Epitope-Specific Antibody Response to HT-1080 Fibrosarcoma Cells by Mimotope Immunization1

Vol. 6, 3629–3635, September 2000 Clinical Cancer Research 3629

Epitope-specific Antibody Response to HT-1080 Fibrosarcoma Cells by Mimotope Immunization1

Mikhail Popkov, Salwa Sidrac-Ghali, genic in humans severely limits their clinical applications, Valery Alakhov, and Rosemonde Mandeville2 especially when repeated administrations are required. It was therefore suggested that passive immunization using human- Institut National Recherche Scientifique, Institute Armand-Frappier, University of Quebec, Laval, Quebec, H7V 1B7 Canada [M. P., ized or reshaped mouse mAbs could have clinical applica- S. S-G., V. A.]; Supratek Pharma Inc., Laval, Quebec, H7V 1B7 tion, and several of these mAbs are now in clinical trials (2). Canada [V. A.]; Department of Biological Sciences, University of Ideally, antibody humanization should not diminish the Quebec in Montreal, Montreal, Quebec, H3C 3P8 Canada [R. M.]; specificity and affinity of the antibody toward the antigen, and Biophage Inc., 6100 Royalmount Avenue, Montreal, Quebec, H4P 2R2 Canada [R. M.] whereas immunogenicity must be completely eliminated. However, it has become apparent that the accomplishment of both aims is usually time-consuming and costly. Another ABSTRACT approach is the use of synthetic peptides (mimotopes) corre- Mouse monoclonal antibody (mAb) BCD-F9, which sponding to the sequence of the native antigen that can induce recognizes an unknown antigen found on the surface of antibodies in vivo (3–5). Furthermore, protective immune many tumor cells, was used to screen a phage display library reactions were elicited by mimotope immunization for sev- expressing random peptide decamers. The phage that was eral infectious agents (6–8). Therefore, the use of peptide selected encoded the unique sequence GRRPGGWWMR, mimotopes might be a valuable approach in developing vac- representing the peptide capable of binding to the BCD-F9 cines for the induction of a defined antibody response to mAb. The peptide was synthesized and found to specifically cancer antigens. inhibit the binding of mAb to HT-1080 fibrosarcoma cells. Phage display technology is a powerful tool for identi- Alanine mutagenesis of the sequence encoding this peptide fying peptide structures that mimic natural epitopes, includ- indicated that three residues, PXXWW, were critical for its ing both linear and conformational epitopes expressed on a binding to the BCD-F9 mAb. Polyclonal antibodies gener- variety of cell types (9–13). Phage peptide libraries consist of ated by immunization of rabbits with the synthetic peptide filamentous phages displaying random peptides of defined GRRPGGWWMR (anti-mimotope antiserum or AM-F9) length on their surface. The peptides are usually fused to the bound specifically to HT-1080 cells and inhibited the bind- phage minor coat protein pIII (14–15), which is expressed at ing of the BCD-F9 mAb to these cells. Using an experimental low density (3–5 molecules/phage particle), or to the major animal model in which CD-1 nude mice are inoculated i.v. coat protein pVIII (16), which is represented at a higher copy with HT-1080 cells, develop lung metastasis, and die within number. Such libraries have been screened successfully with 30 days, we have shown that AM-F9 could significantly a variety of mAbs, and the peptides selected have been shown prolong the life span of these animals. Our results suggest to mimic linear, assembled, and nonpeptidic epitopes (12, 13, that a peptide mimotope can potentially be used as a novel 17). In all of these methodologies, the selection of mimotopes immunotherapy to induce a beneficial antitumor response. (molecules of the repertoire able to bind to the ligate) does not necessarily require that the original ligand be known. The INTRODUCTION mimotopes have been shown to effectively induce a specific immune response directed against the epitope recognized by Since the development of the hybridoma approach (1), a 3 the mAb used for the affinity selection of phage clones (4, 7), large number of rodent mAbs with specificity for antigens of which has suggested a new way to induce epitope-specific therapeutic interest have been generated and characterized. antibody responses against unknown epitopes (5, 18, 19). However, the fact that rodent antibodies are highly immuno- However, there are limited data on whether the antibodies generated by mimotopes can recognize native antigens on tumor cells, and much less is known about the antitumor Received 4/4/00; revised 6/21/00; accepted 6/21/00. activity of anti-mimotope antibodies. The costs of publication of this article were defrayed in part by the Recent findings in a nude xenograft mouse model have payment of page charges. This article must therefore be hereby marked shown that mAb BCD-F9 (20) administrated i.v. is able to advertisement in accordance with 18 U.S.C. Section 1734 solely to reduce the growth and metastasis of human HT-1080 tumor indicate this fact. cells.4 Because of the ability of BCD-F9 to recognize a wide 1 M. P. was supported by scholarships from National Sciences and Engineering Research Council of Canada and Fonds par la Formation de variety of neoplastic cell lines as opposed to normal tissues (21), Chercheurs et l’aide a`la recherche. 2 To whom requests for reprints should be addressed, at Biophage Inc., 6100 Royalmount Avenue, Montreal, Quebec, Canada H4P 2R2. Phone: (514) 496-1488; Fax: (514) 496-1521; E-mail: Rosemonde.Mandeville@ nrc.ca. 4 M. Popkov, S. Sidrac-Ghali, Y. Lusignan, S. Lemieux, and R. 3 The abbreviations used are: mAb, monoclonal antibody; KLH, keyhole Mandeville. Inhibition of tumor growth and metastasis of human fibro- limpet hemocyanin. sarcoma HT-1080 cells by mAb BCD-F9, submitted for publication.

