Blanchet.wyenet.co.uk

A New Generation of Melan-A/MART-1 Peptides That Fulfill
Both Increased Immunogenicity and High Resistance to
Biodegradation: Implication for Molecular Anti-Melanoma
Immunotherapy1
Jean-Se´bastien Blanchet,* Danila Valmori,† Isabelle Dufau,* Maha Ayyoub,†
Christophe Nguyen,* Philippe Guillaume,† Bernard Monsarrat,* Jean-Charles Cerottini,†
Pedro Romero,† and Jean Edouard Gairin2*
Intense efforts of research are made for developing antitumor vaccines that stimulate T cell-mediated immunity. Tumor cells
specifically express at their surfaces antigenic peptides presented by MHC class I and recognized by CTL. Tumor antigenic
peptides hold promise for the development of novel cancer immunotherapies. However, peptide-based vaccines face two major
limitations: the weak immunogenicity of tumor Ags and their low metabolic stability in biological fluids. These two hurdles, for
which separate solutions exist, must, however, be solved simultaneously for developing improved vaccines. Unfortunately, attempts
made to combine increased immunogenicity and stability of tumor Ags have failed until now. Here we report the successful design
of synthetic derivatives of the human tumor Ag Melan-A/MART-1 that combine for the first time both higher immunogenicity and
high peptidase resistance. A series of 36 nonnatural peptide derivatives was rationally designed on the basis of knowledge of the
mechanism of degradation of Melan-A peptides in human serum and synthesized. Eight of them were efficiently protected against
proteolysis and retained the antigenic properties of the parental peptide. Three of the eight analogs were twice as potent as the
parental peptide in stimulating in vitro Melan-specific CTL responses in PBMC from normal donors. We isolated these CTL by
tetramer-guided cell sorting and expanded them in vitro. The resulting CTL efficiently lysed tumor cells expressing Melan-A Ag.
These Melan-A/MART-1 Ag derivatives should be considered as a new generation of potential immunogens in the development
of molecular anti-melanoma vaccines. The Journal of Immunology, 2001, 167: 5852–5861.
T hemolecularcharacterizationoftumorAgsthatcanbe BecauseofthehighfrequencyofMelan-Acirculatingprecur- recognized by CTL has opened new opportunities for the sors (5, 7) and evidence of strong immune responses to this Ag in development of Ag-specific cancer vaccines (1, 2). In the melanoma patients (8), Melan-A represents an attractive candidate case of melanoma, a candidate Ag is Melan-A/MART-1 (Melan-A for generic immunotherapy of HLA-A*0201ϩ melanoma patients.
hereafter), a melanocyte lineage-specific protein expressed in A superagonist variant of the nonameric Melan-A ϳ75–100% of primary and metastatic melanomas depending on been shown to elicit an enhanced anti-melanoma CD8ϩ CTL re- their clinical stage (3). Melan-A-specific CTL have been fre- sponse (9). We have recently undertaken clinical trials of peptide quently isolated from PBMC of both HLA-A*0201 healthy donors vaccination using the decameric analog Melan-A and melanoma patients and from tumor-infiltrating lymphocytes of (ELAGIGILTV). In addition to exhibiting improved HLA-A*0201 melanoma patients (4, 5). These CTL have been shown to primar- binding properties (higher affinity and more stable HLA-A*0201/ tigenicity and immunogenicity than the natural Melan-A peptides(10, 11). Moreover, the large majority of CTL raised either in vitroor in vivo against Melan-A *Laboratoire d’ImmunoPharmacologie Structurale, Institut de Pharmacologie et Bi- the Melan-A parental peptide sequences and able to specifically ologie Structurale, Centre National de la Recherche Scientifique, Toulouse, France; lyse Melan-A-expressing tumor cells (10 –12).
and †Ludwig Institute for Cancer Research, Division of Clinical Onco-Immunology,University Hospital and Lausanne Branch, Lausanne, Switzerland When compared with other vaccination strategies (13), the use Received for publication July 20, 2001. Accepted for publication September 18, 2001.
of antigenic peptides derived from tumor Ags (tumor Ag-derived The costs of publication of this article were defrayed in part by the payment of page peptides) as immunogens offers a number of advantages, including charges. This article must therefore be hereby marked advertisement in accordance low cost and facility of administration of the vaccine, high spec- with 18 U.S.C. Section 1734 solely to indicate this fact.
ificity of elicited immune responses, and low toxicity (14, 15).
