Endocrine Journal 2008, 55 (6), 1043–1049

Ovarian Histological Findings in an Adult Patient with the Steroidogenic Acute Regulatory Protein (StAR) Deficiency Reveal the Impairment of Steroidogenesis by Lipoid Deposition

UIKO KAKU, KAORI KAMEYAMA*, MASAKO IZAWA, MAKOTO YAMADA, JUNKO MIYAMOTO,

TAKASHI SUZUKI**, HIRONOBU SASANO** AND YUKIHIRO HASEGAWA

Department of Endocrinology and Metabolism, Tokyo Metropolitan Kiyose Children’s Hospital, 1-3-1 Umezono, Kiyose, Tokyo

204-8567, Japan

*Division of Diagnostic Pathology, Keio University School of Medicine 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan

**Department of Pathology, Tohoku University School of Medicine, 1-1 Seiryou, Aoba-ku, Sendai 980-8575, Japan

Abstract. Context: The steroidogenic acute regulatory protein (StAR) is essential for the production of steroid hormones.

The mutations in the StAR gene typically cause congenital lipoid adrenal hyperplasia (lipoid CAH), characterized by

severe adrenal insufficiency in both sexes and complete female external genitalia in genetic males. Affected 46, XX

females feminize at puberty and menstruate but have progressive hypergonadotropic hypogonadism. It has been

hypothesized that the cholesterol accumulation in the steroidogenic cells destroys the residual steroidogenic capacity and

progressive ovarian failure occurs (two-hit model). Additionally, ovulation and luteinization in the patients is supposed to

be impaired. However, those hypotheses have not been confirmed histologically. Objective: We examined whether

pathological findings of the ovary in a patient of lipoid CAH corresponded with two-hit model, and whether ovulation and

luteinization occurred or not in the patient. Subject: The ovary in an adult 46, XX female with a homozygous nonsense

mutation (Q258X) in the StAR gene was examined. When the patient was age 22 yr, the ovary was resected because of

enlargement with polycysts and subsequent torsion. Result: The affected ovary demonstrated remarkable lipoid deposition

and changes of the mitochondrial ultrastructure. Immunohistochemical examination showed decrease of steroidogenic

enzymes such as P450 cholesterol side-chain cleavage (P450scc). Additionally, we detected corpus albicans in the

affected ovary. Conclusion: This is the first detailed report on ovarian histology in an adult 46, XX female with a null type

mutation of the StAR gene (Q258X), which indicates the evidence of the impairment of ovarian StAR-independent

steroidogenesis by lipoid deposition.

Key words: StAR, Ovulation, Corpus albicans, Two-hit model

(Endocrine Journal 55: 1043–1049, 2008)

STEROIDOGENIC acute regulatory protein (StAR) is essential for the production of steroid hormones,

mediating the transport of cholesterol from the outer

to the inner mitochondrial membrane in steroidogenic

tissues. A disorder due to StAR mutations, a clinical

entity traditionally called congenital lipoid adrenal

hyperplasia (lipoid CAH), results in severe impairment

of steroid biosynthesis in the adrenal glands and

gonads. In patients with null type mutations, clinical

manifestations of adrenal insufficiency become appar-

ent in the first few weeks of life. 46, XY individuals

have female external genitalia [1], and 46, XX indi-

Received: April 5, 2008

Accepted: August 5, 2008

Correspondence to: Uiko KAKU, M.D., Department of Endo-

crinology and Metabolism, Tokyo Metropolitan Kiyose Chil-

dren’s Hospital, 1-3-1 Umezono, Kiyose, Tokyo 204-8567, Japan

Abbreviation: StAR, steroidogenic acute regulatory protein;

lipoid CAH, congenital lipoid adrenal hyperplasia; P450scc, P450

cholesterol side-chain cleavage; P450c17, 17α-hydroxylase/

17,20-lyase; 3βHSD, 3β-hydroxysteroid dehydrogenase; Ad4BP/

SF-1, adrenal 4 binding protein/steroidogenic factor-1

KAKU et al.1044

viduals can have breast development and menstruation

at puberty spontaneously [2–4]. Despite null StAR

function, enough estrogen is produced in the ovary to

induce puberty in this disorder.