this mAb could potentially be used for antitumor immunother- To test the ability of each phage to be recognized by the apy. In this study, we used phage-displayed peptide libraries to BCD-F9 mAb, approximately 108 virions of a given mutant identify ligands mimicking an epitope for BCD-F9 presented on were incubated for1hatroom temperature with the biotinylated the cancer cells, and we analyzed the anti-cancer activity of antibody previously immobilized on microtiter plates as de- antiserum generated against the discovered ligand. scribed above. After 10 washes, phages were eluted with 0.1 M glycine/HCl (pH 2.2), neutralized, and titered on strain K91 (15). The percentage of attachment was calculated as the num- MATERIALS AND METHODS ber of eluted phages divided by the input phages ϫ 100. Cell Line and mAb. The human fibrosarcoma cell line Synthetic Peptides. The mimotope sequence GRRPGG- HT-1080 was purchased from the American Type Culture Col- WWMR (designated M-F9) was synthesized as linear free pep- lection (Manassas, VA) and maintained in DMEM supple- tides by standard solid-phase method 9-fluorenylmethhoxycar- mented with 10% fetal bovine serum. The cells were cultured at bonyl chemistry and TFA (24). The purity of the peptides was

37°C in 5% CO2 under a humidified atmosphere. assessed by reverse phase high-pressure liquid chromatography The BCD-F9 mAb was obtained from the fusion of NS-1 and mass spectrometry. The complete peptide GRRPGGWW- myeloma cells with spleen cells from BALB/c mice hyperimmu- MRAASYC contains five additional residues at the COOH nized with the human breast carcinoma cell line BT-20 (20). terminus. Three residues (AAS) represent the linker fusing the BALB/c mice were inoculated i.p. with hybridoma, and the peptide and gIII protein as expressed on the phage, and a Tyr BCD-F9 mAb was purified from ascetic fluid using a protein residue was added for possible radiolabeling, and a Cys residue G-Sepharose column (Pharmacia, Baie-Durfe, Quebec, Canada). was added for coupling to KLH. P-␥2 peptide was synthesized Phage Libraries and Biopanning. BCD-F9 ligands as a NVSKEPGGWWKGDYC sequence corresponding to the were selected from random phage libraries expressing linear native PLC-␥2 sequence, except that the COOH-terminal Cys (pIII-10aa) or circular (pIII-10aa.Cys) decapeptides fused to pIII residue was added for coupling. For immunization, the peptides filamentous bacteriophage fd (22). were conjugated to KLH via the COOH terminus as described The BCD-F9 mAb was biotinylated by incubation of 100 previously (3). ␮g of the antibody with 5 ␮g of NHS-LC-biotin (Pierce, Rock- Rabbit Immunization. New Zealand White female rab- ␮ ␮ ford, IL) in 50 lof0.1M NaHCO3 for2hatroom temperature, bits were given a primary i.m. immunization with 100 gofthe followed by dialysis against PBS (23). For panning, Nunc peptide-KLH solution emulsified 1:1 in Freund’s complete ad- Maxisorb microtiter plates were coated with streptavidin at 20 juvant and subsequently boosted with 150 ␮g of antigen emul- ␮ g/ml in 0.1 M NaHCO3 overnight at 4°C and then blocked with sified 1:1 in Freund’s incomplete adjuvant at biweekly intervals. 350 ␮l of blocking solution (1% powdered milk in PBS) for 1 h The rabbits were boosted three times and bled 5 days after the at room temperature. The biotinylated BCD-F9 mAb was di- last boost at week 7. The serum samples from rabbits were luted to 10 ␮g/ml in blocking solution, and 25 ␮l were added to tested by ELISA as described previously (3). The synthetic each well. The mAb was bound to the plate for2hatroom peptides M-F9 and P-␥2 were used as capture antigens (1 ␮g/ temperature, and the wells were washed six times with PBS. well). All ELISA experiments were performed at least twice in Then, 1010 phages were added in 50 ␮l of 0.1% milk/PBS and triplicates. bound to the mAb for1hatroom temperature. The plates were Cell Binding and Inhibition Assays. For the preparation washed 12 times with PBS to remove nonspecific phages, of cells, 10 ml of PBS were added to HT-1080 cell culture, and whereas bound phages were eluted by treatment with 50 ␮lof the cells were washed, harvested by incubation with PBS con- 0.1 M glycine/HCl buffer (pH 2.2) containing 1 mg/ml BSA. taining 0.5 mM EDTA, and transferred to 15-ml centrifuge Neutralization of the eluate, titration, and amplification on agar tubes. Viable cells were then counted using the trypan blue dye medium were carried out essentially as described previously exclusion technique. The concentration of cells was adjusted to (15). The binding and elution steps were repeated four times. 107 cells/ml, and 100 ␮l were used for each sample. The Viral DNA was sequenced with fUSE 32P primer 5Ј- BCD-F9 mAb (1 ␮g) was then added to the sample tubes and TGAATTTTCTGTATGAGG-3Ј (kindly provided by Dr. incubated on ice for 45 min, washed twice with 2 ml of BSA/ George Smith, University of Missouri, Columbia, MO) by using PBS, and centrifuged for 5 min at 1500 rpm. A total of 10 ␮lof the Sequenase T7 kit (Pharmacia) as recommended by the antibody [goat antimouse IgG FITC-labeled antibody (Roche supplier. Diagnostics, Laval, Quebec, Canada) diluted 1:25 with PBS] Alanine Substitutions and Phage Attachment Assay. was added to the sample, incubated on ice for 45 min, and Mutants of the phage selected for its binding to the BCD-F9 washed twice with BSA/PBS. The cells were fixed using 0.5 ml mAb were produced as described previously (18). Briefly, a of 1% paraformaldehyde in PBS, and the percentage of cells series of complementary oligonucleotides in which a given binding the antibody was analyzed on a Epics XL-MCL flow non-Ala residue of DNA encoding the peptide was changed to cytometer (Coulter, Hialeah, FL) equipped with a 488 nm argon encode Ala were used in the construction of mutant phages. laser. Complementary oligonucleotides ␥2for (5Ј-TGGCTTCTAAA- For phage inhibition assay, 0.5 ␮g of the BCD-F9 mAb GAGCCGGGGGGGTGGTGGAAGGGGGCGGCCTCTG-3Ј) was preincubated for2hatroom temperature with 100 ␮lof and ␥2rev (5Ј-AGGCCGCCCCCTTCCACCACCCCCCCGGC- BSA/PBS containing various concentrations of phage particles TCTTTAGAAGCCACGT-3Ј) were used in the construction F9, ␥2, and W7 expressing the VCDWWGWGIC peptide. Ali- of phage ␥2. All mutants were purified and verified by DNA quots were then added to HT-1080 cells (106 cells/sample), and sequencing as described above. the cells were treated as described above.

