Differential Immunogenicity of Two Peptides Isolated by High Molecular Weight-Melanoma-Associated Antigen-Specific Monoclonal Antibodies wit
http://www.100md.com
免疫学杂志 2005年第11期
Abstract
Peptide mimics isolated from phage display peptide libraries by panning with self-tumor-associated Ag (TAA)-specific mAbs are being evaluated as immunogens to implement active specific immunotherapy. Although TAA-specific mAb are commonly used to isolate peptide mimics, no information is available regarding the Ab characteristics required to isolate immunogenic TAA peptide mimics. To address this question, we have used mAb 763.74 and mAb GH786, which recognize the same or spatially close antigenic determinant(s) of the human high m.w.-melanoma-associated Ag (HMW-MAA), although with different affinity. mAb 763.74 affinity is higher than that of mAb GH786. Panning of phage display peptide libraries with mAb 763.74 and mAb GH786 resulted in the isolation of peptides P763.74 and PGH786, respectively. When compared for their ability to induce HMW-MAA-specific immune responses in BALB/c mice, HMW-MAA-specific Ab titers were significantly higher in mice immunized with P763.74 than in those immunized with PGH786. The HMW-MAA-specific Ab titers were markedly increased by a booster with HMW-MAA-bearing melanoma cells, an effect that was significantly higher in mice primed with P763.74 than in those primed with PGH786. Lastly, P763.74, but not PGH786, induced a delayed-type hypersensitivity response to HMW-MAA-bearing melanoma cells. These findings suggest that affinity for TAA is a variable to take into account when selecting mAb to isolate peptide mimics from a phage display peptide library.
Introduction
The identification and molecular characterization of human tumor-associated Ags (TAA) 3 have provided well-defined moieties to implement active specific immunotherapy of malignant diseases (1, 2). At variance with their murine counterparts, most human TAA are shared, nonmutated self-Ags that are weakly or nonimmunogenic in patients with malignant diseases (3, 4). Therefore, one of the challenges facing tumor immunologists is the development of strategies to induce a strong immune response to these self-TAAs, because in animal model systems an association has been found between the magnitude of the elicited TAA-specific immune responses and therapeutic efficacy of active specific immunotherapy (5, 6).
Among the strategies used to overcome unresponsiveness to a self-TAA, one relies on the use of TAA mimics as immunogens (7). The validity of this strategy has been demonstrated by the ability of anti-idiotypic (anti-id) mAb, which mimic TAA, to induce TAA-specific immune responses in patients with malignant diseases when the nominal Ag is not immunogenic (8, 9, 10, 11, 12, 13). In a phase I clinical trial, we have demonstrated that administration of the mouse anti-id mAb MK2–23, which mimics the determinant defined by the mouse anti-high m.w.-melanoma-associated Ag (HMW-MAA) mAb 763.74, induced HMW-MAA-specific Abs in 60% of the immunized melanoma patients (10). The survival of patients who had developed anti-HMW-MAA Abs was significantly longer than that of patients who had not developed this immune response. These clinical findings have prompted us to continue to optimize the immunization strategy with HMW-MAA mimics. In the present study we have investigated the potential of HMW-MAA peptide mimics as an alternative immunogen to anti-id mAb, because peptide mimics have several advantages over anti-id Abs. The relative ease of modifying peptide mimics facilitates the application of strategies to enhance their ability to induce nominal TAA-specific immune responses. Furthermore, peptide mimics can be easily synthesized and standardized, thus expediting the steps required to obtain approval from regulatory agencies for their use in clinical trials.
In the present study we have chosen to pan phage display peptide libraries with anti-HMW-MAA mAb to isolate peptide mimics. This strategy has already been successfully used to isolate peptide mimics of human TAA, including the Lewis Y (LeY) and sialyl Lewis X Ags as well as the oncogenic protein Her-2/neu (14, 15, 16). To date, no information is available about the characteristics of Abs that are required to isolate TAA peptide mimics that are effective in inducing immune responses to the nominal TAA. Therefore, in the present study we have investigated the relationship between the immunogenic properties of HMW-MAA peptide mimics and the affinity of the HMW-MAA-specific mAb used to isolate these peptides from phage display peptide libraries. To address this question, we have used the HMW-MAA-specific mAb 763.74 and anti-anti-id mAb GH786. These mAb recognize the same or spatially close antigenic determinant(s) of HMW-MAA, because mAb 763.74 and mAb GH786 cross-block the binding of each other to HMW-MAA-bearing melanoma cells. However, mAb 763.74 exhibits a markedly higher affinity for HMW-MAA than mAb GH786 (17). Our previous studies have shown that HMW-MAA-binding Abs, elicited by anti-id mAb MK2–23 in mice, rabbits, and melanoma patients, exhibit only a low level of reactivity with melanoma cells (10, 18, 19). This finding does not reflect the poor immunogenicity of mAb MK2–23, because the immunized animals or patients developed high titer anti-anti-id Abs reacting with idiotopes in the Ag-combining site of mAb MK2–23. Because a positive association has been demonstrated between the development of HMW-MAA-binding Abs and clinical outcome (10), we have also investigated in the present study whether HMW-MAA-specific immune responses elicited by the isolated peptide mimics could be enhanced by a booster with HMW-MAA-bearing melanoma cells. This strategy is based on the knowledge that B cell clones undergo somatic hypermutations during the course of an immune response. It is our working hypothesis that after peptide mimic immunization, a booster with HMW-MAA-bearing melanoma cells may selectively expand the mutated B cell clones that secrete Abs with enhanced affinities for HMW-MAA.
