Production of the alpha subunit of guanine nucleotide-binding protein GO by neuroendocrine tumors.

We have found that neuroendocrine tumors (including neuroblastoma, ganglioneuroma, gut carcinoid, pheochromocytoma, medullary thyroid carcinoma, insulinoma, glucagonoma, prolactinoma, carotid body tumor, and small cell lung carcinoma) produce considerable amounts (about 1000-80,000 ng/g tissue) of the alpha subunit of guanine nucleotide-binding protein, GO (GO alpha), whereas nonneuroendocrine tumors contain less than 300 ng of GO alpha/g tissue. GO alpha in the neuroendocrine tumors was present both in the soluble fraction, and cholate-extractable membrane-bound fraction of tissues. Immunoblots of membrane fractions of neuroblastoma and carcinoid tissues confirmed that the immunoreactive substance in the tumor tissues was GO alpha. Immunohistochemically, GO alpha was localized consistently in the cell membrane and occasionally in the cytoplasm of neuroendocrine tumors. GO alpha was also detected in sera of 73% patients with neuroblastoma at diagnosis, whereas serum GO alpha concentrations in control children, or patients with nonneuroendocrine tumors were lower than the detection limit of the immunoassay method employed. Serum GO alpha concentrations in patients with neuroblastoma changed with the clinical course; they fell in patients responding to treatment and increased in patients who relapsed. Since GO alpha, a specific protein in the neural and neuroendocrine cells, was found to be produced in considerable amounts by all types of neuroendocrine tumors but not in nonneuroendocrine tumors, GO alpha might be a useful biomarker for neuroendocrine tumors.

Expression of aldolase C isozyme in renal cell carcinoma.

The authors localized aldolase C in renal tubules and renal cell carcinoma by immunohistochemical study and quantitative analysis by an enzyme immunoassay. Aldolase C was localized in epithelial cells of loops of Henle and collecting ducts and in those of Bowman's capsules. In renal cell carcinoma, aldolase C was immunohistochemically demonstrated in 95% (41 of 43) of cases, including one sarcomatoid variant. The tissue concentrations of aldolase C in the renal cortex (n = 8) were 12.7 +/- 6.2 micrograms/g protein (mean +/- standard deviation), and those of the medulla (n = 8) were 20.3 +/- 6.9 micrograms/g protein. On the other hand, the concentrations in renal cell carcinoma (n = 26) were 93.5 +/- 95.9 micrograms/g protein: about seven times higher than that in renal cortex (P less than 0.001). These findings indicate that aldolase C was first expressed in renal cell carcinoma, which is derived from proximal renal tubules, because proximal renal tubules had aldolase B but not aldolase C.

Immunohistochemical characteristics of proliferative and metaplastic lesions in bronchial mucosa.

To elucidate the characteristics of metaplastic changes of bronchial mucosa, the distribution of four epithelial antigens and two subtypes of enolases was studied immunohistochemically. The authors classified the metaplastic changes into three types: basal cell hyperplasia, stratification, and squamous metaplasia. Secretory component (SC) was detected in all lesions with stratification and in three of ten lesions of squamous metaplasia. Epithelial membrane antigen (EMA) was localized not only on the luminal surfaces but also among the stratified layers of metaplastic epithelium. No carcinoembryonic antigen (CEA) was detectable in normal epithelium, whereas almost all metaplastic lesions had both nonspecific cross-reacting antigen (NCA) and CEA immunoreactivity. In normal mucosa, basal cells were strongly positive for alpha-enolase but negative for gamma-enolase, and columnar cells expressed both enolases. In areas of metaplasia, alpha-enolase was present throughout the layer, but gamma-enolase was absent. These immunohistochemical findings suggest that metaplastic squamous epithelial cells have glandular differentiation and that several biochemical and metabolic aberrations occur during the process of metaplasia.

Highly sensitive enzyme immunoassay for human brain aldolase C.

