ratio of P450scc
to adrenodoxin reductase or adrenodoxin.28,29 In-
cubation with 22R-hydroxycholesterol bypasses the
mitochondrial cholesterol-transport system and pro-
vides a direct index of maximal mitochondrial steroi-
dogenic capacity.30 The ratio of steroidogenic capac-
ity with endogenous cholesterol as substrate to that

Vo l u m e 3 3 5 Nu m b e r 2 5  1873

The New England Journal of Medicine

Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
 The New England Journal of Medicine

t
ec
TABLE 3. ABILITY OF MUTANT STEROIDOGENIC ACUTE

bj
nt A
REGULATORY PROTEINS TO PROMOTE STEROIDOGENESIS.*

su
N
rs

al
r
he

er
ke

ut

m
tie

th
ar

nc

ot

or
Pa

Fa
bp PLASMID VALUE
M

M
U

N
% of wild-type value
300
Control vector 142
Wild-type steroidogenic acute regulatory protein 100
200 Mutant steroidogenic acute regulatory protein
Amino acid changes
Glu169Gly 131
Arg182Leu 111
Deletion of Arg272 111
Glu169Lys 142
Ala218Val 204
Leu275Pro 245
Frame-shift mutations
100 T593 93
C650 102
947/InsA/948 102
Stop-codon mutation
Gln258Stop 164

*COS-1 cells were transfected with a vector expressing the F2 fusion of

the cholesterol-side-chain cleavage system28 and cotransfected with various
vectors expressing steroidogenic acute regulatory protein, and the ability of
A the protein to promote pregnenolone synthesis from endogenous choles-
terol was assayed as described,21,22 except that 5 mg of 22R-hydroxycholes-
terol per milliliter was used as an exogenous-substrate control. The value
for the wild-type protein (control) is set at 100 percent. The values for mu-
6)

tant proteins are the means SE of three separate experiments, each per-
nt

formed in triplicate. Pregnenolone secretion from endogenous cholesterol

ie
at

by COS-1 cells transfected with the F2 vector and the empty control vector
ec
Pa nt A (P

was 11.91.5 ng per dish. In the presence of wild-type protein, preg-

bj
su
tie N

nenolone secretion averaged 81.815.2 ng per dish.

tie 5
th 6
rs
Pa ut D

al
N her
Fa nt
M er
ke

m
ar
nc

ot
or
M
U

bp

600

400
300
200
100
B
Figure 1. Genetic Diagnosis of Congenital Lipoid Adrenal Hy-
perplasia.
Panel A shows the results of amplification of DNA carrying the
mutation commonly found in Japanese patients with congeni-
tal lipoid adrenal hyperplasia Gln258Stop. Genomic DNA
from a homozygous patient, her heterozygous parents, and a
normal subject was amplified with the primers S4 and AS1 and
cut with BstNI. The uncut DNA is 254 bp; the patient has bands
of 133 and 62 bp, the normal subject has bands of 115 and 62
bp, and the parents have both the 133- and 115-bp fragments.
Panel B shows the result of amplification of DNA carrying the
mutation commonly found in Palestinian patients with congen-
ital lipoid adrenal hyperplasia Arg182Leu. DNA from Pa-
tients 5 and 6 (who were siblings), their parents, and a normal
subject was amplified with the primers S3 and Ex5AS and cut
with Tsp45I. The patients have uncut 345-bp fragments, the
normal subject has 223- and 122-bp fragments, and the parents
are heterozygous, having 345-, 223-, and 122-bp fragments.
with 22R-hydroxycholesterol as substrate indicates
the efficiency of mitochondrial cholesterol trans-
port. In cells containing F2 and the control vector,
the level of steroidogenesis from endogenous cho-
lesterol that is independent of steroidogenic acute
regulatory protein was 14 percent of the level with
F2 and the protein. The Ala218Val and Leu275Pro
mutants had minimal activity, and the others had
essentially no detectable activity (Table 3).
To examine the structural effects of the mutations,
transfected cells were assayed by immunoblotting
with rabbit antimurine steroidogenic acute regulato-
ry protein IgG.18 The 37-kd precursor protein was
readily detectable, but not the mature 30-kd form of
the mutants resulting from Glu169Gly and the de-
letion of Arg272 (data not shown). The Arg182Leu
mutant protein was unstable and not detected. Both
the 37-kd precursor and the 30-kd mature form of
the Glu169Gly, Leu275Pro, and Ala218Val mutants
could be detected in about the same ratio as for the
wild-type protein (Fig. 2). Thus, some changes in
amino acids that ablated the activity of the protein
did not alter its mitochondrial processing.