Materials and Methods
Statistical analysis
The statistical significance of differences in the experimental results was analyzed using two-tailed, unpaired, Student’s t test. A difference was considered statistically significant at p 0.05.
Results
Isolation of peptides from phage display peptide libraries LX-8 and X15 by panning with anti-HMW-MAA mAb 763.74 and mAb GH786
Panning of LX-8 and X15 libraries with mAb 763.74 resulted in the isolation of eight phage clones from the LX-8 library, but no clones from the X15 library. All of these clones reacted specifically with mAb 763.74. Sequencing of DNA encoding the peptides displayed by these clones identified the peptide sequence QCTGPNVATNCR (P763.74) in six of eight clones, and QCTGPNFATNCR and TCNGPSVYMNCL in each of the remaining two clones. P763.74 displayed the highest binding reactivity with mAb 763.74 and was chosen for subsequent experiments (data not shown). Panning of LX-8 and X15 libraries with mAb GH786 resulted in the isolation of four and six phage clones from the LX-8 and X15 libraries, respectively. Sequencing of DNA encoding the peptides displayed by these clones identified the peptide sequence TCRLPFQNVACH in two clones and SCLLPFQNIFCS in the other two clones isolated from the LX-8 library. The sequence NQLPQYMGPAPAYMR (PGH786) was identified in all six clones isolated from the X15 library. PGH786 displayed the highest binding reactivity to mAb GH786 and was chosen for subsequent experiments (data not shown).
Reactivity of synthetic peptides P763.74 and PGH786 with mAb 763.74 and mAb GH786
Peptides P763.74 and PGH786 were compared for their specific binding reactivity to their corresponding panning mAb and for their ability to inhibit the binding of these mAb to HMW-MAA. Each peptide reacted with its corresponding panning mAb in a dose-dependent manner (Fig. 1, A and B). Each peptide also specifically inhibited the binding of its corresponding panning mAb to HMW-MAA-bearing melanoma cells in a dose-dependent manner, but did not cross-inhibit the binding of the other mAb to HMW-MAA-bearing melanoma cells (Fig. 1, C and D). These results indicate that peptides P763.74 and PGH786 react specifically with the Ag-combining sites of mAb 763.74 and mAb GH786, respectively.
Immunogenicity of HMW-MAA peptide mimics P763.74 and PGH786 in BALB/c mice
Mice immunized with peptide mimic P763.74 or PGH786 were tested for the development of Abs reacting with the immunizing peptide and with HMW-MAA-bearing melanoma cells in binding assays. The kinetics of the development of Ab populations are shown in Fig. 2. Analysis of sera harvested from mice immunized with peptide P763.74 or PGH786 demonstrated increasing titers of peptide-reactive Abs. The anti-PGH786 Ab titer obtained from PGH786-immunized mice was significantly (p < 0.05) higher than the anti-P763.74 Ab titer obtained from P763.74-immunized mice (Fig. 2A). Reactivity with HMW-MAA-bearing Colo38 melanoma cells, but not with HMW-MAA-negative LG2 B lymphoid cells, was detected in sera from six of the eight mice immunized with peptide P763.74 and in four of the eight mice immunized with peptide PGH786. In contrast to the peptide-reactive Ab titers, the mean titers of HMW-MAA-bearing melanoma cell-reacting Abs were significantly (p < 0.05) higher in sera obtained from P763.74-immunized mice than in sera obtained from PGH786-immunized mice (Fig. 2C). The selective reactivity of these sera with HMW-MAA-bearing melanoma cells was confirmed by their reactivity with M14 clone 5/HMW-MAA cells that express HMW-MAA (Fig. 3), but not with the parental M14 clone 5 cells or empty plasmid-transfected M14 clone 5 cells. The latter cells do not express HMW-MAA. Sera from mice immunized with the control MB1194–208 peptide displayed strong reactivity with MB1194–208 peptide, but no reactivity with HMW-MAA-bearing Colo38 cells. These sera also did not stain M14 clone 5/HMW-MAA cells (Fig. 3). Despite their reactivity with HMW-MAA-bearing melanoma cells in binding assays, sera from P763.74- or PGH786-immunized mice were not able to immunoprecipitate any components from radiolabeled HMW-MAA-bearing Colo38 melanoma cell lysates (data not shown).