1 This work was supported in part by grants from the Center National de la Recherche However, the use of tumor Ag-derived peptides for cancer immu- Scientifique, the Association pour la Recherche sur le Cancer (Contract 5485), and theConseil Re´gional Midi-Pyre´ne´es (Contract 99001129). J.-S.B. is the recipient of a notherapy faces two major limitations: the weak antigenicity and doctoral fellowship from the Association pour la Recherche sur le Cancer and a immunogenicity of tumor Ags (16) and their high susceptibility to European Molecular Biology Organization short-term fellowship (ASF-9575). M.A.
is a recipient of a postdoctoral fellowship from the European Community-Biomed proteolytic degradation by proteases (17–20). Weak immunoge- nicity can partially be ascribed to a certain level of immune tol- 2 Address correspondence and reprint requests to Dr. Jean Edouard Gairin, Institut de erance to self-derived sequences (including suboptimal MHC Pharmacologie et de Biologie Structurale, Unité Mixte de Recherche 5089, Centre binding and/or T cell recognition). Limiting the extent of peptide National de la Recherche Scientifique, 205 route de Narbonne, 31400 Toulouse,France. E-mail address: [email protected] degradation by, for example, delivering the peptide incorporated in Copyright 2001 by The American Association of Immunologists a water/oil emulsion often significantly increases immunogenicity of a free amino group with an F-moc-protected preformed amino aldehyde (21). A more effective approach for avoiding rapid peptide degra- (31). N-terminal hydroxypeptides were synthesized following a previously dation by proteases implies the design of protease-resistant peptide described procedure (32). The retro-inverso bond ⌿(NH-CO) was obtainedby replacement of two sequential amino acids with an (R,S)-2-substituted analogs. Approaches based on structural modifications to inhibit malonate derivative and a gem-diaminoalkyl residue (for example, the proteolytic degradation of bioactive peptides exist (22, 23) and gem-diaminoalkyl residue corresponding to glutamic acid side chain and have been applied to MHC class I-restricted (20, 24, 25) or class 2-substituted malonic acid corresponding to leucine side chain (2(R,S)- II-restricted (26) antigenic peptides. Unfortunately, the structural isobutylmalonic acid) were used for the synthesis of [⌿ A27L) (33, 34). A mixture of two diastereoisomers of the modifications introduced in the antigenic peptide sequence most modified peptide was obtained. The two isomeric forms were not sepa- often result in a dramatic reduction or even in a complete loss of rated. Peptides were purified (purity, Ն98%) by RP-HPLC on a C column peptide binding to MHC and/or T cell recognition (24, 27–29).
(Aquapore Brownlee, PerkinElmer, Norwalk, CT). The identities of the This difficulty may be partly overcome by a detailed knowledge of purified peptides were confirmed by electrospray ionization-mass spec- the degradation pathway of the antigenic peptide in human serum trometry (ESI-MS). Peptide stock solutions were made (10Ϫ3 M in 100%DMSO) and stored at Ϫ20°C.
that allows the rational design of minimally modified, peptidase-resistant, and still biologically active nonnatural analogs (20, 24).
Degradation of Melan-A peptides in human serum It is clear that synthetic derivatives of tumor Ags combining both Peptides were added to preheated (10 min at 37°C before the assay) human high protection against peptidases and higher immunogenicity rep- serum to a final concentration of ϳ0.5 ϫ 10Ϫ4 M and incubated at 37°C.
resent candidates of choice for cancer immunotherapy. However, For the analysis of peptide persistence in serum with Melan-A-specific none has been successfully designed and used until now.
