FAMILIAL CUSHING\'S SYNDROME DUE TO NODULAR ADRENOCORTICAL DYSPLASIA. A PUTATIVE RECEPTOR-ANTIBODY DISEASE?

Clinical Endocrinology ( 1986) 24,299-3 10

FAMILIAL CUSHING’S SYNDROME DUE T O NODULAR ADRENOCORTICAL DYSPLASIA. A PUTATIVE RECEPTOR-ANTIBODY DISEASE? F. T E D I N G VAN BERKHOUT, R. J. M. C R O U G H S , L. KATER*, H. J. S C H U U R M A N * , F. J . H. G M E L I G M E Y L I N G t , C. D. KOOYMANS, R. D. VAN D E R GAAGS;, D . JOLINKS; A N D H . A. DREXHAGES; Department of Endocrinology, University Hospital, Catharijnesingel 101, 351 1 G V Utrecht. The Netherlands, * Division of Immunopathology, Department of Internal Medicine and Institute for Pathology, Uniuersity Hospital, Utrecht, t Department of Clinical Immunology, University Hospital, Utrecht, $ Institute f o r Pathology, University of Utrecht, 9; Laboratory of Clinical Immunology, Department of Pathology, Free University Hospital, de Boelelaan I l l 7, 1007 M B Amsterdam, The Netherlands f

Receiced30 July 1985; rerurnedfor revision 30 ..lugus[ I985;finall.v revised I October 1985; uccepred 29 Ocrober 1985)

SUMMARY

Two sisters aged 13 and 19 years suffering from familial Cushing’s syndrome due to nodular adrenocortical dysplasia are described. Pituitary adrenocortical function tests indicated the presence of adrenal autonomy. Adrenal scintigraphy showed bilateral symmetrical uptake indicating the bilateral character of the autonomous process. Complete adrenalectomy was performed in both girls. The adrenals were of about normal weight showing numerous dark brown pigmented nodules and small perivascular lymphocytic infiltrates. Serum immunoglobulin preparations obtained from both girls stimulated adrenocortical cell growth in a cytochemical bioassay system. It is proposed that circulating growth factors may be involved in the pathogenesis of the disease. Since 1949 some cases of Cushing’s syndrome in children and young adults due to pigmented multinodular adrenocortical dysplasia have been described (Bohm et al., 1983). More recently, four reports have been published of two or more cases within one family or sibship (Arce et al., 1978; Schweizer-Cagianut et al., 1980; Donaldson et al., 1981; Bohm et al., 1983). In the present study we report on two sisters with Cushing’s syndrome due to bilateral nodular pigmented adrenocortical dysplasia with special reference to a possible immune process underlying the pathogenesis of the disease. Autoantibodies against cytoplasmic Correspondence: R. J. M. Croughs M.D.. Department of Endocrinology. University Hospital, Catharijnesingel 101. 351 1 GV Utrecht, The Netherlands.

