Int. J. Cancer: 72, 631–636 (1997) r 1997 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer INHIBITION OF TUMORIGENICITY IN LUNG ADENOCARCINOMA CELLS BY c-erbB-2 ANTISENSE EXPRESSION Patrizia CASALINI1, Sylvie MÉNARD1, Sergio M.I. MALANDRIN1, Cristina M. RIGO1, Maria I. COLNAGHI1*, Constance M. CULTRARO2 and Shoshana SEGAL2 1Division of Experimental Oncology E, Istituto Nazionale Tumori, Milan, Italy 2NCI-Navy Medical Oncology Branch, Naval Hospital Bethesda, MD, USA The lung carcinoma cell line Calu3, which overexpresses the c-erbB-2 oncogene, was stably transfected with antisense (AS) cDNA constructs encompassing different regions of the c-erbB-2 gene. Transfected cells were analyzed for their tumorigenic properties in vitro and in nude mice. Two independent clones, AS F1 (low erbB-2 expressor) and AS B12 (high erbB-2 expressor), as well as the polyclonal Calu3/AS 58, were selected for these analyses. In Calu3/AS 58 transfected cells and in the AS F1 clone, c-erbB-2 RNA and protein levels were lower than those detected in the parental cell line and the AS B12 clone. Anchorage-independent growth and tumor take were also significantly reduced. Furthermore, cells derived from primary tumors of Calu3/AS 58, AS F1 and AS B12 lost the AS c-erbB-2 DNA insert but retained the gene for G418 resistance. Our results suggest that a correlation between c-erbB-2 overexpression and tumorigenicity may exist in the Calu3 lung carcinoma cell line. Int. J. Cancer 72:631–636, 1997. r 1997 Wiley-Liss, Inc. The c-erbB-2/HER-2 proto-oncogene encodes a 185-kDa transmembrane tyrosine kinase receptor (p185c-erbB-2) which is the human counterpart of the rodent c-neu. Abnormalities of p185c-erbB-2 expression which result primarily from transcriptional deregulation and/or gene amplification (Pasleau et al., 1993; Kraus et al., 1987) have been implicated in the pathogenesis of a number of human tumors, including breast, stomach, ovary, bladder and salivary gland carcinomas (Hynes and Stern, 1994). Several retrospective clinical studies of breast adenocarcinoma patients have indicated that p185c-erbB-2 overexpression correlates with reduced overall and disease-free survival (Press et al., 1993; Byers et al., 1990) and a limited response to some adjuvant cytotoxic and hormonal therapies (Gusterson et al., 1992; Wright et al., 1989). A role for p185c-erbB-2 in neoplastic transformation of NIH3T3 cells (Di Fiore et al., 1987) and in the induction of a tumorigenic phenotype in the immortalized MTSV1.7 human mammary epithelial cell line (D’Souza and Papa-Dimitriou, 1994) has also been demonstrated. Decreased expression of p185c-erbB-2, through the use of monoclonal antibodies (MAbs) specific for the extracellular domain of the protein, resulted in growth inhibition and, in some cases, differentiation of breast cancer cell lines (Bacus et al., 1992; Tagliabue et al., 1991; Stancovski et al., 1991). In addition, studies have shown that c-erbB-2 interaction with the adenovirus E1A promoter suppresses tumorigenicity (Yu et al., 1990), although no evidence was found to indicate that the loss of tumorigenicity was the direct consequence of this interaction (Yu et al., 1993). Antisense (AS) RNA has been shown to be an effective tool to directly control the expression of target gene products without affecting other cellular functions (Hélène, 1994). AS oligonucleotides have been used to inhibit the in vitro proliferation of human breast cancer cells in which c-erbB-2 is amplified (Colomer et al., 1994), but it remains unclear whether such inhibition is reflected in decreased tumorigenicity in vivo. Moreover, specific downregulation of c-erbB-2 mRNA and protein has been achieved using phosphorothioate AS oligonucleotides, resulting in accumulation of cells in the G1 phase of the cell cycle (Vaughn et al., 1995). To test whether inhibition of c-erbB-2 expression correlates with reduced tumorigenicity in vivo, the human Calu3 lung carcinoma cell line was transfected with AS cDNA constructs