Immunocytochemical differential diagnosis of adrenocortical neoplasms using the monoclonal antibody D11 *
Sören Schröder1, Axel Niendorf1, Eike Achilles1, Manfred Dietel1, Barbara-Christina Padberg1, Ulrike Beisiegel2, Henning Dralle 3, Max Bressel4, and Günter Klöppel 5
1 Institute of Pathology and 2 Department of Internal Medicine, University Hospital Hamburg-Eppendorf, Hamburg, Federal Republic of Germany
3 Department of Surgery, Hannover School of Medicine, Hannover, Federal Republic of Germany
4 Department of Urology, General Hospital Hamburg-Harburg, Hamburg, Federal Republic of Germany
5 Institute of Pathology, Free University of Brussels, Brussels, Belgium
Received October 23, 1989 / Received after revision February 16, 1990 / Accepted February 21, 1990
Summary. The monoclonal antibody D11 is a valuable aid in the accurate typing of adrenal tumours as, in for- malin-fixed, paraffin-embedding material, strong nucle- ar D11 positivity was observed only in adrenocortical cells in 190 neoplasms (including 100 adrenal tumours). This pattern was demonstrated for all zona glomerulosa cells in 27 normal adrenals and for the neoplastic cells of 15 adrenocortical adenomas derived from that zone, as judged from clinically evident hyperaldosteronism. Normal cells of zona fasciculata and reticularis also showed strong diffuse D11 immunostaining and the same nuclear plus cytoplasmic D11 reactivity was evi- dent in 15 benign and malignant adrenocortical neo- plasms derived from these zones, documented by hyper- cortisolism. Cytoplasmic and/or nuclear D11 staining made topohistogenetic typing possible in 15 non-func- tioning cortical tumours. D11 immunostaining gave neg- ative results in 50 specimens containing normal, hyper- plastic and neoplastic adrenomedullary cells. In addi- tion, absence of D11 reactivity was recorded in 4 adrenal metastases of extra-adrenal carcinomas, 5 paraganglio- mas, 25 primary renal carcinomas and 59 of 60 primary thyroid carcinomas. D11 immunocytochemistry allows the accurate typing of benign and malignant adrenocort- ical neoplasms, irrespective of histology and function. With this method, primary adrenocortical tumours can be separated from carcinomas metastatic to the adrenal gland, including secondary tumours of similar pheno- type (such as renal carcinomas). By exclusion, D11 nega- tivity provides evidence of the medullary origin of prima- ry adrenal tumours even in the absence of clinical, struc-
Offprint requests to: S. Schröder, Institut für Pathologie, Univer- sität Hamburg (UKE), Martinistrasse 52, D-2000 Hamburg 20, FRG
* Dedicated to Prof. Dr. Dr. h.c. mult. Wilhelm Doerr on the occasion of his 75th birthday. This study has been sponsored by the Deutsche Forschungsgemeinschaft and the Hamburger Krebs- gesellschaft and was presented in part at the 80th Annual Meeting of the American Association for Cancer Research, San Francisco, California, 24-27 May 1989 (Schröder et al. 1989)
tural, histochemical and conventional immunohisto- chemical indicators of phaeochromocytoma.
Key words: Adrenal gland - Adrenocortical tumours - Immunocytochemistry
Introduction
Because of their broad microscopical variety, accurate typing of adrenal tumours often poses a major diagnos- tic problem. Conventional histology frequently offers no conclusive evidence of the origin of individual neoplasms (cortical versus medullary, versus extra-adrenal meta- static). Electron microscopy, immunohistology (the de- tection of steroid hormones) and histochemistry (the de- tection of steroid hormone-specific enzymes) have also been found to be of little help with this problem (Page et al. 1986). There is a need for an immunohistological probe which guarantees selective staining of normal and neoplastic cortical cells in the adrenal gland. Our find- ings show that the antibody D11 provides us with the means with which we can solve not only differential di- agnostic problems, but also some questions relating to the histogenesis of adrenocortical tumours.
Materials and methods
The monoclonal antibody D11 was produced using a 59 kDa hu- man liver membrane protein as described elsewhere (Beisiegel et al. 1988). Proteins were analysed on SDS-polyacrylamide gel electro- phoresis using the method of Neville (1971). Immunoblotting was carried out following the procedures as described by Beisiegel et al. (1981) and Hui et al. (1986). The cultivation and passage of HepG2 cells (human hepatoblastoma) and of other freshly established pri- mary cell strains was carried out using the method of Dietel et al. (1987).