Fig. 1 Selection of phages that bind to the BCD-F9 mAb. Phages from Fig. 2 Binding of Ala phage mutants and phage ␥2 to the BCD-F9 linear (F) or circular (E) decapeptide libraries were bound to mAb- mAb. The effect of Ala substitutions of three residues in the peptide coated microtiter wells as described in “Materials and Methods” and GRRPGGWWMR (Pro in the fourth position from the NH2 terminus, Trp in the seventh position, and Trp in the eighth position) is shown. F9 eluted with glycine/HCl buffer (pH 2.2). Enrichment (ni/n1) was calcu- phage served as a positive control for binding to the BCD-F9 mAb. The lated as the total number of phages recovered after elution (ni, number of transducing units recovered after i round of selection) divided by the percentage of attachment is the number of eluted phages divided by number of transducing units recovered after the first round of selection the input phage ϫ 100. Data represent mean values from plating in triplicate. (n1). Data represent mean values from plating in triplicate.

␮ For peptide inhibition assay, 0.5 g of the BCD-F9 mAb protein pIII, whereas the second has additional two cysteines ␮ was preincubated for2hatroom temperature with 100 lof flanking the random decapeptide sequence, leading to formation BSA/PBS containing various concentrations of the synthetic of circular peptides through the creation of a disulfide bridge. ␥ peptides M-F9, P- 2, and VCDWWGWGIC. Aliquots were After four rounds of biopanning using the BCD-F9 mAb as a 6 then added to HT-1080 cells (10 cells/sample), and the cells target, we observed an approximately 33,000-fold increase in were treated as described above. the number of eluted phages in the case of the linear library (Fig. 6 For antisera inhibition assay, HT-1080 cells (10 cells/ 1). The circular library showed a marginal enrichment of about sample) were preincubated for 45 min on ice with different 100-fold after the third round of selection. Therefore, no further ␥ dilutions of antisera: (a) AM-F9; (b) AP- 2; or (c) preimmune manipulations were continued with this library. ␮ rabbit serum. One g of the BCD-F9 mAb was added to the After the fourth round of selection with the linear library, sample tubes, and treatment proceeded as described above. 16 independent phage isolates were amplified and sequenced. Experimental Metastasis. Confluent monolayers of HT- All isolates yielded the identical DNA sequence encoding the 1080 cells were harvested by incubation with PBS containing amino acid sequence GRRPGGWWMR. Eight additional inde- 0.5 mM EDTA. Viable cells were counted by the trypan blue pendent isolates were chosen from the second round and se- Ͼ exclusion dye method. Aliquots containing 95% viable cells quenced. All of the phages from the second round encoded the ϫ 7 were used in this experiment. The cells (2 10 cells/ml) were same sequence obtained from the fourth-round isolates. The suspended in PBS, and 0.1 ml of the suspension was injected i.v. comparison of the identified peptide sequence with the protein into the tail vein of CD-1 nude mice (day 0). Rabbit preimmune sequences available in the GenBank showed the homology of ␮ ␮ serum (100 l) or polyclonal antisera AM-F9 (100 l) was five amino acid residues (PGGWW) to six SH3 domain- administered i.v. in 0.2 ml of PBS on days 1–2, 5–9, 12–16, and containing proteins (National Center for Biotechnology Infor- 19–21, and the lungs were recovered from recently deceased mation accession numbers AAD29953, AAC78610, Q15811, animals. The lungs were fixed in 10% buffered formalin, em- AAD29952, NP_003015, and NP_002652). These proteins are bedded in paraffin, sectioned, and stained with H&E for routine all intracellular signal transducers; therefore, it is unlikely that histological examination by light microscopy. they contain the BCD-F9-reactive epitope. Statistical Analysis. The Cox-Mantel log-rank test was Determination of Critical Amino Acids Necessary for Ͻ performed using INSTAT software. P 0.05 was considered Binding to the BCD-F9 mAb. To determine which residues significant. were critical for binding, a series of mutant phages were constructed in which residues inside the PGGWW block were RESULTS individually changed to Ala. Phage ␥2 expressing the sequence Identification of BCD-F9-specific Mimotopes. To search corresponding to that of PLC-␥2, one of the six proteins iden- for a peptide that is able to bind with the BCD-F9 mAb, we tified in the GenBank (National Center for Biotechnology In- screened two phage peptide libraries. One of these libraries formation accession number NP_002652), was also constructed

