Reduced inducibility of SOCS3 by interferon gamma associates with higher resistance of human breast cancer lines as compared to normal mammary epithelial cells.

The resistance to interferons (IFNs) limits their anticancer therapeutic efficacy. Here we studied the antiproliferative effect of interferon gamma in relation to SOCS3 expression in a panel of breast cancer cell lines and normal mammary epithelial cells. Compared to normal cells most breast cancer lines (7/8) were highly resistant to IFN-gamma. Using Northern blot and real time RT-PCR we investigated transcription of SOCS3 genes. All normal epithelial cells (4/4) showed SOCS3 induction (2-14 fold) while most breast cancer lines did not or weakly activated SOCS3 after the interferon gamma treatment. Among the cancer lines, the MDA-MB-468 cells showed increased sensitivity to IFN-gamma and relatively high level of SOCS3 induction (2-3 fold). Together, there was a good correlation

Interferon-alpha treatment may negatively influence disease progression in melanoma patients by hyperactivation of STAT3 protein.

Interferon-alpha (IFN-alpha) is an important drug used in anti-melanoma therapy. However, metastases eventually reappear in almost 60% of melanoma patients, who have received adjuvant cytokine therapy suggesting that IFN-alpha can paradoxically promote disease progression in some cases, at least. In this study, we have investigated the possibility that a growth-promoting STAT3 protein might be activated by interferon-alpha in melanoma cells. We examined 24 primary cultures established from node metastases of melanoma patients who were monitored in a 5-year clinical follow-up. The patients differed in the course of disease and survival end-points. Using Western blot analyses, we show that interferon-alpha stimulated STAT3 phosphorylation at tyrosine (Y705) residue in 17% of cases. These over-reactive cell populations originated from patients who had the shortest disease-free intervals. A significant correlation was obtained between the length of survival end-points and a lack of STAT3 activation by IFN-alpha. No STAT3 induction was observed in normal melanocytes. The STAT1 activation at tyrosine (Y701) occurred at a similar frequency as that of STAT3 (17%) albeit in different patients, no clear correlation with the clinical status could be made. The interferon-alpha/beta receptors (IRFARs) were expressed irrespective to the signal transducers and activators of transcription (STATs) inducibility suggesting that signalling defects occur downstream from IRFAR. We propose that in some cases the application of IFN-alpha could increase the probability of disease progression via overactive STAT3. The tests for STAT3 inducibility prior to cytokine immunotherapy in the clinic are therefore warranted.

Defective IFNα/γ-induced STAT3 protein activation in human malignant melanoma cells.

Signal transducer and activator of transcription 3 (STAT3) protein has been documented as a significant mediator of interferon (IFN) signaling. Physiological STAT3 phosphorylation involves tyrosine (Y705) and serine (S727) activation. Impairment of STAT3 protein levels and/or of STAT3 phosphorylation after IFN treatment has been found in many pathological conditions such as cancer, immunopathy and inflammatory disease. To analyze tumor-associated defective STAT3 response to IFNs, the induction of S727 and Y705 STAT3 activation after IFN exposure was evaluated in 18 human malignant melanoma cell lines and 68 primary cell cultures established from the lymph node metastases of melanoma patients. STAT3 expression and STAT3 phosphorylated forms were assayed by Western blot analysis employing specific STAT3 antibodies. All melanoma cell lines as well as samples derived from metastatic melanoma patients expressed STAT3 with variable signal intensities depending on the appropriate cell type. Significantly altered IFNγ-induced S727 STAT3 activation was found in both experimental models, with on average 94.1% of patients detected to be non-responders in lymph node cell cultures and 83.3% in melanoma cell lines. Moreover, a deficiency in IFNα-induced S727 induction was detected in 88.9% of melanoma cell lines. Defects in Y705 STAT3 phosphorylation were determined in clinical material (61.8% after IFNγ exposure) as well as in melanoma cell lines (absence of response to IFNα/γ in 83.3 and 55.5%, respectively). Our data clearly confirm STAT3 pathophysiological perturbances in human malignant melanoma cells. Depending on the induction of STAT3-activated phosphoforms by IFNs, three categories of melanoma cells were identified: a) phosphorylation on both the S727 and Y705 amino acid residues; b) STAT3 activation on Y705 only; c) phosphorylation at neither S727 nor Y705. The significance of in vitro STAT3 activation for predicting patient response to immunotherapy will be examined in a prospective clinical study by our group.