A two-hit model of lipoid CAH, which partially

explains these clinical phenotypes, was proposed by

Bose; the first hit is the loss of steroid hormone

synthesis which depends on StAR, resulting in hypo-

steroidogenesis. The second hit is the accumulation of

cholesterol, which damages the steroidogenic cells and

eventually disrupts StAR-independent steroidogenesis

[5]. The fetal testis, which synthesizes testosterone in

early gestation, is severely affected, so that virilization

does not occur in affected 46, XY patients with null

type mutations. On the other hand, the fetal ovary

lacks steroidogenic capacity, so cholesterol esters do

not accumulate until puberty. StAR-independent

estrogen synthesis in the ovaries first leads to breast

development and vaginal bleeding [2–4], followed by

a decrease in that residual synthesis with a progressive

accumulation of cholesterol esters.

Thus, the ovary in adult patient with complete StAR

deficiency (complete first hit) is the most appropriate

tissue to verify the second hit which is the disruption

of StAR-independent steroidogenesis. However, his-

tological analysis of the ovary was reported only by

Tanae et al., who demonstrated two female individuals

with lipoid adrenal hyperplasia had bilateral ovarian

cysts and lipid laden cell tumor after puberty [6].

Ovulation and luteinization in 46, XX patients with

null type mutations is supposed to be impaired because

of the following three reasons. First, in the ovary of

StAR knockout mice at 8 weeks of ages, no corpus

luteum was detected [7]. Second, the basal body tem-

perature was non-biphasic and urinaly pregnanediol

was undetectable in the affected human after puberty

begun [2]. Third, successful pregnancy has not been

reported in either StAR KO mice or affected human

females with null type mutations. However, impair-

ment of the corpus luteum formation and anovulation

have not been confirmed in the affected adult human

female.

Hence, we report in detail on the ovarian of an adult

46, XX female with complete StAR deficiency for the

first time. The findings strongly support the two-hit

model of lipoid CAH and demonstrate ovarian lutein-

ization.

Material and Methods

Case report

The patient, a Japanese female, was born after 41

weeks of an uneventful gestation with a birth weight of

3500 g. At the age of 2 days, pigmentation of her lips

was noted. One week later, the baby was admitted be-

cause of vomiting. Initial laboratory studies revealed

hyponatremia (121 mEq/l) and hyperkalemia (7.0

mEq/l). Serum cortisol level was low (5.2 μg/dl) and

plasma ACTH concentration was extremely high

(1746 pg/ml). On the basis of these findings, lipoid

CAH was suspected. A therapy with hydrocortisone

and fluorocortisone was instituted. A karyotype ob-

tained at this time was 46, XX. Later, we detected a

homozygous nonsense mutation (Q258X) in the StAR

gene. Indeed, the expression of the StAR was not

detected in the ovary of this patient by the immuno-

histochemical analysis in this study (data not shown).

This mutation is the most common one in Japanese pa-

tients with lipoid CAH [8]. Menarche was observed at

the age of 11 years and 3 months and her menstrual

cycle was regular for the first seven years. The basal

body temperature was not checked during her puberty.