299

300

F. Teding van Berkhout et al.

components of adrenal epithelium have occasionally been found in Cushing’s syndrome. Andrada et al. (1979) have found that three of seven patients were positive; adrenal lymphocyte infiltration was present in two of them. Thirteen years beforehand, Wegienka et al. (1966) reported a case of Cushing’s syndrome and adrenal hyperplasia with serum antibodies to adrenocortical tissue and a mononuclear cell infiltration of the cortex, comparable to that in the thyroid of patients with Hashimoto’s disease; the patient also has a sporadic goitre, a disorder nowadays known to be associated with the presence of thyroid growth stimulating immunoglobulins in serum (Drexhage et al., 1980; Valente et al., 1983). In this respect, it is of great interest that a serum containing thyroid stimulating antibodies appeared to be able to induce morphological changes in the adrenal cortex of experimental animals. This suggests the presence of an additional antibody which directly acts on the adrenal cortex (El Kabin & Hockaday, 1969). The present study was prompted by these observations and by the fact that familial Cushing’s syndrome due to pigmented multinodular adrenocortical dysplasia is characterized by bilateral adrenocortical autonomy, suggesting that a factor other than ACTH might stimulate adrenocortical growth and hormone production. We therefore looked particularly for the presence of serum immunoglobulins capable of stimulating the growth of guinea-pig adrenocortical cells in uitro. METHODS AND MATERIALS Endocrine laboratory studies ( I ) Cortisol production rate (CPR) was measured by the method of Thijssen et a f .(1967). Normal values are < 0.8 pmol/kg/d or < 80 jimol/d. ( 2 ) Plasma cortisol was determined after Thijssen et af. (1980). Normal values at 0800 h are 0 . 2 0 4 6 0 pmol/l. (3) Plasma ACTH was measured by radioimmunoassay using a commercial kit (Amersham). Normal values range from 20 to 80 ng/l. (4) Dexamethasone screening test (Nugent et al., 1965) was performed by measuring plasma cortisol at 0900 h after administering 1 mg dexamethasone at 2300 h the preceding night. Normal values < 0.075 jimol/l (Nugent et al., 1965). ( 5 ) Lysine-vasopressin(LVP) test (Croughs, 1970)was performed by measuring plasma cortisol before and 20 and 60 min after the i.m. injection of 10 pressor units of lysinevasopressin. (6) Intravenous dexamethasone test (Croughs et af., 1973)was performed by measuring plasma cortisol before and after a 5 h infusion with dexamethasone 1 mg/h. (7) Synacthen (a 1-24 ACTH) was given as an i.v. infusion of 500 jig in 5 h. Plasma cortisol was measured before starting the infusion and after 3 and 5 h (i.v. synacthen test). (8) Urinary 17-OH-steroids were determined according to Appleby et al. (1955). (9) Twenty-four hour urinary 17-OH-steroid excretion was measured for five consecutive days. On day 3, six doses of 750 mg metyrapone were given orally at 0800 h, 1200 h, 1600 h, 2000 h, 2400 h and 0400 h. Because of variations in urinary creatinine excretion the 17 OHCS/creatinine ratio was calculated (metyrapone test). (10) Twenty-four hour urinary 17-OH-steroid excretion was measured for six consecutive days. On days 3 and 4, four doses of 0-5 mg dexamethasone were given orally at 0600 h, 1200 h, 1800 h and 2400 h. On days 5 and 6 four doses of 2 mg dexamethasone

Nodular adrenocortical dysplasia

30 1

were given orally at the same time. Because of variation in urinary creatinine excretion 17-0HCS/creatinine ratio was calculated (oral dexamethasone test). Histology and immunohistochemistry of adrenals

For histopathology, tissue specimens were fixed in buffered formalin and embedded in paraffin. Sections were stained with haematoxylin-eosin. For immunohistochemistry, tissue was snap frozen in liquid nitrogen and stored at - 70°C. For detection of deposits of immunoglobulin or complement components 6 pm cryostat sections were incubated with fluorescein isothiocyanate labelled antibody to immunoglobulins IgG, IgM, IgA, IgD and IgE and complement components Clq, C4, C3 and properdin. (Reagents were obtained from Behringwerke, Marburg am Lahn, W,Germany; Dakopatts, Copenhagen, Denmark; and Kallestad, Austin, Texas.) For the characterization of mononuclear cell infiltrates monoclonal antibodies to T- or B cell antigens were applied in an indirect immunoperoxidase procedure on frozen tissue sections. After incubation with monoclonal antibody in appropriate dilution, a second incubation was performed with rabbit immunoglobulins to mouse immunoglobulins conjugated with horseradish peroxidase. Staining was performed using 3-amino 9ethylcarbazol and hydrogen peroxide (HlOt) as substrate. The antibodies applied were Leu-1 (pan T-cell reagent), Leu-2 (OKT 8 analogue phenotype associated with Tsuppressor-cytotoxic subpopulation), Leu-3 (OKT 4 analogue phenotype associated with T-inducer-helper subset), anti HLA-DR (all from Becton Dickinson, Mountain View, (California) and BI (Coulter clone, Hialeah, Florida) (Poppema et al., 1981). In addition antibodies to immunoglobulin light and heavy chains were applied in two colour immunofluorescence to detect surface immunoglobulins. Serum autoantibodies