Development of IFN-gamma resistance is associated with attenuation of SOCS genes induction and constitutive expression of SOCS 3 in melanoma cells.

The resistance to interferons (IFNs) limits their anticancer therapeutic efficacy. Here we studied the evolution of an IFN-resistant state in vitro using melanoma cell lines. We found that the cells became less sensitive to antiproliferative effect of IFN-gamma after prolonged cultivation enabling us to isolate sensitive and resistant subclones of the parental line. We investigated transcription of signal transducer and activator of transcription (STAT) 1-6 and suppressor of cytokine signalling (SOCS) 1-3 genes, and phosphorylation of STAT 1 protein. The resistant subline (termed WM 1158R) differed from the sensitive subline (WM 1158S) by a constitutive expression of SOCS 3, lack or weak SOCS 1-3 activation following IFN-gamma, and short duration of cytokine activatory signal. Similar correlations were observed in additional melanoma lines differing in IFN sensitivities. At the protein level, IFN-gamma induced strong and prolonged STAT 1 activation at serine 727 (S727) in WM 1158R while in WM 1158S cells phosphorylation of this amino acid was much less pronounced. On the other hand, phosphorylation of tyrosine 701 (Y701) was stimulated regardless of the sensitivity phenotype. In conclusion, constitutive expression of SOCS 3 is correlated with attenuation of its induction following IFN treatment. These results suggest that progression of melanoma cells from IFN sensitivity to IFN insensitivity associates with changes in SOCS expression.

Lack of STAT 1 phosphorylation at TYR 701 by IFNgamma correlates with disease outcome in melanoma patients.

STAT 1, a member of signal transducer and transcription activator family has been implicated as key downstream mediator of interferon (IFN) signaling. Its functional activation requires phosphorylation at Tyr 701 and Ser 727 residues. Various STAT abnormalities have been found in cancer cells but their relation to oncogenesis, tumor behavior and disease outcome remains mostly unknown. We have examined the inducibility of STAT 1 phosphorylation by IFN alpha/gamma in primary cultures established from melanoma lymph node metastases at first progression and correlated our results with disease outcome and overall survival. Forty-four patients at clinical stage I-III at initial diagnosis entered the study. STAT 1 inducibility of phosphorylation by IFNs was assessed in melanoma cell lysates by means of standard immunoprecipitation and Western blotting using polyclonal and monoclonal antibodies. Lack of STAT 1 phosphorylation at Ser 727 after either IFN was recorded in 75% of patients, however, no correlations with disease evolution could be proved. In contrast, STAT 1 phosphorylation response at Tyr 701 after IFNalpha occurred in 13 (29.5%) and after IFNgamma in 32 (73%) patients. Inducibility of STAT 1 activation at Tyr 701 but not at Ser 727 driven by IFNgamma but not by IFNalpha significantly and unfavorably [corrected] influenced disease- free interval and overall survival. In conclusion, these results show that the absence of IFNgamma inducibility of STAT 1 phosphorylation at Tyr 701 positively correlates with disease outcome in malignant melanoma patients and may represent new independent prognostic marker.

Interferon inducibility of STAT 1 activation and its prognostic significance in melanoma patients.

STAT 1, a member of latent cytoplasmic proteins, plays a pivotal role in mediating biological effects of interferons. Its transducing, DNA binding and transcriptional activity require phosphorylation at both Tyr 701 (Y 701) and Ser 727 (S 727) residues. Deficient phosphorylation or constitutive activation of the STAT 1 protein were observed in some human malignancies. Using immunoprecipitation and Western blots performed with lysates made of melanoma cells derived from patients with clinical stage II/III and employing specific anti-STAT 1 PS 727/PY 701 immunoprobes, we show that STAT 1 activation response induced by IFN-alpha/-gamma is significantly impaired. On average, three quarters of patients were lacking phosphorylation at S 727. STAT 1 PY 701 was not inducible by IFN-alpha in 63% and by IFN-gamma in 34% of samples. However, these STAT 1 activation defects showed no correlation with the disease outcome and immunotherapy response as indicated by progression-free survival profiles in patients treated with IFN-alpha2b.