1874  De c e m b e r 1 9 , 1 9 9 6

The New England Journal of Medicine

Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
 PAT H O P H YS I O LO GY A N D G E N ET I C S O F C O N G E N I TA L L I P O I D A D R E N A L H Y P E R P L AS I A

p
o
ys

to
al
pe

Pr

8S
9L

8V
l

75
ty
ro

16

25
21
u2
nt

ild

lu

ln
la
Co

Le
W

G
kd
Precursor 37
Mature protein 30

Figure 2. Western Blot Analysis of Steroidogenic Acute Regulatory Protein Ex-

pressed in COS-1 Cells.
COS-1 cells were transfected with a control plasmid vector (pSV-SPORT-1),
the vector harboring wild-type cDNA for steroidogenic acute regulatory pro-
tein, or the indicated mutant cDNAs (Glu169Lys, Leu275Pro, Ala218Val, and
Gln258Stop). Whole-cell extracts were subjected to sodium dodecyl sulfatepol-
yacrylamide-gel electrophoresis and Western blotting as previously described21
with a rabbit polyclonal antibody raised against a structurally conserved pep-
tide from mouse steroidogenic acute regulatory protein. The wild-type steroido-
genic acute regulatory precursor protein migrates at 37 kd, and the mature pro-
tein at 32.5 kd.

DISCUSSION lacking a mitochondrial import peptide is fully ac-

Finding mutations in the gene for the steroido-
genic acute regulatory protein in patients with con-
genital lipoid adrenal hyperplasia from various eth-
nic and genetic backgrounds establishes that these
mutations are responsible for most if not all cases of
the syndrome. The basis of the disease in Patient 14,
who had no detectable mutations in the gene for
this protein, is unknown; a mutation in SF-1, a tran-
scription factor involved in the embryonic differen-
tiation of adrenal and gonadal (but not placental)
steroidogenic cells,31,32 is a possibility. The gene for
steroidogenic acute regulatory protein is autoso-
mal,19 yet only 3 of our 21 patients were 46,XX, and
in another study, only 16 of 63 Japanese patients
with known karyotypes were 46,XX.5 These results
suggest that many affected 46,XX fetuses are lost in
early pregnancy, that affected 23,X sperm are less
likely to fertilize an egg, or that 23,X sperm are pro-
duced at disproportionately lower frequencies than
23,Y sperm. However, an ascertainment bias cannot
be ruled out.
Previous studies suggested that the active form of
steroidogenic acute regulatory protein is the 37-kd
precursor, stimulating steroidogenesis by forming
contact sites between the outer and inner membranes
as it enters the mitochondria.18,33 However, the de-
letion of only 28 carboxy-terminal amino acids in
the Gln258Stop mutation commonly found in Jap-
anese patients eliminates all activity,21 and some in-
active mutants undergo normal mitochondrial proc-
essing (Fig. 2). Thus, the carboxy-terminal half of
the protein is crucial for activity. We found missense
mutations only in exons 5, 6, and 7. Thus, either ex-
ons 1, 2, 3, and 4 are less prone to mutation or mis-
sense mutations in this region are phenotypically si-
lent. Because steroidogenic acute regulatory protein
tive,34 we favor the latter explanation.
The clinical descriptions of congenital lipoid
adrenal hyperplasia are remarkably consistent: female
external genitalia, neonatal hyponatremia, hyperkale-
mia, and dehydration.1-4,6-8 Minimal degrees of pos-
terior labial fusion or of clitorimegaly, which would
reflect androgen action in early or late gestation, re-
spectively, were not seen in our patients; thus, all
46,XY patients had a profound impairment of testos-
terone synthesis. By contrast with this consistent
genital phenotype, the severity, manifestations, and
age at onset of clinically apparent mineralocorticoid
and glucocorticoid deficiency varied considerably.
Most infants had vomiting, dehydration, hypoten-
sion, failure to thrive, hyponatremia, and hyperkale-
mia within two weeks after birth, but Patients 3, 5,
12, and 13 survived for three months or more with-
out hormone-replacement therapy. Hyperpigmenta-
tion, a sign of corticotropin hypersecretion, was seen
in two thirds of the newborns we studied, and about
one fourth had neonatal hypoglycemia and compro-
mised pulmonary development, both associated with
glucocorticoid deficiency. Patient 13, who had a
46,XY karyotype and was a compound heterozygote
for two amino acid replacements that allowed mini-
mal steroidogenic acute regulatory protein activity,
survived for four months without hormone-replace-
ment therapy but had no evidence of fetal testos-
terone production. Thus, the testicular lesion in
congenital lipoid adrenal hyperplasia appears to be
substantially more severe than the adrenal lesion. In
sharp contrast with this profound reduction in testic-
ular steroidogenesis, some affected 46,XX patients
undergo feminization and have vaginal bleeding at
puberty.5
We hypothesize that there are two lesions in con-

