119D5F1, reactive with lamin B1, kindly provided by Dr. Y. Raymond. 9. LN43, directed against lamin B2 (Bridger et al. 1993), kindly provided by Dr. E.B. Lane ...
Histochem Cell Biol (1997) 108:155–165
© Springer-Verlag 1997
O R I G I N A L PA P E R
&roles:Nicole Senden · Ilona Linnoila · Erika Timmer Helgi van de Velde · Anton Roebroek Wim Van de Ven · Jos Broers · Frans Ramaekers
Neuroendocrine-specific protein (NSP)-reticulons as independent markers for non-small cell lung cancer with neuroendocrine differentiation An in vitro histochemical study &misc:Accepted: 27 March 1997
&p.1:Abstract Neuroendocrine-specific protein (NSP)-reticulons have recently been discovered and were shown to exhibit a restricted, neuroendocrine/neural-specific expression pattern. These protein aggregates are anchored to the membranes of the endoplasmic reticulum and occur in small cell lung cancer (SCLC), but not in typical non-SCLC. In the current study we have examined the occurrence of NSP-reticulons in non-SCLC cell lines known to express neuroendocrine features (non-SCLCNE). NSP-reticulon expression was observed in all three non-SCLC-NE cell lines studied, albeit with variable intensity and in varying numbers of cells. Western blot analysis confirmed the presence of NSP-reticulon expression in these non-SCLC-NE cell lines, and showed that they were predominantly of the NSP-A type. When compared to conventional neuroendocrine markers, NSPreticulons revealed a distinct staining profile, showing only partial overlap with the other markers. The nonSCLC-NE cell lines combined these neuroendocrine characteristics with some features of non-SCLC. We conclude that NSP-reticulon expression is restricted to lung carcinoma cells with a neuroendocrine phenotype and predict that these constituents may become clinically relevant markers for the detection of neuroendocrine differentiation in solid tumours.&bdy:
N. Senden · E. Timmer · J. Broers · F. Ramaekers (✉) Department of Molecular Cell Biology and Genetics, Universiteit Maastricht, P.O. Box 616, NL-6200 MD Maastricht, The Netherlands Tel. +31-43-3881351; fax +31-43-3670948 I. Linnoila Biomarkers and Prevention Research Branch, National Cancer Institute, Rockville, Maryland, USA H. van de Velde · A. Roebroek · W. Van de Ven Laboratory for Molecular Oncology, Center for Human Genetics, University of Leuven, Campus Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium&/fn-block:
Introduction It is of clinical importance to distinguish lung cancers into small cell lung carcinoma (SCLC) and non-SCLC because of differences in their therapeutic response and prognosis (Ginsberg et al. 1994; Ihde et al. 1994). However, it is well accepted that carcinoma of the lung may present as a variety of phenotypes within one lung carcinoma subtype (Gazdar et al. 1981; Linnoila and Aisner 1995). In recent years, various groups have reported that a subset of 10–20% of histologically non-SCLC, particularly adenocarcinoma, express a variety of neuroendocrine markers, such as L-DOPA decarboxylase (DDC), neuron-specific enolase (NSE), chromogranin A, synaptophysin, neural cell adhesion molecule (NCAM) and Leu-7 (Berendsen et al. 1989; Gazdar et al. 1988, Graziano et al. 1989; Kibbelaar et al. 1991; Linnoila et al. 1988; Skov et al. 1991). Patients presenting with nonSCLC containing neuroendocrine characteristics (nonSCLC-NE) may exhibit different clinical behaviour from patients with non-SCLC without these properties, including responsiveness to chemotherapy (Graziano et al. 1994; Linnoila et al. 1994; Shaw et al. 1993). Neuroendocrine-specific protein (NSP)-reticulons have recently been described to be novel indicators of neuroendocrine tumour cells of the lung (van de Velde et al. 1994c) and were identified as a novel cluster of neuroendocrine markers during the Third International IASLC Workshop on Lung Tumour and Differentiation Antigens (Stahel et al. 1994). The protein family of NSP-reticulons, with the main members being NSP-A and NSP-C, is encoded by a single gene and associated with the endoplasmic reticulum (Roebroek et al. 1993; Senden et al. 1994b; van de Velde et al. 1994b). Immunohistochemical studies in solid lung tumours revealed expression of NSP-A in most of the SCLC and carcinoids, whereas non-SCLC tumours were negative. Interestingly, a group of non-SCLC-NE tumours showed NSPA expression (Senden et al. 1997; van de Velde et al. 1994c). In the present study we compared the expression profiles of NSP-reticulons in three non-SCLC-NE
156
cell lines in relation to conventional neuroendocrine markers and examined to what extent these neuroendocrine markers are coincident with typical non-SCLC characteristics.