Malignant melanoma associates with deficient IFN-induced STAT 1 phosphorylation.

STAT 1, a member of signal transducers and transcription activators of STAT family proteins, has been implicated as important mediator of IFN signaling. Functional activation of STAT 1 requires tyrosine and serine phosphorylation. Defects in its expression or activation in response to IFNs were observed in numerous pathological conditions including cancer. To further explore cancer-associated impaired STAT 1 response to IFNs, the inducibility of serine (S 727) and tyrosine (Y 701) phosphorylation by IFN-alpha/-gamma was assessed in 21 melanoma cell lines and in 35 primary cultures derived from melanoma patients. STAT 1 levels and inducibility of its activated phospho-forms were detected by Western analysis using specific polyclonal and monoclonal antibodies. All cell lines as well as patient melanoma samples expressed STAT 1 with variable signal intensity. Significant impaired IFN-induced STAT 1 S 727 phosphorylation was observed in both model systems with average of 77% of non-responders recorded in patient melanoma cells and 76% in melanoma cell lines. Failure of PY 701 induction occurred in patient samples (63% after IFN-alpha and 34% after IFN-gamma induction) and to a lesser degree in cell lines (i.e. response absence to IFN-alpha in 5 and to IFN-gamma in 2 melanoma lines). Our study demonstrates STAT 1 functional abnormalities in melanoma cells. On the basis of detailed analyses of patient melanoma cells with respect to the inducibility of STAT 1 phosphorylation by IFNs, four categories of patients could be distinguished: a) activation on both S 727 and Y 701, b) not inducible response, c) activation on Y 701 but not on S 727, d) heterogeneous response. Clinical study is now in progress to establish the significance of in vitro STAT 1 activation for predicting the response to IFN-based therapy and to explore biological consequences in cases responding in vitro to IFN-induced STAT 1 activation on only one of the critical amino acid residues.

Malignant melanoma associates with Th1/Th2 imbalance that coincides with disease progression and immunotherapy response.

The immunological dysfunction associated with human cancer is well known phenomenon. It comprises number of pathological changes within immune network including imbalance in cytokines of Th1/Th2 origin. The objectives of our study were (i) to evaluate the abnormalities in serum levels of selected cytokines in malignant melanoma patients with regional lymph node metastases as compared to normal values, (ii) to examine the relationship between postoperative cytokine levels and disease progression and (iii) to correlate cytokine profile changes during IFN-alpha therapy with the disease progression and potential therapeutical response. Nine Th1/Th2 related cytokines and sIL-2R were determined in 26 malignant melanoma patients at clinical stage III prior and during adjuvant immunotherapy. Control group consisted of 26 healthy persons. Patients were treated with rIFN-alpha according to EORTC Melanoma group protocol 18952. Cytokines were quantified in patients sera using commercial ELISA kits. Majority of melanoma patients showed significantly lower values of IL-2 and IFN-gamma and pathologically elevated levels of IL-4, IL-6, IL-10 as compared to healthy subjects what indicates disease associated Th1/Th2 imbalance. In addition increased IL-12 and IL-15 values were noted in some patients (54% and 27%, respectively). All patients who manifested early relapse during immunotherapy had significantly lower pretreatment levels of IL-2 and IL-12 than those remaining without progression and probably benefiting from the treatment. Cytokine changes during immunotherapy disclosed that decreases in IL-2 and IL-12 and raises in IL-6 and IL-10 values occurred at least one month prior to relapse. Moreover, the continuous elevation of TNF-alpha and sIL-2R could be observed in patients who remained without progression during 10 months lasting immunotherapy. Our data illustrate that malignant melanoma associates with Th1/Th2 perturbances which are directed towards extended Th2 responses and that measurement of selected cytokines of Th1/Th2 category may be used as an early signal of disease deterioration and for evaluation of immunotherapy response.