Routine autoantibody screening included indirect immunofluorescence on tissue sections of adrenal, ovary, testis and several other organs (including thyroid, stomach, liver). Also, antibodies against thyroglobulin and thyroid microsomal antigens were tested for using the ‘Thymune’ haemagglutination kit (Wellcome, Dartford, Kent, Great Britain). Isolation of serum immunoglobulins

Immunoglobulins from serum of the two patients and four healthy controls (laboratory staff) were prepared by precipitation with 1.64 mol/l ammonium sulphate and dialysed extensively against phosphate buffered saline. The protein content was estimated spectrophotometrically. The final preparation contained 80% immunoglobulins and did not contain ACTH. The cytochemical bioassay developed for the detection of adrenal growth stimulating immunoglobulins (nucleic acid cytophotometry )

The method used is based on the measurement of the amount of DNA in individual cell nuclei of a guinea-pig adrenal segment. The DNA is visualized by the Feulgen reaction and is quantified per nucleus in relative arbitrary units by means of a microdensitometer

302

F. Teding van Berkhout et a!.

'"II 2o

2C

2.5C 2 8 C Colour densty

Fig. 1. Population histograms obtained with the Feulgen densitometric method for the measurement of adrenal growth in uifro. The histograms show the colour density of the Feulgen reaction per nucleus ( = the relative DNA content) o n the horizontal axis and the number with a particular colour density on the vertical axis. Upper panel: in the presence of 10 fg ACTH, ?./ml culture fluid ( > 2 X : loo/:); lower panel: in the absence of any addition (>2.8C:2%) (see text).

or cytophotometer (Sandritter, 1979). A hundred nuclei are measured and the results are plotted as a population histogram, with the DNA-content along the horizontal axis and the number of nuclei showing a particular content of D N A on the vertical axis (Fig. 1). The cell cycle forms the basis of the assessments. In differentiated tissues, there are only cells with the diploid (2C)content of DNA; but in proliferating tissues some cells will be in S-phase and hence have DNA content values intermediate between the 2C and 4C amount. From the population histograms the percentage of cells in S-phase can easily be seen and these high values together with 4C values are taken as evidence for proliferation. For the detection of immunoglobulins stimulating in uitro adrenal growth the assay system is as follows. Adrenals were removed from Hartley strain guinea-pigs weighing 250-350 g, after killing the animals by asphyxiation in nitrogen. The organs were divided in four segments and each segment was cultured under non-proliferative conditions at 37"C,in Trowell's T8 culture fluid (Flow Laboratories) supplemented with lo-' mol/l ascorbic acid (Chayen, 1980). Adrenal segments were cultured for 5 h in the presence of 5-50 fg/ml ACTH al-24 (Synacthen) or 30-1000pg Ig preparation/ml culture fluid for the entire culture period. In each assay one segment was cultured in the absence of ACTH and used as control. Thereafter the segments were chilled in n-hexane and stored a t -70°C overnight. The chilled segments were sectioned at 16 pm in an automatically driven cryostat and the Feulgen reaction was carried out (Chayen, 1980). The amount of reaction product was measured in 100 randomly selected individual cell nuclei of the zona reticularis and zona fasciculata of the adrenal cortex. A Vickers N85a 550 nm, with a 100 x oil scanning and integrating microdensitometer was used (i= immersion objective and 0.2 pm scanning spot).

