Epithelial membrane antigen: partial purification, assay and properties

A simple diagnostic index comprising epithelial membrane antigen and fibronectin for hepatocellular carcinoma

Background and rationale for the study. Continuing search for suitable tumor-markers is of clinical value in managing patients with various malignancies. These markers may be presented as intracellular substances in tissues or may be released into the circulation and appear in serum. Therefore, this work is concerned with identification and quantitative determination of epithelial membrane antigen (EMA) and fibronectin and estimating their performances as surrogate markers for identifying hepatocellular carcinoma (HCC).

RESULTS

A total of 627 individuals constituted this study [fibrosis (F1-F3) = 217; cirrhosis = 191; HCC = 219]. Western-blot was used for identifying EMA and fibronectin in sera. As a result, a single immunoreactive band was shown at 130-kDa and 90-kDa corresponding to EMA and fibronectin, respectively. They were quantified using ELISA providing values in HCC higher than fibrosis or cirrhosis with a significant difference (P < 0.0001). For identifying HCC, EMA showed 0.82 area under receiver-operating characteristic curve (AUC) with sensitivity = 70% and specificity = 78% while fibronectin yielded AUC = 0.70 with sensitivity = 67% and specificity = 82%. FEBA-Test comprising fibronectin and EMA together with total-bilirubin and AFP was constructed yielding AUC = 0.92 for identifying HCC from cirrhosis with sensitivity = 89% and specificity = 85%. FEBA-Test was then tested for differentiating HCC from fibrosis showing AUC = 0.97 with sensitivity = 90% and specificity = 89%. FEBA-Test enabled the correct identification of HCC patients with CLIP 0-1 and size ≤ 3 cm with AUC = 0.80 and AUC = 0.84, respectively, indicating its ability in identifying early HCC.

CONCLUSIONS

A four-marker index may improve the early detection of HCC with a high degree of accuracy.

Combined use of Epithelial Membrane Antigen and Nuclear Matrix Protein 52 as Sensitive Biomarkers for Detection of Bladder Cancer

Background

The advent of noninvasive urine-based markers as well as other novel modalities has yielded improved diagnostic accuracy. However, the new markers failed to reach higher sensitivity and specificity. We therefore evaluated the potential role of epithelial membrane antigen (EMA) and nuclear matrix protein 52 (NMP-52) singly and combined as noninvasive biomarkers for the detection of bladder cancer (BC).

Methods

A total of 160 individuals including 66 patients with BC, 54 patients with benign urologic disorders and 40 healthy volunteers were investigated. Urinary EMA at 130 kDa and NMP at 52 kDa were identified, purified and quantified by Western blot, electroelution and enzyme-linked immunosorbent assay (ELISA). The diagnostic performance of each biomarker and their combination were compared using area under receiver operating characteristic curves (AUC).

Results

Mean urinary EMA, 2.42 µg/mL, and NMP-52, 17.85 µg/mL, were significantly elevated in patients with BC compared to controls, 1.18 and 3.44 µg/mL, respectively (p<0.0001). The combined use of these markers yielded values which were increased 4.4- and 13.7-fold in the benign and malignant disease groups, respectively, with respect to the normal group. The values of EMA and NMP-52 were significantly higher in patients with higher-grade tumors than those with lower-grade tumors (p<0.0001). Moreover, this combination could predict all BC stages and grades with 0.91 AUC, 94% sensitivity and 80% specificity.

Conclusions

EMA and NMP-52 in combination could be promising noninvasive biomarkers for BC detection.

Non-invasive predictive score of fibrosis stages in chronic hepatitis C patients based on epithelial membrane antigen in the blood in combination with routine laboratory markers

AIM

  The epithelial membrane antigen (EMA) could detect small deposits of liver malignant cells. However, no information exists regarding the use of EMA in patients with chronic hepatitis C (CHC). Therefore, we attempted to evaluate the diagnostic performance of EMA to distinguish patients with different liver fibrosis stages.

METHODS

  Epithelial membrane antigen was identified in sera of 154 CHC patients using Western blot and enzyme linked immunosorbent assay (ELISA). Multivariate discriminant analysis (MDA) and receiver operating characteristic (ROC) curve analyses were used to create a predictive model including EMA in addition to a panel of routine blood markers. A combination algorithm was developed and validated prospectively in 170 CHC additional patients.

RESULTS

  Epithelial membrane antigen at 130 kDa was identified, purified and quantified in sera of CHC patients using ELISA. Based on these encouraging results, we purified and developed a direct ELISA for the quantitation of EMA in sera of CHC. MDA selected a score for the prediction of significant liver fibrosis patients based on measurements of EMA, aspartate aminotransferase to platelet ratio index and albumin. Areas under the ROC curves (AUC) of the score for the three biomarkers were 0.82 for patients with liver fibrosis (F1-F4), 0.86 for significant liver fibrosis (F2-F4), 0.87 for advanced liver fibrosis (F3-F4) and 0.86 for liver cirrhosis (F4). The results of the validation study demonstrated that (74%) of patients could have avoided liver biopsy.

CONCLUSION

  This score was validated for the prediction of liver fibrosis stages and may minimize the need for liver biopsy.

Structure and function of human sweat glands studied with histochemistry and cytochemistry

The basic structure and the physiological function of human sweat glands were reviewed. Histochemical and cytochemical techniques greatly contributed the elucidation of the ionic mechanism of sweat secretion. X-ray microanalysis using freeze-dried cryosections clarified the level of Na, K, and Cl in each secretory cell of the human sweat gland. Enzyme cytochemistry, immunohistochemistry and autoradiography elucidated the localization of Na,K-ATPase. These data supported the idea that human eccrine sweat is produced by the model of N-K-2Cl cotransport. Cationic colloidal gold localizes anionic sites on histological sections. Human eccrine and apocrine sweat glands showed completely different localization and enzyme sensitivity of anionic sites studied with cationic gold. Human sweat glands have many immunohistochemical markers. Some of them are specific to apocrine sweat glands, although many of them stain both eccrine and apocrine sweat glands. Histochemical techniques, especially immunohistochemistry using a confocal laser scanning microscope and in situ hybridization, will further clarify the relationship of the structure and function in human sweat glands.

MUC1 and MUC-X, epithelial mucins of breast and milk

While reasons for gross variations in milk composition of the various mammalian species are well established, we will be revealing minor, subtle, yet highly important differences for years to come. One of the distinctive differences of human milk concerns its mucins, MUC1 and MUC-X. These are high molecular mass glycoproteins that occur on the apical surface of the lactating cell and are transferred to the milk fat globule upon its secretion from the cell. Among mammals, it seems that the human has developed the largest, most extended versions of these mucin molecules. Evidence is accumulating that this greater size benefits the human in terms of protection against infections and injurious environmental agents.