encompassing various regions of the c-erbB-2 gene. Growth properties of the transfectants as well as their ability to induce tumors in nude mice were examined. MATERIAL AND METHODS Preparation of constructs The LTR-2/erbB-2 expression vector, kindly provided by Dr. P.P. Di Fiore (Di Fiore et al., 1987), was used as a template to obtain the following fragments: 1) 151-bp PCR product corresponding to the 58 end (nts 1–151) of the human c-erbB-2 coding region; 2) 291-bp PCR product corresponding to the transmembrane region (TM) (nts 2110–2400) of the human c-erbB-2 coding region; and 3) 415-bp PCR product corresponding to the 38 end (nts 3241–3655) of the human c-erbB-2 coding region. cDNA fragments were amplified using synthetic primers (Fig. 1a) containing HindIII restriction sites (Fig. 1a, boldface). Amplifications were performed in a 100-µl volume containing 0.3 mM of each primer and 2.5 units of Taq polymerase (Perkin Elmer Cetus, Norwalk, CT); the amplification program used was 1 min at 94°C, 1 min at 65°C and 2 min at 72°C for 30 cycles followed by 10 min at 72°C. The amplified cDNA fragments were ligated into HindIII-digested pRC/CMV vector (Invitrogen, San Diego, CA), and one clone of each construct in the antisense orientation (pRC/CMV-AS 58, pRC/CMV-AS TM and pRC/CMV-AS 38) was selected. Sequence analysis of the 3 selected clones showed 100% homology with the published sequence of c-erbB-2 (Coussens et al., 1985) (GenBank A.N. M11730). Transfection and selection The human lung adenocarcinoma cell line Calu3 (ATCC, Rockville, MD) with the amplified and overexpressed c-erbB-2 gene (Ménard et al., 1993) was used. The cell line was maintained in RPMI-1640 (Sigma, St. Louis, MO) medium supplemented with 10% FCS (HyClone, Logan, UT) and glutamine at 37°C in 5% CO2 incubator. Cells were plated in 100-mm dishes, and when 80% confluent, transfected using the Lipofectin (GIBCO BRL, Paisley, UK) procedure; cells were treated either with 50 µl of Lipofectin reagent alone for selection control, or with Lipofectin plus 20 µg of pRC/CMV or pRC/CMV-AS 58 in serum-free medium. The resulting transfectants were designated as Calu3/neo and Calu3/AS 58, respectively. Cells were then incubated for 16–18 hr at 37°C and, after fresh medium replacement, were incubated for an additional 24 hr. To select for clones, cells were trypsinized and plated in the presence of 500 µg/ml of G418 (GIBCO BRL). When all control cells died, medium was replaced and transfectants maintained in medium supplemented with 300 µg/ml of G418. Surviving resistant colonies were harvested, grown and pooled. An aliquot of the Calu3/AS 58 pool was used for single-cell cloning by limiting dilution. Contract grant sponsor: Associazone Italiana per la Ricerca sul Cancro; Contract grant sponsor: CNR-ACRO. *Correspondence to: Division of Experimental Oncology E, Istituto Nazionale Tumori, Via Venezian 1, 20133 Milan, Italy. Fax: 139-22362692. E-mail: [email protected] Received 19 December 1996; revised 18 March 1997 632 CASALINI ET AL. FIGURE 1 – Construction and integration analysis strategies. (a) Schematic representation of the construction of pRC/CMV-AS 58, pRC/CMV-AS TM and pRC/CMV-AS 38 expression vectors. The AS fragments were amplified from the LTR-2/erbB-2 expression vector using synthetic primers (HindIII recognition sequence in bold). (b) Schematic of primer annealing sites for construct integration analysis by PCR and antisense mRNA detection by RT-PCR. Construct integration analysis Genomic DNA was extracted from cultured cells using standard procedures (Ausubel et al., 1987). For construct integration analysis, a 180-bp fragment was amplified by PCR (see above) using a vector-specific T7 oligonucleotide as an upstream primer and a c-erbB-2-specific oligonucleotide as a downstream primer (Fig. 1b). Genomic DNAs (0.5 µg) obtained from parental or transfected cells or 50 ng of control plasmid were used as templates. The amplification program used was an initial denaturation step for 5 min at 