Right multiple ovarian cysts were detected by ultra-

sonography at the age of 12 yr, when, basal levels of

luteinizing hormone (LH) and follicle-stimulating

hormone (FSH) were 17.6 mIU/ml and 7.5 mIU/ml,

respectively (LH, normal range 1.8–7.6 mIU/ml, FSH,

normal range 5.2–14.8 mIU/ml at pubertal stage) and

serum estradiol concentration was 17 pg/ml. After

that, basal LH levels were elevated to 40 mIU/ml,

whereas serum FSH levels remained within the normal

range. As she menstruated irregularly, replacement

therapy with progesterone was started at the age of

18 yr. By this age, breast development was at Tanner

stage 5, pubic hair was at Tanner stage 2, and axillary

hair was not found. Following the initiation of the

treatment, menstrual cycle became regular again. At

the age of 19 yr, she complained of acute pelvic and

abdominal pain and ultrasonography demonstrated an

enlarged right ovary (8.5 × 6.0 × 8.0 cm) with many

cysts. At laparotomy, we found complete torsion of

the right adnexa, and the left ovary was also enlarged

(about 5.0 × 2.8 × 4.0 cm) with multiple cysts. A right

oophorectomy was performed. One month later, treat-

ment with estrogen and progesterone were initiated to

prevent further enlargement of the left ovarian cysts.

OVARIAN HISTOLOGY IN StAR DEFICIENCY 1045

After the initiation of the treatment, basal LH and FSH

levels determined at 3-month intervals ranged from 9.5

to 22.2 mIU/ml and from 4.5 to 12.8 mIU/ml, respec-

tively and the sizes of the cysts and ovary were not

diminished. Serum estradiol concentration, which had

fluctuated from 17 to 60 pg/ml until the age of 19 yr,

were decreased and remained below 12 pg/ml after the

age of 19 years and 10 months. By the age of 22 yr,

she was admitted to the hospital four times because of

abdominal pain and adrenal crisis due to torsion of the

left enlarged ovary. Thus, left oophorectomy and its

histological analysis were performed after obtaining

informed consent.

Histological examination

Ovarian tissue was fixed in Bouin’s solution and

embedded in paraffin, according to routine techniques,

and cut at 4–7 μm. Sections were deparaffinized and

stained with hematoxylin and eosin.

Immunohistochemical analysis of P450 cholesterol

side-chain cleavage (P450scc), 17α-hydroxylase/17,20-

lyase (P450c17), 3β-hydroxysteroid dehydrogenase

(3βHSD) and adrenal 4 binding protein/steroidogenic

factor-1 (Ad4BP/SF-1) was performed as described

previously elsewhere [9–11]. For normal control, im-

munohistochemical analysis of these enzymes was

performed in a morphologically normal adult human

ovary.

For electron microscopic analyses, ovary was

trimmed free of fat, fixed in 0.1 M sodium phosphate

(pH 7.4) containing 1% paraformaldehyde and 2.5%

glutaraldehyde, prestained by use of 1% osmium

tetroxide in the above buffer, embedded in epoxy res-

in, and cut in 70-nm sections with an Ultratome 3.

Thin sections stained with uranyl acetate and lead

citrate were examined with a JEOL-1200EX trans-

mission electron microscope (Nihon Denshi, Tokyo).

Results

The left ovary was the size of 5.7 × 5.0 × 2.8 cm. At

least 10 cysts of 2–3 mm in diameter and some hemor-

rhagic regions were observed in the ovary. In spite of

several episodes of torsion, no remarkable necrotic

change was detected.

Histological findings support the two-hit model

The ovary contained abundant lipid deposits, prima-

rily in the theca cells and the stromal cells (Fig. 1).

First, to assess the lipoid accumulation mediated

damage to organelles, we examined the intracellular

ultrastructure of the theca cells. Electron microscopy

demonstrated many lipid droplets of approximately 1–

2 μm diameter accumulated within the cytosol, and the

number of mitochondria was decreased as compared

with normal cells (Fig. 2). Additionally, in some mito-

chondrias their bodies were swollen and their ultra-

structures were perturbed. Remarkable abnormality of

the ultrastructure was not seen in other organelles.

Next, we examined expression of steroidogenic

enzymes and Ad4BP/SF-1 by immunostaining of the

ovarian tissue. P450c17 were detected within the theca

interna cells and the stromal cells with lower expres-

sion level compared with normal ovary. The expres-

sion of 3βHSD and P450scc, which were detected in

the theca cells of normal ovary, were not observed in

the cells of the affected ovary. The immunohisto-

chemical localization and amount of Ad4BP/SF-1

were the same as normal ovary (Fig. 3).