Purification of MUC1 from bovine milk-fat globules and characterization of a corresponding full-length cDNA clone

The highly glycosylated protein MUC1 was purified from bovine milk-fat globule membranes by a procedure involving detergent extraction, ion-exchange chromatography and reverse-phase chromatography. The identity of the purified mucin protein was confirmed by N-terminal sequencing and partial amino acid sequences obtained by peptide mapping. The complete amino acid sequence of MUC1 was determined by cloning and sequencing the corresponding bovine mammary gland cDNA, which was shown to encode a protein of 580 amino acid residues comprising a cleavable signal peptide of 22 residues. The deduced amino acid sequence demonstrated structural features characteristic for mucins, including an extracellular tandem repeat region with 11 partially conserved repeats (20 amino acids each), a membrane-proximal SEA module, a transmembrane domain, and a cytoplasmic C-terminal region. Monosaccharide composition determinations suggested significant structural differences between O-linked glycans of MUC1 originating from either bovine or human milk. Interspecies differences of the consensus repeat sequence in MUC1 and the physiological functions are discussed.

Histochemical and immunohistochemical markers for human eccrine and apocrine sweat glands: an aid for histopathologic differentiation of sweat gland tumors

Apocrine and eccrine sweat glands are distinct in function, although they are closely related to each other developmentally and morphologically. In certain sweat gland tumors, it is difficult to differentiate between eccrine or apocrine sweat glands. Therefore, this paper reviews histochemical and immunohistochemical markers to differentiate apocrine and eccrine sweat glands with the aim of better understanding the structural and functional characteristics of these sweat glands. Specific markers for apocrine sweat glands are as follows: neuraminidase sensitive anionic sites detected by cationic colloidal gold at pH 2.0, and mitochondrion-like secretory granules that have epidermal growth factor-like antigenicity. The following antibodies react with apocrine sweat glands but not with eccrine sweat glands; the antibodies raised against 70 kDa glycoprotein purified from human milk fat globule membranes, and HMFG-1 (1.10.F3) monoclonal antibody produced by immunizing mice with defatted human milk fat globule membranes. Markers for eccrine sweat glands are as follows: dark cell granules that have chondroitinase ABC sensitive anionic sites detected by cationic gold at pH 2.0 after pretreatment with EGTA, and intercellular canaliculi with high activity of alkaline phosphatase. CEA and GCDFP-15 are expressed in both eccrine and apocrine sweat glands. Anti-EMA monoclonal antibody (E29) stains both eccrine and apocrine sweat glands.

A comparison of epithelial membrane antigen overexpression in benign and malignant endometrium

OBJECTIVE

The aim of this study was to analyze the value of epithelial membrane antigen (EMA) overexpression in benign and malignant endometrium and its prognostic significance.

METHODS

EMA immunostaining was performed in 178 paraffin-embedded specimens including 105 endometrial cancers, 40 endometrial hyperplasias, and 33 benign endometriums. EMA immunostaining was correlated with traditional prognostic factors and progression-free survival in endometrial cancer specimens.

RESULTS

EMA overexpression was observed more frequently in adenocarcinomas (60%) than in hyperplasias (15%) or benign endometrium (9.1%). EMA overexpression was observed in two patients with endometrial hyperplasia who progressed to carcinoma. In adenocarcinomas, EMA overexpression had a positive correlation with nonendometrioid subtypes (P = 0.012). In multivariate analysis, FIGO stage (P = 0.025) and EMA overexpression (P = 0.017) were independent prognostic factors for progression-free survival.

CONCLUSIONS

EMA overexpression appears to be a marker of malignant transformation in the endometrium and it is an independent predictor of recurrent disease in endometrial cancer.

A review and proposed nomenclature for major proteins of the milk-fat globule membrane

The characteristics and possible functions of the most abundant proteins associated with the bovine milk-fat globule membrane are reviewed. Under the auspices of the Milk Protein Nomenclature Committee of the ADSA, a revised nomenclature for the major membrane proteins is proposed and discussed in relation to earlier schemes. We recommend that proteins be assigned specific names as they are identified by molecular cloning and sequencing techniques. The practice of identifying proteins according to their Mr, electrophoretic mobility, or staining characteristics should be discontinued, except for uncharacterized proteins. The properties and amino acid sequences of the following proteins are discussed in detail: MUC1, xanthine dehydrogenase/oxidase, CD36, butyrophilin, adipophilin, periodic acid Schiff 6/7 (PAS 6/7), and fatty acid binding protein. In addition, a compilation of less abundant proteins associated with the bovine milk-fat globule membrane is presented.

Establishment and characterization of an immortalized human sebaceous gland cell line (SZ95)

Human facial sebaceous gland cells were transfected with a PBR-322-based plasmid containing the coding region for the Simian virus-40 large T antigen. The resulting proliferating cell cultures have been passaged over 50 times to date, have been cloned, and show no signs of senescence after 4&DF;1 2 y in vitro, whereas normal human sebocytes can only be grown for three to six passages. The immortalized transfected cells, termed SZ95, expressed the Simian virus-40 large T antigen and presented an hyper-diploid-aneuploid karyotype with a modal chromosome number of 64.5. The SZ95 cell line exhibited epithelial, polymorphous characteristics with different cell sizes of up to 3.25-fold during proliferation and 6-fold at confluence, showing numerous cytoplasmic lipid droplets. The cells showed large cytoplasm profiles with abundant organelles, including vacuoles and myelin figures which indicated lipid synthesis. Lack of or only few desmosomal areas were observed. SZ95 cells expressed molecules typically associated with human sebocytes, such as keratins 7, 13, and 19, and several proteins of the polymorphous epithelial mucin family. Functional studies revealed synthesis of the sebaceous lipids squalene and wax esters as well as of triglycerides and free fatty acids, even after 25-40 passages; active lipid secretion; population doubling times of 52.4 +/- 1.6 h; reduced growth but maintenance of lipid synthesis under serum-free conditions; and retrieval of cell proliferation after addition of 5alpha-dihydrotestosterone. Retinoids significantly inhibited proliferation of certain SZ95 cell clones in the expected magnitude 13-cis-retinoic acid > all-trans-retinoic acid > > acitretin. Thus SZ95 is an immortalized human sebaceous gland cell line that shows the morphologic, phenotypic and functional characteristics of normal human sebocytes.

Expression of MUC-1 glycoprotein in plasma cells, follicular dendritic cells, myofibroblasts and perineurial cells: immunohistochemical analysis using three monoclonal antibodies

Normal and malignant plasma cells (PC), follicular dendritic cells (FDC), myofibroblasts (MFB) and perineurial cells (PNC) were investigated for the expression of MUC-1 glycoprotein (MUC-1gp) by immunohistochemical and immunoelectron microscopic techniques using monoclonal antibodies E29, 115D8, DF3 and a combination of the three. MUC-1 glycoprotein-positive PC detected by the combined antibodies were frequently seen in a variety of pathological lesions tested, including chronic cervicitis, chronic synovitis, Hodgkin's disease, allergic rhinitis and sinusitis, tuberculous lymphadenitis, foreign body granuloma, multiple myeloma, and chronic tonsillitis. In the lesions containing MUC-1gp-positive PC, the infiltration of immunoglobulin (Ig) E PC and/or IgE-bound mast cells was significantly increased, but MUC-1gp-positive PC did not contain any specific immunoglobulin heavy or light chains. The findings suggest that the expression of MUC-1 gp in PC, although not restricted to IgE-class cells, may be induced in an allergic status. Plasma cells and PNC mainly reacted with the antibodies E29 and 115D8, while FDC and MFB were principally reactive with the antibody DF3. In some cases of multiple myeloma, the neoplastic PC were predominantly immunoreactive with DF3. The results indicate: (i) the epitopic variability of MUC-1gp molecules expressed on the non-epithelial cells; and (ii) the epitopic alterations during malignant transformation. It should also be noted that the expression of MUC-1gp in the non-epithelial cells represents a pitfall in histopathological diagnosis.