Relation of prenephrectomy CD profiles and serum cytokines to the disease outcome and response to IFN-alpha/IL-2 therapy in renal cell carcinoma patients.

Immune parameters, including cytokine levels and CD profiles were determined in 78 renal cell carcinoma patients (RCC) prior to nephrectomy. The values were correlated with the outcome of disease and response to cytokine-based treatment during a 3-year follow-up. Significantly lower frequency of progressions and higher proportion of survivors were recorded in 24 treated patients compared to 43 untreated ones (22.9% vs. 53.5% and 82.9% vs. 55.8%) illustrating the beneficial effect of immunotherapy on the course of RCC at localized stage. RCC-related immune changes are demonstrated by reduced proportion of CD19+, CD28+, HLA-DR+, CD19+/80+ and CD8+/28+ subsets, by increased serum levels of IL-6, sIL-2R, CRP and by impaired production of IL-2 and TNF-alpha released by in vitro stimulated PBMC. Only increased CRP, IL-6 serum values, decreased CD8+ and increased CD122+ were significantly related to patients' prognosis. Comparisons of preoperative CD profiles and cytokine values with the response to IL-2/IFN-alpha based therapy disclosed significant correlation in only CD80+ and CD19+/80+ subsets. Treated patients who relapsed during the 3-year follow-up exhibited at the diagnosis significantly reduced proportion of CD80+ and CD19+/80+ cells (CD80+ means - 0.79 vs. 1.69 and CD19+/80+ means - 0.32 vs. 0.61) comparing to those surviving disease-free. In addition initial proportion of CD3+, CD8+ and CD19+ cells was reduced in treated patients who manifested progression but statistical difference from those remaining disease-free was not proved.

Renal cell carcinoma-associated immune impairment that may interfere with the response to cytokine therapy.

This prospective study was carried out to explore cytokine-related immune alterations in 69 renal cell carcinoma patients (RCC) and to look for changes which might potentially serve as a reliable predictors of response to cytokine-based therapy. Interleukin-2 (IL-2), its soluble receptor (sIL-2R) and tumor necrosis factor (TNF-alpha) levels produced in vitro by PHA activated and intact mononuclear cells (PBMC) were determined. Concentrations of IL-2, IL-4, IL-6, sIL-2R, TNF-alpha and CRP were measured in sera. Cytokine level was evaluated by enzyme-linked immunoadsorbent assay (ELISA) and CRP was determined by means of turbidimetric method. All measurements were performed in patients without any prior treatment. PHA activated PBMC of RCC patients were significantly defective in producing IL-2 and TNF-alpha comparing to controls (p < 0.03 and p < 0.001). The difference of sIL-2R was noted in metastatic stage only (p < 0.03). Unstimulated PBMC manifested decrease in IL-2 (p < 0.03) and increased level of TNF-alpha in advanced disease (p < 0.02). This impairment reflected tumor size and differentiation stage. Serum concentrations of IL-2, sIL-2R and TNF-alpha were within normal range. However, in relation to the clinical stage, significantly increased serum IL-2 was noted in combined Stage I and II as compared to controls (p = 0.012). IL-6 and CRP showed markedly elevated levels with a significancy which allowed to distinguish samples from metastatic patients. In conclusion careful comparisons of these data with clinical course of cytokine treated patients will disclose which of those tests may possess predictive power in the individual patients who are likely to respond to cytokine-based treatment.

Monoclonal antibody against DNA adducts with osmium structural probes.