Formalin-fixed, paraffin-embedded material of the 103 adre- nalectomy specimens was collected from the surgical pathology files of various Institutes of Pathology. Information relating to the preoperative symptoms were noted from the patients’ medial
+
67-
43-
0 1 2 3 4 5 6 7
-59 kDa
Fig. 1. Biochemical characterisation of D11-positive antigens. Top: Immunoblot of a two-dimensional electrophoresis. Approximately 10 different dots of molecular weight 59 kDa react as antigens of D11.
Bottom: Immunoblot of an SDS gel electrophoresis using membrane homogenates of various rabbit organs: The 59 kDa antigens are detectable in all organs and in HepG2 cells, with the highest concentration in the liver and in the brain and with only a very low concentration in the adrenal gland (1, HepG2 cells; 2, liver; 3, intestine; 4, adrenal gland; 5, testicle; 6, brain; 7, adipose tissue)
a
b
records. In each case the development of the illness following oper- ation was documented until spring 1989. For the purpose of com- parison, 25 renal cell carcinomas, 60 thyroid carcinomas, 5 para- gangliomas and various normal human tissues were also brought into the study.
The immunohistological examinations used the avidin-biotin- peroxidase-complex (ABC) method (Hsu et al. 1981). The anti- bodies employed were directed against neuron specific enolase (NSE) (polyclonal/rabbit; Dakopatts, Copenhagen, Denmark, di- lution 1:3000), chromogranin A (Chr A: monoclonal; Hybritech, San Diego, Calif., 1:5000), synaptophysin (SYN: polyclonal/rab- bit; Dr. Jahn, Max-Planck-Institut, Martinsried, FRG, 1:1000). In addition, the monoclonal antibody D11 (culture supernatant; 1:5) and Grimelius’ method for the demonstration of argyrophilia (Grimelius and Wilander 1984) were applied.
Results
In two-dimensional electrophoresis, approximately ten different dots of the same molecular weight showed im-
munoreactivity for D11 (Fig. 1a). The presence of D11- positive 59 kDa antigens were observed in the immuno- blot analysis of an SDS-polyacrylamide gel electrophor- esis in solubilisates of the membrane fractions of various rabbit organs as well as in cultivated HepG2 cells (Fig. 1 b).
When a non cell-membrane permeabilising fixative (Bouin’s solution) was used, no D11 reaction was noted in HepG2 cells, fibroblasts or in cultivated cells of differ- ent gastrointestinal carcinomas (stomach, pancreas and colon). Incubation of the primary antibody before prep- aration gave the same result. In contrast, a diffuse cyto- plasmic fluorescence in HepG2 cells was observed after permeabilisation of the cell membrane by methanol/ace- tone fixation (Fig. 2a). Co-cultivated fibroblasts and other cultivated tumour cells, however, remained nega- tive (Fig. 2b). In the liver, formalin-fixed specimens showed intense staining of hepatocytes and of occasional bile duct epithelium. In the kidney, epithelial positivity
a
b
C
was recorded for some tubular segments. In the pancre- as, positive results were shown for duct epithelia and for acini of the exocrine parenchyma. In all of the organs examined, smooth muscle cells of arterial and venous vessel walls and fibrocytes within the organ capsules were D11-positive; D11 staining was always confined to the cytoplasm.
Three adrenalectomy specimens from healthy multi- organ donors and histologically inconspicuous residual parenchyma from the periphery of 24 tumours-bearing adrenals gave the immunocytochemical results recorded in Fig. 3a and b. D11 positivity was demonstrated in the nuclei of all three cortical zones. Faint granular cyto- plasmic immunostaining was demonstrated in the zona glomerulosa and in the peripheral two-thirds of the zona fasciculata. In the central third of this zone, as in the entire zona reticularis, there was intense diffuse cytoplas- mic staining. Medullary tissue was completely devoid of D11 reactivity, while argyrophilia and positivity for NSE, Chr A and SYN were only seen in this region, thus producing a virtually complementary picture to D11 immunostaining.