Expression of a glycoprotein of the carcinoembryonic antigen family in normal and neoplastic sebaceous glands. Limited role of carcinoembryonic antigen as a sweat gland marker

BACKGROUND

Carcinoembryonic antigen (CEA) is a well-known marker for sweat gland differentiation in adnexal neoplasms.

OBJECTIVE

The aim of this study was to examine the expression of glycoproteins of the CEA family, that is, CEA-180, nonspecific cross-reacting antigens (NCAs), and biliary glycoprotein (BGP), in sebaceous glands and in neoplasms with sebaceous differentiation.

METHODS

Normal adult and fetal skin, hyperplasias, hamartomas, and neoplasms with sebaceous or follicular differentiation were stained immunohistochemically with a panel of polyclonal and monoclonal antibodies highly specific for CEA-180, NCAs, and BGP. Double immunostaining was performed to correlate the CEA expression with that of epithelial membrane antigen (EMA), a glycoprotein consistently found in differentiating sebocytes.

RESULTS

Whereas sweat glands coexpressed CEA, NCAs, BGP, and EMA, sebaceous glands were exclusively labeled with the antibodies recognizing BGP or EMA. Staining of the sebaceous glands was restricted to mature sebocytes, sparing immature cells. At the ultrastructural level immunoreactivity for BGP and EMA was demonstrable in the golgi area, in small vesicles, and along the cell membranes. During fetal development BGP was not found until the sebaceous glands matured. The expression of BGP and EMA was highly conserved in reactive, hamartomatous, and neoplastic proliferations of adnexal structures with sebaceous differentiation.

CONCLUSION

The expression of BGP, a CEA glycoprotein, in differentiating sebocytes accounts for the reactivity of many anti-CEA antibodies with sebaceous glands and thus disqualifies the CEA family as a monospecific marker for sweat gland differentiation.

Immunohistochemistry in diagnostic dermatopathology

Immunopathology continues to be important in diagnostic dermatopathology. Immunopathology is an invaluable tool for assessing the tissue of origin or direction of differentiation of cells. In some cases this can result in a more precise diagnosis. This article reviews the role of immunopathology in determining the biologic behavior of hematolymphoid infiltrates. It explores the methodology of immunoperoxidase, discusses the most commonly used antibody reagents, and presents a series of diagnostic dilemmas in which immunopathology can be useful. In each case a strategy is established that maximizes the likelihood of making a definitive diagnosis.

The epithelial mucin, MUC1, of milk, mammary gland and other tissues

MUC1 is a mucin-type glycoprotein that is integrally disposed in the apical plasma membrane of the lactating epithelial cell and protrudes from the cell surface into the alveolar lumen where milk is stored. Envelopment of milk fat globules by this membrane accomplishes their secretion and conveys MUC1 into milk. The human form of this mucin has been detected in many other organs, tissues and body fluids. It projects from the cell surface as long filaments. In the human and a number of other species, MUC1 is polymorphic due to variable numbers of a tandemly repeated segment 20 amino acids in length. The individual codominantly expresses two alleles for the mucin so that differences in its size among individuals and between the two forms of an individual are observed. The tandem repeats are rich in serines and threonines which serve as O-glycosylation sites. Carbohydrate content of MUC1, as isolated from milk of human, bovine and guinea pig, is approximately 50%. The oligosaccharides carry substantial sialic acid at their termini and this accounts for two putative functions of this mucin, i.e., to keep ducts and lumens open by creating a strong negative charge on the surface of epithelial cells which would repel opposite sides of a vessel, and to bind certain pathogenic microorganisms. MUC1 is protease resistant (trypsin, chymotrypsin and pepsin) and large fragments of it can be found in the feces of some but not all breast-fed infants. MUC1 has a highly varied structure because of its polymorphism, qualitative and quantitative variations in its glycosylation between tissues, individuals and species, and differences due to divergence in the nucleotide sequences among species. Sequencing of the MUC1 gene for various species is showing promise of revealing unique evolutionary relationships and has already indicated conserved aspects of the molecule that may be functionally important. Among these are positions of serine, threonine and proline in the tandem repeats and a high degree of homology in the transmembrane and cytoplasmic segments of the molecule.

Mucin expression by transitional cell carcinomas of the bladder

OBJECTIVE

To examine the presence of a membrane-associated and secreted mucin (MUC1) and a secreted gel-forming mucin (MUC2) in normal and malignant urothelium.

MATERIALS AND METHODS

Sections were obtained from archival paraffin blocks from 11 patients with nonmalignant urological conditions and 89 patients with transitional cell carcinomas (TCC). Mucin expression was examined by immunohistochemistry using monoclonal antibodies BC2 and 4F1, reactive with epitopes on the protein core of MUC1 and MUC2 respectively.

RESULTS

In normal urothelium MUC1 was limited predominantly to the apical membranes of the umbrella cell layer. MUC1 was present in all cases of TCC, and the pattern of expression divided into three categories: luminal membrane staining only, luminal plus cytoplasmic staining of intermediate +/- basal layers, or staining of only isolated cells or cell groups. These staining patterns were significantly associated with both tumour grade and stage (P < 0.001), with cytoplasmic staining more prevalent in higher grade and stage tumours. MUC2 was not detected in normal urothelium, and was present in 40% of cases of TCC, characterized by intense granular cytoplasmic staining. No association between MUC2 expression and either tumour grade or stage was demonstrated.

CONCLUSION

MUC1 mucin was expressed by both normal and malignant urothelium, with increased expression characteristic of higher grade and stage tumours. MUC2 expression was found in 40% of tumours but not in normal urothelium. The role of these mucins in the biology of the bladder requires further investigation.

Characterization of normal breast epithelial cells in primary cultures: differentiation and growth factor receptors studies

The growth and differentiation of normal human mammary epithelial cells (HMEC) were studied after propagation of serial cultures from breast tissue biopsies from 42 mammoplasty patients. Cells were grown for up to 7 mo. in low calcium medium. HMEC cultures displayed heterogeneous growth patterns, according to the average doubling time of 44 +/- 6 h for 32 generations. Proliferation peaked at Day 30. HMEC maintained a normal karyotype and were organized in ductlike structures when cultured in collagen gel matrix. The cultures retained several phenotype traits of the epithelial lineage, including the expression of cytokeratins 18 and 19, specific mammary gland antigens, as shown by indirect HMEC immunostaining by the monoclonal antibodies DF3, EMA, 7B10, and 1BE12. Estrogen receptors were undetectable, whereas progesterone receptors were present at very low density. High-affinity cell surface receptors for epidermal growth factor (EGF) (Kd = 1.1 x 10(-10) M) were observed at a density of 50,000 to 100,000 sites per cell. Accordingly, [3H]thymidine incorporation in HMEC was optimally stimulated by EGF at concentrations of 10(-11) to 10(-10) M. HMEC were also seen to possess functional VIP receptors linked to the adenylate cyclase system, as we previously observed in seven human breast cancer cell lines. These results show that long-term cultures of HMEC provide useful models for studying the growth and differentiation of the normal human mammary gland, and the role of growth factors and hormones in these functions.