Osmium tetroxide complexes with nitrogen ligands (Os,L) have been widely used as probes of the DNA structure. A monoclonal antibody OsBP7H8 against DNA adducts with Os,L was produced in mice. OsBP7H8 does not bind to proteins or total yeast RNA modified with Os,2,2'-bipyridine (bipy) nor to the unmodified nucleic acids and proteins. The antibody recognizes DNA modified with Os,bipy (DNA-Os,bipy) or with OsO4,1,10-phenanthroline (DNA-Os,phen) but it does not cross-react with oxidized DNA and with DNA adducts of osmium tetroxide complexes with other ligands (such as pyridine, TEMED and bathophenanthroline disulfonic acid). The affinity of OsBP7H8 to DNA-Os,phen is about five-fold higher as compared to DNA-Os,bipy. The antibody can be thus applied either for recognition of single-stranded and distorted regions in DNA (after DNA modification with Os,bipy) or for detection of both single-stranded and double-stranded DNAs (after DNA modification with Os,phen). A new simplified procedure for the dot-blot analysis is proposed, not requiring the purification of DNA-osmium adduct prior to its application to the membrane.

Significance of pre-treatment immunological parameters in colorectal cancer patients with unresectable metastases to the liver.

BACKGROUND/AIMS: In this study, we have compared the profiles of peripheral blood lymphocyte (PBL) subsets and serum cytokine levels of healthy individuals with those of patients with unresectable liver metastases from colorectal carcinoma before starting regional chemoimmunotherapy. Since the therapeutic responses are limited only to a subset of patients, we hypothesize that the initial status of immunity and individual immune response to a tumor might be significant to the therapeutic outcome. METHODOLOGY: Cellular and humoral immunological parameters were compared between 10 patients with colorectal cancer metastases to the liver responding and non-responding to regional intra-arterial chemo-immunotherapy, and 5 healty individuals. Analyses included a flow cytometric immunophenotyping of peripheral blood mononuclear cells (CD3, CD4, CD8, CD19, CD25, CD28, CD56, CD57, CD80 and HLA.DR), estimation of serum cytokine levels of interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-alpha), and other immunological parameters are soluble IL-2 receptor (sIL-2), carcinoembryonic antigen (CEA), gastrointestinal cancer-associated antigen (CA 19-9), and C-reactive acute phase protein (CRP). A significantly lower proportion of CD8 lymphocytes and a trend for decreased CD19, CD28 and CD80 was detected among colorectal cancer patients before liver-directed chemotherapy compared to healthy controls. RESULTS: The cancer patients showed a significantly increased population of peripheral NK cells as detected by both CD56+ and CD57+ phenotypes. Elevated serum levels of CRP, IL-4 and TNF-alpha, sIL-2R, but not IL-2, were also demonstrated in cancer patients as compared to controls. Activated CD25+ lymphocytes correlated negatively with CD28+ lymphocytes (r = -0.68, p < 0.01) and less significantly with CD4+ lymphocytes (r = -0.56, p < 0.05). The CD8+ cytotoxic cell subset might be negatively influenced by serum IL-4 (r = -0.57, p < 0.05). Positive correlation was found between sIL-2R and CRP (r = -0.78, p < 0.01), and between sIL-2R and TNF-alpha (r = 0.64, p < 0.05) serum levels in patients with progressive disease during the course of therapy, the initial proportions of CD4+, CD19+ and CD28+ lymphocytes were significantly lower than those among responders. Among humoral parameters, only sIL-2R showed a marginal correlation with therapeutic response, being more elevated among non-responding patients. Pre-treatment serum levels of CEA and CA 19-9 showed correlation with neither therapeutic response nor with any of the cellular or humoral immunological parameters analyzed. CONCLUSIONS: The results may serve as an initial guideline to open a discussion on the rationale of such a panel of tests, hopefully leading to standardized laboratory pre-selection and monitoring of patients treated with regional chemoimmunotherapy.

Conformational changes in p53 analysed using new antibodies to the core DNA binding domain of the protein.