CTL recognition assay, aliquots (100 ␮l) were removed at different timesand put in liquid nitrogen to stop the enzymatic reaction. 51Cr-labeled T2 The aim of the present study was to design such derivatives of cells were then pulsed with serial dilutions of each degradation sample at the Melan-A tumor Ag. For that, we merged complementary ap- 4°C, and chromium release experiments were performed as previously de- proaches based on manipulation of the structural and pharmaco- logical properties of MHC class I-restricted Ags and of the im- For on-line RP-HPLC/ESI-MS analysis, aliquots of the degradation so- mune response against cancer. First we dissected the mechanism of lution (100 ␮l) were removed at different times, and the enzymatic reactionwas stopped by addition of 11 ␮l of trifluoroacetic acid (TFA). Precipitated Melan-A peptide degradation in human serum. Second, on the ba- serum proteins were pelleted by centrifugation at 15,000 rpm for 10 min at sis of these results, we designed a series of 36 nonnatural deriva- 4°C. The supernatants were frozen and kept at Ϫ20°C until analysis. We checked that 1) the Melan-A peptides did not precipitate and were stable in them, eight were fully resistant to proteolysis by serum proteases the presence of 10% TFA, and 2) DMSO did not affect serum proteaseactivities at the concentration used. The HPLC profile of serum alone pre- and retained the antigenicity and immunogenicity of the parental cipitated by 10% TFA was recorded to detect nonprecipitated peptides peptide. When used to stimulate in vitro Melan-specific CTL re- present in the serum. Melan-A peptides and their degradation products sponses in PBMC from normal donors, three of these sequences were separated, and their sequences were determined and quantified by were more potent activators of tumor-reactive CTL than the pa- rental peptide. These analogs represent a new generation of Melan- Coulter, Palo Alto, CA), a Waters 600 MS chromatograph (Waters, Mil-ford, MA), and a TSQ-700 Finnigan-MAT mass spectrometer (Thermo A/MART-1 Ag derivatives that now combine enhanced immuno- Finnigan, San Jose, CA) as previously described (19, 24). Quantitative genicity with proteolysis resistance. Therefore, they should be determination of peptides and their degradation products were obtained considered potential immunogens in the development of molecular operating in the selected ion monitoring mode.
Materials and Methods
Peptide binding to HLA-A*0201 was assessed in a functional competition assay based on inhibition of recognition of the antigenic peptide tyrosi-nase (YMDGTMSQV) by the HLA-A*0201-restricted CTL clone The tumor cell lines Me 290 and Me 275 were established from surgically 156/34. Various concentrations of competitor peptides (50 ␮l) were incu- excised melanoma metastasis from patients LAU203 and LAU50, respec- bated with 51Cr-labeled T2 cells (50 ␮l; 1000 cells/well) for 15 min at room tively, and maintained in culture as previously described (10). T2, an HLA- temperature. A suboptimal dose (1 nM) of the antigenic peptide tyrosi- A*0201ϩ human T/B cell hybrid peptide transporter-deficient cell line (50 ␮l) was then added together with specific CTL (5000 cells/ (30), was cultured in DMEM/10% FCS supplemented with 0.55 mM Arg, well; 50 ␮l). Chromium release was measured after a 4-h incubation at 0.24 mM Asn, and 1.5 mM Gln. The tyrosinase-specific CTL clone 156/34 37°C. The concentration of each competitor peptide required to achieve was derived after in vitro stimulation of tumor-infiltrating lymphocytes 50% inhibition of target cell lysis was then determined (IC ).
from patient LAU156 with the natural tyrosinase MSQV) (12). The polyclonal monospecific CTL line used for analog rec- Peptide recognition by Melan-A specific CTL ognition experiments was induced in vitro after stimulation of PBMC from Ag recognition was assessed using chromium release assays. Target cells (T2 cells) were labeled with 51Cr for 1 h at 37°C and washed three times.
51 Cr-labeled target cells (1000 cells/50 ␮l) were then added to various concentrations of antigenic peptide (50 ␮l) in V-bottom 96-well plates for Peptides were synthesized by the solid phase method using F-moc chem- 15 min before addition of effector cells. A polyclonal CTL line specific for istry and the N,NЈ-diisopropylcarbodiimide/N-hydroxyaza-benzotriazole A27L was used as effector cells and added (5000 cells/100 coupling procedure. Peptide with ␤ amino acids, d amino acids, ␣-meth- ␮l) at a defined E:T cell ratio. Chromium release was measured in 100 ␮l ylated amino acids (␣Meaa),3 or N-methylated amino acids (NMeaa) were of supernatant harvested after 4 h of incubation at 37°C. The percentage of synthesized by incorporation of the corresponding modified amino acids specific lysis was calculated as follows: % specific lysis ϭ [(experimental commercially available (Bachem, Budendorf, Switzerland (NMeaa); Fluka, release Ϫ spontaneous release)/(total release Ϫ spontaneous release)] ϫ Buchs, Switzerland (␤ amino acids); and Acros Organic, Noisy-Le-Grand, 100. The concentration of each peptide required to achieve 50% maximal France (␣Meaa)). Peptides with ␣Meaa were obtained as racemics, because lysis of target (EC ) was then determined.