95°C, followed by 1 min at 94°C, 1 min at 50°C and 2 min at 72°C for 30 cycles and a final extension step for 10 min at 72°C. PCR products (30 µl) were fractionated on a 1.5% INHIBITION OF TUMORIGENICITY BY c-erbB-2 ANTISENSE agarose gel, transferred to a nitrocellulose membrane (Schleicher and Schuell, Keene, NH) and hybridized to a [32P]dCTP (Amersham, Aylesbury, UK) random-primed (Boehringer Mannheim, Mannheim, Germany) c-erbB-2 probe encompassing nt 1–151. Reverse transcription and RT-PCR Total RNA (0.1 µg) (see Northern Blot Analysis) was reversetranscribed at 50°C for 15 min in a volume of 20 µl using 50 U of M-MLV reverse transcriptase (Perkin Elmer Cetus), 1 mM of each dNTP and 4 mM of SP6 primer (vector-specific) (Fig. 1b). The enzyme was then heat-inactivated and cDNAs were amplified using T7 and the specific c-erbB-2 primers as described in the procedure for construct integration analysis (except that annealing was done at 50°C). Amplified material (30 µl) was fractionated on a 1.5% agarose gel, and Southern blot analysis was performed as described above. Northern blot analysis Total RNA was purified from cells using the Ultraspec RNA isolation system (BioTecx, Houston, TX). Aliquots of 30 µg were fractionated on a 1% agarose-formaldehyde gel, transferred to nitrocellulose membranes (Schleicher and Schuell), and hybridized to a [32P]-labeled-c-erbB-2 coding region probe. After stripping, membranes were rehybridized to a [32P]-labeled-b-actin probe (Amersham) (Gunning et al., 1983). Densitometric analysis of membranes was performed by scanning on a phosphorimager using the Image Quant System (Molecular Dynamics, Sunnyvale, CA). Flow cytometric analysis Cells were trypsinized, washed with fluorescence-activated cell sorting (FACS) buffer (PBS containing 0.03% of BSA) and resuspended in 100 µl FACS buffer containing 10 µg/ml of anti-p185c-erbB-2 MAb MGR6 (Centis et al., 1992). Following incubation on ice for 45 min, cells were washed with FACS buffer and incubated with fluorescein-labeled goat anti-mouse Ig (Kierkegard and Perry, Gaithersburg, MD) for 45 min on ice. After washing with FACS buffer, cells were resuspended in 1 ml FACS buffer and analyzed for fluorescence using a Becton Dickinson FACScan (Mountain View, CA). Fluorescence intensity for the c-erbB-2 oncoprotein was calibrated using quantitative radio-immunoassay of different cell lines with known protein expression (Pupa et al., 1993). 633 RESULTS Effects of different c-erbB-2 antisense constructs on cell proliferation Three constructs, one consisting of 151 bp derived from the 58 end (AS 58-pRC/CMV), another of 300 bp corresponding to the transmembrane region (AS TM-pRC/CMV) and the third of 450 bp encompassing the 38 region (AS 38-pRC/CMV) of the c-erbB-2 c-DNA, were cloned independently in the antisense orientation into the pRC/CMV expression vector (Fig. 1). To test the effect of c-erbB-2 AS expression on cell proliferation, Calu3 cells, which constitutively overexpress c-erbB-2, were transfected with the 3 recombinant plasmids and with an empty pRC/CMV vector as a control, and colony formation in each of the transfected cell populations was evaluated. Two days after transfection, cells were plated at 3.5 3 105 (cells/plate) and grown in the presence of G418 for 3–4 weeks. Colonies consisting of at least 10 cells were then counted. The colony-forming ability of Calu3 cells transfected with each of the 3 AS vectors was markedly reduced as compared to cells transfected with the empty vector, but the largest reduction observed was in cells transfected with the AS 58-pRC/CMV construct (Fig. 2), as repeatedly confirmed in other experiments (data not shown). Thus, all remaining experiments were performed using this construct. Construct integration and its effect on c-erbB-2 expression in Calu3 cells Colonies of Calu3 cells stably transfected with the AS 58-pRC/ CMV vector (Calu3/AS 58) were pooled and grown for further studies. This Calu3/AS 58 polyclonal cell line was also subcloned by