Bone marrow micrometastases in primary breast cancer: prognostic significance after 6 years' follow-up

Using an antiserum to epithelial membrane antigen we have screened multiple bone marrow aspirates from 350 patients with primary breast cancer taken at the time of initial surgery. 89 (25%) patients were found to have micrometastases and their presence was related to pathological size (P less than 0.01), the presence of peritumoral vascular invasion (P less than 0.001), and positive lymph nodes (P less than 0.005) but not menopausal status. At a median follow-up of 76 months (range 34-108) 107 patients had relapsed with distant metastases. 48% (43 of 89) of these patients had micrometastases initially compared with 25% (64 of 261) who did not (P less than 0.005). The test predicts for relapse in bone (P less than 0.01) and other distant sites excluding bone (P less than 0.001) and is associated with a shorter overall survival (P less than 0.005). We conclude that the detection of micrometastases signals a high likelihood of early relapse and decreased survival in breast cancer.

Micrometastatic tumour cells in bone marrow of patients with gastric cancer: methodological aspects of detection and prognostic significance

Monoclonal antibodies (Mab) are potent probes to identify individual tumour cells or small tumour cell clusters in bone marrow. In the present study, various antibodies directed against either cell surface or intracytoplasmic antigens of epithelial cells were assessed for their ability to detect such cells in bone marrow of patients with breast, colorectal and gastric cancer. According to the presented data, monoclonal antibodies against intracellular cytokeratin (CK) components are superior in terms of specificity and sensitivity to antibodies reacting with epitopes of the cell membrane. Using a monoclonal antibody against the cytokeratin polypeptide 18 in connection with the alkaline phosphatase anti-alkaline phosphatase detection system (APAAP), we could detect tumour cells in bone marrow of 34 out of 97 patients with gastric cancer examined at the time of primary surgery. The incidence of positive findings was correlated to established risk factors, such as histological classification and locoregional lymph node involvement. Clinical follow-up studies on 38 patients demonstrated a significantly increased relapse rate in patients presenting with CK-positive cells in their bone marrow at the time of primary surgery. Thus the described technique may help to identify patients with gastric cancer carrying a high risk of early relapse.

The effect of circulating antigen and radiolabel stability on the biodistribution of an indium labelled antibody

This study has investigated two of the main problems with radiolabelled antibody imaging, the formation of circulating immune complexes (I.C.) and the non specific binding of radiolabel to the antibody molecule. Patients undergoing immunoscintigraphy with 111In labelled monoclonal antibody ICR2 were divided into three groups who received either the radiolabelled antibody alone (control, n = 12), the radiolabelled antibody which was incubated with the chelating agent diethylene triamine pentacetic acid (DTPA) prior to size exclusion chromatography (n = 6) or whose injectate was treated with DTPA and cold MAb administered intravenously prior to radiolabelled MAb administration (n = 6). Radiolabelled antibody uptake in abdominal organs was measured by region of interest analysis using a gamma camera with online computer and that in tumour and normal tissues by gamma well counting of biopsies. Circulating antigen and immune complex was measured by high pressure liquid chromatography (HPLC). The sensitivity of tumour imaging and the tumour uptake of radiolabelled antibody was not significantly different between the groups. Patients with high circulating antigen levels developed high levels of circulating immune complex but also had high tumour uptakes of radiolabelled antibody. Administration of cold MAb increased the splenic, but did not effect the tumour uptake of radiolabelled antibody and only minimally reduced levels of circulating immune complex. Chelate administration reduced the urinary excretion of radioactivity but increased the liver uptake of radioactivity. These results have shown that successful antibody imaging can be carried out despite high levels of circulating antigen, that large doses of unlabelled antibody are required to prevent immune complex formation and that removal of non specifically bound 111In does not reduce the liver uptake of radioactivity.

Intraoperative localization of colorectal cancers using radiolabelled monoclonal antibodies

Radiation detectors may allow the intraoperative localization of small cancer deposits following administration of radiolabelled tumour-associated antibodies. This technique was evaluated in 16 patients with colorectal tumours (14 cancers, one adenoma, one lipoma) with the 111In-labelled monoclonal antibody (MAb) ICR2 which recognizes the tumour-associated epithelial membrane antigen (EMA). At operation counting was carried out (3 x 20 s per site) using a hand-held radiation probe over the primary lesions and any palpable lymph nodes in the mesocolon. The tumour to normal colon (T/NC) ratio of counts recorded at operation was more than 1.5:1 in eight of the 14 patients with cancer (mean(s.d.), 1.54(0.41):1) and 0.91:1 and 1.06:1 respectively in the two patients with benign tumours. Node to normal colon ratios were higher in lymph nodes containing metastases. The uptake of radiolabelled antibody (T/NC ratio) was higher in EMA-expressing cancers than in those not expressing the target antigen (mean(s.d.), 2.45(0.65):1 versus 1.40(0.20):1, P = 0.019). An abdominal tumour model was also developed. Radioactively filled containers of 0.5-10 ml representing tumour deposits were suspended in a tank of 111In solution representing the background activity found in normal tissues. The ratio of radioactivity in the 'tumour' to that of background varied from 2:1 to 8:1. The 'tumour' was considered to be detectable if the mean counts recorded over the 'tumour' exceeded the mean of counts recorded over background by three standard deviations. At a ratio of 2:1 only 'tumours' greater than 5 ml could be detected with a sodium iodide probe and those over 10 ml could be detected with a cadmium telluride (CdTe) probe. At a ratio of 8:1, 'tumours' of 0.5 ml could be detected with either probe. At all ratios and counting periods the NaI probe was more sensitive than the CdTe.

Micrometastases in breast cancer: long-term follow-up of the first patient cohort

"Micrometastases" can be identified in the bone marrow of patients with apparently localised breast cancer using an immunocytochemical stain for epithelial membrane antigen (EMA). Of 39 women who had marrow samples examined at the time of initial presentation (37), or with locally recurrent disease (2), 13 (33%) had samples which contained small numbers of EMA positive cells. 10 out of 23 (44%) lymph-node positive patients were marrow positive, compared to 1 out of 14 (7%) lymph node negative cases (P = 0.03). Long-term follow-up (median 9.5 years) has shown that 11 out of 13 (85%) patients with micrometastases have developed metastatic disease compared to 8 out of 26 (31%) with negative bone marrow aspirates (P less than 0.05). The small number of EMA positive cells detected in bone marrow samples probably reflects the high metastatic potential of primary or recurrent cancers rather than established microscopic deposits; it is not yet clear whether the finding of such micrometastases will act as an independent variable compared to established prognostic factors.