Vo l u m e 3 3 5 Nu m b e r 2 5  1875

The New England Journal of Medicine

Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
 The New England Journal of Medicine

A Normal B Early Congenital Lipoid C Late Congenital Lipoid

Adrenal Hyperplasia Adrenal Hyperplasia

Low-density Corticotropin Low-density Low-density

Corticotropin lipoprotein lipoprotein Corticotropin lipoprotein

ATP
cAMP cAMP
cAMP ATP
ATP Steroid Steroid
Lysosome Lysosome
Lipid
droplet
StAR
Lipid
droplet

StAR StAR

Lipid
Mitochondrion ? droplet
? StAR-independent
StAR-independent cholesterol
cholesterol flow
flow
Endoplasmic
reticulum
Nucleus Nucleus Nucleus

Endoplasmic
reticulum

Figure 3. Model of Congenital Lipoid Adrenal Hyperplasia in an Adrenal Cell.

In the normal cell (Panel A), cholesterol is derived by endogenous synthesis from acetylcoenzyme A in the endoplasmic reticulum,
from cholesterol esters stored in lipid droplets, and from low-density lipoprotein cholesterol, which after receptor-mediated en-
docytosis, is processed in lysosomes before it is used or stored in lipid droplets. Cholesterol is transported to the outer mitochon-
drial membrane by ill-defined processes involving the cytoskeleton.35 The rate-limiting step in steroidogenesis is the movement of
cholesterol from the outer to the inner mitochondrial membrane; this can be promoted by steroidogenic acute regulatory protein
(StAR), but may also be mediated by mechanisms independent of this protein. Thus, the net synthesis of steroid is due to mech-
anisms dependent on, as well as independent of, the protein. In Panel B, in the absence of steroidogenic acute regulatory protein,
as in early congenital lipoid adrenal hyperplasia or in a placental cell, mechanisms independent of the protein can still move some
cholesterol into the mitochondria, resulting in a low level of steroidogenesis. In patients with affected adrenal cells this results in
increased corticotropin secretion, stimulating further production of cholesterol and its accumulation as cholesterol esters in lipid
droplets. In Panel C, as lipid droplets accumulate they engorge the cell, damaging its cytoarchitecture through both physical dis-
placement and the chemical action of cholesterol auto-oxidation products. Steroidogenic capacity is destroyed, and consequently
tropic stimulation continues. In the ovary, follicular cells remain unstimulated and, hence, undamaged until they are recruited at
the beginning of each menstrual cycle. Small amounts of estradiol are produced, as shown in Panel B, effecting phenotypic femi-
nization and vaginal bleeding, but the cycles are anovulatory, resulting in infertility and progressive hypergonadotropic hypogo-
nadism. cAMP denotes cyclic AMP.

genital lipoid adrenal hyperplasia (Fig. 3). First, a
mutant steroidogenic acute regulatory protein pre-
vents the acute steroidogenic response in the fetal
testis and adrenal gland. This destroys the steroidal
responses to tropic stimulation that are dependent
on the protein but permits basal steroidogenesis that
is independent of the protein, as occurs in the pla-
centa15 and in COS-1 cells transfected only with the
cholesterol-side-chain cleavage system (Table 3). Sec-
ond, the accumulation of cholesterol esters and
sterol auto-oxidation products in affected cells dam-
ages the cell and eventually disrupts the basal steroi-
dogenesis that is independent of steroidogenic acute
regulatory protein. Thus, the fetal testis, which is
stimulated by chorionic gonadotropin in early gesta-
tion, is severely affected early in gestation, so that
virilization does not occur; this was the case even in
Patient 13, who had some steroidogenic acute regu-
latory protein activity. Similarly, the fetal zone of the
adrenal gland, which secretes large amounts of de-