limiting dilution. Analysis of 10 independent clones obtained by this procedure revealed a wide range of c-erbB-2 expression at both the mRNA and protein levels. Two clones, AS F1 (c-erbB-2 lowest expressor) and AS B12 (c-erbB-2 highest expressor), together with the original polyclonal cell line Calu3/AS 58 and the empty vector-transfected Calu3 (Calu3/neo), were used for further analyses. Construct integration was examined by PCR analysis of Soft agar assay Cells were plated in triplicate wells of 6-well plates (Costar, Celbio, Milan, Italy) in semi-solid medium containing 0.3% Bacto-Agar (DIFCO, Detroit, MI) supplemented with 30% FCS. Medium for transfectants was supplemented with 300 µg/ml G418. Plates were incubated at 37°C in 5% CO2 and colonies were scored after 3 weeks. Statistical analysis was performed using a two-tailed t-test. Tumor growth assay The tumorigenic capacity of transfected and parental Calu3 cells was assayed in 2 series of experiments following s.c. injection of 1 3 105 cells into the flanks of nude mice. In the first series, parental Calu3, AS F1 and AS B12 cells were compared, and in the second series Calu3, Calu3/neo and AS 58 cells were compared. Tumor volume was calculated using the formula 0.5 3 d21 3 d2, where d1 is the smaller and d2 the larger diameter of the tumor as measured with calipers. Animals were killed when the average tumor size reached 1 cm3, or earlier if the tumor became necrotic. Statistical analysis on tumor take results was performed for every series using the chi-square test. After sacrificing the mice, one of each randomly selected primary tumors was re-established in culture and then tested for G418 resistance and c-erbB-2 expression. The remaining tumors were snap-frozen by immersion in liquid nitrogen immediately following removal and stored at 280°C prior to RNA isolation. RNAs were isolated from tumor tissue using the Ultraspec isolation system (BioTecx). FIGURE 2 – Colony-forming potential of Calu3 cells transfected with pRC/CMV-AS 58, pRC/CMV-AS TM or pRC/CMV-AS 38. After transfection, cells were plated at 3 3 105 cells/well in the presence of 300 µg/ml G418. After 3 weeks, colonies were stained with methylene blue. Data are reported as percentage of colonies obtained in control cells transfected with the empty vector. Error bars represent standard deviation of 3–6 replicates. 634 CASALINI ET AL. genomic DNA using a 58 T7 vector-specific oligonucleotide and a 38 c-erbB-2 oligonucleotide spanning nucleotides 115–141 (Fig. 1b). Southern analysis with a c-erbB-2 cDNA probe supported the specificity of the PCR product obtained (Fig. 3). The intensity of the hybridization signal indicated that the level of construct integration in the 2 AS clones was similar to that of the polyclonal Calu3/AS 58. The set of primers specifically amplified transfected AS c-erbB-2 and, as expected, no PCR product was generated with parental Calu3 genomic DNA. To determine whether this AS inhibits c-erbB-2 expression, total RNA obtained from polyclonal Calu3/neo, Calu3/AS 58 and clones AS F1 and AS B12 was analyzed by Northern blotting, with densitometry normalized to actin levels (Fig. 4). The polyclonal Calu3/AS 58 showed only a slightly lower level of c-erbB-2 mRNA than parental Calu3 or control Calu3/neo cells. The AS F1 clone demonstrated a 2.2-fold reduction as compared to the Calu3/neo cells, whereas c-erbB-2 expression in the AS B12 clone was unaffected. AS c-erbB-2 mRNA was undetectable by Northern blot analysis (data not shown). Because antisense transcripts may not accumulate to levels detected by Northern blot analysis due to rapid degradation, we utilized the more sensitive RT-PCR assay followed by Southern blot analysis. As shown in Figure 5, AS mRNA expression was lower in the AS B12 clone than in the AS F1 clone. FACScan analyses to determine whether the reduction in c-erbB-2 mRNA was paralleled by a reduction in p185c-erbB-2 membrane protein expression (Fig. 6) revealed a wide peak for the Calu3/AS 58 polyclonal