Materials and methods Lung cancer cell lines Human lung cancer cell lines used in this study included the nonSCLC-NE cell lines NCI-H460, NCI-810, NCI-H1155 derived from large cell lung carcinomas (LCLC) (Brower et al. 1986; Lan et al. 1993), cell line LCLC-103H (Bepler et al. 1988), also derived from a LCLC, and the adenocarcinoma cell line NCI-H125 (Brower et al. 1986). Furthermore, a classic SCLC (SCLC-C) cell line, NCI-H69, and a variant SCLC (SCLC-V) cell line, NCI-H82, were used for comparison (Carney et al. 1985; Gazdar et al. 1985). All cell lines were cultured as previously described (Broers et al. 1988b) Protein and RNA assays DDC assays were performed as described in Gazdar et al. (1988), while radioimmunoassays for NSE, chromogranin A and calcitonin were performed and evaluated as described by Carney et al. (1985) and Deftos et al. (1988). For cell line LCLC-103H, the DDC activity and NSE concentration were deduced from Bepler et al. (1988). For protein gel electrophoresis, immunoblotting and sample preparation the same procedures were followed as previously described (Senden et al. 1994b), while RNA in situ hybridization of surfactant protein-A (SP-A) and Clara cell protein was done as described by Broers et al. (1992)iii and Gu et al. (1988).
Antibodies to intermediate filament proteins used in immunofluorescence microscopy 1. RCK 102, which recognizes cytokeratins 5 and 8 (Ramaekers et al. 1987) 2. RCK 105, reactive with cytokeratin 7 (Broers et al. 1988a; Ramaekers et al. 1987) 3. MNF, reactive with the 68-kDa and 200-kDa neurofilament subunits (Klück et al. 1984) (Euro-Diagnostica, Arnhem, The Netherlands) 4. RNF403, reactive with the 160-kDa neurofilament subunit 5. 41CC4, directed to lamin A and lamin C, kindly provided by Dr. G.Warren (Heidelberg, Germany) (Burke et al. 1983) 6. R27, reactive with lamins A and C, and a gift from Dr. G. Krohne (Heidelberg, Germany) (Zatloukal et al. 1992) 7. 133A2, reactive only with lamin A (Hozák et al. 1995), kindly provided by Dr. Y. Raymond (Montreal, Canada) 8. 119D5F1, reactive with lamin B1, kindly provided by Dr. Y. Raymond 9. LN43, directed against lamin B2 (Bridger et al. 1993), kindly provided by Dr. E.B. Lane (Dundee, UK) Antibodies to NSP-reticulons used for immunofluorescence microscopy and immunoblotting 1. MON 160, MON 161 and MON 162, recognizing specifically NSP-A and used as a 1:1:1 culture supernatant mix (MON 160–162) (Roebroek et al. 1993; van de Velde et al. 1994a, b). 2. RNL-2 and RNL-3, recognizing epitopes in NSP-A and NSP-B (Broers et al. 1991; Senden et al. 1994a, b) 3. RNL4, raised against a synthetic peptide encompassing the unique 20 amino-terminal amino acids of NSP-C (Senden et al. 1996) 4. Polyclonal rabbit antiserum POL-1, recognizing epitopes present in NSP-A and in NSP-B (Senden et al. 1994a; van de Velde et al. 1994a, b)
Electron microscopy and immunofluorescence microscopy The methodology for transmission electron microscopy is described elsewhere (Gazdar et al. 1988). For the indirect immunofluorescence technique, floating cells were placed onto glass slides using cytocentrifugation, while cells which grew as adherent cultures were detached from the culture dish with 0.125% trypsin (Difco Laboratories, Detroit, Mich., USA), 0.02% EDTA and 0.02% glucose (Merck, Darmstadt, Germany) in PBS, pH 7.4 and allowed to grow on