␣Meaa were purchased in a racemic form. Whenever possible, stereoiso-mers were isolated by reverse phase HPLC (RP-HPLC) and named p1 and Generation of Melan-A specific CTL p2. The reduced bond ⌿(CH -NH) was formed by the reductive alkylation PBMC from HLA-A*0201ϩ healthy donors were isolated by centrifuga-tion in Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden). CD8ϩ lym- phocytes were isolated using a miniMACS device (Miltenyi Biotec, Ber- Abbreviations used in this paper: ␣Meaa, ␣-methylated amino acid; ESI-MS, elec- gisch Gladbach, Germany). The resulting population routinely contained trospray ionization-mass spectrometry; DC, dendritic cell; NMeaa, N-methylatedamino acid; NOHG, N-hydroxylated glycine; RP-HPLC, reverse phase HPLC; TFA, Ͼ75% CD8ϩ T cells and was used as the responder cell population. Pu- trifluoroacetic acid; RIC, reconstructed ion current.
rified CD8ϩ T cells were plated at 1 ϫ 106 cells/well together with 2 ϫ 106 A NEW GENERATION OF MELAN-A/MART-1 Ag DERIVATIVES stimulator cells/well in 24-well plate in a total volume of 2 ml of Iscove’s 24-h treatment in serum (Fig. 1A), although it appeared almost medium supplemented with 10% human serum, Asn, Arg, and Gln (com- completely degraded after 2 h (Fig. 1B). This observation can be plete medium) in the presence of IL-7 (10 ng/ml; R&D Systems, Oxon, explained by a 2- to 3-log difference in the sensitivity of the assays U.K.), IL-2 (10 U/ml; Glaxo Wellcome, Geneva, Switzerland), and stim-ulating peptide (1 ␮M). Stimulator cells were prepared as follows. Cells (biological vs physicochemical) as we previously noted (24). It (2 ϫ 106) derived from the CD8Ϫ population after miniMACS CD8ϩ lym-phocyte isolation were irradiated (3000 rad) and adjusted to the appropriatevolume before addition to the CD8ϩ-enriched responder cell population.
On day 7 cells were restimulated with peptide-pulsed T2 cells. T2 cellswere incubated for 2 h at 37°C in serum-free medium (X-VIVO 10; Bio-Whittaker, Walkersville, MD) with the appropriate stimulating peptide (1 ␮M) and human ␤ -microglobulin (3 ␮g/ml; Sigma-Aldrich, St. Louis, MO). Peptide-pulsed T2 cells were washed, irradiated (10,000 rad), ad-justed to the appropriate volume of complete medium supplemented withIL-7 (10 ng/ml) and IL-2 (10 U/ml), and added to the lymphocyte culture(2 ϫ 105 cells/well). Subsequent restimulations were performed weeklywith peptide-pulsed T2 cells. CTL activity was first tested at the end of thefirst restimulation using an ELISPOT assay for IFN-␥ production bypeptide-reactive CTL.
Flow cytometric analysis and Melan-A tetramer-guided cellsorting of specific CD8ϩ CTL The specificity of CTL recognition was monitored by flow cytometric anal-ysis after Melan-A A27L HLA-A*0201 tetramer staining (5, 8). Cells were stained with tetramers (200 ng/sample) in 20 ␮l of PBS/2% FCS for20 min at room temperature, then 20 ␮l of anti-CD8-FITC mAb (BD Bio-sciences, Basel, Switzerland) was added, and cells were incubated for anadditional 30 min at 4°C. Cells were washed once in the same buffer andanalyzed on a FACScan (BD Biosciences, San Jose, CA) flow cytometer.
Data analysis was performed using CellQuest software (BD Biosciences).
The CD8ϩtetramerϩ and CD8ϩtetramerϪ lymphocyte populations weresorted using a FACSVantage (BD Biosciences) cell sorter. After cell sort-ing the polyclonal monospecific CTL population was expanded by restimu-lation in a nonspecific fashion using PHA (1 ␮g/ml; Sigma-Aldrich), andIL-2 (150 U/ml).