Structures of acidic O-linked polylactosaminoglycans on human skim milk mucins

O-Linked glycans were isolated from human skim milk mucins or mucin-derived high-molecular weight glycopeptides and fractionated by anion exchange chromatography into neutral and acidic alditols. Major oligosaccharides contained in the acidic fraction were purified by high performance liquid chromatography and structurally characterized by a combination of fast atom bombardment mass spectrometry, methylation analysis and 500 MHz 1H-nuclear magnetic resonance spectroscopy. The structural aspects exhibited by these major species in the acidic fraction resemble those established previously for the neutral oligosaccharides from human skim milk mucins: 1) the size of the alditols varies from tri- to decasaccharides, 2) the core structure is of the ubiquitous type 2, 3) the backbone sequences are of the poly-N-acetyllactosamine type with a particular preponderance of linearly extended GlcNAc beta(1-3)Gal (major) or GlcNAc beta(1-6)Gal units (minor).

Selective immortalization of a phenotypically distinct epithelial cell type by microinjection of SV40 DNA into cultured human milk cells

An immortal cell line, MMSV-1, has been developed which exhibits many features of the common luminal epithelial cell of the human mammary gland. The cell line was developed by microinjection of SV40 DNA into individual cells in selected colonies in cultures of human milk epithelial cells. Immunohistochemical staining shows that the MMSV-1 cells express keratins 7, 8, 18 and 19 homogeneously in organized filaments which lead into well-developed desmosomes. They do not express vimentin or keratins found in stratified epithelia or keratin 14 found in basal cells in the mammary gland. The PEM mucin, recognized by the antibody HMFG-I, is also expressed and appears to be processed normally. Fibronectin is detected but shows the punctate pattern typical of cultured normal milk epithelial cells. MMSV-1 cells show a reduced requirement for added growth factors, including cyclic AMP-elevating agents, but do not grow in agar or form tumours in nude mice. Since the transfected cells could be selected on the basis of an extended in vitro life span, antibiotic resistance markers were not introduced and the cells remain sensitive to hygromycin and neomycin.

Inhibition of the hepatocyte uptake of radiolabelled monoclonal antibodies by chelating agents

The imaging of small abdominal tumours with indium 111 labelled monoclonal antibodies (MAbs) is often obscured by the uptake of activity into the heptocytes of normal liver tissue. A model has therefore been developed to analyse reagents which may inhibit the hepatocyte uptake of 111In-MAb whilst preserving tumour uptake. Primary rat hepatocyte cultures and an epithelial membrane antigen (EMA) expressing tumour cell line (MCF7), recognised by the EMA-specific MAb ICR2, were obtained in tissue culture. Monolayers of both cells were incubated with the 111In-MAb with or without the additional reagents and the cell uptake then measured and expressed per milligram of cell protein using a Lowry protein assay. No preferential reduction in hepatocyte uptake was noted by incubating cells with either saturated or unsaturated transferrin. The chelating agent, diethylene triamine penta-acetic acid (DTPA), however, significantly reduced the uptake of activity in hepatocytes but not the tumour cell line (P less than 0.05). An optimum concentration and time period for incubating DTPA with labelled MAb was established. The mean hepatocyte uptake was reduced by 80% with a 1 h incubation with 1 mM DTPA. These results suggest that DTPA may have a role in reducing the liver uptake of radioactivity in patient studies using 111In-MAb.

Critical evaluation of monoclonal antibody staining in breast carcinoma

The immunoperoxidase staining of 84 primary invasive breast carcinomas with four monoclonal antibodies (BRST-1, HMFG1, EMA, B72.3) was evaluated by semiquantitative light microscopical examination and quantitative image analysis. Major differences in the staining of the tumours for each of the monoclonal antibodies was observed. Correlation between monoclonal antibody staining and patient age, survival, histological grade, tumour diameter and cellularity was also carried out. This showed a significant association between histological grade and staining with BRST-1 and EMA.

Binding of anti-EMA, AGF 4:48 and the lectin UEA-1 to human ovarian carcinomas: histological and clinical correlations

The staining reactions of the monoclonal antibodies anti-EMA, AGF 4:48 and the lectin UEA-1 from Ulex europaeus were investigated in formalin-fixed, paraffin-embedded sections of 36 primary ovarian carcinomas. The staining patterns were graded and related to tumour type, differentiation, FIGO stage and patient survival. Both antibodies and UEA-1 showed variations in their staining between and within tumours. The staining of mucinous tumours using anti-EMA was significantly less than in non-mucinous tumours. No other significant associations were found between staining grade and the different variables under study. The variations in staining were not demonstrated to have any prognostic significance.

B-cell lymphoma marking only with anti-epithelial membrane antigen

A large cell malignancy identified as a B-lymphoma only by virtue of mu and kappa chain gene rearrangement additionally displayed profuse plasmalemma marking with anti-epithelial membrane antigen (EMA). Affinity absorption of the antibody with N-acetyl-galactosamine but not with N-acetylglucosamine or d-galactose effectively removed the lymphoma reactive antibodies. Soy bean agglutinin but not wheat germ agglutinin demonstrated lectin activity towards an epitope on the lymphoma membrane. Finally, digestion of cryostat sections of the lymphoma with N-acetylgalactosaminidase effectively removed the sugar epitope involved in EMA marking. Such data taken together would tend to indicate a membrane positioned complex carbohydrate with galactosamine dominancy. By analogy to the purported "late differentiation antigen secretory function" of EMA this lymphoma may be close to plasma cell stage.

Monoclonal antibodies reactive with mucin expressed in breast cancer

Three murine monoclonal antibodies (BC1, BC2 and BC3) were developed against human milk fat globule membrane (HMFGM). By immunoperoxidase staining, it was found that the antigenic determinants had a predominant distribution in breast cancer tissue. In addition, the antibodies reacted preferentially with mucin derived from human milk rather than that derived from the breast cancer cell line ZR75; they also recognized polymorphic high molecular weight components (MW greater than or equal to 230,000) in serum and in human milk fat globule membrane. Thus the antibodies appear to react with a component of the family of mucins found in breast cancer and human milk and it appears likely that at least part of each epitope is protein in nature. Antibodies BC1, BC2 and BC3 recognized related but not identical epitopes, and they appear to be co-expressed on the same molecules as 3E1.2-defined antigen (mammary serum antigen, MSA) which is also a member of the family of breast cancer-related mucin. However, the 3E1.2 epitope is distinct and non-cross-reactive with those described for BC1, BC2 and BC3. The BC2 and BC3 defined epitopes were examined for their value in serum assays. Immunoassay was developed with a combination of two antibodies, using antibody BC3 for antigen capture and antibody BC2 or 3E1.2 for antigen detection and gave reasonable sensitivity (approximately 85%) and specificity (approximately 95%) in such serum tests for breast cancer. In a limited study, these tests appeared to complement the MSA test in the detection of breast cancer.