A sensitive sandwich-type enzyme immunoassay for brain-type isozyme of human aldolase C4 was developed using purified antibodies specific to the C subunit. The antibodies were raised in rabbits by injecting the purified aldolase C4, and purified by means of immunoaffinity chromatography on a column of aldolase C4-coupled Sepharose. The assay system consisted of polystyrene balls with immobilized antibody F(ab')2 fragments and the same antibody Fab' fragments labelled with beta-D-galactosidase from Escherichia coli. The assay was highly sensitive and the minimum detection limit of aldolase C4 was 3 pg/tube. The assay was specific to the C subunit of aldolase (aldolase C). It cross-reacted about 60% with aldolase AC3, 30% with aldolase A2C2, and 4% with aldolase A3C, but showed no cross-reactivity with aldolase A4, the muscle-type isozyme. Coefficients of variation in within-run and between-run precision studies for serum aldolase C were less than or equal to 11%. Serum aldolase C levels in healthy adults of various ages (16-59 yr old) and both sexes ranged from 8.74-18.9 ng/ml. Immunoreactive aldolase C in the extracts of various human tissues was determined. It was distributed at high concentrations in the central nervous tissue and heart and at significant levels in liver, adrenal glands and testis. The assay of aldolase C in cerebrospinal fluid or serum by employing this sensitive immunoassay might be useful in the diagnosis of neurological disorders or acute myocardial damage.

Sensitive enzyme immunoassay for human aldolase B.

A highly sensitive enzyme immunoassay system for measurement of aldolase B subunit (aldolase B) was established. Antisera were raised in rabbits by injecting aldolase B4 purified from human liver, and specific antibodies to aldolase B were purified by the use of a column of aldolase B4-coupled Sepharose. The purified antibody IgG was digested with pepsin to obtain the F(ab)' fragments. The antibody F(ab)' fragments were immobilized noncovalently on polystyrene balls, and the same antibody Fab' fragments were labeled with beta-D-galactosidase from Escherichia coli. The sandwich-type assay system using these reagents was sensitive and specific to aldolase B, showing no cross-reactivity with aldolase-A or aldolase-C. The minimum detection limit of the assay was 3 pg aldolase B4/assay tube. The immunoreactive aldolase B was present at high levels in the liver and kidney, and considerably in the small intestine. It was detected in all the tissues examined. Immunohistochemically, aldolase B is localized in hepatocytes, proximal renal tubular cells and epithelial cells of small intestine. Serum levels of aldolase B in healthy subjects were ranged from 33 to 202 ng/ml.

[Immunohistochemical localization of epithelial membrane antigen, carcinoembryonic antigen and secretory component in urinary bladder cancer].

To clarify the relationship between the immunohistochemical distribution pattern of epithelial antigens in transitional cell carcinomas and their histopathological grading and staging, epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA) and secretory component (SC) were localized. Formalin-fixed paraffin-embedded sections from 55 patients with bladder carcinoma were stained by the indirect immunoperoxidase method. In normal transitional epithelium, EMA was found on the luminal side of the plasma membrane of a few surface layers, and in the cytoplasm of the superficial cells. In the lower grade and stage of transitional cell carcinoma, only the luminal surface of superficial cells was positively stained. Membrane and cytoplasmic staining of EMA was frequently found in the intermediate and basal layers of the carcinoma, and the incidence of cytoplasmic staining increased with the higher grade and stage. CEA was not detected in normal epithelium. Cytoplasmic staining of CEA was progressively more frequent in the higher grade and stage of transitional cell carcinoma. In normal epithelium SC was observed on the apical surface plasma membrane and in the cytoplasm of the superficial cells, as shown in the immunohistochemical staining for EMA. The correlation of immunohistochemical detection of SC with the grade or stage was not as good as the correlations for EMA or CEA. These findings suggest that immunohistochemical examination for EMA, CEA and SC in bladder carcinoma could provide valuable information for grading or staging in pathological diagnosis.

[Localization of enolase in the cancer nest of digestive organs].

Gamma-enolase, otherwise called neuron specific enolase (NSE), is specific for neural and neuro-endocrine systems and is present in the tumors of these systems. Gamma-enolase is also reported to be present, however, in a localized form in the tissues of other nonneural systems in some cases of lung cancer in relatively large number of incidences. We examined immunohistologically specimens of digestive organ cancers. Two principal conclusions were obtained. First, gamma-enolase is not specific for neural and neuro-endocrine systems. Second, gamma-enolase is localized in the epithelium of cancers of some cases following the transformation into cancer.

S100a0 (alpha alpha) protein in cardiac muscle. Isolation from human cardiac muscle and ultrastructural localization.