Results
Melan-A Ag-related peptides are rapidly degraded in human
serum by amino- and dipeptidyl-carboxy-peptidases
The kinetics as well as the mechanism of Melan-A peptide pro-teolytic degradation were determined by incubation of syntheticpeptides in human serum. We studied different peptides related tothe Melan-A Ag: the nonamer Melan-A A27L (ELAGIGILTV). We first determined the persistence of the antigenic peptide over time in serum as monitored by spe-cific CTL recognition using a chromium release assay. After dif-ferent incubation times in serum, serial dilutions of the samplescontaining the synthetic peptide were pulsed onto T2 cells andincubated with Melan-A-specific CTL. Lysis of peptide-pulsed T2cells represents the persistence of intact peptide in human serumover time. Clearly, Melan-A antigenic peptides incubated in serum FIGURE 1.
Degradation of the Melan-A peptides in human serum. A, very rapidly lost the ability to sensitize T2 target cells for lysis by Melan-A peptides were incubated in human serum, and the persistence ofthe Ag was evaluated using a functional cytolytic assay. At each incubation specific CTL (Fig. 1A), suggesting that they were degraded by time T (F, 0 min; E, 30 min; f, 60 min; Ⅺ, 120 min; Œ, 240 min; ‚, 24 h), peptidases in human serum. Of note, the Melan-A nonamer AA- 51Cr-labeled T2 cells were pulsed with serial dilutions of each degradation GIGILTV seemed to be more sensitive than the two Melan-A sample and used as target cells in the chromium release assay. The CTL used was a polyclonal monospecific CTL line raised against Melan-A26– The exact mechanism of antigenic peptide degradation in serum 35A27L peptide. The E:T cell ratio used was 10:1. B, Degradation kinetics was addressed by on-line RP-HPLC/ESI-MS that allows unambig- of Melan-A Ags analyzed by HPLC/ESI-MS. Melan-A peptides were in- uous identification and precise quantification of the degradation cubated in human serum, and the presence of the peptide was addressed by products resulting from tumor Ag proteolysis (19, 24). As shown direct detection using on-line HPLC/ESI-MS. Peptide quantification was in Fig. 1B (left panels), the Melan-A achieved by analysis of the mass spectrometry data. For a given molecular , 5 min), in agreement with a previous study (20).
species, the reconstructed ion current (RIC) intensity characteristic of the identified peptide or degradation product was determined. B, left panels, The percentage of peptide remaining was calculated for each sample with A27L (ELAGIGILTV) were degraded less rapidly the amount of peptide detected at T0 as a reference and plotted against the than the nonamer, thus confirming the initial observation, and dis- time: (RIC intensity)T ϫ 100)/(RIC intensity)T0). B, middle and right panels, Kinetics of appearance of the amino- and carboxyl-terminal deg- radation products of Melan-A Ags. The amount of each product is shown could sensitize T2 cells for lysis by Melan-A-specific CTL after a as the RIC intensity plotted against time.
further suggests that few copies of Melan-A Protection against amino-peptidases was obtained by substitution able by MS remained in the medium and were still able to stim- of glutamic acid residue at position 1 (E1) by ␤aa (␤E1, ␤A1, or ulate, even though very weakly, the Melan-A-specific CTL.
␤D1) (35), NMeaa (NMeE1), a pyro-glutamic acid (pE1), N-acety- Identification of the amino- and carboxyl-terminal fragments al- lated E, or an N-hydroxylated glycine (NOHG1) (32). Modifica- lowed us to determine the mechanism of peptide degradation. The tion of leucine residue at position 2 (L2) was achieved by replacing degradation process of the Melan-A peptides in human serum fol- it with methylated aa (␣MeL2, NMeL2) or dL2. To prevent deg- lowed two main pathways involving amino- and dipeptidyl-car- radation by dipeptidyl-carboxy-peptidase we replaced L8 by ␣MeL8 boxy-peptidase activities (Fig. 1B, middle and right panels, and or dL8 or T9 by NMeT9 or dT9. To avoid carboxy-peptidase degra- Fig. 2), as previously observed for the MAGE-1.A1 tumor Ag dation, we also synthesized carboxy-amidated (CO-NH ) peptide. Fi- (24). Degradation experiments using selective peptidase inhibitor nally, peptide bond alterations, such as reduced bond ⌿(CH -NH) or bestatine (amino-peptidase inhibitor) or captopril (dipeptidyl-car- retro-inverso bond ⌿(NH-CO), were included to prevent amino- or boxy-peptidase inhibitor) confirmed this. In both cases peptide carboxyl-terminal degradation. All the analogs are listed in Table I.