Purification and biochemical characterisation of a novel breast carcinoma associated mucin-like glycoprotein defined by antibody 3E1.2

A member of the high molecular weight glycoproteins of human milk and breast cancer was isolated from the sera, ascites and breast carcinoma tissue of patients with breast cancer using monoclonal antibody 3E1.2. The 3E1.2 defined antigen, termed mammary serum antigen (MSA) was obtained by immunoaffinity chromatography and a solid phase immuno-precipitation technique (SPIT) from serum of patients with metastatic breast cancer. MSA was found to be a high molecular weight glycoprotein with a Mr greater than 300,000 by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and a native Mr approximately 1 x 10(6) by gel filtration chromatography; in accord with the published Mr of other high molecular weight glycoproteins obtained from human milk and breast cancer. A high degree of glycosylation of MSA molecule was shown by its poor staining with Coomassie blue but good staining in a PAS-silver stain. In addition, MSA contained N-acetyl neuraminic acid and N-acetyl glucosamine as indicated by its binding to wheat-germ agglutinin. The epitope defined by antibody 3E1.2 is sensitive to treatment by sodium periodate and neuraminidase, implying that both carbohydrate and sialic acid are required for binding of antibody 3E1.2. Sandwich immunoassays demonstrated that MSA+ molecules are likely to express repeated 3E1.2 defined epitopes. Furthermore, MSA was susceptible to degradation by pronase, subtilisin and proteinase K and gave a different peptide profile from that of the PAS-O glycoprotein of human milk. MSA+ molecules were found to carry epitopes for a number of other monoclonal antibodies which were reactive with the PAS-O glycoprotein. It is suggested that MSA has the same core protein as is recognised by antibody DF3 which has been used to clone the same cDNA as was cloned with antibodies HMFG-1, HMFG-2 and SM-3. However, the epitope detected by the 3E1.2 antibody is either absent or weakly expressed on human milk, human milk-fat globule membrane (HMFGM) or deglycosylated HMFGM--all of which react strongly with various anti-HMFG antibodies. The antibody 3E1.2 thus recognises a unique epitope of the high molecular weight glycoproteins of human milk and breast cancer, being found in cancer tissue, serum and ascitic fluid of patients with breast cancer but weakly expressed or absent in human milk.

Immunohistochemical demonstration of epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA), and keratin on mammary and extramammary Paget's disease

We examined the presence and the distribution of epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA), and keratin in 4 cases (6 specimens) of mammary Paget's disease and 9 cases (18 specimens) of extramammary Paget's disease utilizing immunoperoxidase technique. Paget's cells in 23 out of 24 specimens were demonstrable for EMA. Positive staining was observed for CEA in every specimen. Among the 24 specimens, 16 showed positivity for keratin. In addition, CEA was positive in Paget's cells and the staining for CEA was stronger than that of EMA and keratin. On the other hand, some Paget's cells were negative for EMA of keratin although other positive cells were observed in the same sections. This findings indicates that Paget's cells might express CEA and/or EMA and/or keratin depending on their differentiation. EMA is present in most organs showing glandular differentiation, and anti-keratin antibody used in this study recognizes not only keratinocytes but glandular cells. Thus, our study suggests Paget's cells are of glandular origin.

Epithelial membrane antigen (EMA) distribution in various biological fluids

Different human biological fluids, namely breast cyst fluids (five), milks (four), sera (five), were submitted to molecular sieving chromatography on Sepharose CL6B. Global protein contents of the eluted fractions were estimated by the Bradford method. Epithelial membrane antigen (EMA) was assayed by two different ELISA techniques using polyclonal and monoclonal antibodies. Various molecular species reacting with EMA (15) were found in the chromatographies with molecular weights ranging from 35 to 1500 kd. But the total amount of antigens detected using polyclonal or monoclonal antibodies was quite similar. Moreover no significant difference was found between the sera from two lactating women and the sera from three women with adenocarcinoma with respect to the molecular distribution of different molecular species of EMA.

Automated screening for micrometastases in bone marrow smears

It is possible to detect micrometastases in primary breast cancer using immunocytochemical staining of bone marrow smears. However, using the light microscope the procedure is time-consuming and laborious because such cells occur rarely (less than 1 in 10,000). Using an image analysis system, the Leytas machine, and a specially prepared reproducible slide it has been possible to automate the technique. A 100% concordance was found between the machine and the light microscope in the identification of slides containing moderate to high numbers of tumour cells in bone marrow, and in those containing no tumour cells. However, in those slides containing low numbers of tumour cells (1-10 tumour cells/10(6) normal bone marrow cells) the sensitivity was decreased to 91%. In the presence of non-specific staining the false positive rate was increased from 0% to 22%. This method represents a potential improvement in the assessment of an important clinical staging procedure.

Comparison of monoclonal antibodies for the detection of occult breast carcinoma metastases in bone marrow

Twenty percent (n = 6) of Stage III or IV breast cancer patients (n = 30) had bone marrow metastases detected in bilateral bone marrow biopsy/aspiration preparations using standard histologic preparations. Each metastasis was also detected by four separate monoclonal antibodies (MAbs) which recognize breast carcinoma associated antigens (DF3, anti-EMA, HMFG-2, and CAM5.2). These MAbs were then utilized to stain other bone marrow preparations (n = 81) to determine their utility for the detection of micrometastatic breast carcinoma. MAbs HMFG-2, anti-EMA, and DF3 were each strongly reactive with bone marrows containing histologically-evident metastatic breast carcinoma (18/18). These anti-epithelial membrane antigen MAbs, however, were also reactive with rare plasma cells and immature cells (as well as cell clusters) in some of the control bone marrow samples tested, including those from normal patients and patients with hematologic disorders. They also reacted with some of the preparations from patients with leukemia and lymphoma, and with uninvolved marrows from patients with non-epithelial malignancies. The anti-keratin MAb CAM5.2, in contrast, reacted with 83% (15/18) breast cancer metastases and failed to stain any cells in the various categories of control marrow preparations. These data suggested that MAb CAM5.2 might be utilized to immunohistochemically differentiate micrometastatic breast carcinoma from immature myeloid or erythroid elements. Each MAb was then reacted with histologically uninvolved marrow preparations from the remaining 24 of 30 breast cancer patients in an attempt to identify occult breast carcinoma metastases. While MAbs HMFG-2, DF3, and anti-EMA demonstrated reactive cells in some of these marrows, this reactivity was similar to that seen with control preparations. MAb CAM5.2, in contrast, was negative with all specimens. These data suggest that MAb CAM5.2 may be a useful immunologic probe for the detection and confirmation of metastatic breast carcinoma in bone marrow, while more caution must be employed in the interpretation of results obtained using MAbs anti-EMA, DF3, and HMFG-2.