1876  De c e m b e r 1 9 , 1 9 9 6

The New England Journal of Medicine

Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
 PAT H O P H YS I O LO GY A N D G E N ET I C S O F C O N G E N I TA L L I P O I D A D R E N A L H Y P E R P L AS I A
hydroepiandrosterone for placental conversion to
estriol, is also profoundly affected, leading to low
plasma estriol concentrations in women carrying af-
fected fetuses.
By contrast, the definitive zone of the fetal adrenal
gland makes relatively small amounts of steroids but
probably develops into the adrenal zonae glomeru-
losa and fasciculata after birth.36 These zones are
variably affected by the accumulation of cholesterol
esters, leading to the synthesis of small but detect-
able amounts of steroid hormones in affected new-
borns and permitting survival for a brief time, until
the accumulation of cholesterol esters finally leads to
the destruction of residual adrenal steroidogenesis.
In contrast to the testis and adrenal gland, the fetal
ovary lacks steroidogenic enzymes and steroidogenic
capacity.37 Since it remains unstimulated until puber-
ty, it does not accumulate cholesterol esters. The on-
set of puberty then stimulates the maturation of in-
dividual ovarian follicles, leading to some estrogen
synthesis by the ovaries that is independent of steroi-
dogenic acute regulatory protein in affected 46,XX
females. The accumulation of ovarian cholesterol es-
ters eventually destroys the steroidogenic capacity of
the stimulated follicles that is independent of steroi-
dogenic acute regulatory protein, but previously un-
stimulated follicles are recruited in subsequent cycles,
permitting a low level of steroidogenesis associated
with anovulatory cycles, progressive ovarian failure,
and hypergonadotropic hypogonadism.
Thus, elucidation of the genetics and cell biology
of the mutant steroidogenic acute regulatory protein
has suggested a model to explain the phenotypic
variations in congenital lipoid adrenal hyperplasia.
Supported by grants from the National Institutes of Health (DK37922
and DK42154 to Dr. Miller and HD06274 to Dr. Strauss) and a grant
from the March of Dimes (to Dr. Miller).
We are indebted to Dr. Takuma Kondo, Kondo Clinic, Osaka, Ja-
pan, for the clinical data on Patients 1, 2, and 3; to Dr. Songya
Pang, Department of Pediatrics, University of Illinois, Chicago, for
the samples from and data on Patient 10; to Drs. David C.L. Sav-
age and John Barton, Bristol Royal Hospital for Sick Children, Bris-
tol, United Kingdom, for the data on and samples from Patient 13;
and to Dr. Douglas M. Stocco, Department of Biochemistry, Texas
Tech University, Lubbock, for the antibody against mouse steroido-
genic acute regulatory protein.
APPENDIX
Other members of the International Congenital Lipoid Adre-
nal Hyperplasia Consortium are as follows: Kenji Fujieda, M.D.,
Hokkaido University, Sapporo, Japan; Ziva Ben-Neriah, M.D.,
and Ariel Rsler, M.D., Hebrew University Hadassah Medical
Center, Jerusalem, Israel; Jorn Mller, M.D., Marianne Schwartz,
Ph.D., and Niels E. Skakkebaeck, M.D., National University Hos-
pital, Copenhagen, Denmark; Marlin Nino Nawas, M.D., Khalidi
Hospital, Amman, Jordan; Anastasios Papadimitriou, M.D., Pen-
teli Childrens Hospital, Athens, Greece; Jeremy S.D. Winter,
M.D., University of Alberta, Edmonton, Canada; Christopher
T. Cowell, M.D., Royal Alexandria Hospital for Sick Children,
Paramatta, Australia; and Gary Warne, M.D., Royal Childrens
Hospital, Melbourne, Australia.
The New England Journal of Medicine
Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.
REFERENCES
1. Sandison AT. A form of lipoidosis of the adrenal cortex in an infant.
Arch Dis Child 1955;30:538-41.
2. Prader A, Gurtner HP. Das Syndrom des Pseudohermaphroditismus
masculinus bei kongenitaler Nebennierenrinden-Hyperplasie ohne Andro-
genberproduktion (adrenaler Pseudohermaphroditismus masculinus).
Helv Paed Acta 1955;10:397-412.
3. Kirkland RT, Kirkland JL, Johnson CM, Horning MG, Librik L, Clay-
ton GW. Congenital lipoid adrenal hyperplasia in an eight-year-old pheno-
typic female. J Clin Endocrinol Metab 1973;36:488-96.
4. Hauffa BP, Miller WL, Grumbach MM, Conte FA, Kaplan SL. Congen-
ital adrenal hyperplasia due to deficient cholesterol side-chain cleavage ac-