Nodular adrenocortical dysplasia

303

The relative absorption values (relative DNA content per nucleus) were plotted as population histograms, from which the diploid value (2C) could be defined. Cells with a content of more than 2.8C were considered as being in or having completed the S-phase, and results are expressed as the percentage of cells in S-phase. The 2.8 value was chosen as decision level as this value differs 2 SD from the mean of 2C in histograms of normal adrenal segments.

RESULTS Case 1 A 13-year-old girl, had a 2-year history of weight gain with growth arrest, muscle weakness and the development of striae. She also complained of lack of concentration and increased irritability. She was seen elsewhere because of the development of backache and colic in both groins. No renal stones could be demonstrated at the time of investigation but the classical appearance of Cushing's syndrome was recognized and the patient was referred for evaluation. Physical examination revealed a girl with a height of 1.47 m and a weight of 57 kg. Blood pressure was 140/100. She exhibited centripetal obesity with a moonface and buffalo hump. There were numerous dark-red striae at the lower abdomen and upper part of arms and legs and bruises at the legs. Breast development and pubic hair: Tanner stage 11. Routine analysis of blood and urine was normal. X-rays of the skull showed a normal sella turcica. Detailed analysis of pituitary adrenocortical function confirmed the presence of Cushing's syndrome with a CPR of 142 pmo1/24 h. Administration of lysine-vasopressin and metyrapone did not stimulate pituitary-adreno-corticalfunction and no suppression could be demonstrated by oral or intravenous administration of high doses of dexamethasone. No ACTH was detectable in plasma. Intravenous administration of synacthen was followed by a slight plasma cortisol increment. (In normal subjects the plasma cortisol increment is at least loo%.) The pattern is characteristic of adrenal autonomy. However, a CT scan did not show an adrenal mass. Both adrenals were of about normal size. Adrenal scintigraphy using "'I- 19-nor-cholesterol showed symmetrical uptake (Table 1). Bilateral total adrenalectomy was performed. Twelve days later the patient was discharged with glucocorticoid and rnineralocorticoid substitution therapy. Six months later there was a marked remission of clinical signs of Cushing's syndrome. At that time she had a height of 1-52 m and a weight of 59 kg. Blood pressure was 120/75 and the patient felt well.

Case 2

An 18-year-old girl, the older sister ofcase 1, was evaluated too, as the mother had noticed that she had the same physical appearance as her sister. She had a 3-year history of weight gain and slight muscle weakness. She also complained of increased irritability. Menarche was at the age of 15, but menses were irregular and for 1; years she had been amenorrhoeic. For the previous two months she had been treated for hypertension. Physical examination revealed a girl with a height of 1.57 m and a weight of 60 kg. Blood pressure was 160/115. She exhibited centripetal obesity with a moonface and a

F. Teding van Berkhout et al.

304

Table I . Laboratory findings in two sisters with familial Cushing’s syndrome ~~

Case I

Case 2

I42

129

pmo1124 h

Cortisol 0600 h 0900 h 1700 h 2300 h

0.66 0.69 0.68 0.74

0.55 0.87 0.51 0.51

pmol/l pmol/l pmol/l pmol/l

Dexamethasone screening (Nugent) (cortisol N c 0.075)

Cortisol

0.74

0.68

pmol/l

Lysine-vasopressin test (I0pU i.m.)

Cortisol

0 min 30min 60min

0.61 0.65 0.65

0.54 0.55 0.49

pmol/l pmol/l pmolll

i.v. Dexamethasone test ( 5 mg/5 h)

Cortisol

0h 5h

0.46 0.41

0.41 0.42

pmol/l pmol/l

0h 3h 5h

0.72 0.97 1.01

0.40 0.59 0.54

pmol/l pmol/l pmoljl

< 12.5

c 12.5

I7-OHCS/ day I creatinine day 2 day 3 day 4 day 5

13.1 14.4 14.7 10.6 11.1

11.4

Oral dexamethasone suppression (Liddle) I7-OHCS/ day I (4 x 0.5 mg on days 3 and 4) creatinine day 2 (4 x 2 mg on days 5 and 6) day 3 day4 day 5 day 6