In 15 patients, benign adrenocortical adenomas (mean weight 12 g, range 7-35 g) were classified as al-
dosteronomas as they were associated with clinically evi- dent Conn’s syndrome. On NSE immunostaining, 1 showed doubtful positivity. The remaining lesions proved NSE-negative and, like all of the other subse- quently described cortical tumours, also gave negative results when tested for Chr A and argyrophilia. In each of the 15 cases, D11 produced the same staining pattern as described above for the normal zona glomerulosa (Fig. 4a). In 6 lesions, increased cytoplasmic D11 stain- ing was seen focally. This never reached the intensity of the nuclear reaction.
Seven cortical tumours (mean weight 23 g, range 11- 48 g) associated with Cushing’s syndrome were classified as being adenomas owing to their intact capsular con- finement, the lack of mitotic activity and the absence of any post-operative symptoms (mean observation peri- od 36 months). All cases exhibited intense nuclear and cytoplasmic D11 staining (Fig. 4b, c). NSE immunocy- tochemistry gave negative results each time.
In 2 tumours associated with Cushing’s syndrome (weights 125 g and 450 g), malignancy could not be en- tirely discounted because of the presence of moderate nuclear pleomorphism and of focal capsular invasion, although neither patient showed any post-operative
a
b
symptoms over a period of 6 and 13 months respectively. D11 and NSE staining provided results identical to those of the 7 adenomas described before.
In 6 patients, adrenocortical carcinomas (mean weight 900 g, range 130-2250 g) associated with Cush- ing’s syndrome were observed. Four of these were wom- en, in whom elevated plasma levels of testosterone was accompanied by virilisation. Three of the patients died not more than 8 months after diagnosis. The remaining patients are alive 4, 5 and 62 months later, respectively, all suffering from metastatic tumours. Upon NSE immu- nostaining, questionable reactivity was seen with 4 of the carcinomas. In one case, this finding initially led to the mistaken diagnosis of phaeochromocytoma when the hormonal symptoms were not recognised. In con- trast, D11 gave constantly clear positive nuclear and cytoplasmic results with all 6 tumours (Fig. 4d).
In 7 cases solitary, well-defined adrenal tumours (mean weight 18 g, range 8-34 g) were found which were classified as endocrinologically silent cortical adenomas from their unremarkable histological appearances, the absence of endocrine symptoms and an uneventful post- operative course (mean observation period 32 months). All the tumours were NSE negative. On D11 immunocy- tochemistry strong nuclear positivity was found in each. Five lesions demonstrated intense cytoplasmic staining in addition (as described above for adrenocortical tu- mours associated with Cushing’s syndrome), while the remaining 2 neoplasms lacked significant cytoplasmic positivity (as noted for aldosteronomas).
The biological potential of 5 larger non-functioning cortical tumours (mean weight 209 g, range 35-750 g) could not be determined by morphological means. The patients had hitherto remained free from recurrent dis- ease over a mean post-operative observation period of 33 months. On NSE immunocytochemistry, 2 neoplasms
showed doubtful staining while the others gave negative results. Conversely, all 5 tumours showed distinct nucle- ar staining as well as an intense diffuse cytoplasmic reac- tivity for D11 in 4 lesions.
Cortical carcinomas lacking endocrine symptoms were found in 3 patients, leading to death within not more than 8 months after diagnosis. In 2 cases, adrenal- ectomy specimens weighing 995 g and 3250 g, respective- ly, were investigated. In the third case, only a needle biopsy from a para-aortic lymph node metastasis was available for study. The first 2 tumours showed doubtful NSE reactivity. In contrast, all 3 carcinomas exhibited clear nuclear and cytoplasmic positivity for D11.
In the 1 case of bilateral primary adrenocortical mi- croadenomatosis, strong cytoplasmic and nuclear D11 positivity of the hyperplastic cortical nodules was ob- served. The same intense D11 reactivity was demon- strated in sparsely scattered cortical cells in 2 myelolipo- mas and in 1 adrenal which showed tumour-like enlarge- ment by extramedullary myelopoiesis, due to thalassae- mia minor. In all 4 lesions, negative results were re- corded when tested for NSE, Chr A, SYN and argyro- philia.