S100 protein, an acidic and calcium-binding protein, was believed to be localized in the nervous tissue, but recently it has been reported to be mainly present in the cardiac and the skeletal muscles of various mammals in the alpha alpha form (S100a0) at much higher levels than the nervous tissues. We isolated here S100 protein from human cardiac muscles. The isolated cardiac muscle S100 protein showed a single band on electrophoresis at the same position as that of human skeletal muscle S100a0. The amino acid composition of the purified S100 protein was quite similar to that of human skeletal muscle S100a0 or bovine brain S100a0. The immunohistochemical study by use of antibodies monospecific to the alpha subunit of S100 protein (S100-alpha) revealed that S100-alpha was strongly labeled in human myocardial cells, whereas the beta subunit of S100 protein (S100-beta) was not detected in the cells. These results suggest that a predominant form of S100 protein in human myocardial cells is not S100a (alpha beta) or S100b (beta beta), but S100a0 (alpha alpha). In order to determine the ultrastructural localization of S100a0 in mouse cardiac muscle, the direct peroxidase-labeled antibody method was employed. S100a0 was mainly localized in the polysomes in the interfibrillar spaces, the fine filamentous structure of the Z line and fascia adherens of the intercalated disc and in the lumen of junctional sarcoplasmic reticulum.

S100a0 (alpha alpha) protein, a calcium-binding protein, is localized in the slow-twitch muscle fiber.

We previously showed that, in contrast to the distribution of S100b (beta beta), S100a0 (alpha alpha) is mainly present in human skeletal and heart muscles at the level of 1-2 micrograms/mg of soluble protein and is universally distributed at high levels in skeletal and heart muscles of various mammals. To elucidate cellular and ultrastructural localizations of the alpha subunit of S100 protein (S100-alpha) in skeletal muscle, we used immunohistochemical and enzyme immunoassay methods. The immunohistochemical study revealed that S100-alpha is mainly localized in slow-twitch muscle fibers, whereas the beta subunit of S100 protein (S100-beta) was not detected in both types of muscle fibers, an observation indicating that the predominant form of S100 protein in the slow-twitch muscle fiber is not S100a or S100b, but S100a0. The quantitative analysis using enzyme immunoassay corroborates the immunohistochemical finding: The S100-alpha concentration of mouse soleus muscle (mainly composed of slow-twitch muscle fibers) is about threefold higher than that of mouse rectus femoris muscle (mainly composed of fast-twitch muscle fibers). At the ultrastructural level, S100-alpha is associated with polysomes, sarcoplasmic reticulum, the plasma membrane, the pellicle around lipid droplets, the outer membrane of mitochondria, and thin and thick filaments, by immunoelectron microscopy.

Differential distribution of immunoreactive S100-alpha and S100-beta proteins in normal nonnervous human tissues.

Differential distribution of alpha-subunit (S100-alpha) and beta-subunit (S100-beta) of S100 protein in nonnervous tissues was studied by employing the indirect peroxidase-labeled antibody method with monospecific antibodies to human S100-alpha or bovine S100-beta. It became clear that S100-alpha and S100-beta were more widespread than previously reported. In light of this study, together with the previous report describing quantitative differential distribution of S100-alpha and S100-beta (Kato K, Kimura S: Biochim Biophys Acta, 842:146, 1985), S100-alpha was localized at much higher levels in myocardial cells and slow-twitch muscle fibers, at high levels in renal tubules, chondrocytes, follicular cells of thyroid, exocrine cells of salivary, mammary, and eccrine sweat glands, and centroacinar and acinar cells of the pancreas. On the other hand, S100-beta was mainly localized in Schwann cells, chondrocytes, and adipocytes. There were several tissues showing a remarkable differential localization of S100-alpha and S100-beta; heart, skeletal muscle, exocrine tissues such as mammary gland, salivary gland, and pancreas, liver, kidney, testis, epididymidis, and lymphoreticular system. Based on the wide distribution and the characteristic differential localization, the possible biologic significance of S100 proteins and their application for diagnostic pathology were discussed.

The ultrastructural changes of S-100 protein localization during lipolysis in adipocytes. An immunoelectron-microscopic study.