21.7 20.5 18.5

11.0

Cortisol production rate (N c 80) Diurnal rhythm of cortisol

i.v.al.:4ACTH test (500 pg/5 h)

Cortisol

Plasma ACTH Metyrapone-test (6 x 750 mg orally during day 3)

19.3 14.7 23.5

ng/l

9.8 8.9 643 7.9 9.4 12.2 12.2 13.6

13.8

buffalo hump. Purple striae were visible on both breasts. Routine analysis of blood and urine was normal. X-rays of the skull showed a normal sella turcica. Results of pituitary adrenocortical function tests were virtually identical to those of her sister (Table 1). A CT scan showed two adrenals of about normal size and adrenal scintigraphy showed symmetrical uptake. She was subjected to bilateral total adrenalectomy on the same day as her sister and she was also discharged 12 days later with glucocorticoid and mineralocorticoid substitution therapy. Six months later there was a marked remission of clinical signs of Cushing’s syndrome. At that time she had a weight of49-3kg with a height of 1.57 m. Blood pressure was 120/80 without medical treatment and the patient felt excellent. Menses had resumed one month postoperatively. Morphology and histology of the adrenals

See Figs 2 4 . The gross appearance and histology of the adrenals were similar in both cases. Adrenal weights after removal of periadrenal fatty tissue were 4.4and 3.7 g in case

Nodular adrenocortical dysplasia

305

Fig. 2. Case 2. Adrenal gland at examination after bilateral adrenalectomy. Nodules and pigmentation arc present.

Fig. 3. Case I. A sharply demarcated nodule with adrenal cortex derived cells. Formalin fixed paraffin-embedded material, haematoxylin-eosin staining, x 70.

1, 4.3 and 3.9 g in case 2. There were numerous dark brown pigmented nodules with a diameter ranging from 1 to 4 mm (Fig. 2). Microscopically these nodules were localized

throughout all layers of the adrenal cortex. They were sharply demarcated and consisted oflarge cells with abundant eosinophilic cytoplasm and round chromatin-rich nuclei (Fig. 3). A granular pigment (lipofuscin) was present in many cells. Lymphocyte infiltrates of

306

F. Teding van Berkhout et al.

Fig. 4. Case 2. Mononuclear cell infiltrate at the periphery of a nodule. Fonnalin-fixed. paraffinembedded material, haematoxylin-eosin staining, x 180. Inset: immunohistochemistry with antibody leu 2 (directed to cells with the phenotype associated with suppressor-cytotoxic T cell function). The majority of mononuclear cells in the infiltrate is leu 2-positive. Frozen tissue section, immunoperoxidase method, x 290.

varying size were noted mainly around vessel walls but also incidentallyinside the nodules (Fig. 4). Zmmunohistochemistry

In direct immunofluorescenceon cryostat sections of the adrenal cortex of both sisters no deposits of immunoglobulins or of complement C4,C3 and properdin were detectable. However, we observed complement component Clq in the cytoplasma of the nodular cells. The lymphoid cell infiltrates in the adrenals of case 1 contained mainly T-lymphocytes with the phenotype Leu 1 and Leu 3 associated with T helper cells. A minority of cells expressed the T-cell marker Leu 2 associated with T suppressor/cytotoxic cells. There were some B-cells detectable by expression of the B-cell antigen B 1 and of S-IgM. HLA-DR was expressed on a minor part of the lymphoid cell infiltrates, presumably B lymphocytes. In the adrenals of case 2 the majority of lymphoid cells in the infiltrate were T cells expressing Leu 1 and Leu 2 (see inset of Fig. 4). There was almost no expression of Leu 3. In some of the infiltrates an occasional B lymphocyte could be identified by its positivity for B1, S-IgM and S-IgD. In both cases there was no expression of HLA-DR in the nodules, except for positivity of blood vessels and some stromal components.