displays linear decapeptides that are fused to the minor coat to determine the effect of the remaining residues NH2-or

Fig. 4 Inhibition of binding of the BCD-F9 mAb to HT-1080 cells by polyclonal antisera. The binding of BCD-F9 to HT-1080 cells was studied in the presence of the competing polyclonal antisera AM-F9 (f), AP-␥2(`), and pre-immune serum (Ⅺ) as described in “Materials and Methods.” The Y axis shows inhibition (percentage) of binding of BCD-F9 to HT-1080 cells as a function of the rabbit polyclonal antiserum dilution (X axis). The data represent the means Ϯ SD from three independent experiments.

were used as controls mainly because they show the highest and the lowest binding to the BCD-F9 mAb, respectively (Fig. 3). BCD-F9 was incubated with the serial dilutions of phages and then assayed for its ability to bind to HT-1080 cells. The results shown in Fig. 3A demonstrate that phage F9 induced a 47% inhibition at a concentration of 1013 virions/ml, whereas phage ␥2 was less potent, inducing a 38% inhibition at the same virion Fig. 3 Inhibition of BCD-F9 binding to HT-1080 cells by phages and concentration. The phage W7 did not show any specific binding by synthetic peptides. Binding of BCD-F9 to HT-1080 cells was studied to the BCD-F9 mAb. in the presence of the competing phages (A)F9(F), ␥2(f), and W7 (E) A peptide corresponding to the sequence presented on F ␥ f or in the presence of the competing peptides (B) M-F9 ( ), P- 2( ), phage F9 (M-F9 peptide) was then synthesized and tested for its and VCDWWGWGIC irrelevant peptide (E) as described in “Materials and Methods.” The Y axis shows inhibition (percentage) of binding of ability to bind to the BCD-F9 mAb. We also synthesized a BCD-F9 to HT-1080 cells as a function of the phage (A) or of the free peptide corresponding to phage ␥2 (P-␥2 peptide) and an unre- peptide (B) concentration (X axis). The data represent the means Ϯ SD lated (VCDWWGWGIC) peptide. These three peptides were from three independent experiments. tested for their ability to bind to the BCD-F9 mAb. Briefly, BCD-F9 was incubated with the peptides at various concentrations of the synthetic peptides and then assayed for its ability to bind to HT-1080 cells. Fig. 3B shows that M-F9 bound specif- COOH-terminal to this block. Each of the resulting phages was ically to the antibody and blocked its subsequent binding to then tested for its ability to bind to the BCD-F9 mAb. Results HT-1080 cells. The highest potency of the M-F9 peptide was ϭ ϫ Ϫ6 ␥ shown in Fig. 2 indicate that mutations of three residues, Pro IC50 2 10 M, as compared with the P- 2 peptide that (fourth position), Trp (seventh position), and Trp (eighth posi- showed an inhibitory effect only at concentrations higher than tion), led to a significant reduction in binding of the peptide to 100 ␮M. The peptide VCDWWGWGIC had no inhibitory effect the BCD-F9 mAb. The remaining residues had no effect on the (Fig. 3B). peptide binding when changed all at once in phage ␥2. We did Induction of an Epitope-specific Immune Response. not change two Gly residues inside the PGGWW block mainly To investigate whether the mimotope represents an immuno- because it has been shown previously that Gly residues do not genic structure corresponding to the natural epitope expressed usually contribute directly to the binding but rather serve as on cancer cells, rabbits were immunized with either the M-F9 structural linkers to position the critical residues (18). peptide or the P-␥2 peptide conjugated with KLH. After a Binding of Phage F9 and Mimotope (M-F9) to the primary immunization and three boosts, the sera from individual BCD-F9 mAb. To confirm the specificity of the peptide rep- rabbits were collected and tested for anti-peptide antibodies resented on phage F9, we analyzed its ability to inhibit binding using an ELISA technique. All of the sera collected showed an of the BCD-F9 mAb to HT-1080 cells i.e., cells that express a ability to bind to peptides at 1:2000 dilution, whereas the pre- high level of the BCD-F9-specific antigen. Phages ␥2 and W7 immune serum showed no anti-peptide reactivity. We then in-