degradation was partially inhibited (data not shown). Thus, the first As exemplified in Fig. 3 for analogs [␤D1]-Melan-A and [⌿ two peptide bonds to be sensitive to peptidases are Glu1-Leu2 (or NH)]-Melan-A, the amino- and carboxyl-terminal structural mod- Ala2) and Leu8-Thr9. On-line RP-HPLC/ESI-MS analysis also ifications of the Melan-A Ag protected the peptidase-sensitive gave us an estimation of the quantity of a given degradation prod- bond. However, one-site protection did not result in peptides with uct present during the time course of degradation. As shown in Fig.
significantly improved stability, indicating that protection at both 1B, the main amino-terminal degradation products of the Melan- ends is needed to stabilize the peptide. Indeed, all the two-site nonamer (AGIGILTV and GIGILTV) were more abundant protected analogs were rendered resistant to peptidase activities values varying from Ͼ10 to ϾϾ24 h, as exemplified in Fig. 3 (analog [␤D1, ␣MeL8]-Melan-A) and summarized in Table amino-peptidases than the Melan-A decamers. This sensitivitycould explain the very short half-life of the nonameric peptide.
I. Among the doubly protected analogs, we noted that those bear- Analysis of carboxyl-terminal degradation of the Melan-A pep- ing either amidation or ␣-methylation at the C terminus displayed tides also revealed differences between the nonamer and the the shortest or longest half-lives, respectively.
decamers. The first carboxyl-terminal product from Melan-A27–35degradation (AAGIGIL) was detected very rapidly, peaked at 20 Nonnatural Melan-A derivatives are efficiently recognized by min, and was totally degraded within 60 min. In the case of the two decamers, the quantity of the first carboxyl-terminal degradationproducts (EAAGIGIL or ELAGIGIL) was maximal at 60 min.
We and others have previously shown that structural modification These products were then degraded, but were still present in a of an antigenic peptide may considerably affect the efficiency of detectable quantity at 240 min. The scheme depicted in Fig. 2 peptide-MHC binding and/or CTL recognition (24, 25, 27). We summarizes the degradation mechanism of Melan-A first determined the HLA-A*0201 binding properties of the 36 The slower rate of degradation may reflect that glutamic acid at the newly designed nonnatural analogs in a functional competition as- amino terminus of the peptide is not optimal for amino-peptidase say (10). As reported in Table I, a dramatic reduction in peptide- activity. As a consequence, these peptides were less susceptible to MHC binding was observed for about half the peptides. However, amino-terminal degradation and were more available for carboxy- for the other half, the peptide affinity for HLA-A*0201 was either peptidase activity. Altogether, the degradation data suggest that the nature of the amino-terminal residues of peptide has a direct or reduced. Indeed, the structural alteration of the anchor residue L2 indirect effect on both amino- and carboxy-peptidase activities.
was often deleterious on MHC binding ([dL2]- or [NMeL2]-Melan-A with an IC Designing synthetic antigenic peptides resistant to degradation shown by others (27), peptide backbone alteration in reduced orretro-inverso analogs dramatically reduced the peptide affinity for To identify Melan-A peptides resistant to proteolysis, we synthe- MHC. Other modifications showed limited or no impact on peptide sized a series of peptide analogs with structural modifications at affinity for MHC even if they concerned anchor residue ([␤E1]- one or both peptidase-sensitive positions and tested their degrada- Melan-A IC , 2 nM; [␣MeL2]-Melan-A IC , 6 nM). Finally, doubly protected Melan-A analogs such as [␤A1, ␣MeL8]( p1), starting sequence because of its enhanced immunogenicity (10).
[NOHG1, ␣MeL8]( p1), or [␣MeL2, ␣MeL8] displayed HLA-A*0201 binding affinity very similar to if not identical with that ofthe parental Melan-A Next, we tested the ability of the nonnatural Melan-A analogs to be recognized by specific CTL (exemplified in Fig. 4and summarized in Table I). As expected, CTL recognition of thenonnatural Melan-A analogs correlated with MHC binding formost of the analogs, with poor HLA-A2 binders being weaklyrecognized by CTL. However, strikingly, most of the structurallymodified peptides were able to efficiently sensitize target cells tolysis by Melan-A-specific CTL, with EC molar range. Interestingly, some doubly protected analogs dis-played efficiencies comparable to or even higher than that of theparental Melan-A FIGURE 2.
man serum. Full and dotted arrows correspond to dipeptidyl-carboxy-pep- tidases and amino-peptidases, respectively.