The p53 protein contains a protease resistant core section that binds to DNA in a sequence specific manner and whose crystal structure has been determined. This core is flanked at the N-terminus by the transcriptional transactivation domain and at the C-terminus by sequences involved in the oligomerisation of the protein. Extensive immunochemical analysis of p53 has shown that dominant antigenic sites lie within these N- and C-terminal domains while few antibodies to the central core have been identified. One of these, PAb240, has been extensively characterised as its epitope is cryptic in the native DNA binding core structure but is exposed by denaturation. This epitope is also exposed on many p53 proteins that contain point mutations in the core domain suggesting that these mutations may have a common affect on the structure of the core. To investigate this further we have generated several new antibodies to novel sites on p53 and mapped their epitopes using synthetic peptides. We find that antibodies to two other discrete sites in the core can also, like PAb240, recognize cryptic epitopes and distinguish mutant from wild-type conformations implying that the point mutations found in p53 in human tumours have widespread effects on the folding pattern of the DNA binding domain.

p53 protein overexpression associates with growth patterns rather than with metastasizing in operable breast cancer.

We have analyzed p53 protein expression in 121 primary breast cancer biopsies by immunohistochemistry using the monoclonal antibody DO-1 and polyclonal serum CM-1. p53 protein overexpression has correlated in our study with mitotic activity (p=0.001), nuclear atypia (p=0.002), less favorable histological type of tumor and in a lesser extent with tumor size. The inverse, but highly significant, correlation (p=0.007) has been observed with lymph node involvement. There was also a trend for higher p53 positivity among DNA aneuploid tumors as compared with DNA diploid cases, but this was not significant. Our study suggests that p53, at least in some patients, may not be directly involved in the process of metastatic progression in breast cancer. Preliminary data would suggest that the detection of p53 protein overexpression could be a useful additional prognostic parameter in breast cancer.

Detection of keratin subtypes in routinely processed cervical tissue: implications for tumour classification and the study of cervix cancer aetiology.

We investigated the expression of keratin subtypes 7, 8, 10, 13, 14, 17, 18 and 19 in the normal cervix, in cervical intraepithelial neoplasia (CIN) lesions and in cervical carcinomas, using a selected panel of monoclonal keratin antibodies, reactive with routinely processed, formalin fixed paraffin embedded tissue fragments. The reaction patterns derived for each keratin antibody were compared with known expression patterns of the various epithelia, previously examined in frozen tissues. Although the reactivity of the antibodies was generally acceptable, considerable modifications to the manufacturers' staining instructions were often necessary. For some antibodies, which were previously thought to be reactive with fresh frozen tissue only, we developed staining protocols rendering them reactive with routinely processed material. As with previous findings in frozen sections we observed increasing expression of keratins 7, 8, 17, 18 and 19 with increasing grade of CIN. In cervical carcinomas the differences in keratin detectability between the main categories were more pronounced than in frozen sections, probably due to fixation and processing. For routine pathology, keratin phenotyping of cervical lesions may be of value in classification. The fact that keratin 7 was detected for the first time in reserve cells, and that this keratin was also found to be expressed in a considerable number of CIN lesions and cervical carcinomas supports the suggestion that reserve cells are a common progenitor cell for these lesions.

Fusion-induced malignancy? A preliminary study. (a challenge to today's common wisdom).

Five fusions between mouse embryonic cells and syngeneic adult peritoneal macrophages were performed. The resulting hybrids as well as both parental cells (6 cultures of embryonal cells and 6 cultures of adult macrophages) were grown in vitro under the same culture conditions. All populations of explanted macrophages died during the second month in primary culture and five populations of cultured embryonic cells were lost within six months under in vitro conditions as well. One embryonic cell line survived and acquired transformed and/or malignant phenotype: When inoculated into either newborn or adult syngeneic mice, progressive growth of tumors with 100% take (6/6), histologically classified as poorly differentiated fibrosarcoma with areas of metaplastic bone and osteoid, was observed. Two out of five wild hybrid strains died within six months of cell culture. The resulting three hybrid cultures adapted themselves to in vitro conditions and finally permanent lines were established with all features of transformed phenotype in vitro and with the capacity to grow as undifferentiated fibrosarcomas with 100% take (6/6) when inoculated into syngeneic mice either s.c. or i.p. Cytogenetic studies were performed and phenotypic characteristics of these lines were explored as well. Biological assays performed for the presence of oncogenic viruses were negative and none of the malignant cell lines showed positive staining with the monoclonal antibody specific for large T-antigen. It is suggested that cell fusion of two normal partners may switch on the cascade of abnormal processes which may culminate in neoplastic conversion. Cell fusion might play also a significant role in the so called "spontaneous" transformation.