cells, indicating a high degree of heterogeneity in p185c-erbB-2 expression, consistent with the similarity of a majority of cells to the Calu3/neo and parental cells, and with a small fraction of cells exhibiting reduced oncoprotein expression. Based on the linear relationship between fluorescence intensity and amount of oncoprotein (Pupa et al., 1993), we estimated a 70% decrease in c-erbB-2 expression in this fraction. In contrast, a homogeneous peak was obtained for the AS F1 cells, with a lower level of p185c-erbB-2 expression as compared with those in AS B12 cells. The mean intensity decreased from 300 in the control to 100 in F1 cells, corresponding to a 67% reduction. The protein:mRNA ratio was therefore lower in the F1 clone (0.6) than in the B12 clone (1) and the parental cells (1). Reduction in tumor development by antisense c-erbB-2 The tumorigenic capacity of the AS clones and polyclonal Calu3/AS 58 was evaluated by assaying anchorage-independent growth and/or subcutaneous growth in nude mice (Table I). Both the clonogenic potential and the tumor take of the AS F1 cells were significantly lower than the values obtained for parental Calu3 or FIGURE 3 – Southern blot analysis of PCR products derived from genomic DNAs. PCR products, obtained as described in Figure 1b, were fractionated on a 1.5% agarose gel, transferred to nitrocellulose filters and hybridized with a [32P]-dCTP-labeled c-erbB-2 probe encompassing nucleotides 1–151. Lane 1, parental Calu3 cell line; lane 2, clone AS B12; lane 3, clone AS F1; lane 4, polyclonal Calu3/AS 58; lane 5, pRC/CMV-AS 58 control plasmid. AS B12 cells. Indeed, the AS F1 cells exhibited a 3-fold reduction in anchorage-independent growth as compared to AS B12 and Calu3 cells. After subcutaneous injection of 105 AS F1 cells or Calu3/AS 58 cells, only 6 of 13 and 9 of 15 nude mice developed measurable nodules, respectively. Nude mice injected with AS B12 or Calu3/neo control cells displayed 85% and 100% tumor take, respectively, whereas parental Calu3 cells induced tumors in 93–100% of the mice. The growth properties of tumors developing from AS F1 or AS B12 cells were very similar to those of the parental Calu3 tumors, except that measurable tumors from AS F1 arose 10 days later than those from parental cells. Northern blot analysis of tumor RNAs derived from antisense transfectants after in vivo growth showed that endogenous levels of c-erbB-2 expression were identical to those in controls (data not shown). RT-PCR analysis clearly demonstrated that tumor cells derived from primary tumors of Calu3/AS 58, AS F1 and AS B12 lost the AS c-erbB-2 DNA insert but retained the gene for G418 resistance (Fig. 7). DISCUSSION The present study was designed to test whether the c-erbB-2 oncogene plays a role in determining the transformed phenotype of FIGURE 4 – Northern blot analysis of parental and transfected Calu3 cells. Total RNA (30 µg) was fractionated on a 1% agaroseformaldehyde gel, transferred to a nitrocellulose filter and hybridized with [32P]dCTP-full-length erbB-2 and a [32P]dCTP-labelled b-actin control. Histograms show results of densitometric analysis expressed in arbitrary units and normalized to b-actin expression. Lane 1, parental Calu3 cell line; lane 2, clone AS B12; lane 3, clone AS F1; lane 4, polyclonal Calu3/AS 58; lane 5, Calu3/neo. FIGURE 5 – Southern blot analysis of RT-PCR products derived from total RNA. Total RNAs (1 µg) were reverse-transcribed and amplified by PCR as described in Figure 1b. The amplified products were fractionated on a 1.5% agarose gel, transferred to a nitrocellulose filter and hybridized with the same [32P]-dCTP-labeled c-erbB-2 probe used for Figure 3. Lane 1, Calu3/neo cell line; lane 2, polyclonal Calu3/AS 58; lane 3, clone AS F1; lane 4, clone AS B12. INHIBITION OF TUMORIGENICITY BY c-erbB-2 ANTISENSE 635 FIGURE 6 – FACScan analysis of parental and transfected Calu3 cells. (a) (1) parental Calu3, (2) Calu3/neo, (3) Calu3/AS 58. (b) (1) parental Calu3, (4) AS F1, (5) AS B12. TABLE I – CLONOGENICITY AND TUMOR TAKE OF TRANSFECTED CELLS1 Target cells Clonogenicity Calu3 AS B12 AS F1 Calu3 Calu3/neo Calu3/AS 58 112 6 31.82 140 6 16.97 38 6 4.24 N.D. N.D. N.D. p2 0.442 0.016 Tumor take % 13/14 11/13 6/13 14/14 9/9 9/15 93 85 46 100 100 60 p3 0.01 0.005 1In clonogenicity experiments, cells were plated in triplicate in 0.3% Bacto-Agar and scored after 3 weeks. In tumorigenicity experiments, 105 cells were injected subcutaneously in the flank of 9–15 nude mice. Tumor take was scored when control tumors reached 1,000 mm3.