glass coverslips for 1–2 days. All cells were fixed by dipping in methanol for 5 s (−20° C), followed by three rinses in acetone (Merck) for 5 s (4° C). Subsequently, the cells were incubated for the single- and double-label immunofluorescence assay, as described (Senden et al. 1994b). Antibodies Antibodies to conventional neuroendocrine markers used in immuofluorescence- or radio-immunoassays 1. RNL-1 recognizing NCAM (Broers et al. 1991) 2. A polyclonal rabbit antiserum to synaptophysin (DAKO A/S, Glostrup, Denmark) 3. LK2H10 to chromogranin A (available from BioGenex, San Ramon, Calif., USA) 4. HNK-1, recognizing Leu-7 (Abo and Balch 1981; Bunn et al. 1985) (Becton Dickinson, Mountain View, Calif., USA) 5. Polyclonal rabbit antiserum to NSE (Accurate Chemical, Westbury, N.Y., USA) 6. Polyclonal rabbit antiserum to calcitonin (Linnoila et al. 1984), kindly provided by Dr. K.Becker (Washington, D.C., USA), diluted 1:5000 for radioimmunoassays
Results The expression patterns of the differentiation markers tested in the lung carcinoma cell lines are summarised in Table 1. The results of the immunofluorescence studies are further illustrated in Figs. 1–3. Several differentiation characteristics of the two SCLC cell lines have been extensively described and are used here as reference cell lines for neuroendocrine differentiation. The same holds true for the two non-SCLC cell lines, LCLC-103H and NCI-H125, which serve as references for non-neuroendocrine differentiation. For comparison, phenotype characteristics of these SCLC and non-SCLC cell lines were extracted from the literature (Bepler et al. 1988; Broers et al. 1988b, 1991, 1993; Brower et al. 1986; Carney et al. 1985; Gazdar et al. 1990; Lan et al. 1993) and are included in Table 1. NSP-reticulons In both SCLC cell lines NSP-A is present, in NCI-H82 in most cells and in NCI-H69 in some of the cells. NSPC immunofluorescence results were negative in both SCLC cell lines. In the non-SCLC cell lines LCLC103H and NCI-H125, no expression of NSP-A or NSP-
157 ron-specific enolase, CgA chromogranin A, Syn synaptophysin, NCAM neural cell adhesion molecule, SP-A surfactant protein-A, NF neurofilament, CK cytokeratin, NSP neuroendocrine-specific protein, ND not done, + positive, − negative)&/tbl.c:&
Table 1 Phenotypes of the lung cancer cell lines. (SCLC-V Variant-type SCLC, SCLC-C classic-type SCLC, non-SCLC-NE, nonSCLC with neuroendocrine features, EM electron microscopy, DCV dense core vesicle, DDC L-DOPA decarboxylase, NSE neuProperty/ phenotype Type Origin Histology
Cell line NCI-H82
NCI-H69
LCLC-103H
NCI-H125
NCI-H460
NCI-H810
NCI-H1155
SCLC-V Pleural effusion Small cell/ large cell
SCLC-C Pleural effusion Small cell with giant cells Floating
Non-SCLC Pleural effusion Large cell
Non-SCLC Solid lung tumour Adenosquamous
Non-SCLC-NE Pleural effusion Large cell
Non-SCLC-NE Solid lung tumour Large cell
Non-SCLC-NE Mediastinal bronchoscopy Large cell
Adhering
Adhering
Adhering
Present
ND
None
Rare or none
Floating, partly adhering Clusters of small DCVs
Growth morphology EM/DCVs
Floating, partly adhering Occasionally 2– 3 small DCVs
DDC (U/mg) NSE (ng/mg protein) CgA (ng/mg protein) Syn NCAM Leu-7 Calcitonin (ng/mg protein) SP-A mRNA Clara cell protein mRNA NF 68/200 NF 160 CK 5,8 CK 7 Vimentin Lamin A/C Lamin A Lamin B1 Lamin B2 NSP-A NSP-B NSP-C
0.6 317
149 817