A NEW GENERATION OF MELAN-A/MART-1 Ag DERIVATIVES Table I. Sequences, HLA-A*0201 binding properties, Melan-A-specific CTL recognition, and stability in human serum of Melan-A peptides andnonnatural Melan-A derivatives a Competitor activity was measured on the basis of the inhibition of recognition of the tyrosinase (YMDGTMSQV) antigenic peptide in the context of HLA-A*0201 by the tyrosinase peptide-specific CTL clone 156/34.
b Peptide half-life was determined by HPLC-ESI-MS analysis. n.t., Not tested.
c Melan-A designation corresponds to Melan-A d The values correspond to the biological activities of each stereoisomer ( p , p ) of the ␣-methylated peptide purified from resolved HPLC peaks.
e The value corresponds to the biological activity of the mixture of the stereoisomers when HPLC peaks were not resolved and did not allow the separation of the stereoisomers Selected nonnatural Melan-A analogs efficiently stimulate the were tested by flow cytometry 7 days after the second stimulation expansion of Melan-A-specific CTL precursors in the circulation (MC2) for the presence of CD8ϩA2/Melan-A tetramerϩ cells. In cul- The eight doubly substituted derivatives displaying the highest ef- tures stimulated with the nonnatural Melan-A analogs, A2/Melan-A ficiencies were selected for further studies. They are listed in Table II.
tetramerϩ cells were detected in the CD8ϩ cell population, indicating To assess their immunogenicity, we used them to stimulate CD8ϩ- that the A2/Melan-A tetramer was fully or partially cross-reactive enriched cells isolated from PBMC. The Melan-A with the analog-specific CTL, as shown in Fig. 5. Cross-reactivity of was used as a positive control. For the second stimulation, irradiated the A2/Melan-A tetramer allowed us to directly evaluate the efficacy T2 cells pulsed with the appropriate peptide were used. All cultures of the nonnatural analogs to generate Melan-A-specific CTLs from FIGURE 3.
and doubly protected nonnatural Melan-A analogs. Exam-ples of stability of mono-protected ([␤D]LAGIGILTV:[␤D1]-Melan-A, left panel; ELAGIGIL-[␺(CH -NH)]- (CH2-NH]-Melan-A, middle panel) and doubly protected ([␤D]-LAGIGI-[␣MeL]-TV: [␤D1, ␣MeL8]-Melan-A, right panel) analogs in human serum areshown. Degradation was addressed using HPLC/ESI-MS detection as described for Melan-A peptides(see Fig. 1B). The amount of each peptide was deter-mined after analysis of MS data and is presented as apercentage of the maximum amount of the parentalpeptide at time zero (T0). f, Amount of peptide re-maining; F and E, amounts of degradation productsgenerated from amino- and dipeptidyl-carboxy-pepti-dase activities, respectively.
PBMC of healthy donors. The percentage of Melan-A-specific cells out specific stimulation, the antigenic specificity of the sorted cells was tested in a chromium release assay. As shown in Fig. 5, B and donor to another, related to the frequency of Melan-A-reactive pre- C, the sorted CTL specific for a nonnatural Melan-A analog were cursors in the initial CD8ϩ population (Table II; HD224, 2.3%; able to recognize and efficiently lyse not only T2 target cells pre- HD410, 2.5%; HD220, 0.3%). Interestingly, three analogs were able to induce twice as many Melan-A-specific cells superior as the pa- cell lines Me 290 and Me 275 that naturally express the Melan-A [␤A1, ␣MeL8]-, [␤E1, ␣MeL8]-, and [␤D, ␣MeL8]-Melan-A (Table Ag on their surfaces were also recognized efficiently by nonnatural II and Fig. 5). Thus, these three analogs are more immunogenic in Melan-A-specific CTL. The Na8 cell line that does not express the Melan-A gene was used as a control. It was not susceptible oronly weakly susceptible to lysis by the nonnatural Melan-A-spe- A2/Melan-A tetramer-guided isolation and functional cific CTL, indicating the absence of nonspecific lytic activity, but characterization of in vitro stimulated nonnatural Melan-A became efficiently lysed when pulsed with the Melan-A ural peptide. In conclusion, CTL induced with the nonnatural an- To further document the Ag specificity of CTL generated upon alogs are fully competent to recognize the parental Melan-A pep- stimulation with the nonnatural Melan-A analogs, we sorted the tides and lyse tumor cells that naturally express the natural CD8ϩA2/Melan-A tetramerϩ cells. After in vitro expansion with- FIGURE 4.