Sporadic adrenomedullary hyperplasia was found in 3 patients. The morphological and clinical features of these cases are described in detail elsewhere (Dralle et al. 1990). In 12 patients, benign hereditary phaeochromocy- tomas (PCCs) (mean weight 29 g, range 8-85 g) were found to be present in the setting of the multiple endo- crine neoplasia syndrome, type IIa. In 3 cases, bilateral tumours were observed. In each of the patients, the dis- ease was accompanied by the typical clinical symptoms of PCC (hypertensive crises, increased levels of catechol- amines in plasma and urine and increased excretion rates for vanillylmandelic acid). In 24 patients (bilaterally in 1 of these), benign sporadic PCCs (mean weight 71 g,
a
b
C
d
Fig. 5a-d. D11 immunocytochemistry in adrenalcortical tumours with Cushing’s syndrome. a Adenoma showing strong D11 positivity: atrophy of the darkly coloured central cortical parts in the residual parenchyma (left side of the picture). x 3.8. b Complementary aspects of the same specimen upon immunocytochemistry for Chr A. × 3.8.
c Magnification from a: nuclear and cytoplasmic staining of tumour cells, the stroma being negative. x290. d Carcinoma showing same results as in c. × 150
range 13-140 g) were found. Follow-up study disclosed all patients to be symptom-free at the end of the observa- tion period (mean 34 months). In 4 patients with malig- nant sporadic PCC, the formation of metastases was discovered during operation or became apparent 2 and 24 months later, respectively. The 4 primary tumours weighed 110-350 g (mean 245 g). Three of these patients died 3-13 months after adrenalectomy; 1 is still alive suffering from metastatic disease. Apart from 2 benign and 2 malignant non-functioning tumours, all other pa- tients with the sporadic type of adrenomedullary neopla- sia had shown typical PCC symptoms before surgery.
Chr A staining gave positive results in connection with the hyperplastic and neoplastic adrenomedullary lesions, except in the case of 2 malignant PCCs. In addi- tion to these exceptions, NSE immunohistochemistry
and silver staining were negative in 1, or where appli- cable, 2 benign PCCs. All of the 47 lesions were positive for SYN, while D11 stainig invariable produced negative results.
Two patients showed adrenal metastases from renal cell carcinomas which, in both cases, represented the primary manifestation of an undiscovered kidney tu- mour. Both tumours, weighing 32 g and 34 g, respective- ly, were initially thought to be functionally inactive corti- cal adenomas on account of their clear-cell appearance and the absence of any capsular infiltration. In one case, a contralateral nephrectomy was carried out 5 months later because of a highly differentiated renal cell carcino- ma (G1). The other patient died 12 months after adre- nalectomy. The post-mortem revealed a hitherto clini- cally undiagnosed (ipsilateral) renal cell carcinoma (also
G1) with lung and osseous metastases. Immunocyto- chemistry with D11 and for NSE, Chr A and SYN in both neoplasms gave negative results.
In 2 patients, adrenal metastases, weighing 90 g and 195 g, respectively, offered the first indication of a large cell lung carcinoma. One tumour was initially mistaken for a non-functioning PCC as a result of a questionable (false-positive) NSE immunostaining. The patient died 2 months after adrenalectomy. Post-mortem confirmed the presence of a metastatic lung carcinoma. Immuno- histological reclassification revealed doubtful NSE reac- tivity in the adrenal metastasis and in the primary tu- mour (post-mortem histology) with no staining for Chr A and SYN or D11. In the second case, the adrenal neoplasm was typed initially as non-functioning adreno- cortical tumour of indeterminate biological potential. Twelve months later, biopsies carried out on enlarged cervical lymph nodes and on a lung tumour which had, in the meantime, been located radiologically, showed the same histological appearance. Immunocytochemi- cally, D11 and antibodies against NSE, Chr A and SYN gave negative results in all three specimens.
Two parasympathetic and 3 extra-adrenal sympa- thetic paragangliomas all gave negative results with D11 and positive results when tested for NSE, Chr A and SYN. Of 25 primary renal cell carcinomas (10 graded as G1 and G2 respectively and 5 graded as G3), no tu- mour showed D11 immunoreactivity. Out of 60 primary thyroid carcinomas (15 examples of each papillary, fol- licular, medullary and anaplastic tumours), only in 1 papillary neoplasm was there a focal diffuse-cytoplas- mic D11 positivity, accompanied by the absence of any nuclear staining.
Discussion
Our findings show that the monoclonal antibody D11 is capable of immunocytochemical identification of ad- renocortical cells, irrespective of their function or biolog- ical potential. All 27 specimens of normal cortex and all 49 adrenocortical tumours or tumour-like lesions were D11 positive, whereas all 50 specimens of normal, hyperplastic of neoplastic adrenomedullary tissue and 4 secondary adrenal tumours, 5 paragangliomas, 25 re- nal and 59 of 60 thyroid carcinomas were D11 negative. The significance of D11 for tumour diagnosis lies initial- ly, therefore, in its ability to distinguish clearly cortical from medullary or secondary adrenal tumours. More- over, our results shows that D11 also provides evidence of the histogenesis in individual adrenocortical tumours.