A comparison of carcinoembryonic antigen (CEA) and epithelial membrane antigen (EMA) in human colorectal cancer

The development of monoclonal antibodies (MAbs) to tumour-associated antigens has allowed the successful radioimmunolocalization of a variety of tumours and has provided a basis for targeted therapy. In patients with colorectal cancer, antibodies to CEA have been the most widely used for imaging, but their role in targeted therapy may be inhibited by their reaction with normal tissues and with circulating CEA. Epithelial-membrane antigen (EMA) is expressed by most epithelial tumours, including colorectal cancers, and antibodies to EMA may provide a satisfactory alternative. We therefore compared two high-affinity MAbs, one to CEA (C46) and one to EMA (ICR2), in tissues obtained from 31 patients with cancer (18 primary colonic, 6 nodal metastases and 7 liver metastases), 14 patients with adenomatous polyps and 17 with normal colon. The indirect immunoperoxidase staining reaction was used and the results classified as either positive or negative. A heterogeneous pattern of staining was found for both antibodies. ICR2, the anti-EMA, reacted with slightly fewer colonic cancers than C46, the anti-CEA antibody (83% vs. 100%) and a similar number of metastases. Most noticeable was the minimal reaction of anti-EMA with normal colon (12% vs. 71%) and benign polyps (7% vs. 79%) in comparison to anti-CEA. This would suggest a possible role for ICR2 in the radioimmunolocalization and targeting of colorectal cancer.

Micrometastases in bone marrow in patients with primary breast cancer: evaluation as an early predictor of bone metastases

The bone marrow of 307 patients with primary breast cancer was examined for tumour cells by immunocytochemistry using an antiserum to epithelial membrane antigen. Micrometastases were found in 81 cases (26.4%) and their presence was related to various poor prognostic factors: spread to lymph nodes, vascular invasion, T stage, and pathological size. The median duration of follow up was 28 months. Seventy five patients relapsed, 60 at distant sites. Of these 60 patients, 26 had micrometastases detected at presentation and 34 were free of micrometastases initially. The relapse free interval was significantly shorter for patients with micrometastases, and these patients had a shorter survival. Analysis of the sites of relapse showed that the test predicted bone metastases only. Thus 10 out of 19 patients (53%) who developed bone metastases at first relapse had micrometastases at presentation compared with only 41 out of 288 patients (14%) who remained free of bone metastases or relapsed in non-skeletal sites. The presence of micrometastases detected at the time of initial surgery in a patient with primary breast cancer is a useful predictor of early relapse in bone and may help in selecting patients for subsequent systemic treatment.

Immunological analysis of mucin molecules expressed by normal and malignant mammary epithelial cells

Many existing MAbs raised against the human milk fat globule or against carcinoma cells recognise epitopes on high-molecular-weight glycoproteins. In a comparative ELISA assay a number of these antibodies have been shown to react with an extract of the human milk fat globule. In comparative immunoblots of cultured normal milk cells and breast cell lines, all were found to bind to large molecules which show some variation in molecular weight depending on the cell source. The HMFG-2 antibody, which is widely used in cancer diagnosis, also recognises epitopes on lower-molecular-weight components. In T47D cells these may be as small as 80,000 Mr, and with electron microscopy this cell line can be shown to accumulate HMFG-2-reactive components in the Golgi apparatus. Using an HMFG-2 affinity column we have immunopurified HMFG-2-reactive material from the 2 breast cancer cell lines MCF-7 and T47D and shown that all of the above antibodies react in a solid-phase ELISA with the purified material. In addition to high-molecular-weight components, the immunopurified material was found to contain lower-molecular-weight components including a glycoprotein of 68,000 Mr that was not, however, recognised by the HMFG-2 antibody on a Western blot. We have used this immunopurified material to generate new MAbs. All of these recognise the high-molecular-weight bands seen with the other antibodies, but 2 of them also recognise a band at 68,000 Mr in blots of MCF-7 and T47D. The second-generation antibodies show a spectrum of reactivity on tissues similar to HMFG-2 and one reacts at least as strongly as HMFG-2 with methanol-acetone-fixed sections of breast cancers.

Cloning of partial cDNA encoding differentiation and tumor-associated mucin glycoproteins expressed by human mammary epithelium

Human mammary epithelial cells secrete and express on their cell surfaces complex mucin glycoproteins (Mr greater than 250,000) that are developmentally regulated, tumor-associated, and highly immunogenic. Studies using monoclonal antibodies directed to these glycoproteins suggest that their molecular structures can vary with differentiation stages in the normal gland and in malignancy. To analyze the molecular nature of these glycoproteins, milk mucin was affinity-purified and deglycosylated with hydrogen fluoride, yielding bands at 68 and 72 kDa on silver-stained gels. Polyclonal and monoclonal antibodies to the stripped core protein were developed and used to screen a lambda gt11 expression library of cDNA made from mRNA of the mammary tumor cell line MCF-7. Seven cross-reacting clones were isolated, with inserts 0.1-1.8 kilobases long. RNA blot analysis, using as a probe the 1.8-kilobase insert subcloned in plasmid pUC8 (pMUC10), revealed transcripts of 4.7 and 6.4 kilobases in MCF-7 and T47D mammary tumor cells, whereas normal mammary epithelial cells from pooled milks have additional transcripts. The expression of mRNA correlates with antigen expression as determined by binding of two previously characterized anti-mucin monoclonal antibodies (HMFG-1 and HMFG-2) to seven cell lines. Restriction enzyme analysis detected a restriction fragment length polymorphism when human genomic DNA was digested with EcoRI or HinfI.

Keratins versus epithelial membrane antigen in tumor diagnosis: an immunohistochemical comparison of five monoclonal antibodies

Among the monoclonal antibodies capable of detecting epithelial lineage, some recognize keratin and others identify antigens of epithelial membranes. Of the latter, the most commonly used is directed against an antigen present in cell membranes derived from milk fat globules, epithelial membrane antigen (EMA). To determine their relative sensitivity and specificity and hence their diagnostic value, we compared four commercially available monoclonal antibodies to low-molecular-weight keratins--AE1, CAM 5.2, PKK1, and 35 beta H11--with the monoclonal antibody to EMA (anti-EMA). We studied 383 samples of human neoplasms of diverse histogenetic types and degrees of differentiation. Anti-EMA was found to be less sensitive than the monoclonal antibodies to keratin in several epithelial tumors, including tumors of the prostate (11 of 13 negative), gastrointestinal tract (13 of 34 negative), and thymus (seven of eight negative). Anti-EMA was also less sensitive in mesotheliomas (nine of 24 negative). Of the embryonal carcinomas, all stained with the monoclonal antibodies to keratin, whereas none stained with anti-EMA. False-positive staining with anti-EMA was seen in two of 14 T-cell lymphomas. We conclude that the monoclonal antibodies to low-molecular-weight keratins are more sensitive and specific for the identification of epithelial origin of neoplasms than is monoclonal anti-EMA. Anti-EMA should not be used as the sole marker of epithelial differentiation in tumor diagnosis.

Epithelial membrane antigen as a marker of human salivary gland acinar and ductal cell function

Monoclonal antibody to epithelial membrane antigen (EMA) from human milk fat globule membrane was used with the immunoperoxidase method to examine the distribution of EMA in normal human salivary glands and in their obstructive lesions. EMA staining in the normal salivary gland was limited to luminal and lateral borders of acinar cells. The antigen was also present in high concentration on the luminal side of ductal cells. In obstructive lesions or sialoadenitis, lateral border and luminal positive staining in atrophic acinar cells was reduced or absent, and the EMA staining in duct-like structures was also markedly decreased on even absent in chronic stage. The degree of reduction in EMA staining was dependent on the extent of degeneration. Expression of immunohistochemically detectable EMA in paraffin sections suggests the antigen to be useful as a new marker of salivary gland function, of either acinar compartments or ductal segments.