if not all, anti-carbohydrate mAbs (28–30). Peptide epitopes have also been shown to mimic glycosphingolipids and oligonucleotide structures (30–32). In the case of mimotope immunization, several investigators showed that synthetic peptides could induce an epitope-specific immune response in vivo (3–5). Furthermore, protective immune reactions by mimotope immu- nizations have been shown for several infectious agents (6–8), viral antigens (33), and pollen allergens (34). The larger issue of whether such mimotopes can be exploited to elicit functional antibody response against tumor cells has yet to be fully re- solved. In this study, we chose a well-characterized functional mAb, BCD-F9, and used it to identify peptides that mimic the epitope recognized by this antibody. The BCD-F9 mAb is a murine mAb (IgG2a) that recognizes an antigen present on the surface of many tumor cell lines. Because of its high selectivity Fig. 5 Survival analysis of HT-1080 tumor-bearing nude mice. HT- toward tumor cells as opposed to normal tissues, this mAb could 1080 cells were inoculated i.v. into CD-1 nude mice on day 0. Rabbit potentially be used for killing tumor cells (35). Whereas the preimmune serum (E) or AM-F9 (F) was administered i.v. on days 1–2, 5–9, 12–16, and 19–21. Each group comprised five mice. Statistical tumor antigen that is recognized by BCD-F9 has not been significance (P Ͻ 0.05 versus pre-immune serum) was calculated using isolated, our unpublished data demonstrate that this mAb rec- 4 the Cox-Mantel log-rank test. ognizes a conformational epitope on Mr 57,000 glycoprotein. Although we expected to identify several mimotopes, only one sequence was selected from a linear random decapeptide library, whereas a circular library did not generate any specific vestigated the specificity of the rabbit anti-mimotope antisera by phages. Despite the great number of independent clones in these inhibiting the binding of the BCD-F9 mAb to HT-1080 cells. As libraries (Ͼ109), they cover only a small percentage of all shown in Fig. 4, serum from the rabbit immunized with M-F9 theoretically possible decapeptides. Therefore, it is possible that peptide (AM-F9) inhibited the binding of BCD-F9 to HT-1080 the circular library does not contain peptides mimicking the cells by 75%, suggesting that the rabbit anti-mimotope anti- most optimal structure of the native epitope. It is also possible serum was directed against the epitope of these tumor cells. In that there is a bias against particular sequences because of the ␥ contrast, the serum from the rabbit immunized with the P- 2 need to maintain phage infectivity, resulting in the absence of peptide, as well as rabbit pre-immune serum, did not inhibit the those sequences from the library. binding of the BCD-F9 mAb to HT-1080 cells. Mutational analysis indicated that specificity is likely to Effect of AM-F9 on Experimental Lung Metastasis. reside in the PGGWW block of the amino acids (Fig. 2). This Finally, we examined the antitumor activity of AM-F9 in an block is common to several SH3 domain-containing regulatory experimental metastasis model. In this model, HT-1080 cells proteins involved in signal transduction (36). The intracellular injected i.v. into CD-1 nude mice induce the formation of localization of these signaling proteins suggests that the F9 metastatic foci in the lung, and 100% of the animals die within peptide mimics the native tumor-specific epitope despite the fact 30 days after inoculation. After 50 days of observation, the that they have different amino acid sequences. Binding studies mean survival time values were calculated for groups of mice and immunization experiments confirmed the specificity of the injected with AM-F9 and mice treated with pre-immune serum. mimotope for the BCD-F9 mAb. In addition, the mimotope was Ϯ Ϯ These values were 47 2 and 25 1, respectively (Fig. 5). The specifically recognized by BCD-F9 without the phage carrier, animals treated with AM-F9 had significantly prolonged life indicating that the mimotope alone is responsible for the inter- ϭ span (T/C 188%), suggesting that the antitumor activity of action with BCD-F9, without the involvement of structure en- AM-F9 was not dependent on the nonspecific action of the tities from the phage particle. Another indication for the correct rabbit serum. Histological examination of the lungs showed mimicking of the epitope by F9 peptide was obtained by the diffuse metastasis in the case of control animals and more immunization experiments. We demonstrated that rabbits im- localized metastatic foci in the case of animals treated with munized with peptide-KLH could elicit an IgG response to the anti-mimotope antisera. In-life observations showed that mice natural antigen on tumor cells. treated with AM-F9 appeared as healthy as the control mice On the other hand, results obtained with the peptide ␥2 during the duration of the experiment, suggesting that no severe were surprising, and we cannot reasonably explain their poor side effects were caused by AM-F9. binding. However, we cannot exclude the possibility that the amino acid residues flanking the PGGWW block interfere with DISCUSSION binding when the peptide is isolated from the phage carrier. A Several authors have demonstrated previously that peptide similar situation has been reported previously in which the mimotopes can be obtained from phage epitope libraries (9–11) peptides were able to mimic the natural epitope only when or by chemical synthesis (25–27) and that these sequences bind presented on the phage surface (19, 34). antibodies raised against native structures. In fact, several stud- In summary, this report describes the isolation of a mimo- ies have shown that peptide mimotopes can be found for most, tope corresponding to the epitope of the BCD-F9 mAb. Inhibi-