To elucidate the changes of ultrastructural localization of S-100 protein during lipolysis in adipocytes, an immunoelectron-microscopic study was performed. Epididymal fat pads from Wistar rats were incubated in the buffer with or without 10 microM epinephrine. Before incubation with epinephrine, S-100 protein was found to be associated with closely packed polysomes, the membrane of microvesicles, plasma membranes, the outer membrane of mitochondria, and the pellicle around fat droplets. In the epinephrine-treated tissues, however, S-100 protein-positive polysomes decreased drastically. S-100 protein-positive microvesicles increased in number, lined up below the plasma membranes, and fused with the plasma membrane, frequently opening into the interstitium. These microvesicles were also found around the lipid droplets. These findings, together with those of a previous report on ultrastructural changes of adipocytes during lipolysis, suggest that S-100 protein molecules interact with free fatty acids (FFAs) on their hydrophobic portions on the membrane of microvesicle and then are translocated through the cytoplasm and discharged from the surface of plasma membranes with FFAs into the interstitium. That is, S-100 protein might serve as one of carrier proteins of FFAs in adipocytes.

Expression of enolases in T cell tumors and Hodgkin's disease.

Frozen biopsy specimens taken from 30 cases with T cell tumors (8 with T cell acute lymphoblastic leukemia, 8 with T cell lymphoblastic lymphoma, and 14 with peripheral T cell lymphomas), and from 12 with Hodgkin's disease, were investigated using a direct immunohistochemical method to detect alpha-, beta- and gamma-enolases. Normal thymus and lymph node specimens with reactive lymphadenitis were also investigated. Subcortical thymocytes and the majority of deep cortical thymocytes showed reactivity of alpha-/beta-/gamma- approximately +/- -enolases, and medullary thymocytes and small lymphocytes in T zone areas of lymph node showed reactivity of alpha-/beta+/gamma- approximately +/- -enolases. Seven of the 8 cases with T cell acute lymphoblastic leukemia showed reactivity of alpha-/beta-/gamma(-)-enolases or alpha+/beta-/gamma(-)-enolases in leukemic lymphocytes, 7 of the 8 cases with T cell lymphoblastic lymphoma showed reactivity of beta(+)-enolase, and all 14 cases with peripheral T-cell lymphomas showed reactivity of alpha-/beta- approximately +/gamma(+)-enolases in lymphoma cells. All the 12 cases with Hodgkin's disease showed reactivity of alpha-/beta+/gamma(+)-enolases in Reed-Sternberg and Hodgkin's cells. These results indicate the following: (a) The neoplastic cells of T cell acute lymphoblastic leukemia, T cell lymphoblastic lymphoma and peripheral T cell lymphomas present different expressions in each of these three categories. This may imply a difference of maturation and differentiation or activation among neoplastic T lymphocytes. (b) T lymphocytes may switch from alpha- to beta-enolase and from alpha- to gamma-enolase in the course of differentiation and activation. (c) It is worth noting that the Reed-Sternberg and Hodgkin's cells of Hodgkin's disease present an identical expression of enolases.

Aldolase C is localized in neuroendocrine cells.

To elucidate the localization of the subunit C of aldolase (aldolase C) in peripheral neuroendocrine cells, we made an immunohistochemical study with monospecific antibodies against human aldolase C. Aldolase C was found to be localized in various types of neuroendocrine cells; in the pituitary gland, thyroid, pancreas, adrenal gland, bronchus, and gastrointestinal tract.

Expression of enolases in B cell tumors.

Frozen lymph node biopsy specimens from 38 patients with B cell tumors, including 5 with childhood non-T-ALL and 3 with reactive lymphadenitis, were investigated using a direct immunohistochemical method to detect alpha-, beta- and gamma-enolases. alpha-Enolase-positive cells were not observed in reactive lymphadenitis. On the contrary, almost all the lymphocytes including germinal center cells were positive for beta-enolase. Small lymphocytes in the mantle zones were negative, centrocytes were negative or weakly positive, the majority of centroblasts were strongly positive and the remaining were weakly positive for gamma-enolase. In all 5 patients with childhood non-T-ALL, leukemic lymphocytes were strongly positive only for alpha-enolase. In all 33 patients with B cell lymphoma, lymphoma cells were positive for beta-enolase. In many patients with follicular lymphoma, lymphoma cells were positive only for beta-enolase. Four of five patients with malignant lymphoma, diffuse, small cleaved cell, showed the reactivity of alpha-, beta+, gamma+-enolases in lymphoma cells. Our results suggest the possibility of the two isoenzyme switches from alpha- to beta-enolase and from alpha- to gamma-enolase in the B lymphocyte lineage accompanying differentiation, similar to those of skeletal muscles and neurons.