Histological findings demonstrate luteinization

Oocytes with follicles were located in the cortex of

the ovary, and the amount of them was decreased com-

Fig. 1. Hematoxylin and eosin staining of ovarian tissue. The

theca cells had lipoid deposits (white arrow) and the

granulosa cells (arrow head) had no lipid. (magnifica-

tion, ×400)

KAKU et al.1046

pared with those of normal adult female. Although the

development up to secondary follicle with insufficient

development of the theca cells and the granulosa cells

was seen, graafian follicles were not detected in the

ovary. Similarly to StAR KO mice, no corpus luteum

was detected. However, we found corpus albicans,

which is a regressed form of corpus luteum. The size

and the number of corpus albicans were slightly small-

er and less than these in normal ovary (Fig. 4).

Discussion

The histological findings of the ovary in our patient

with complete StAR deficiency support the second hit

of two-hit model theory, that is, lipoid deposition in

the theca cells causes disruption of the StAR-indepen-

dent steroidogenesis. Similarly to past reports about

ovarian histology of human patients [6] and StAR KO

mice [7], abundant lipoid deposition was detected in

the adult female ovary, primarily in the theca cell,

where steroidogenesis begins during preovulatory

phase. In the theca cells mitochondrial ultrastructures

were changed. This change is suspected to result from

not ovarian torsion but lipoid deposition because the

granulosa cells had no deformed mitochondria (date

not shown).

Immunohistochemical analysis in this study demon-

strated that lipoid accumulation damaged not only

mitochondria but also other organelles. Diminished

expression of P450scc represented mitochondrial dys-

function, which was consistent with the findings by

electron microscopy. Furthermore, the expressions of

P450c17 and 3βHSD were decreased. These enzymes

are generally expressed in the microsome of the ste-

roidogenic cells. Because the expression of Ad4BP/

SF-1 was normal, it was speculated that the nucleus

was not severely damaged.

One study exhibited high levels of the P450scc in

the ovaries of StAR KO mice compared with the wild-

type ovaries at 8 weeks of age [7]. The reason the ex-

pression of P450scc, which was not observed in our

patient, was detected in StAR KO mice at 8 weeks is

currently unknown. One possible explanation is that

mitochondria damage by the accumulation of choles-

terol was more severe in the ovary of our patient than

StAR KO mice at 8 weeks. Indeed, serum estradiol

had been below the measurable level for half a year

before the ovary was resected in our patient, whereas

estradiol levels did not differ significantly from wild-

type values in StAR KO mice at 8 weeks [7]. Addi-

tionally, we speculated that the reason of the increased

expression of P450scc in KO mice was due to stimula-

tion by LH [12, 13].

The presence of the corpus luteum or corpus albi-

cans in patients with congenital lipoid CAH has not

been described in the past. Hasegawa et al. reported

the ovaries of StAR KO mice exhibited impaired folli-

cular maturation and absent corpus luteum at 8 weeks

of age, when sexual maturation normally has occurred.

In this case, we revealed the StAR deficient ovary had

corpus albicans. Corpus albicans is a regressed form

of corpus luteum, which normally develops from an

ovarian follicle following ovulation. Thus, the pres-

Fig. 2. Mitochondrial ultrastructure of the theca cell. Panel A indicates lower magnification view of the theca cell, whereas panel B

shows a high-power view of mitochondrial ultrastructure. L, lipid droplet; M, mitochondria. Multiple small lipid droplets of

approximately 1 to 2 μm diameter were detected in the theca cell. Mitochondria was swollen and the ultrastructure was

distorted.