Enhancement of immune responses elicited with HMW-MAA peptide mimics by a booster with HMW-MAA-bearing melanoma cells
To investigate whether HMW-MAA-specific reactivity induced by peptides P763.74 and PGH786 can be augmented by a booster with the original Ag, mice immunized with these peptide mimics were boosted s.c. with HMW-MAA-bearing Colo38 melanoma cells (5 x 105 cells/mouse) on day 132. HMW-MAA-bearing melanoma cells were used for this purpose due to the practical difficulties in isolating or producing sufficient quantities of purified HMW-MAA. The booster with Colo38 melanoma cells resulted in a markedly enhanced reactivity of the sera with HMW-MAA-bearing melanoma cells; the mean Ab titer increased from 1/60 to 1/250 in P763.74-immunized mice and from 1/40 to 1/80 in PGH786-immunized mice, with a significant (p < 0.05) difference between these two groups (Fig. 2). However, this booster did not increase the reactivity with HMW-MAA-bearing melanoma cells of sera from mice that had failed to develop HMW-MAA-binding Ab responses after immunization with peptide P763.74 or PGH786. The enhanced reactivity is specific to HMW-MAA-bearing melanoma cells, because the same sera did not react with HMW-MAA-negative LG2 B-lymphoid cells. Flow cytometric analysis also showed that these sera strongly stained M14 clone 5/HMW-MAA cells, but did not stain the parental M14 clone 5 cells (Fig. 3) or empty plasmid-transfected M14 clone 5 cells. This finding is in contrast to the lack of reactivity with M14 clone 5/HMW-MAA cells of sera from mice immunized with the control peptide MB1194–208 and boosted with HMW-MAA-bearing Colo38 melanoma cells (Fig. 3). The specificity of the Ab response against HMW-MAA in mice immunized with peptide P763.74 or PGH786 and boosted with HMW-MAA-bearing melanoma cells was further analyzed by immunochemical assays. SDS-PAGE analysis showed that sera from mice that had been immunized with peptide P763.74 and boosted with HMW-MAA-bearing Colo38 melanoma cells immunoprecipitated the two subunits of HMW-MAA from radiolabeled Colo38 cell lysate (Fig. 5). Somewhat surprisingly, sera from mice immunized with peptide PGH786 and boosted with Colo38 cells failed to immunoprecipitate any HMW-MAA components from radiolabeled Colo38 cell lysate. However, as expected, neither sera from mice immunized with the control MB1194–208 peptide and boosted with Colo38 cells nor sera from mice immunized with either P763.74 or PGH786 and boosted with HMW-MAA-negative LG2 cells were able to immunoprecipitate HMW-MAA from radiolabeled Colo38 cell lysate.
Discussion
Several lines of evidence indicate that peptides P763.74 and PGH786, isolated by panning phage display peptide libraries with mAb 763.74 and mAb GH786, respectively, are mimics of HMW-MAA. First, each peptide reacted in vitro with the Ag-combining site of the respective anti-HMW-MAA mAb used to carry out the panning process. Second, each peptide induced selective Ab responses (albeit at low titers) directed against HMW-MAA-bearing melanoma cells. Interestingly, the sequences of the two peptides displayed no homology with the published amino acid sequence of HMW-MAA core protein. These findings suggest that mimicry of the HMW-MAA determinant(s) by these two peptides reflects structural homology. In addition, although mAb 763.74 and mAb GH786 share 90% amino acid sequence identity in their VH and VL regions (S. Ferrone, unpublished observations), peptides P763.74 and PGH786 reacted only with the binding sites of their corresponding panning mAb and did not cross-react with the other mAb.
The immune responses elicited by peptide mimics P763.74 and PGH786 resemble that elicited by anti-id mAb MK2–23, which also mimics the HMW-MAA antigenic determinant defined by mAb 763.74. First, a marked variability in the level of HMW-MAA-specific immunity has been found in mice immunized with peptide P763.74 or PGH786. This variability ranged from lack of detectable HMW-MAA-specific immune response to relatively strong immunity, although the immunized hosts were syngeneic mice obtained from one colony. This variability is also frequently found in mice that have been immunized with other TAA mimics (18, 31). The mechanism(s) underlying this observation has not been elucidated. Furthermore, consistent with our previous findings in mice immunized with anti-id mAb MK2–23 (18), these two peptides elicited two major Ab populations: one subset reacts with only the immunizing peptide, and the other reacts with the immunizing peptide and cross-reacts with HMW-MAA. The titer of Abs reacting with HMW-MAA-bearing melanoma cells induced by HMW-MAA peptide mimics is low, although that of Abs reacting with the immunizing peptides is high. The high titer of the latter Abs indicates that the immunization schedule used is effective in generating a robust humoral immune response and cannot account for the low titer of anti-HMW-MAA Abs.
The present study has also shown that HMW-MAA peptide mimics could induce a DTH reaction to HMW-MAA-bearing melanoma cells in BALB/c mice. This finding is consistent with the induction of a DTH reaction to HMW-MAA-bearing melanoma cells after immunization with anti-id mAb MK2–23 (18) and raises the possibility that a peptide mimic can elicit both humoral and cellular immune responses to the nominal TAA.