In all normal and neoplastic cortical cells we found nuclear D11 positivity hitherto unencountered in any other organ. An additional intense cytoplasmic D11 re- activity appeared only in the central zones of the cortex, which histochemical and experimental findings showed to be the site of glucocorticoid, androgen and oestrogen synthesis (Dhom 1981; Page et al. 1986). The aldoster- one-producing peripheral cortex, however, showed only a faint granular cytoplasmic staining at most. Both of these patterns of D11 reactivity proved to be the same in all functioning cortical tumours in which the sites
of origin could be deduced by means of the clinical symptoms - either Cushing’s or Conn’s syndrome. The constantly obvious correlation between the D11 pattern and the endocrine cell capacity found in functioning ad- renocortical tumours is noteworthy, since electron mi- croscopically there is no continuous homology in the ultrastructure of normal and neoplastic cells of the var- ious cortical zones (Probst 1965; Mackay 1969; Som- mers and Terzakis 1970; Mitschke et al. 1973). Even the recently described detection of P-450 cytochromes in ad- renocortical adenomas (Sasano et al. 1988a, b) does not allow an exact topographical classification, since this en- zyme system is present in many cells from all three corti- cal zones (Geuze et al. 1987).
Since D11 staining also gave a uniformly positive result in non-functioning adrenocortical tumours, this method does not allow conclusions to be drawn about the functional capacity of individual neoplasms. The de- tection of the two distinct D11 patterns described above in different cortical layers and in both functioning neo- plasms and tumours not presenting with endocrine ab- normalities suggests that in these particular zones of the adrenal cortex, structural proteins exist which are not associated with specific hormonal synthetic capacity. These proteins also appear to be expressed after neoplas- tic transformation and make histogenetic typing possi- ble.
The proteins responsible for the D11 staining of ad- renocortical cells have not been identified. From our biochemical experiments, D11 recognises several 59 kDa antigens which possess the same antigenetic determi- nants and are present in various organs. In vitro findings show that these proteins are capable of binding apolipo- protein E (Apo E), which is present in trigylceride-rich chylomicron remnants in high concentrations (Beisiegel et al. 1988). One of the 59 kDa proteins was identified by cDNA-sequencing as being protein disulphide- isomerase (PDI) (Ihrke and Stoffel, unpublished results). The cytoplasmic D11 positivity of cultivated HepG2 cells after membrane permeabilisation agrees with the results reported for the immunolocalisation of PDI by Kaetzel et al. (1987). Further analogies exist in the D11 positivity of fibrocytes in various organ capsules, since, according to the findings of Myllylä et al. (1983), PDI activity and rates of procollagen synthesis are positively correlated. No plausible explanation can be found, how- ever, for the presence of PDI and/or of an Apo E-bind- ing protein in the adrenal cortex. The organ-specific nu- clear and the cytoplasmic D11 immunoreactivities of ad- renocortical cells are thus presumably based on two dif- ferent 59 kDa proteins which correspond physiologically neither to the PDI nor to an Apo E-binding protein and need to be more accurately determined in further tests.
The diagnosis relevance of the antibody D11 (as con- firmed through the certain exclusion of PCC) is indepen- dent of the biochemical characterisation of the D11 im- munoreactive adrenocortical proteins. Our observation of cortical neoplasms with phenotypic similarity to med- ullary tumours (and vice versa) emphasises the well- known likelihood of confusion between the two tumour
forms, as has been described authoritively in mono- graphs (Page et al. 1986) as well as in individual case reports (Ramsay et al. 1987). Furthermore, such confu- sion cannot always be avoided by applying commercial antibodies. According to our own and others experience (Heitz 1987), the diagnostic usefulness of NSE immuno- staining is hampered by an occasionally inconclusive re- sult. Thus in 9 lesions which were clearly related to the cortex (and which were D11 positive), we observed an NSE reaction which could not clearly be defined as nega- tive. Conversely, amongst the PCCs in this series, in one case a questionable and in another case a negative NSE reaction was observed. As shown by ourselves and by others (Hacker et al. 1988; Kimura et al. 1988) PCCs exhibiting either doubtful or no NSE reactivity can, in the majority of cases, be defined with certainty by Chr A or SYN immunocytochemistry or by silver staining. The absence of argyrophilia and Chr A reactivity in re- spectively, 5 and 2 of the PCCs in our material and in several cases described in other series (Hamid et al. 1987; Hacker et al. 1988) indicate the limitations of us- ing these methods. It should be added that in about 5% of the adrenomedullary tumours - 9% (4/44), in our study - there is no clinical indication of PCC symp- toms (Scott et al. 1982; Shapiro et al. 1984; Plouin et al. 1987; Samaan and Hickey 1987a; Krause et al. 1988). Thus, in such cases, the correct diagnosis can only be determined by the pathologist.