Nodular adrenocortical dysplasia

1

307

I

25 50 Synacthen concentration ( f q / d ) 0 10

Iq concantrotion (pq/rnc)

Fig. 5 . The cytochemical bioassay for the detection of adrenal growth stimulating immunoglobulins. Left panel: effect of ACTH,., (Synacthen). In-vitro growth is expressed as the percentage of cells in S-phase (see Fig. 1) in the presence of 0, 10. 25 and 50 fg of the polypeptide hormone per ml culture fluid. Note the bell-shaped dose-response. Right panel: effect of the immunoglobulin-preparations of both girls. In-vitro growth was induced in the presence ofas little as 30-60 p g ammonium-sulphate precipitate per ml culture fluid (=optimum of the bell-shaped curve). The effect of four preparations obtained from healthy laboratory staff is shown as control ( 0 ) . Familial Cushing‘s syndrome due to pigmented multinodular adrenocortical dysplasia (0).

Serum auto-antibodies and immunoglobulins stimulating in-vitro growth of guinea-pig adrenal

Serum samples were taken from both girls a few days before bilateral adrenalectomy. Autoantibodies to adrenal, ovary, testis, thyroid, stomach, mitochondria, smooth muscle and nuclei were not demonstrable in the immunofluorescence tests. Rheumatoid factors as well as haemagglutinating antibodies to thyroid microsomes and thyroglobulin were also not detectable. Figure 5 shows the data obtained with the nucleic acid cytophotometric method for the detection of immunoglobulins stimulating growth in uitro. The growth promoting effect of synthetic ACTH was used as a control. As can be seen the hormone is capable of inducing a raised percentage of cells in S-phase at concentrations of 10-25 fg/ml culture fluid. The ‘bell-shaped’ curve obtained is a well-known feature of the cytochemical bioassay of polypeptide hormone (Chayen, 1980). The relatively low optimal dose, i.e. 10-25 fg/rnl is comparable to the low dosages of ACTH needed in the ascorbic depletion bioassay for this pituitary hormone (Chayen, 1980). Interestingly, the serum Ig-preparations of both girls had similar growth stimulating capabilities at concentrations ranging from 30 to 150 pg Ig per ml culture fluid. None of the four serum Igs of healthy laboratory staff had any noteworthy effect on DNAsynthesis in guinea-pig adrenal explants. DISCUSSION The present report describes the biochemical, pathological and immunological characteristics of two sisters with familial Cushing’s syndrome due to nodular adrenocortical dysplasia. Endocrinologically the disease was characterized by adrenal autonomy with suppression of the hypothalamic-pituitary system as in Cushing’s syndrome due to an

F. Teding van Berkhout et al.

308

Table 2. Results of endocrinological investigation of familial adrenal nodular dysplasia-a review of literature

Sex

Arce (1978) Schweizer (1980)

Age

M 13 years M 27 years F I5 years M I8 years F 15 years

Donaldson M 10 weeks (1981) F I week Bohm (1983)

F 14years M 14 years

Metyrapone-test ACTH-stimulation

No effect No effect

No effect No effect

No effect

Slight response

Dexamethasone-suppression (Liddle) Plasma ACTH No effect No effect No effect

No effect

No effect (2 mg)