In the present study we have demonstrated that HMW-MAA-specific Ab titers induced by peptides P763.74 and PGH786 can be specifically and significantly enhanced by boosting with HMW-MAA-bearing melanoma cells. This prime-boost immunization strategy has previously been used with TAA mimics in several carbohydrate TAA systems (15, 32, 33). However, the results reported in the literature are conflicting. Chapman et al. (33) have reported that melanoma patients immunized with anti-id mAb BEC2, which mimics GD3 ganglioside, followed by GD3-L-KLH did not develop GD3-specific Ab responses. In contrast, Kieber-Emmons et al. (15) have shown that anti-LeY responses in mice primed with the minigene encoding LeY mimics is significantly enhanced by boosting with LeY. Our findings are consistent with those from the latter report and support the hypothesis that hypersomatic mutations in the variable regions of Abs elicited by a TAA mimic may enhance the association constant of the mutated Abs for the nominal TAA. Selective expansion of a B cell population(s) producing mutated Abs by a booster with the nominal TAA may eventually result in a marked increase in Abs that are highly reactive to TAA in the immunized host.
Several lines of evidence indicate that peptide P763.74 is more effective in inducing HMW-MAA-specific immune responses than peptide PGH786. First, although P763.74 induced a lower titer of peptide-specific Abs than PGH786, P763.74 induced a significantly higher titer of Abs reacting with HMW-MAA-bearing melanoma cells than PGH786. Second, the enhancement of anti-HMW-MAA Ab responses after a booster with HMW-MAA-bearing melanoma cells was much stronger in mice immunized with peptide P763.74 than in mice immunized with peptide PGH786. Third, P763.74, but not PGH786, was able to specifically induce a DTH reaction to HMW-MAA-bearing melanoma cells.
Characterization of the immune response elicited by peptide mimics in several TAA systems has shown that the level of Abs reacting with the nominal TAA is lower than that of the Abs reacting with the immunizing peptide mimics. Similar results have been reported for other types of TAA mimics, including xenogeneic Ags, which display homology in their amino acid sequence with human TAA (34, 35), and anti-id Abs, which bear the internal image of human TAA (19, 36). To overcome this limitation, investigators are exploring several approaches to augment the TAA-specific immune response elicited by TAA mimics. These approaches include 1) prime-boost immunization strategies, which we have used in this study; 2) administration of TAA mimics with adjuvants (37) or cytokines (38); and 3) enhancement of the ability of peptide mimics to mimic the nominal TAA, based on information derived from structural analysis of the interactions of TAA peptide mimics with the corresponding Abs (39, 40).
Our present study demonstrates that the affinity for nominal TAA of the Ab used to isolate TAA peptide mimics from phage display peptide libraries may play an important role in the identification of TAA mimics that can effectively induce TAA-specific immune responses. This finding may suggest a new strategy to enhance the immunogenicity of TAA mimics. However, caution must be exercised in the interpretation of these results, because only one peptide dose, one immunization schedule, and one Ag-Ab system were used. Nevertheless, the results we have obtained argue in favor of the possibility that an Ab with higher affinity to the nominal Ag may be more effective for isolating TAA peptide mimics with a strong ability to induce TAA-specific immune responses than one with lower affinity to the nominal Ag.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by U.S. Public Health Service Grants P01CA89480 and R01CA105500 awarded by the National Cancer Institute, Department of Health and Human Services, and by a grant from the Harry J. Lloyd Charitable Trust.
2 Address correspondence and reprint requests to Dr. Soldano Ferrone, Department of Immunology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263. E-mail address: soldano.ferrone{at}roswellpark.org
3 Abbreviations used in this paper: TAA, tumor-associated Ag; anti-id, anti-idiotypic; DTH, delayed-type hypersensitivity; HMW-MAA, high m.w.-melanoma-associated Ag; KLH, keyhole limpet hemocyanin; LeY, Lewis Y Ag.
Received for publication December 29, 2004. Accepted for publication March 25, 2005.
References
Reisfeld, R. A., D. A. Cheresh. 1987. Human tumor antigens. Adv. Immunol. 40: 323-377.
Rosenberg, S. A.. 2001. Progress in human tumour immunology and immunotherapy. Nature 411: 380-384.
Boon, T., P. G. Coulie, B. Van den Eynde. 1997. Tumor antigens recognized by T cells. Immunol. Today 18: 267-268.
Perales, M. A., N. E. Blachere, M. E. Engelhorn, C. R. Ferrone, J. S. Gold, P. D. Gregor, G. Noffz, J. D. Wolchok, A. N. Houghton. 2002. Strategies to overcome immune ignorance and tolerance. Semin. Cancer Biol. 12: 63-71.[
Perez-Diez, A., P. J. Spiess, N. P. Restifo, P. Matzinger, F. M. Marincola. 2002. Intensity of the vaccine-elicited immune response determines tumor clearance. J. Immunol. 168: 338-347.
Nanni, P., L. Landuzzi, G. Nicoletti, C. De Giovanni, I. Rossi, S. Croci, A. Astolfi, M. Iezzi, E. Di Carlo, P. Musiani, et al 2004. Immunoprevention of mammary carcinoma in HER-2/neu transgenic mice is IFN- and B cell dependent. J. Immunol. 173: 2288-2296.
Wang, X., W. Luo, K. A. Foon, S. Ferrone. 2001. Tumor associated antigen (TAA) mimicry and immunotherapy of malignant diseases from anti-idiotypic antibodies to peptide mimics. Cancer Chemother. Biol. Response Modif. 19: 309-326.