OVARIAN HISTOLOGY IN StAR DEFICIENCY 1047

ence of the corpus albicans gave an indication of the

possibility of ovulation in the patient with StAR defi-

ciency. However, the indication is not infallible. A

rare disease, luteinized unruptured follicle syndrome,

is known to undergo luteinization without ovulation

[14]. In addition, progesterone receptor-null mice are

able to form corpus luteum containing trapped oocyte

[15]. Thus, because we had not monitored ovarian

follicle directly or ultrasonographically, we could not

confirm whether or not ovulation had actually oc-

curred in our patient.

Low level of progesterone does not necessarily

demonstrate the failure of luteinization. Despite the

formation of corpus albicans and presumably corpus

luteum, in our patient all of serum progesterone levels

had always been undetectable (date not shown) from

early in puberty as in past reports [2, 7]. Similarly, no-

biphasic basal body temperature does not necessarily

Fig. 3. Immunohistochemistry for human p450c17, 3βHSD, p450scc, and Ad4BP/SF-1 of the ovaries. A–D, normal ovary; E–H,

ovary of the patient with StAR deficiency. A and E, immunohistochemistry for p450c17; B and F, for 3βHSD; C and G, for

p450scc; D and H, for Ad4BP/SF-1. In the normal ovary, p450c17 remarkably expressed in the theca cells (A), but in the

ovary of the patient, the expression was decreased (E). Expression of 3βHSD and p450scc were detected mainly in the theca

cells and slightly in granulosa cells in the normal ovary (B and C), but not detected in the theca cells of affected ovary (F and

G). Expression of Ad4BP/SF-1 in the theca cells was comparable in normal ovary (D) and affected ovary (H). (magnification,

×200)

KAKU et al.1048

deny luteinization.

Increased secretion of LH is speculated to be re-

sponsible for the development of ovarian cyst in the

patient of lipoid CAH. The transgenic mice over-

expressing LH showed cysts formation and marked

enlargement of ovaries [16]. Additionally, hormonal

replacement initiated at the age of 14 yr diminished or

suppressed ovarian cysts [2, 4]. However, in our pa-

tient the progesterone and estrogen treatments initiated

at the age of 18 yr were unable to suppress her LH se-

cretion fully and to prevent the progressive ovarian

cyst formation. If we had started the hormonal treat-

ment earlier, the cyst enlargement and torsion of the

ovary would have been prevented.

In summary, our report about the ovarian histology

of an adult female with a complete StAR deficiency

(Q258X), which demonstrated markedly lipoid depos-

its, decreased expression of both mitochondrial and

microsomal steroidogenic enzymes, and deformation

of mitochondrial ultrastructure, strongly supports two-

hit model of lipoid CAH.

Acknowledgements

We thank Dr. Tomonobu Hasegawa (Department of

Endocrinology and Metabolism, Keio University

School of Medicine), and Dr. Masahiko Morikawa

(Department of Pathology, Tokyo Metropolitan

Kiyose Children’s Hospital) for supporting the histo-

logical analysis.

References

1. Korsch E, Peter M, Hiort O, Sippell WG, Ure BM,

Hauffa BP, Bergmann M (1999) Gonadal histology

with testicular carcinoma in situ in a 15-year-old

46,XY female patient with a premature termination in

the steroidogenic acute regulatory protein causing con-

genital lipoid adrenal hyperplasia. J Clin Endocrinol

Metab 84: 1628–1632.

2. Shima M, Tanae A, Miki K, Katsumata N, Matsumoto

S, Nakajima S, Harada T, Shinagawa T, Tanaka T,

Okada S (2000) Mechanism for the development of

ovarian cysts in patients with congenital lipoid adrenal

hyperplasia. Euro J Endocrinology 142: 274–279.

3. Matsuo N, Ogata T (1988) Characteristics of gonadal

dysfunction in congenital lipoid adrenal hyperplasia.

Acta Paedeatr Jpn 30: 243.

4. Fujieda K, Tajima T, Nakae J, Sageshima S, Tachibana

S, Suwa S, Sugawara T, Strauss III JF (1997) Sponta-

neous puberty in 46,XX subjects with congenital lipoid

adrenal hyperplasia. J Clin Invest 99: 1265–1271.