No effect

No effect Slight response

Not detectable

adrenocortical tumour. This conclusion confirms and extends the results of previous studies (Table 2). The aetiology of the disorder is unknown. The familial occurrence supports the idea of a common and possibly genetic origin. Schweizer et af.(1980) proposed that the disease is due to an inherited defect within the adrenals. Bohm et al. (1983) proposed that the basic morphological defect is a hamartomatous malformation of the adrenal cortex. This view was supported by the finding of varying amounts of mature fat cells within the nodules and the occurrence of dysplastic tissue in the periadrenal fatty tissues. We also found adipocyte infiltration of the nodules but no dysplastic tissue outside the adrenal capsule. In line with Bricaire er al. ( 1 970) and Aron er af.( 198l), Arce et al. ( 1978)have proposed that the disease is due to hypothalamic pituitary dysfunction leading initially to overstimulation and nodule formation of the adrenal cortex and finally to adrenal autonomy with suppression of pituitary ACTH secretion. However, the patients of Schweizer et al. (1980) both had low plasma ACTH levels many years after bilateral adrenalectomy at a time when plasma cortisol levels were low. Furthermore, Nelson’s syndrome has not been reported after bilateral adrenalectomy. To test for the presence of an unidentified adrenocorticotrophic stimulus, Donaldson et al. (198 1) exposed a culture of normal adult adrenocortical cells to serum of an affected neonate, but no changes in the in-vitro steroid synthesis was observed. Our study, using a crude serum-immunoglobulin preparation and cultured adrenal segments from the guinea-pig, indicates that adrenocorticotrophic stimuli of non pituitary origin exist in familial Cushing’s syndrome. The data indicate that the serum immunoglobulins of both girls have in-vitro growth stimulating capacity for adrenocortical cells. This finding is compatible with the present knowledge of thyroid-immunoendocrinopathy in which serum immunoglobulins from patients with Graves disease and sporadic euthyroid goitre affect in-vitro growth of thyroid cells; the mechanism of this action may be by competition with TSH for the TSH receptor (Valente et al., 1983). As in our cases the adrenal growth stimulating property of serum immunoglobulins was similar to that of synthetic ACTH, it can be speculated that the serum immunoglobulins contain autoantibody activity to the ACTH receptor. It should be noticed, however, that other

Nodular adrenocortical dysplasia

309

signs of autoimmunity were not detectable in our patients and that there was no positive family history of autoimmune disease. In search of an in-vivo relevance to our in-vitro findings, we did not observe the presence of immunoglobulins in situ in the nodules of the adrenals by tissue section analysis. This may be due to lack of sensitivity of the immunohistochemical procedure or to the fact at the time of adrenalectomy tissue bound immunoglobulins were no longer present. The finding of Clq in the cytoplasm of nodular cells is remarkable in this respect; it may be the representative of formerly ingested immunoglobulins after complement fixation. A final argument for an immunological component in the aetiology of the disorder comes from the presence of lymphocytic infiltrates in and around the adrenal nodules. These cells were identified as T cells mainly with the phenotype associated with inducer helper T cells in case 1 and mainly with the phenotype associated with suppressorcytotoxic T cells in case 2. In conclusion, this is the first report on the presence of circulating immunoglobulins with in-vitro adrenal cortex growth promoting properties in familial Cushing's syndrome due to nodular adrenocortical dysplasia. It remains to be established whether this phenomenon is of more general relevance by in-depth study of a larger series of similar patients. ACKNOWLEDGEMENTS T h i s study was s u p p o r t e d in p a r t by t h e Division for H e a l t h Research TNO (project no. 13.45.40).