tivity (20, 22 desmolase) in a patient treated for 18 years. Clin Endocrinol
(Oxf ) 1985;23:481-93.
5. Matsuo N, Tsuzaki S, Anzo M, Ogata T, Sato S. The phenotypic defi-
nition of congenital lipoid adrenal hyperplasia: analysis of the 67 Japanese
patients. Horm Res 1994;41:Suppl:106. abstract.
6. Camacho AM, Kowarski A, Migeon CJ, Brough AJ. Congenital adrenal
hyperplasia due to a deficiency of one of the enzymes involved in the bio-
synthesis of pregnenolone. J Clin Endocrinol Metab 1968;28:153-61.
7. Degenhart HJ, Visser HKA, Boon H, ODoherty NJ. Evidence for de-
ficiency of 20a-cholesterol-hydroxylase activity in adrenal tissue of a pa-
tient with lipoid adrenal hyperplasia. Acta Endocrinol 1972;71:512-8.
8. Koizumi S, Kyoya S, Miyawaki T, Kidani H, Funabashi T. Cholesterol
side-chain cleavage enzyme activity and cytochrome P-450 content in adre-
nal mitochondria of a patient with congenital lipoid adrenal hyperplasia
(Prader disease). Clin Chim Acta 1977;77:301-6.
9. Miller WL. Molecular biology of steroid hormone synthesis. Endocr
Rev 1988;9:295-318.
10. Lin D, Gitelman SE, Saenger P, Miller WL. Normal genes for the cho-
lesterol side chain cleavage enzyme, P450scc, in congenital lipoid adrenal
hyperplasia. J Clin Invest 1991;88:1955-62.
11. Lin D, Chang YJ, Strauss JF III, Miller WL. The human peripheral
benzodiazepine receptor gene: cloning and characterization of alternative
splicing in normal tissues and in a patient with congenital lipoid adrenal
hyperplasia. Genomics 1993;18:643-50.
12. Matteson KJ, Chung BC, Urdea MS, Miller WL. Study of cholesterol
side-chain cleavage (20,22 desmolase) deficiency causing congenital lipoid
adrenal hyperplasia using bovine-sequence P450scc oligodeoxyribonucle-
otide probes. Endocrinology 1986;118:1296-305.
13. Sakai Y, Yanase T, Okabe Y, et al. No mutation in cytochrome P450
side chain cleavage in a patient with congenital lipoid adrenal hyperplasia.
J Clin Endocrinol Metab 1994;79:1198-201.
14. Fukami M, Sato S, Ogata T, Matsuo N. Lack of mutations in P450scc
gene (CYP11A) in six Japanese patients with congenital lipoid adrenal hy-
perplasia. Clin Pediatr Endocrinol 1995;4:39-46.
15. Saenger P, Klonari Z, Black SM, et al. Prenatal diagnosis of congenital
lipoid adrenal hyperplasia. J Clin Endocrinol Metab 1995;80:200-5.
16. Epstein LF, Orme-Johnson NR. Regulation of steroid hormone bio-
synthesis: identification of precursors of a phosphoprotein targeted to the
mitochondrion in stimulated rat adrenal cortex cells. J Biol Chem 1991;
266:19739-45.
17. Stocco DM, Sodeman TC. The 30-kDa mitochondrial proteins in-
duced by hormone stimulation in MA-10 mouse Leydig tumor cells are
processed from larger precursors. J Biol Chem 1991;266:19731-8.
18. Clark BJ, Wells J, King SR, Stocco DM. The purification, cloning, and
expression of a novel luteinizing hormone-induced mitochondrial protein
in MA-10 mouse Leydig tumor cells: characterization of the steroidogenic
acute regulatory protein (StAR). J Biol Chem 1994;269:28314-22.
19. Sugawara T, Holt JA, Driscoll D, et al. Human steroidogenic acute
regulatory protein: functional activity in COS-1 cells, tissue-specific expres-
sion, and mapping of the structural gene to 8p11.2 and a pseudogene to
chromosome 13. Proc Natl Acad Sci U S A 1995;92:4778-82.
20. Sugawara T, Lin D, Holt JA, et al. Structure of the human steroido-
genic acute regulatory protein (StAR) gene: StAR stimulates mitochondrial
cholesterol 27-hydroxylase activity. Biochemistry 1995;34:12506-12.
21. Lin D, Sugawara T, Strauss JF III, et al. Role of steroidogenic acute
regulatory protein in adrenal and gonadal steroidogenesis. Science 1995;
267:1828-31.
22. Tee MK, Lin D, Sugawara T, et al. TA transversion 11 bp from a
splice acceptor site in the gene for steroidogenic acute regulatory protein
causes congenital lipoid adrenal hyperplasia. Hum Mol Genet 1995;4:
2299-305.
23. Kondo T, Saito H, Nagahara S, Kobayashi Y. Clinical findings and go-
nadal pathology in three cases of congenital lipoid adrenal hyperplasia
(Prader syndrome). Clin Endocrinol 1986;34:322-5. (In Japanese.)
24. Kondo T, Saito H. Hormonal status in a patient with congenital lipoid
adrenal hyperplasia and hyperpigmentation. Jpn J Pediatr 1986;34:Suppl:
1223-8. (In Japanese.)
Vo l u m e 3 3 5 Nu m b e r 2 5  1877
 The New England Journal of Medicine