Lo Gerfo, P., F. P. Herter, S. J. Bennett. 1972. Absence of circulating antibodies to carcinoembryonic antigen in patients with gastrointestinal malignancies. Int. J. Cancer 9: 344-348(Wei Luo, Jeff Chi-feng Hs)
Peptide mimics isolated from phage display peptide libraries by panning with self-tumor-associated Ag (TAA)-specific mAbs are being evaluated as immunogens to implement active specific immunotherapy. Although TAA-specific mAb are commonly used to isolate peptide mimics, no information is available regarding the Ab characteristics required to isolate immunogenic TAA peptide mimics. To address this question, we have used mAb 763.74 and mAb GH786, which recognize the same or spatially close antigenic determinant(s) of the human high m.w.-melanoma-associated Ag (HMW-MAA), although with different affinity. mAb 763.74 affinity is higher than that of mAb GH786. Panning of phage display peptide libraries with mAb 763.74 and mAb GH786 resulted in the isolation of peptides P763.74 and PGH786, respectively. When compared for their ability to induce HMW-MAA-specific immune responses in BALB/c mice, HMW-MAA-specific Ab titers were significantly higher in mice immunized with P763.74 than in those immunized with PGH786. The HMW-MAA-specific Ab titers were markedly increased by a booster with HMW-MAA-bearing melanoma cells, an effect that was significantly higher in mice primed with P763.74 than in those primed with PGH786. Lastly, P763.74, but not PGH786, induced a delayed-type hypersensitivity response to HMW-MAA-bearing melanoma cells. These findings suggest that affinity for TAA is a variable to take into account when selecting mAb to isolate peptide mimics from a phage display peptide library.
Introduction
The identification and molecular characterization of human tumor-associated Ags (TAA) 3 have provided well-defined moieties to implement active specific immunotherapy of malignant diseases (1, 2). At variance with their murine counterparts, most human TAA are shared, nonmutated self-Ags that are weakly or nonimmunogenic in patients with malignant diseases (3, 4). Therefore, one of the challenges facing tumor immunologists is the development of strategies to induce a strong immune response to these self-TAAs, because in animal model systems an association has been found between the magnitude of the elicited TAA-specific immune responses and therapeutic efficacy of active specific immunotherapy (5, 6).
Among the strategies used to overcome unresponsiveness to a self-TAA, one relies on the use of TAA mimics as immunogens (7). The validity of this strategy has been demonstrated by the ability of anti-idiotypic (anti-id) mAb, which mimic TAA, to induce TAA-specific immune responses in patients with malignant diseases when the nominal Ag is not immunogenic (8, 9, 10, 11, 12, 13). In a phase I clinical trial, we have demonstrated that administration of the mouse anti-id mAb MK2–23, which mimics the determinant defined by the mouse anti-high m.w.-melanoma-associated Ag (HMW-MAA) mAb 763.74, induced HMW-MAA-specific Abs in 60% of the immunized melanoma patients (10). The survival of patients who had developed anti-HMW-MAA Abs was significantly longer than that of patients who had not developed this immune response. These clinical findings have prompted us to continue to optimize the immunization strategy with HMW-MAA mimics. In the present study we have investigated the potential of HMW-MAA peptide mimics as an alternative immunogen to anti-id mAb, because peptide mimics have several advantages over anti-id Abs. The relative ease of modifying peptide mimics facilitates the application of strategies to enhance their ability to induce nominal TAA-specific immune responses. Furthermore, peptide mimics can be easily synthesized and standardized, thus expediting the steps required to obtain approval from regulatory agencies for their use in clinical trials.
In the present study we have chosen to pan phage display peptide libraries with anti-HMW-MAA mAb to isolate peptide mimics. This strategy has already been successfully used to isolate peptide mimics of human TAA, including the Lewis Y (LeY) and sialyl Lewis X Ags as well as the oncogenic protein Her-2/neu (14, 15, 16). To date, no information is available about the characteristics of Abs that are required to isolate TAA peptide mimics that are effective in inducing immune responses to the nominal TAA. Therefore, in the present study we have investigated the relationship between the immunogenic properties of HMW-MAA peptide mimics and the affinity of the HMW-MAA-specific mAb used to isolate these peptides from phage display peptide libraries. To address this question, we have used the HMW-MAA-specific mAb 763.74 and anti-anti-id mAb GH786. These mAb recognize the same or spatially close antigenic determinant(s) of HMW-MAA, because mAb 763.74 and mAb GH786 cross-block the binding of each other to HMW-MAA-bearing melanoma cells. However, mAb 763.74 exhibits a markedly higher affinity for HMW-MAA than mAb GH786 (17). Our previous studies have shown that HMW-MAA-binding Abs, elicited by anti-id mAb MK2–23 in mice, rabbits, and melanoma patients, exhibit only a low level of reactivity with melanoma cells (10, 18, 19). This finding does not reflect the poor immunogenicity of mAb MK2–23, because the immunized animals or patients developed high titer anti-anti-id Abs reacting with idiotopes in the Ag-combining site of mAb MK2–23. Because a positive association has been demonstrated between the development of HMW-MAA-binding Abs and clinical outcome (10), we have also investigated in the present study whether HMW-MAA-specific immune responses elicited by the isolated peptide mimics could be enhanced by a booster with HMW-MAA-bearing melanoma cells. This strategy is based on the knowledge that B cell clones undergo somatic hypermutations during the course of an immune response. It is our working hypothesis that after peptide mimic immunization, a booster with HMW-MAA-bearing melanoma cells may selectively expand the mutated B cell clones that secrete Abs with enhanced affinities for HMW-MAA.