REFERENCES ANDRADA. J.A., MURRAYF.T.. ANDRADA. E.C. & EZRIN,C. (1979) Cushing's syndrome and autoimmunity. Archives of Parhological Laborarory Medicine, 103, 244-246. J.K. ( 1955) Urinary excretion of 17-hydroxy-20-oxosteroidsin normal health APPLEBY,J.I. & NORYMBERSKI, and in rheumatoid arthritis. Annuls of rhe Rheumatic Diseuses, 14, 172-175. ARCE.B.. LICEA.M., HUNG,S.& PADRON.R. (1978) Familial Cushing's syndrome. Acra Endocrinologica, 87, 139- 147. ARON.D.C.. FINDLING, J.W.. FITZGERALD, P.A. el 01. (1981) Pituitary ACTH dependency of nodular adrenal hyperplasia in Cushing's syndrome. ,4mericun Journal of Medicine. 71. 302-306. BOHM.N., LIPPMANN-GROB, B. & PITRYKOWSKI. W.V. (1983) Familial Cushing's syndrome due to pigmented multinodular adrenocortical dysplasia. Acfu Endocrinologica, 102,428-535. BRICAIRE,H.. LUTON,J.P..GHOZLAND, M. & FOREST, M . (1970) La polymicroadenomatose de la corticosurrenale dans le syndrome de Cushing. A propos de 15 observations. Annales de Medicine Inrerne;121, 75 5-777. CHAYEN, J. (1980) The cytochemical bioassay of polypeptide hormones. In Monographs on Endocrinology, vol. I7 pp. 80-94. Springer Verlag, Berlin. CROUGHS,R.J.M. ( 1970) Use of lysine-vasopressin in the differential diagnosis o f Cushing's syndrome. Acru Endocrinologica. 65, 595-607. CROL'GHS. R.J.M., DOCTER.R. & JONG,F.H. DE (1973) Comparison of oral and intravenous dexamethasone suppression tests in the differential diagnosis of Cushing's syndrome. Acra Endocrinologica, 72, 54-62. DONALDSON. M.D.C., GRANT,D.B., O'HARA. M.J. & SHACKELTON, C.H.L. (1981) Familial congenital Cushing's syndrome due to bilateral adrenal hyperplasia. Clinical Endocrinology, 14, 519-526. DREXHAGE, H.A., BOTAZZO. G.F.. DONIACH. D., BITENSKY, L. & CHAYEN, J. (1980) Evidence for thyroid-growth stimulating immunoglobulins in some goitrous thyroid diseases. Lancer, i, 287-29 I , EL KABIN,D.J. & HOCKADAY. T.D.R. (1969) Morphological changes in the adrenal cortex of experimental animals after injection of sera containing the long activity thyroid stimulator. iVarure, 224, 608.

F. Teding van Berkhout et al.

310

NUGENT,C.A., NICHOLS,T. & TYLER,F.H. (1965) Diagnosis of Cushing's syndrome. Single dose dexamethasone suppression test. Archives of Internal Medicine, 116, 172-1 16. POPPEMA, S., BAHN,A.K.. RAINHERZ, E.L., MCCLUSKY. R.T. & SCHLOSSMAN S.F. (1981) Distribution ofT-cell subsets in human lymph nodes. Journal of Experimental Medicine, 153.3042. SANDRIRER. W. (1979) Review of Nucleic Acid Cytophotometer in General Pathology. Quantitative Cytochemistry and its Applications, pp. 1-8. Academic Press. London. C. (1980) Familial Cushing's syndrome with primary SCHWZER-CAGIANUT. M.. FROESCH,E.R. & HEDINGER. adrenocortical microadenomatosis (primary adrenocortical nodular dysplasia). Acta Endocrinologica, 94, 529-535.

THIJSSEN, J.H.H.. DE WAARD,F. & WIERSINGA, A. (1967) Hydrocortisone production rates in Africans. Lancet, i, 8 14-8 I 6. THUSSEN. J.H.H., VAN DEN BERG.J.H.M., ADLERCREUTZ, H.etal. (1980)Thedeterminationofcortisol in human plasma: evaluation and comparison of seven assays. Clinica Chimica Acta, 100.39-46. VALENTE,W.A., V ~ IP.,, ROTELLA, C.M. el al. (1983) Antibodies that promote thyroid growth. A distinct population of thyroid stimulating auto-antibodies. New England Journal of Medicine, 309, 1028-1034. WEGIENKA. L.C.. WUEPPER. K.D., KOMARMY, L.E. & FORSHAM, P.H.(1966) Cushing's syndrome with adrenal medullary insufficiency and adrenal auto-antibodies. Lancet, i, 741-743.