5. Bose S, Sugawara T, Strauss JF 3rd, Miller WL (1996)

The pathophysiology and genetics of congenital lipoid

adrenal hyperplasia. New Eng J Med 335: 1870–1878.

6. Tanae A, Miki Y, Hibi I (1988) Pubertal presentation

in patients with congenital lipoid adrenal hyperplasia

(Prader’s Syndrome). Acta Paedeatr Jpn 30: 236–238.

7. Hasegawa T, Zhao L, Caron KM, Majdic G, Suzuki T,

Shizawa S, Sasano H, Parker KL (2000) Developmen-

tal roles of the steroidogenic acute regulatory protein

(StAR) as revealed by StAR KO mice. Mol Endocrinol

14: 1462–1471.

8. Nakae J, Tajima T, Sugawara T, Arakane F, Hanaki K,

Hotsubo T, Igarashi N, Igarashi Y, Ishii T, Koda N,

Kohno H, Nakagawa Y, Tachibana K, Takeshima Y,

Tsubouchi K, Strauss JF 3rd , Fujieda K (1997) Analy-

sis of the steroidogenic acute regulatory protein (StAR)

gene in Japanese patients with congenital lipoid adre-

nal hyperplasia. Hum Mol Genet 6: 571–576.

9. Suzuki T, Sasano H, Kimura N, Midori T, Fukuya T,

Yajima A, Nagura H (1994) Immunohistochemical

distribution of progesterone, androgen and oestrogen

receptors in the human ovary during the menstrual

cycle: relationship to expression of steroidogenic

enzyme. Hum Reprod 9: 1589–1595.

10. Suzuki T, Sasano H, Tamura M, Aoki H, Fukaya T,

Fig. 4. Detection of corpus albicans in the ovary. Corpus albi-

cans (arrow) were detected in the ovary by hematoxylin

and eosin staining. (magnification, ×100)

OVARIAN HISTOLOGY IN StAR DEFICIENCY 1049

Yajima A, Nagura H (1993) Temporal and spatial

localization of steroidogenic enzymes in premenopausal

human ovaries: in situ hybridization and immunohisto-

chemical study. Mol Cell Endocrinol 97: 135–143.

11. Sato Y, Susuki T, Hidaka K, Sato H, Ito K, Ito S,

Sasano H (2003) Immunolocalization of nuclear tran-

scription factors, DAX-1 and COUP-TF II, in the nor-

mal human ovary: correlation with adrenal 4 binding

protein/steroidogenic factor-1 immunolocalization dur-

ing the menstrual cycle. J Clin Endocrinol Metab 88:

3415–3420.

12. Oonk RB, Parker KL, Gibson JL, Richards JS (1990)

Rat cholesterol side-chain cleavage cytochrome P-450

(P-450scc) gene; structure and regulation by cAMP in

vitro. J Biol Chem 265: 22392–22401.

13. Natraj U, Richards JS (1993) Hormonal regulation, lo-

calization, and functional activity of the progesterone

receptor in granulose cells of rat preovulatory follicles.

Endocrinology 133: 761–769.

14. Qublan H, Amarin Z, Nawasreh M, Diab F, Malkawi S,

Al-Ahmad N, Balawneh M (2006) Luteinized unrup-

tured follicle syndrome: incidence and recurrence rate

in infertile women with unexplained infertility under-

going intrauterine insemination. Hum Reprod 121:

2110–2113.

15. Lydon JP, DeMayo FJ, Conneely OM, O’Malley BW

(1996) Reproductive phenotypes of progesterone re-

ceptor null mutant mouse. J Steroid Biochem Mol Biol

56: 67–77.

16. Risma KA, Clay CM, Nett TM, Wagner T, Yun J,

Nilson JH (1995) Targeted overexpression of luteiniz-

ing hormone in transgenic mice leads to infertility,

polycystic ovaries, and ovarian tumors. Proc Natl Acad

Sci USA 92: 1322–1326.