tion studies with either mimotope-displaying phage particles or 6. Motti, C., Nuzzo, M., Meola, A., Galfre`, G., Felici, F., Cortese, R., synthetic peptides proved the specificity of the isolated mimo- Nicosia, A., and Monaci, P. Recognition by human sera and immuno- tope. Immunization of rabbits with the mimotope induced poly- genicity of HbsAg mimotopes selected from an M13 phage display library. Gene (Amst.), 146: 191–198, 1994. clonal antibodies capable of blocking the binding of the 7. Stoute, J. A., Ballou, W. R., Kolodny, N., Deal, C. D., Wirtz, R. A., BCD-F9 mAb to HT-1080 tumor cells. Finally, we assessed the and Lindler, L. E. Induction of humoral immune response against effect of AM-F9 serum on tumor metastasis in an experimental Plasmodium falciparum sporozoites by immunization with a synthetic animal model and showed that i.v. treatment with anti-mimo- peptide mimotope whose sequence was derived from screening a fila- tope antisera (15 injections of 100 ␮l/mouse) significantly pro- mentous phage epitope library. Infect. Immun., 63: 934–939, 1995. longed the life span of nude mice (Fig. 5). However, the bene- 8. Prezzi, C., Nuzzo, A., Meola, A., Delmastro, P., Galfre`, G., Cortese, ficial effects of this passive immunotherapy are only achieved if R., Nicosia, A., and Monaci, P. Selection of antigenic and immunogenic relatively large amounts of antibody are applied, making this mimics of hepatitis C virus using sera from patients. J. Immunol., 156: 4504–4513, 1996. treatment very expensive. Thus, it would be an advantage to 9. Scott, J. K., and Smith, G. P. Searching for peptide ligands with an replace or combine the passive treatment with an active immu- epitope library. Science (Washington DC), 249: 386–390, 1990. nization using a mimotope. The success of such an approach has 10. Devlin, J. J., Panganiban, L. C., and Devlin, P. E. Random peptide already been demonstrated with the use of mimotopes as exper- libraries: a source of specific protein binding molecules. Science (Wash- imental oral anti-IgE vaccines (37). ington DC), 249: 404–406, 1990. The properties of AM-F9 were further investigated in a 11. Cwirla, S. E., Peters, E. A., Barrett, R. W., and Dower, W. J. separate study.5 It was found that (a) treatment with AM-F9 Peptides on phage: a vast library of peptide for identifying ligands. Proc. significantly inhibited the growth of HT-1080 tumor cells in- Natl. Acad. Sci. USA, 87: 6378–6382, 1990. jected s.c. into CD-1 nude mice, (b) AM-F9 elicited antibody- 12. Luzzago, A., Felici, F., Tramontano, A., Pessi, A., and Cortese, R. dependent cellular cytotoxicity by splenic natural killer cells and Mimicking of discontinuous epitopes by phage-displayed peptides. I. Epitope mapping of human H ferritin using a phage library of con- by peritoneal cells, and (c) AM-F9 mediated complement- strained peptides. Gene (Amst.), 128: 51–57, 1993. dependent cytotoxicity. 13. Felici, F., Luzzago, A., Folgori, A., and Cortese, R. Mimicking of The results of this study suggest that phage epitope librar- discontinuous epitopes by phage-displayed peptides. II. Selection of ies may have broad application for the design of anti-cancer clones recognized by a protective monoclonal antibody against the vaccines. We believe that this strategy may allow the design of Boretella pertussis toxin from phage peptide libraries. Gene (Amst.), immunogenic peptides without prior knowledge of the target 128: 21–27, 1993. antigen. A new generation of anti-cancer vaccines could consist 14. Smith, G. P. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science (Washington of a mixture of mimotopes, including peptides that mimic car- DC), 228: 1315–1316, 1985. bohydrates and complex epitopes, selected using tumor-specific 15. Parmley, S. F., and Smith, G. P. Antibody-selectable filamentous fd mAbs. phage vector: affinity purification of target genes. Gene (Amst.), 73: 305–318, 1988. ACKNOWLEDGMENTS 16. Felici, F., Castagnoli, L., Musacchio, A., Jappelli, R., and Cesareni, We thank Dr. G. P. Smith for providing the fUSE 32P primer, Dr. G. J. Selection of antibody ligands from a large library of oligopeptides expressed on a multivalent exposition vector. Mol. Biol., 222: 301–310, F. Shareck for the oligonucleotide synthesis, Dr. J. Hu for peptide 1991. synthesis, and L. Labrie and L. Forget for DNA sequence analysis. 17. Balass, M., Heldman, Y., Cabilly, S., Givol, D., Katchalski Katzir, E., and Fuchs, S. Identification of a hexapeptide that mimics a confor- REFERENCES mation-dependent binding site of acetylcholine receptor by use of a 1. Kohler, G., and Milstein, C. Continuous cultures of fused cells phage-epitope library. Proc. Natl. Acad. Sci. USA, 90: 10638–10642, secreting antibody of predefined specificity. Nature (Lond.), 256: 495– 1993. 497, 1975. 18. Hoess, R., Brinkmann, U., Handel, T., and Pastan, I. Identification 2. Hollinger, P., and Hoogenboom, H. Antibodies come back from the of a peptide which binds to the carbohydrate-specific monoclonal anti- brink. Nat. Biotechnol., 16: 1015–1016, 1998. body B3. Gene (Amst.), 128: 43–49, 1993. 3. Willis, A. E., Perham, R. N., and Wraith, D. Immunological prop- 19. Chargelegue, D., Obeid, O. E., Hsu, S-C., Shaw, M., Denbury, erties of foreign peptides in multiple display on a filamentous bacteri- A. N., Taylor, G., and Steward, M. W. A peptide mimic of a protective ophage. Gene (Amst.), 128: 79–83, 1993. epitope of respiratory syncytial virus selected from a combinatorial library induces virus-neutralizing antibodies and reduces viral load in 4. Meola, A., Delmastro, P., Monaci, P., Luzzago, A., Nicosia, A., vivo. J. Virol., 72: 2040–2046, 1998. Felici, F., Cortese, R., and Galfre`, G. Derivation of vaccines from mimotopes. Immunologic properties of human hepatitis B virus surface 20. Mandeville, R., Lecomte, J., Dumas, F., Giroux, L., Chausseau, antigen mimotopes displayed on filamentous phage. J. Immunol., 154: J-P., Ajdukovic, I., and Boury, F. Production and characterization of 3162–3172, 1995. monoclonal antibodies showing a different spectrum of reactivity to human breast tissue. Cancer Detect. Prev., 10: 89–100, 1987. 5. Rudolf, M., Vogel, M., Kricek, F., Ruf, C., Zurcher, A., Reuschel, R., Auer, M., Miescher, S., and Stadler, B. M. Epitope-specific antibody 21. Zelechowska, M. G., and Mandeville, R. Immunogold and immu- responce to IgE by mimotope immunization. J. Immunol., 160: 3315– nogold/silver staining in the ultrastructural localization of target mole- 3321, 1998. cules identified by monoclonal antibodies. Anticancer Res., 9: 53–58, 1989. 22. Popkov, M., Lussier, I., Medvedkine, V., Este`ve, P-O., Alakhov, V., and Mandeville, R. Multidrug-resistance drug-binding peptides generated by using a phage display library. Eur. J. Biochem., 251: 155–163, 5 M. Popkov, Y. Lusignan, S. Lemieux, and R. Mandeville. Immuno- 1998. therapy with antimimotope polyclonal antibodies in a nude mouse 23. Gretch, D. R., Suter, M., and Stinski, M. F. The use of biotinylated xenograft model: mechanisms of action, submitted for publication. monoclonal antibodies and streptavidin affinity chromatography to iso-