Materials and Methods
Statistical analysis
The statistical significance of differences in the experimental results was analyzed using two-tailed, unpaired, Student’s t test. A difference was considered statistically significant at p 0.05.
Results
Isolation of peptides from phage display peptide libraries LX-8 and X15 by panning with anti-HMW-MAA mAb 763.74 and mAb GH786
Panning of LX-8 and X15 libraries with mAb 763.74 resulted in the isolation of eight phage clones from the LX-8 library, but no clones from the X15 library. All of these clones reacted specifically with mAb 763.74. Sequencing of DNA encoding the peptides displayed by these clones identified the peptide sequence QCTGPNVATNCR (P763.74) in six of eight clones, and QCTGPNFATNCR and TCNGPSVYMNCL in each of the remaining two clones. P763.74 displayed the highest binding reactivity with mAb 763.74 and was chosen for subsequent experiments (data not shown). Panning of LX-8 and X15 libraries with mAb GH786 resulted in the isolation of four and six phage clones from the LX-8 and X15 libraries, respectively. Sequencing of DNA encoding the peptides displayed by these clones identified the peptide sequence TCRLPFQNVACH in two clones and SCLLPFQNIFCS in the other two clones isolated from the LX-8 library. The sequence NQLPQYMGPAPAYMR (PGH786) was identified in all six clones isolated from the X15 library. PGH786 displayed the highest binding reactivity to mAb GH786 and was chosen for subsequent experiments (data not shown).
Reactivity of synthetic peptides P763.74 and PGH786 with mAb 763.74 and mAb GH786
Peptides P763.74 and PGH786 were compared for their specific binding reactivity to their corresponding panning mAb and for their ability to inhibit the binding of these mAb to HMW-MAA. Each peptide reacted with its corresponding panning mAb in a dose-dependent manner (Fig. 1, A and B). Each peptide also specifically inhibited the binding of its corresponding panning mAb to HMW-MAA-bearing melanoma cells in a dose-dependent manner, but did not cross-inhibit the binding of the other mAb to HMW-MAA-bearing melanoma cells (Fig. 1, C and D). These results indicate that peptides P763.74 and PGH786 react specifically with the Ag-combining sites of mAb 763.74 and mAb GH786, respectively.
Immunogenicity of HMW-MAA peptide mimics P763.74 and PGH786 in BALB/c mice
Mice immunized with peptide mimic P763.74 or PGH786 were tested for the development of Abs reacting with the immunizing peptide and with HMW-MAA-bearing melanoma cells in binding assays. The kinetics of the development of Ab populations are shown in Fig. 2. Analysis of sera harvested from mice immunized with peptide P763.74 or PGH786 demonstrated increasing titers of peptide-reactive Abs. The anti-PGH786 Ab titer obtained from PGH786-immunized mice was significantly (p < 0.05) higher than the anti-P763.74 Ab titer obtained from P763.74-immunized mice (Fig. 2A). Reactivity with HMW-MAA-bearing Colo38 melanoma cells, but not with HMW-MAA-negative LG2 B lymphoid cells, was detected in sera from six of the eight mice immunized with peptide P763.74 and in four of the eight mice immunized with peptide PGH786. In contrast to the peptide-reactive Ab titers, the mean titers of HMW-MAA-bearing melanoma cell-reacting Abs were significantly (p < 0.05) higher in sera obtained from P763.74-immunized mice than in sera obtained from PGH786-immunized mice (Fig. 2C). The selective reactivity of these sera with HMW-MAA-bearing melanoma cells was confirmed by their reactivity with M14 clone 5/HMW-MAA cells that express HMW-MAA (Fig. 3), but not with the parental M14 clone 5 cells or empty plasmid-transfected M14 clone 5 cells. The latter cells do not express HMW-MAA. Sera from mice immunized with the control MB1194–208 peptide displayed strong reactivity with MB1194–208 peptide, but no reactivity with HMW-MAA-bearing Colo38 cells. These sera also did not stain M14 clone 5/HMW-MAA cells (Fig. 3). Despite their reactivity with HMW-MAA-bearing melanoma cells in binding assays, sera from P763.74- or PGH786-immunized mice were not able to immunoprecipitate any components from radiolabeled HMW-MAA-bearing Colo38 melanoma cell lysates (data not shown).