Monoclonal antibodies recognizing different epitopes of cytokeratin No.18.

A comparative study of six mouse monoclonal antibodies against human 45 kDa keratin polypeptide (keratin No. 18) was undertaken using three experimental approaches: immunohistochemistry on normal human tissues, examination of interspecies cross-reactivity and identification of the target polypeptides in 1-D and 2-D immunoblots. The data suggest that at least five different antigenic sites of keratin 18 are recognized by this panel of reagents. The C-04 epitope is keratin 18-specific and widely conserved among mammalian species, while the antibodies DA7 and DC10 also react specifically with the 45 kDa keratin but stain simple epithelia of human origin only. Two antibodies, C-11 and C-66, decorate simple as well as stratified epithelia in human and in all seven animal species tested, but their respective target epitopes are shared by different groups of keratin polypeptides, which indicates their non-identity. In contrast to keratin specificity of the five above mentioned antibodies, the C-08 antibody cross-reacts with a 70 kDa nuclear lamina protein found in human and bovine tissues. The results of the present study provide the necessary basis for future applications of these antibodies in both routine immunodiagnostic work and as probes to study the biology of epithelial cells in general and the significance of keratin intermediate filaments in particular.

Novel monoclonal antibodies defining epitope of human cytokeratin 18 molecule.

Two monoclonal antibodies, DA7 and DC10, were obtained from fusions of mouse myeloma cells with splenic lymphocytes from mice immunized with human breast cancer cells of PMC 42 line. The indirect immunofluorescence studies performed on established tumor cell lines together with immunoperoxidase staining of normal human tissues showed that the components reacting with the antibodies were cytokeratins. Positive reaction was noted in all epithelia derived cultured cells and in all simple epithelial tissues known to express keratin 18. Immunoblotting performed on various cytoskeletal preparations demonstrated strong staining of a single band with a mobility corresponding to that of cytokeratin 18 (45 kD). The negative immunoperoxidase reaction found in different epithelial tissues of seven animal species suggests that both antibodies are specific for human keratin 18. It was shown that DA7 and DC10 antibodies exhibited strong reaction in paraffin embedded tissues fixed in either methacarn or standard formalin. These characteristics predetermine both antibodies as suitable reagents for the specialized histopathological work.

Monoclonal antibodies against individual cytokeratins in the detection of metastatic spread.

A panel of 17 monoclonal antibodies (MAbs) recognizing various keratin polypeptides has been used to define their binding on non-epithelial elements in 28 bone-marrow samples and 14 lymph nodes, in order to establish their limitations for use as a possible tool for immunodiagnosis of carcinoma spread. Immunocytochemical studies have shown that only 8 antibodies consistently exhibited no false-positive staining of marrow cells. All the remaining MAbs labelled (mostly in a non-specific manner) a few cells of marrow samples derived from patients with either haematological disorders or malignant lymphomas. Fine granules and droplet-like cytoplasmic inclusions were predominant patterns of positive reactions. Homogeneous cytoplasmic staining reminiscent of specific keratin immunolabelling was occasionally seen as well. The positive cells could be also identified in some lymph nodes free of tumour infiltration. All antibodies visualized cytoplasmic droplets in scattered cells of lymph nodes taken from a patient with non-Hodgkin lymphoma. This type of positivity was mostly associated with positive histochemical reactions for iron. Quite significant was the detection of fibrillar positivity in the extrafollicular reticular cells in all nodes examined. Such a specific type of staining was exclusively induced by antibodies directed against epitopes of keratin 8 and 18, whereas those MAbs recognizing keratin 7 and 19 always gave negative results. Our data indicate that caution is required when such MAbs, considered as markers of specific cell types, are being used as an immunodiagnostic tool to identify single carcinoma cells. A series of criteria, including morphological ones, must be utilized in order to obtain meaningful results.