– 2Evaluated by two-tailed t-test.–3Evaluated by chi-square test. Calu3 cells. The results suggest that a correlation between c-erbB-2 overexpression and tumorigenicity may exist in these cells. Reduced expression of the oncoprotein, due to expression of AS c-erbB-2, resulted in reduced clonogenicity and reduced tumor take in vivo. Inhibition of c-erbB-2 expression by the AS construct indicates that c-erbB-2 overexpression is an important factor required for the growth of this human lung adenocarcinoma cell line. This requirement was evidenced by the reduced ability of AS transfectants to form colonies. Therefore, it appears that inhibition of c-erbB-2 overexpression is detrimental to cells that normally overexpress the gene. Our findings are consistent with recent reports demonstrating that inhibition of c-erbB-2 expression by the introduction of c-erbB-2-specific antibody into cells leads to apoptosis (Deshane et al., 1996). It is thus likely that an intermediate expression of antisense constructs in the polyclonal cells and those derived from the selected clones is able to reduce but not abolish oncoprotein expression allowing cell survival and growth. In fact, the growth of these transfectants in soft agar and in nude mice was reduced but not completely inhibited. The possibility that the decrease in malignancy observed in the AS F1 clone was due to the selection of a pre-existing poorly malignant cell present in the parental Calu3 cannot be completely ruled out; however, random selection of cells with low proliferation potential from rapidly growing cells seems unlikely. Moreover, all clones isolated from the parental Calu3 cells by limiting dilution overexpressed FIGURE 7 – Southern blot analysis of RT-PCR products derived from total RNA. Total RNA (1 µg) from transfected cells and tumors was processed as described in Figure 5. Lane 1, Calu3/neo control cell line; lane 2, polyclonal Calu3/AS 58 line; lane 3, clone AS F1; lane 4, clone AS B12; lanes 5–8 correspond, respectively, to lanes 1–4, except for using RNA extracted from tumors derived from the transfected cell lines. c-erbB-2 at levels similar to those observed in the parental cells. The various patterns of behavior observed in the polyclonal transfected population further support the relevance of the antisense transfection in reducing c-erbB-2 expression. The role c-erbB-2 plays in tumorigenicity was confirmed by RT-PCR of tumor mRNAs, since only the cells that lost the AS insert were able to establish and develop tumors (Fig. 7). In fact, the delay in tumor development reflects a lag in time due to the selection of cells that lost the AS insert. Nevertheless, these cells remained neomycin-resistant. Immunohistochemical analysis of the frozen specimens for c-erbB-2 expression revealed homogeneous and comparable staining of the tumors despite the origin of the cells or the size of the nodule (data not shown). Ideally, the targeted mRNA and protein are most effectively inhibited in clones with the highest level of construct integration and antisense RNA expression. Indeed, AS F1 and B12 cells display comparable levels of construct integration, but inhibition of c-erbB-2 expression was observed only in clone F1, which expresses high levels of AS RNA. Neither the inability to detect AS RNA by Northern blot analysis nor the loose correlation between 636 CASALINI ET AL. the level of AS RNA expression/construct integration and the degree of reduction