More important than the exclusion of PCC would appear to be the possibility of differentiating between primary and secondary adrenal neoplasms. In this series 3 secondary lesions were mistaken initially for adrenal tumours, 2 of which were thought to be benign. The D11 negativity of 2 clear-cell renal tumours metastatic to the adrenal was confirmed by identical results in 25 primary renal carcinomas. This result is of great value; it has barely been possible to distinguish between adrenal infiltrations of such renal secondaries and pri- mary adrenocortical neoplasms because of their pheno- typic similarity in the absence of a specific means of identification. The single case of one D11 positive papil- lary neoplasm occurring amongst 60 primary thyroid carcinomas does not cast doubt on this evidence, but underlies a demand for further systematic analyses of the expression patterns of D11-positive antigens in other normal and neoplastic tissues. In so far as such studies sustain our assumption of a specificity of the nuclear D11 reactivity type for adrenocortical tumours, it fol- lows that this antibody must be introduced into the pan- el of immunocytochemical probes used to type occult neoplasms. The reason for this is that up to 60% of adrenocortical carcinomas do not produce endocrine manifestations; 50% of the cases have developed metas- tasis at the time of clinical diagnosis (Samaan and Hick- ey 1987b); this may represent the first signs of the illness (Page et al. 1986).
References
Beisiegel U, Kita T, Anderson RGW, Schneider WJ, Brown MS, Goldstein JL (1981) Immunologic cross-reactivity of the low
density lipoprotein receptor from bovine adrenal cortex, human fibroblasts, canine liver and adrenal gland, and rat liver. J Biol Chem 1256:4071-4078
Beisiegel U, Weber W, Havinga JR, Ihrke G, Hui DY, Wernette- Hammond ME, Turck CW, Innerarity TL, Mahley RW (1988) Apolipoprotein E-binding proteins isolated from dog and hu- man liver. Arteriosclerosis 8:288-297
Dhom G (1981) Die Nebennierenrinde. In: Doerr W, Seifert G, Uehlinger E (eds) Spezielle pathologische Anatomie, vol 14/ii. Pathologie der endokrinen Organe II. Springer, Berlin Heidel- berg New York, pp 729-970
Dietel M, Arps H, Gerding D, Trapp M, Niendorf A (1987) Estab- lishment of primary cell cultures: experiences with 155 cell strains. Klin Wochenschr 65:507-512
Dralle H, Schröder S, Gratz KF, Grote R, Padberg B, Hesch RD (1990) Sporadic unilateral adrenomedullary hyperplasia with hypertension cured by adrenalectomy. World J Surg (in press)
Geuze HJ, Slot JW, Yanagibashi K, McCracken JA, Schwartz AL, Hall PF (1987) Immunogold cytochemistry of cytochromes P- 450 in porcine adrenal cortex. Two enzymes (side-chain cleav- age and 11 beta-hydroxylase) are co-localized in the same mito- chondria. Histochemistry 86:551-557
Grimelius L, Wilander E (1984) Silver impregnation and other non-immunocytochemical staining methods. In: Polak JM, Bloom SR (eds) Endocrine tumours: the pathobiology of regu- latory peptide-containing tumours. Churchill-Livingstone, Edinburgh, pp 95-115
Hacker GW, Bishop AE, Terenghi G, Varndell IM, Aghahowa J, Pollard K, Thurner J, Polak JM (1988) Multiple peptide production and presence of general neuroendocrine mar- kers detected in 12 cases of human phaeochromocytoma and in mammalian adrenal glands. Virchows Arch [A] 412:399- 411
Hamid Q, Varndell IM, Ibrahim NB, Mingazzini P, Polak JM (1987) Extraadrenal paragangliomas. An immunocytochemical and ultrastructural report. Cancer 60:1776-1781