S100a0 (alpha alpha) protein: distribution in muscle tissues of various animals and purification from human pectoral muscle.

When the concentrations of alpha-S100 (alpha subunit of S100 protein) and beta-S100 (beta subunit) proteins in various tissues of human and rat were determined by the immunoassay method, immunoreactive beta-S100 was present at high levels in the CNS, adipose tissue, and cartilaginous tissue. In contrast, the alpha-S100 was found in the heart and skeletal muscles at concentrations much higher than in the CNS. The concentration of alpha-S100 protein was also high in the heart and skeletal muscles of bovine, porcine, canine, and mouse. Since beta-S100 protein levels in those tissues were low, it was suggested that S100 protein in the muscle tissues is present mainly as the alpha alpha form (S100a0 protein). To confirm the above findings, immunoreactive alpha-S100 protein was purified from human pectoral muscle by employing column chromatographies with butyl-Sepharose, diethylaminoethyl (DEAE)-Sepharose, Sephadex G-75, and finally with an anion-exchange Mono Q column in a HPLC system. The elution profile of alpha-S100 protein from the Mono Q column suggested some heterogeneity of the final preparation. However, each of these fractions traveled with a single band at a position similar to that of bovine S100a0 protein on slab-gel electrophoresis. The amino acid composition of the final preparation was very similar to the composition of bovine S100a0 protein. The purified alpha-S100 protein was eluted from a gel-filtration column (Superose 12) in the same fraction as bovine S100a0 protein.(ABSTRACT TRUNCATED AT 250 WORDS)

Enolase isozymes in seminoma.

We determined concentrations of alpha and gamma-enolases in normal testis and in seminoma tissues by enzyme immunoassay. Concentrations of alpha-enolase were 4,170 +/- 2,040 ng/mg protein in normal testis (n = 8) and 8,140 +/- 4,480 ng/mg protein in seminoma (n = 8). Concentrations of gamma-enolase in seminoma (460 +/- 571 ng/mg protein) were significantly higher than those of normal testis (59 +/- 15 ng/mg protein). Immunohistochemistry showed positive tumor cells for gamma-enolase in 6 of 8 seminoma cases (75%). Serum gamma-enolase levels were elevated (greater than 6.0 ng/ml) in 9 of 12 patients (75%) with seminoma: 60% of stage I, and 100% of stages II and III. In 10 patients treated by surgical excision and chemotherapy, serum gamma-enolase was significantly reduced after the treatment. These findings indicate that elevated serum gamma-enolase is derived from enhanced gamma-enolase in seminoma tissues, and that serum gamma-enolase could be a useful biomarker for staging and monitoring clinical course in patients with seminoma.

An immunochemical and immunohistochemical study of S100 protein in renal cell carcinoma.

The authors localized S100 protein in renal tubules and renal cell carcinoma by immunohistochemical study and quantitative analysis by enzyme immunoassay. The alpha subunit of S100 protein (S100-alpha) was localized in epithelial cells of proximal tubules, thin limbs of loops of Henle, collecting tubules, and a few of Bowman's capsules. The beta subunit (S100-beta) was immunostained in the distal tubules, thick and thin loops of Henle, collecting tubules, and a few proximal tubules. In renal cell carcinoma, S100-alpha was immunohistochemically demonstrated in 82% (41/50) of patients including sarcomatoid variants, whereas S100-beta was detected in 46% (23/50). Both the number of positively stained tumor cells and the staining intensity were greater in S100-alpha than in S100-beta. Concentrations of S100-alpha in the cortex were 80.3 +/- 22.5 ng/mg protein (n = 7), whereas those of renal cell carcinoma were 387 +/- 533 ng/mg protein (n = 19), i.e., about five times higher. Concentrations of S100-beta in both normal kidney (1.96 +/- 0.74 ng/mg protein) and renal cell carcinoma (2.05 +/- 2.16 ng/mg protein) were much lower than those of S100-alpha. The authors also localized S100-alpha and S100-beta in tissues of other renal tumors and tumors arising from other organs. S100a0 appears to be a useful immunohistochemical marker for renal cell carcinoma.