in c-erbB-2 mRNA was totally unexpected, since these discrepancies have been observed in other models (Kook et al., 1994; Neckers et al., 1992; Kasid et al., 1989; van der Krol et al., 1988). Possible explanations for these findings include positional effects that may influence the level of transgene transcription, high instability of the RNA duplexes and a greater instability of the antisense RNA relative to the mRNA such that antisense RNA is less easily detectable (Kook et al., 1994). In fact, we were unable to detect AS RNA by Northern analysis even with polyA RNA, or by RT-PCR using polyT oligonucleotides. Furthermore, RT-PCR of cytoplasmic RNA with the same oligonucleotide primers used in RT-PCR of total RNA was also negative, suggesting that AS c-erbB-2 functions in part via transcriptional inhibition by the formation of DNA/RNA duplexes and/or by inhibition of mRNA transport to the cytoplasm, forming RNA/RNA hybrids. Our results, together with similar reports on c-raf-1, EGFR and uPAR antisense inhibition (Kook et al., 1994; Moroni et al., 1992; Kasid et al., 1989), suggest the potential utilization of AS approaches for therapeutic purposes to downmodulate expression of tumor-promoting genes. However, technical problems such as antisense instability and low transfection efficiency must first be solved. ACKNOWLEDGEMENTS We thank Mrs. D. Labadini for preparing the manuscript and Mr. M. Azzini for photographic reproduction. This work was partially supported by a grant from the Associazione Italiana per la Ricerca sul Cancro and by CNR-ACRO. REFERENCES AUSUBEL, F.M., BRENT, R., KINGSTON, R.E., MOORE, D.D., SEIDMAN, J.G., SMITH, J.A. and STRUHL, K., Current Protocols in Molecular Biology, Green and Wiley-Interscience, New York (1987). BACUS, S.S., STANCOVSKI, I., HUBERMAN, E., CHIN, D., HURWITZ, E., MILLS, G.B., ULLRICH, A., SELA, M. and YARDEN, Y., Tumor-inhibitory monoclonal antibodies to the HER-2/Neu receptor induce differentiation of human breast cancer cells. Cancer Res., 52, 2580–2589 (1992). BYERS, V.S., LEVIN, A.S., WAITES, L.A., STARRETT, B.A., MAYER, R.A., CLEGG, J.A., PRICE, M.R., ROBINS, R.A., DELANEY, M. and BALDWIN, R.W., A Phase-I/II study of Trichosanthin treatment of HIV disease. AIDS, 4, 1189–1196 (1990). CENTIS, F., TAGLIABUE, E., UPPUGUNDURI, S., PELLEGRINI, R., MARTIGNONE, S., MASTROIANNI, A., MÉNARD, S. and COLNAGHI, M.I., p185 HER2/neu epitope mapping with murine monoclonal antibodies. Hybridoma, 11, 267–276 (1992). COLOMER, R., LUPU, R., BACUS, S.S. and GELMANN, E.P., erbB-2 antisense oligonucleotides inhibit the proliferation of breast carcinoma cells with erbB-2 oncogene amplification. Brit. J. Cancer, 70, 819–825 (1994). COUSSENS, L., YANG-FENG, T.L., LIAO, Y.-C., CHEN, E., GRAY, A., MCGRATH, J., SEEBURG, P.H., LIBERMANN, F.A., SCHLESSINGER, J., FRANCKE, U., LEVINSON, A. and ULLRICH, A., Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene. Science, 230, 1132–1139 (1985). DESHANE, J., GRIM, J., LOECHEL, S., SIEGAL, G.P., ALVAREZ, R.D. and CURIEL, R.D., Intracellular antibody against erbB-2 mediates targeted tumor cell eradication by apoptosis. Cancer Gene Therapy, 3, 89–98 (1996). DI FIORE, P.P., PIERCE, J.H., KRAUS, M.H., SEGATTO, O., KING, C.R. and AARONSON, S.A., erbB-2 is a potent oncogene when overexpressed in NIH/3T3 cells. Science, 237, 178–182 (1987). D’SOUZA, B. and TAYLOR-PAPADIMITRIOU, J., Overexpression of ERBB2 in human mammary epithelial cells signals inhibition of transcription of the E-cadherin gene. Proc. nat. Acad. Sci. (Wash.), 91, 7202–7206 (1994). GUNNING, P., PONTE, P., OKAYAMA, H., ENGEL, J., BLAU, H. and KEDES, L., Isolation and characterization of full-length cDNA clones for human a-, b-, and gamma-actin mRNAs: skeletal but not cytoplasmic actins have an amino-terminal cysteine that is subsequently removed. Mol. cell. Biol., 3, 787–795 (1983). GUSTERSON, B.A. and 20 OTHERS, Prognostic importance of c-erbB-2 expression in breast cancer. J. clin. Oncol., 10, 1049–1056 (1992). HÉLÈNE, C., Control of