Cross reactivity of anti-epithelial membrane antigen monoclonal for reactive and neoplastic glial cells

An anti-epithelial membrane antigen monoclonal (Clone E29) was used as part of a panel of antisera in the investigation of primary CNS and metastatic tumours. In contrast to others, definite and often strongly positive cross-reactivity has been demonstrated on glial cells. The material consisted of fourteen astrocytomas, two ependymomas, six cerebral metastases and two cases of gliosis.

Assay of epithelial membrane antigen (EMA) in human serum by ELISA

An enzyme linked immuno sorbent assay (ELISA) for an epithelial membrane antigen (EMA) was described. Possible cross reactions with various antigens were investigated. In sera, EMA has been found in all the subjects studied. In normal population, levels were in the range of 500 ng/ml +/- 125 (S.D.) A significant increase was observed at the end of pregnancy and during lactation. A large number of patients suffering various benign and cancerous diseases were studied. The elevated levels found in breast and pulmonary pathology indicated that this assay could be useful in the follow-up of patients suffering from these diseases.

Epithelial markers in thyroid carcinoma: an immunoperoxidase study

Ten cases each of papillary, follicular, anaplastic and medullary carcinoma of the thyroid were stained for thyroglobulin, calcitonin, epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA) and cytokeratin (CAM 5.2). Monoclonal or affinity purified polyclonal antibodies, and an indirect immunoperoxidase technique were used. All the papillary and follicular tumours, 5/10 anaplastic and 3/10 medullary carcinomas contained thyroglobulin. Only the 10 medullary carcinomas stained positively for calcitonin. Three out of 10 papillary, 1/10 follicular, 0/10 anaplastic and 10/10 medullary carcinomas were positive for CEA. Nine out of ten papillary, 7/10 follicular, 2/10 anaplastic and 3/10 medullary carcinomas were positive for EMA. Ten out of 10 papillary, 10/10 follicular, 5/10 anaplastic and 10/10 medullary carcinomas were positive for cytokeratin. The presence of calcitonin and CEA is of value in the diagnosis of medullary carcinoma, and enable its distinction from anaplastic thyroid carcinoma. Thyroglobulin is a useful marker in thyroid carcinomas.

High-Mr glycoprotein profiles in human milk serum and fat-globule membrane

Gradient-polyacrylamide-gel electrophoresis of human milk serum separated three high-Mr glycoprotein bands. The properties of the components corresponding to the three bands (tentatively termed 'Components C, B and A' in their order of migration) were compared by staining with four monoclonal antibodies and lectins. Components B and C both reacted with the four antibodies, but Component A did not. Components B and C were stained with peanut (Arachis hypogaea) agglutinin (PNA) and wheat (Triticum)-germ agglutinin (WGA), Component A being stained with soya-bean (Glycine max) agglutinin as well as PNA and WGA. These results suggest that Components B and C were related molecules, whereas Component A was markedly different from them. The reactivities of Components B and C were the same as those of PAS-0, a high-Mr periodate/Schiff (PAS)-positive glycoprotein previously isolated from human milk fat-globule membrane (MFGM). Component C, whose electrophoretic mobility was the same as PAS-0, could have been a soluble form of PAS-0. A high-Mr glycoprotein having the same properties as Component A was also observed in MFGM. The amino acid composition of the isolated Component A was significantly different from that of Component C and PAS-0, high threonine and serine contents being characteristic of Component A. The carbohydrate content of Component A was 65-80%, and was much higher than that of Component C and PAS-0. Fucose, galactose, N-acetylglucosamine, N-acetylgalactosamine and sialic acid were each detected as constituent sugars of Component A. Component A represents, therefore, a new high-Mr glycoprotein species in human milk serum and MFGM. Since these glycoproteins were high-Mr mucin-like glycoproteins, the names 'HM glycoprotein-A' and 'HM glycoprotein-C' were proposed for Component A and Component C (PAS-O) respectively.

Epithelial membrane antigen in normal and proteinuric glomeruli and in damaged proximal tubules

Epithelial membrane antigen was detected in normal glomeruli by a polyclonal antiserum to the antigen and by the monoclonal antibodies Ca 1, DAKO-EMA and HMFG 2, but not HMFG 1, using an indirect immunoperoxidase method. The antigen was in the form of a thin ring or collar at the junction of glomerulus and tubule. In a series of 47 renal biopsies from patients with proteinuria, the antigen could still be seen in glomeruli, provided that there were adequate numbers of glomeruli in the sections. The main object of study was the glomerular tip lesion, in which tip adhesions were seen to be just adjacent to the patch of epithelial membrane antigen. This suggested that the antigen may be important in pathogenesis of the lesion. Normal proximal tubules did not express epithelial membrane antigen but it was detected on the luminal border of acutely damaged proximal tubules. Thus the distribution of epithelial membrane antigen in the kidney is more complex than was previously thought.

Immunohistochemical reactivity of monoclonal antibodies to human milk fat globule with breast carcinoma and with other normal and neoplastic tissues

The immunohistochemical reactivity of three monoclonal antibodies, generated against human milk fat globule membranes (HMFG), were studied using peroxidase conjugated secondary antibody or avidin-biotin-peroxidase complex method. Two of the antibodies, III E 8 and III H 2 showed rather broad reactivity, staining all mammary, ovarian, endometrial, cervical and bronchial adenocarcinomas. They also stained all pancreatic adenocarcinomas but only 30% of the carcinomas of the colon. Reaction with prostatic and renal adenocarcinomas varied. No reaction was seen with squamous or transitional cell carcinomas. Antibody III D 5, which previously has been shown to recognize an estrogen receptor activation associated antigen in mammary carcinomas, stained 60% of mammary or gynecological adenocarcinomas. This antibody had a more limited reactivity with other tissues as well, as all intestinal carcinomas were negative. It stained all bronchial adenocarcinomas, however, and 50% of the prostatic carcinomas. The three monoclonal antibodies, although not specific, show a more restricted reactivity than the so far described HMFG antibodies and they may therefore be of value in histopathological diagnosis.

Production of monoclonal antibodies against human epithelial membrane antigen for use in diagnostic immunocytochemistry

Two monoclonal murine antibodies have been raised against a delipidated extract of human cream. These antibodies were detected by immunohistological screening of hybridoma culture supernatants on sections of human breast tissue. One of those antibodies (E29) was subsequently screened against a range of normal and neoplastic human tissues and shown to react with a wide variety of human epithelia and with mesothelial cells. Antibody E29 was unreactive with other cell types, with the exception of occasional plasma cells. Antibody E29 is suitable for use on paraffin embedded tissue and represents a valuable reagent for the identification of tumours of epithelial origin.