Heitz PU (1987) Neuroendocrine tumor markers. In: Seifert G (ed) Morphological tumor markers. Springer, Berlin Heidelberg New York, pp 279-306
Hsu SM, Raine L, Fanger H (1981) Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a compari- son between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 29:577-580
Hui DY, Brecht WJ, Hall EA, Friedman G, Innerarity TL, Mahley RW (1986) Isolation and characterization of the apolipoprotein E receptor from canine and human liver. J Biol Chem 261:4256-4267
Kaetzel CS, Rao CK, Lamm ME (1987) Protein disulphide-isomer- ase from human placental and rat liver. Purification and immu- nological characterization with monoclonal antibodies. Bio- chem J 241: 39-47
Kimura N, Sasano N, Yamada R, Satoh J (1988) Immunohisto- chemical study of chromogranin in 100 cases of pheochromocy- toma, carotid body tumour, medullary thryoid carcinoma and carcinoid tumour. Virchows Arch [A] 413:33-38
Krause M, Reinhardt D, Kruse K (1988) Phaeochromocytoma without symptoms: desensitization of the a- and ß-adrenocep- tors. Eur J Pediatr 147:121-122
Mackay AM (1969) Atlas of human adrenal cortex ultrastructure. In: Symington T (ed) Functional pathology of the human adre- nal gland, part IV. Williams and Wilkins, Baltimore, pp 346- 489
Mitschke H, Saeger W, Breustedt JJ (1973) Zur Ultrastruktur der Nebennierenrindentumoren beim Cushing-Syndrom. Virchows Arch [A] 360:253-264
Myllylä R, Koivu J, Pihlajaniemi T, Kivirikko KI (1983) Protein disulphide-isomerase activity in various cells synthesizing colla- gen. Eur J Biochem 134:7-11
Neville DM (1971) Molecular weight determination of protein- dodecyl sulfate complexes by gel electrophoresis in a discontin- uous buffer system. J Biol Chem 246:6328-6334
Page DL, DeLellis RA, Hough AJ (1986) Tumors of the adrenal. Atlas of tumor pathology, 2nd series, fascicle 23. Armed Forces Institute of Pathology, Washington, DC
Plouin PF, Chatellier G, Delahousse M, Rougeot MA, Duclos JM, Pagny JY, Corvol P, Ménard J (1987) Recherche, diagno- stic et localisation du phéochromocytome. 77 case dans une population de 21 420 hypertendus. Presse Med 16:2211- 2215
Probst A (1965) Elektronenmikroskopie der Nebennierenrinde bei primärem Aldosteronismus. Beitr Pathol Anat 131:1-21
Ramsay JA, Asa SL, Nostrand AWP van, Hassaram ST, Harven EP de (1987) Lipid degeneration in pheochromocytomas mi- micking adrenal cortical tumors. Am J Surg Pathol 11:480- 486
Samaan NA, Hickey RC (1987a) Pheochromocytoma. Semin On- col 14:297-305
Samaan NA, Hickey RC (1978b) Adrenal cortical carcinoma. Semin Oncol 14:292-296
Sasano H, Okamoto M, Sasano N (1988a) Immunohistochemical study of cytochrome P-450 11 ß-hydroxylase in human adrenal
cortex with mineralo- and glucocorticoid excess. Virchows Arch [A] 413:313-318
Sasano H, White PC, New MI, Sasano N (1988b) Immunohisto- chemical localization of cytochrome P-450 C21 in human adre- nal cortex and its relation to endocrine function. Hum Pathol 19:181-185
Schröder S, Niendorf A, Dietel M, Beisiegel U (1989) Immunocyto- chemical differential diagnosis of adrenocortical neoplasm through new monoclonal antibody D11 (A 1193). Proc Am As- soc Cancer Res 30:300
Scott HW, Reynolds V, Green N, Page D, Oates JA, Robertson D (1982) Clinical experience with malignant pheochromocyto- mas. Surg Gynecol Obstet 154:801-818
Shapiro B, Sisson JC, Lloyd R, Nakajo M, Satterlee W, Beierwaltes WH (1984) Malignant pheochromocytoma: clinical, biochemi- cal and scinigraphic characterization. Clin Endocrinol 20:189- 203
Sommers SC, Terzakis JA (1970) Ultrastructural study of aldoster- one-secreting cells of the adrenal cortex. Am J Clin Pathol 54:303-308