Enhancement of immune responses elicited with HMW-MAA peptide mimics by a booster with HMW-MAA-bearing melanoma cells
To investigate whether HMW-MAA-specific reactivity induced by peptides P763.74 and PGH786 can be augmented by a booster with the original Ag, mice immunized with these peptide mimics were boosted s.c. with HMW-MAA-bearing Colo38 melanoma cells (5 x 105 cells/mouse) on day 132. HMW-MAA-bearing melanoma cells were used for this purpose due to the practical difficulties in isolating or producing sufficient quantities of purified HMW-MAA. The booster with Colo38 melanoma cells resulted in a markedly enhanced reactivity of the sera with HMW-MAA-bearing melanoma cells; the mean Ab titer increased from 1/60 to 1/250 in P763.74-immunized mice and from 1/40 to 1/80 in PGH786-immunized mice, with a significant (p < 0.05) difference between these two groups (Fig. 2). However, this booster did not increase the reactivity with HMW-MAA-bearing melanoma cells of sera from mice that had failed to develop HMW-MAA-binding Ab responses after immunization with peptide P763.74 or PGH786. The enhanced reactivity is specific to HMW-MAA-bearing melanoma cells, because the same sera did not react with HMW-MAA-negative LG2 B-lymphoid cells. Flow cytometric analysis also showed that these sera strongly stained M14 clone 5/HMW-MAA cells, but did not stain the parental M14 clone 5 cells (Fig. 3) or empty plasmid-transfected M14 clone 5 cells. This finding is in contrast to the lack of reactivity with M14 clone 5/HMW-MAA cells of sera from mice immunized with the control peptide MB1194–208 and boosted with HMW-MAA-bearing Colo38 melanoma cells (Fig. 3). The specificity of the Ab response against HMW-MAA in mice immunized with peptide P763.74 or PGH786 and boosted with HMW-MAA-bearing melanoma cells was further analyzed by immunochemical assays. SDS-PAGE analysis showed that sera from mice that had been immunized with peptide P763.74 and boosted with HMW-MAA-bearing Colo38 melanoma cells immunoprecipitated the two subunits of HMW-MAA from radiolabeled Colo38 cell lysate (Fig. 5). Somewhat surprisingly, sera from mice immunized with peptide PGH786 and boosted with Colo38 cells failed to immunoprecipitate any HMW-MAA components from radiolabeled Colo38 cell lysate. However, as expected, neither sera from mice immunized with the control MB1194–208 peptide and boosted with Colo38 cells nor sera from mice immunized with either P763.74 or PGH786 and boosted with HMW-MAA-negative LG2 cells were able to immunoprecipitate HMW-MAA from radiolabeled Colo38 cell lysate.
Discussion
Several lines of evidence indicate that peptides P763.74 and PGH786, isolated by panning phage display peptide libraries with mAb 763.74 and mAb GH786, respectively, are mimics of HMW-MAA. First, each peptide reacted in vitro with the Ag-combining site of the respective anti-HMW-MAA mAb used to carry out the panning process. Second, each peptide induced selective Ab responses (albeit at low titers) directed against HMW-MAA-bearing melanoma cells. Interestingly, the sequences of the two peptides displayed no homology with the published amino acid sequence of HMW-MAA core protein. These findings suggest that mimicry of the HMW-MAA determinant(s) by these two peptides reflects structural homology. In addition, although mAb 763.74 and mAb GH786 share 90% amino acid sequence identity in their VH and VL regions (S. Ferrone, unpublished observations), peptides P763.74 and PGH786 reacted only with the binding sites of their corresponding panning mAb and did not cross-react with the other mAb.
The immune responses elicited by peptide mimics P763.74 and PGH786 resemble that elicited by anti-id mAb MK2–23, which also mimics the HMW-MAA antigenic determinant defined by mAb 763.74. First, a marked variability in the level of HMW-MAA-specific immunity has been found in mice immunized with peptide P763.74 or PGH786. This variability ranged from lack of detectable HMW-MAA-specific immune response to relatively strong immunity, although the immunized hosts were syngeneic mice obtained from one colony. This variability is also frequently found in mice that have been immunized with other TAA mimics (18, 31). The mechanism(s) underlying this observation has not been elucidated. Furthermore, consistent with our previous findings in mice immunized with anti-id mAb MK2–23 (18), these two peptides elicited two major Ab populations: one subset reacts with only the immunizing peptide, and the other reacts with the immunizing peptide and cross-reacts with HMW-MAA. The titer of Abs reacting with HMW-MAA-bearing melanoma cells induced by HMW-MAA peptide mimics is low, although that of Abs reacting with the immunizing peptides is high. The high titer of the latter Abs indicates that the immunization schedule used is effective in generating a robust humoral immune response and cannot account for the low titer of anti-HMW-MAA Abs.
The present study has also shown that HMW-MAA peptide mimics could induce a DTH reaction to HMW-MAA-bearing melanoma cells in BALB/c mice. This finding is consistent with the induction of a DTH reaction to HMW-MAA-bearing melanoma cells after immunization with anti-id mAb MK2–23 (18) and raises the possibility that a peptide mimic can elicit both humoral and cellular immune responses to the nominal TAA.
In the present study we have demonstrated that HMW-MAA-specific Ab titers induced by peptides P763.74 and PGH786 can be specifically and significantly enhanced by boosting with HMW-MAA-bearing melanoma cells. This prime-boost immunization strategy has previously been used with TAA mimics in several carbohydrate TAA systems (15, 32, 33). However, the results reported in the literature are conflicting. Chapman et al. (33) have reported that melanoma patients immunized with anti-id mAb BEC2, which mimics GD3 ganglioside, followed by GD3-L-KLH did not develop GD3-specific Ab responses. In contrast, Kieber-Emmons et al. (15) have shown that anti-LeY responses in mice primed with the minigene encoding LeY mimics is significantly enhanced by boosting with LeY. Our findings are consistent with those from the latter report and support the hypothesis that hypersomatic mutations in the variable regions of Abs elicited by a TAA mimic may enhance the association constant of the mutated Abs for the nominal TAA. Selective expansion of a B cell population(s) producing mutated Abs by a booster with the nominal TAA may eventually result in a marked increase in Abs that are highly reactive to TAA in the immunized host.