oncogene expression by antisense nucleic acids. Europ. J. Cancer [A], 30A, 1721–1726 (1994). HYNES, N.E. and STERN, D.F., The biology of erbB-2/neu/HER-2 and its role in cancer. Biochim. biophys. Acta, 1198, 165–184 (1994). KASID, U., PFEIFER, A., BRENNAN, T., BECKETT, M., WEICHSELBAUM, R.R., DRITSCHILO, A. and MARK, G.E., Effect of antisense c-raf-1 on tumorigenicity and radiation sensitivity of a human squamous carcinoma. Science, 243, 1354–1356 (1989). KOOK, Y.H., ADAMSKI, J., ZELENT, A. and OSSOWSKI, L., The effect of antisense inhibition of urokinase receptor in human squamous cell carcinoma on malignancy. EMBO J., 13, 3983–3991 (1994). KRAUS, M.H., POPESCU, N.C., AMSBAUGH, S.C. and KING, C.R., Overexpression of the EGF receptor-related proto-oncogene erbB-2 in human mammary tumor cell lines by different molecular mechanisms. EMBO J., 6, 605–610 (1987). MÉNARD, S., PELLEGRINI, R., TOSI, E., SRINIVAS, U., POMPEN, M., CALDERA, M., CAMPIGLIO, M., MANENTI, G. and COLNAGHI, M.I., Oncogene products as targets for immunotherapy. In A.A. Epenetos (ed.): Monoclonal Antibodies 2: Applications in Clinical Oncology, pp. 227–231, Chapmann and Hall, London (1993). MORONI, M.C., WILLINGHAM, M.C. and BÉGUINOT, L., EGF-R antisense RNA blocks expression of the epidermal growth factor receptor and suppresses the transforming phenotype of a human carcinoma cell line. J. biol. Chem., 267, 2714–2722 (1992). NECKERS, L., WHITESELL, L., ROSOLEN, A. and GESELOWITZ, D.A., Antisense inhibition of oncogene expression. Crit. Rev. Oncogenes, 3, 175–231 (1992). PASLEAU, F., GROOTECLAES, M. and GOL-WINKLER, R., Expression of the c-erbB2 gene in the BT474 human mammary tumor cell line: measurement of c-erbB2 mRNA half-life. Oncogene, 8, 849–854 (1993). PRESS, M.F., PIKE, M.C., CHAZIN, V.R., HUNG, G., UDOVE, J.A., MARKOWICZ, M., DANYLUK, J., GODOLPHIN, W., SLIWKOWSKI, M., AKITA, R., PATERSON, M.C. and SLAMON, D.J., Her-2/neu expression in node-negative breast cancer: direct tissue quantitation by computerized image analysis and association of overexpression with increased risk of recurrent disease. Cancer Res., 53, 4960–4970 (1993). PUPA, M.S., MÉNARD, S., MORELLI, D., POZZI, B., DE PALO, G. and COLNAGHI, M.I., The extracellular domain of the c-erbB-2 oncoprotein is released from tumor cells by proteolytic cleavage. Oncogene, 8, 2917–2923 (1993). STANCOVSKI, I., HURWITZ, E., LEITNER, O., ULLRICH, A., YARDEN, Y. and SELA, M., Mechanistic aspects of the opposing effects of monoclonal antibodies to the ERBB2 receptor on tumor growth. Proc. nat. Acad. Sci. (Wash.), 88, 8691–8695 (1991). TAGLIABUE, E., CENTIS, F., CAMPIGLIO, M., MASTROIANNI, A., MARTIGNONE, S., PELLEGRINI, R., CASALINI, P., LANZI, C., MÉNARD, S. and COLNAGHI, M.I., Selection of monoclonal antibodies which induce internalization and phosphorylation of p185HER2 and growth inhibition of cells with HER2/neu gene amplification. Int. J. Cancer, 47, 933–937 (1991). VAN DER KROL, A.R., MOL, J.N.M. and STUITJE, A.R., Modulation of eukaryotic gene expression by complementary RNA or DNA sequences. Biotechniques, 6, 958–976 (1988). VAUGHN, J.P., IGLEHART, J.D., DEMIRDJI, S., DAVIS, P., BABISS, L.E., CARUTHERS, M.H. and MARKS, J.R., Antisense DNA downregulation of the ERBB2 oncogene measured by a flow cytometric assay. Proc. nat. Acad. Sci. (Wash.), 92, 8338–8342 (1995). WRIGHT, C., NICHOLSON, S., ANGUS, B., CAIRNS, J., SAINSBURY, J.R.C., GULLICK, W.J., HARRIS, A.L. and HORNE, C.H.W., Association of c-erbB-2 oncoprotein expression with lack of response to endocrine therapy in recurrent breast-cancer. J. Pathol., 158, 350 (1989). YU, D., SHI, D., SCANLON, M. and HUNG, M.-C., Reexpression of neu-encoded oncoprotein counteracts the tumor-suppressing but not the metastasis-suppressing function of E1A. Cancer Res., 53, 5784–5790 (1993). YU, D., SUEN, T.-C., YAN, D.-H., CHANG, L.-S. and HUNG, M.-C., Transcriptional repression of the neu protooncogene by the adenovirus 5 E1A gene products. Proc. nat. Acad. Sci. (Wash.), 87, 4499–4503 (1990).