25. Kondo T, Nagahara S, Kobayashi Y. A case of congenital lipoid adrenal
hyperplasia complicated by Wilsons disease. Clin Endocrinol 1993;41:
Suppl:163-7. (In Japanese.)
26. Mller J, Torsson A, Damkjaer Nielsen M, Petersen KE, Christoffersen
J, Skakkebaek NE. Gonadal development and growth in 46,XX and 46,XY
individuals with P450scc deficiency (congenital lipoid adrenal hyperplasia).
Horm Res 1991;36:203-8.
27. Saenger P, Lin D, Gitelman SE, Miller WL. Congenital lipoid adrenal
hyperplasia genes for P450scc, side chain cleavage enzyme, are normal.
J Steroid Biochem Mol Biol 1993;45:87-97.
28. Harikrishna JA, Black SM, Szklarz GD, Miller WL. Construction and
function of fusion enzymes of the human cytochrome P450scc system.
DNA Cell Biol 1993;12:371-9.
29. Zuber MX, Mason JI, Simpson ER, Waterman MR. Simultaneous
transfection of COS-1 cells with mitochondrial and microsomal steroid hy-
droxylases: incorporation of a steroidogenic pathway into nonsteroidogenic
cells. Proc Natl Acad Sci U S A 1988;85:699-703.
30. Toaff ME, Schleyer H, Strauss JF III. Metabolism of 25-hydroxycho-
lesterol by rat luteal mitochondria and dispersed cells. Endocrinology
1982;111:1785-90.
31. Luo X, Ikeda Y, Parker KL. A cell-specific nuclear receptor is essential
for adrenal and gonadal development and sexual differentiation. Cell 1994;
77:481-90.
32. Hatano O, Takayama K, Imai T, et al. Sex-dependent expression of a
transcription factor, Ad4BP, regulating steroidogenic P-450 genes in the
gonads during prenatal and postnatal rat development. Development
1994;120:2787-97.
33. Stocco DM, Clark BJ. Regulation of the acute production of steroids
in steroidogenic cells. Endocr Rev 1996;17:221-44.
34. Arakane F, Sugawara T, Nishino H, et al. Steroidogenic acute regula-
tory protein (StAR) retains activity in the absence of its mitochondrial tar-
geting sequence: implications for the mechanism of StAR action. Proc Natl
Acad Sci U S A 1996;93:13731-6.
35. Hall PF. The roles of microfilaments and intermediate filaments in the
regulation of steroid synthesis. J Steroid Biochem Mol Biol 1995;55:601-
5.
36. Mesiano S, Coulter CL, Jaffe RB. Localization of cytochrome P450
cholesterol side-chain cleavage, cytochrome P450 17a-hydroxylase/
17, 20 lyase, and 3b-hydroxysteroid dehydrogenase isomerase steroido-
genic enzymes in human and rhesus monkey fetal adrenal glands:
reappraisal of functional zonation. J Clin Endocrinol Metab 1993;77:
1184-9.
37. Voutilainen R, Miller WL. Developmental expression of genes for the
steroidogenic enzymes P450scc (20,22-desmolase), P450c17 (17a-
hydroxylase/17,20-lyase), and P450c21 (21-hydroxylase) in the human
fetus. J Clin Endocrinol Metab 1986;63:1145-50.

1878  De c e m b e r 1 9 , 1 9 9 6

The New England Journal of Medicine

Downloaded from nejm.org on February 27, 2015. For personal use only. No other uses without permission.
Copyright 1996 Massachusetts Medical Society. All rights reserved.