Several lines of evidence indicate that peptide P763.74 is more effective in inducing HMW-MAA-specific immune responses than peptide PGH786. First, although P763.74 induced a lower titer of peptide-specific Abs than PGH786, P763.74 induced a significantly higher titer of Abs reacting with HMW-MAA-bearing melanoma cells than PGH786. Second, the enhancement of anti-HMW-MAA Ab responses after a booster with HMW-MAA-bearing melanoma cells was much stronger in mice immunized with peptide P763.74 than in mice immunized with peptide PGH786. Third, P763.74, but not PGH786, was able to specifically induce a DTH reaction to HMW-MAA-bearing melanoma cells.
Characterization of the immune response elicited by peptide mimics in several TAA systems has shown that the level of Abs reacting with the nominal TAA is lower than that of the Abs reacting with the immunizing peptide mimics. Similar results have been reported for other types of TAA mimics, including xenogeneic Ags, which display homology in their amino acid sequence with human TAA (34, 35), and anti-id Abs, which bear the internal image of human TAA (19, 36). To overcome this limitation, investigators are exploring several approaches to augment the TAA-specific immune response elicited by TAA mimics. These approaches include 1) prime-boost immunization strategies, which we have used in this study; 2) administration of TAA mimics with adjuvants (37) or cytokines (38); and 3) enhancement of the ability of peptide mimics to mimic the nominal TAA, based on information derived from structural analysis of the interactions of TAA peptide mimics with the corresponding Abs (39, 40).
Our present study demonstrates that the affinity for nominal TAA of the Ab used to isolate TAA peptide mimics from phage display peptide libraries may play an important role in the identification of TAA mimics that can effectively induce TAA-specific immune responses. This finding may suggest a new strategy to enhance the immunogenicity of TAA mimics. However, caution must be exercised in the interpretation of these results, because only one peptide dose, one immunization schedule, and one Ag-Ab system were used. Nevertheless, the results we have obtained argue in favor of the possibility that an Ab with higher affinity to the nominal Ag may be more effective for isolating TAA peptide mimics with a strong ability to induce TAA-specific immune responses than one with lower affinity to the nominal Ag.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by U.S. Public Health Service Grants P01CA89480 and R01CA105500 awarded by the National Cancer Institute, Department of Health and Human Services, and by a grant from the Harry J. Lloyd Charitable Trust.
2 Address correspondence and reprint requests to Dr. Soldano Ferrone, Department of Immunology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263. E-mail address: soldano.ferrone{at}roswellpark.org
3 Abbreviations used in this paper: TAA, tumor-associated Ag; anti-id, anti-idiotypic; DTH, delayed-type hypersensitivity; HMW-MAA, high m.w.-melanoma-associated Ag; KLH, keyhole limpet hemocyanin; LeY, Lewis Y Ag.
Received for publication December 29, 2004. Accepted for publication March 25, 2005.
References
Reisfeld, R. A., D. A. Cheresh. 1987. Human tumor antigens. Adv. Immunol. 40: 323-377.
Rosenberg, S. A.. 2001. Progress in human tumour immunology and immunotherapy. Nature 411: 380-384.
Boon, T., P. G. Coulie, B. Van den Eynde. 1997. Tumor antigens recognized by T cells. Immunol. Today 18: 267-268.
Perales, M. A., N. E. Blachere, M. E. Engelhorn, C. R. Ferrone, J. S. Gold, P. D. Gregor, G. Noffz, J. D. Wolchok, A. N. Houghton. 2002. Strategies to overcome immune ignorance and tolerance. Semin. Cancer Biol. 12: 63-71.[
Perez-Diez, A., P. J. Spiess, N. P. Restifo, P. Matzinger, F. M. Marincola. 2002. Intensity of the vaccine-elicited immune response determines tumor clearance. J. Immunol. 168: 338-347.
Nanni, P., L. Landuzzi, G. Nicoletti, C. De Giovanni, I. Rossi, S. Croci, A. Astolfi, M. Iezzi, E. Di Carlo, P. Musiani, et al 2004. Immunoprevention of mammary carcinoma in HER-2/neu transgenic mice is IFN- and B cell dependent. J. Immunol. 173: 2288-2296.
Wang, X., W. Luo, K. A. Foon, S. Ferrone. 2001. Tumor associated antigen (TAA) mimicry and immunotherapy of malignant diseases from anti-idiotypic antibodies to peptide mimics. Cancer Chemother. Biol. Response Modif. 19: 309-326.
Lo Gerfo, P., F. P. Herter, S. J. Bennett. 1972. Absence of circulating antibodies to carcinoembryonic antigen in patients with gastrointestinal malignancies. Int. J. Cancer 9: 344-348(Wei Luo, Jeff Chi-feng Hs)