late herpesvirus hydrophobic proteins or glycoproteins. Anal. Biochem., 31. Taki, T., Ishikawa, D., Hamasaki, H., and Handa, S. Preparation of 163: 270–277, 1987. peptides which mimic glycosphingolipids by using phage peptide library 24. Atherton, E., Clive, D. L., and Sheppard, R. C. Letter. Polyamide and their modulation on ␤-galactosidase activity. FEBS Lett., 418: supports for polypeptide synthesis. J. Am. Chem. Soc., 97: 6584–6585, 219–223, 1997. 1975. 32. Sibille, P., Ternynck, T., Nato, F., Buttin, G., Strosberg, D., and 25. Geysen, H. M., Rodda, S. J., and Mason, T. J. A priori delineation Avrameas, A. Mimotopes of polyreactive anti-DNA antibodies identi- of a peptide which mimics a discontinuous antigenic determinant. Mol. fied using phage-display peptide libraries. Eur. J. Immunol., 27: 1221– Immunol., 23: 709–715, 1986. 1228, 1997. 26. Houghten, R. A., Pinilla, C., Blondelle, S. E., Appel, J. R., Dooley, 33. D’Mello, F., Partidos, C. D., Steward, M. W., and Howard, C. R. C. T., and Cuervo, H. Generation and use of synthetic peptide combi- Definition of the primary structure of hepatitis B virus (HBV) pre-S natorial libraries for basic research and drug discovery. Nature (Lond.), hepatocyte binding domain using random peptide libraries. Virology, 354: 84–86, 1991. 237: 319–326, 1997. 27. Lam, K. S., Salmon, S. E., Hersh, E. M., Hruby, V. J., Kazmierski, 34. Jensen-Jarolim, E., Leitner, A., Kalchhauser, H., Zurcher, A., Gan- W. M., and Knapp, R. J. A new type of synthetic peptide library for glberger, E., Bohle, B., Scheiner, O., Boltz-Nitulescu, G., and Bre- identifying ligand-binding activity. Nature (Lond.), 354: 82–84, 1991. iteneder, H. Peptide mimotopes displayed by phage inhibit antibody 28. Bonnycastle, L. L., Mehroke, J. S., Rashed, M., Gong, X., and binding to Bet v 1, the major birch pollen allergen, and induce specific Scott, J. K. Probing the basis of antibody reactivity with a panel of IgG response in mice. FASEB J., 12: 1635–1642, 1998. constrained peptide libraries displayed by filamentous phage. J. Mol. Biol., 258: 747–762, 1996. 35. Mandeville, R., Schatten, C., Pateisky, N., Dicaire, M-J., Barbeau, B., and Grouix, B. Immunolymphoscintigraphy with BCD-F9 mono- 29. Harris, S. L., Craig, L., Mehroke, J. S., Rashed, M., Zwick, M. B., Ј Kenar, K., Toone, E. J., Greenspan, N., Auzanneau, F. I., Marino- clonal antibody and its F(ab )2 fragments for the preoperative staging of Albernas, J. R., Pinto, B. M., and Scott, J. K. Exploring the basis of breast cancers. In: R. L. Ceriani (ed.), Breast Cancer Immunodiagnosis peptide-carbohydrate cross-reactivity: evidence for discrimination by and Immunotherapy, pp. 203–210. New York: Plenum Press, 1989. peptides between closely related anti-carbohydrate antibodies. Proc. 36. Cohen, G. B., Ren, R., and Baltimore, D. Modular binding domains Natl. Acad. Sci. USA, 94: 2454–2459, 1997. in signal transduction proteins. Cell, 80: 237–248, 1995. 30. Moe, G. R., Tan, S., and Granoff, D. M. Molecular mimetics of 37. Zuercher, A. W., Miescher, S. M., Vogel, M., Rudolf, M. P., polysaccharide epitopes as vaccine candidates for prevention of Neis- Stadler, M. B., and Stadler B. M. Oral anti-IgE immunization with seria meningitidis serogroup B disease. FEMS Immunol. Med. Micro- epitope-displaying phage. Eur. J. Immunol., 30: 128–135, 2000. biol., 26: 209–226, 1999.

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2000 American Association for Cancer Research. Epitope-specific Antibody Response to HT-1080 Fibrosarcoma Cells by Mimotope Immunization

Mikhail Popkov, Salwa Sidrac-Ghali, Valery Alakhov, et al.

Clin Cancer Res 2000;6:3629-3635.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/6/9/3629

Cited articles This article cites 33 articles, 10 of which you can access for free at: http://clincancerres.aacrjournals.org/content/6/9/3629.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/6/9/3629.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/6/9/3629. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.