Recognition of the nonnatural, doubly protected Melan-A analogs by Melan-A A27L-specific CTL. Recognition of the doubly protected analogs [␤A1, ␣MeL8]-Melan-A, [␤E1, ␣MeL8]-Melan-A, [␤D1, ␣MeL8]-Melan-A, and [␤A1, V10-CONH ]-Melan-A (F, upper panels) by CTL was addressed in a chromium release assay using a polyclonal monospecific CTL line raised against Melan-A A27L. The target cells were T2 cells pulsed with increasing concentrations of peptide, and the E:T cell ratio was 10:1. The Melan-A peptides Melan-A E26A were tested as controls (E, lower panels).
A NEW GENERATION OF MELAN-A/MART-1 Ag DERIVATIVES Table II. In vitro immunogenicity of Melan-A less, the use of such immunogenic peptide analogs as a vaccine may be restricted due to their high susceptibility to protease deg-radation. Thus, the next step was to introduce structural modifica- tions to obtain proteolysis-resistant antigenic peptides. In the present study we first elucidated the degradation pathway of the Melan-A antigenic peptides. Then, according to these results wedesigned structurally modified nonnatural Melan-A analogs that were potent activators of tumor-reactive CTL.
Our results are in accord with previous findings that antigenic peptides can be rapidly degraded in human serum by proteases (17, [␤A1, ␣MeL8]-Melan-A (p )
2
18, 20, 24). The analysis of antigenic peptide persistence in serum [␤E1, ␣MeL8]-Melan-A (p )
2
using a Melan-A-specific CTL recognition assay indicates that the [␤D1, ␣MeL8]-Melan-A (p )
2
nonamer AAGIGILTV is degraded more rapidly than the two Melan-A decamers. This observation could be ex- plained in part by the ability of the Melan-A a Melan-A-specific cells (A2/ELA-tetramerϩCD8ϩ) were directly detected 7 days CTL used in the test to recognize shorter peptides that could be after the second stimulation by flow cytometry by costaining of the culture with intermediate degradation products of the decamers. Indeed, the A2/ELAtetramersPE and CD8FITC mAb (see Materials and Methods for more details).
nonapeptides AAGIGILTV and LAGIGILTV resulting from re- Analogs displaying an efficiency higher than that of the parental Melan-A moval of the first amino acid from Melan-A b PBMC from three different healthy donors (HD224, HD410, HD220) were stim- 35A27L, respectively, are efficiently recognized by the Melan-A- ulated in vitro with the indicated peptide as described in Materials and Methods. specific CTL used in the degradation test (data not shown) (10). Thislimitation was overcome by the analytical on-line HPLC-ESI/MS ap- Discussion
proach (19). We confirmed the initial observation indicating that the Improvement of immunogenicity of TDAP is critical for the de- Melan-A nonamer is degraded more rapidly than the two decamers.
velopment of efficient cancer vaccines. Introducing favorable an- More importantly, we identified and quantified the degradation frag- chor residue in the peptide sequence to increase peptide affinity for ments leading to outlining the exact mechanism of Melan-A peptide MHC is a first step to improved immunogenicity (36). Neverthe- degradation and identification of protease-sensitive bonds within the FIGURE 5.
Antigenic specificity of, and tumor recognition by, CTL induced with the nonnatural Melan-A analogs. After in vitro stimulation of PBMC from healthy donor HD224 with the Melan-A peptides, the tetramerϩCD8ϩ lymphocyte population was sorted on a FACSVantage cell sorter (BDBiosciences). Cell sorting of the tetramerϩCD8ϩ population and subsequent in vitro expansion of lymphocytes were realized for all stimulations with thedifferent analogs under the same conditions. Cytolytic CTL activity was measured after in vitro expansion of the different CTL populations (see Materialsand Methods). A, Dot plot representation of cytometry analysis of culture after in vitro stimulation with Melan-A analogs. The gate defined for cell sorting of tetramerϩCD8ϩ population is represented. B, Melan-A peptide recognition by the sorted tetramerϩCD8ϩ CTLfrom the corresponding gate shown in A (‚, Melan-A A27L; F, Melan-A doubly modified analog). The chromium release assay was performed with T2 cells as target cells pulsed with increasing peptide concentrations and at an E:T cell ratio of 10:1. C, Tumor cellrecognition by the sorted tetramerϩCD8ϩ CTL from the corresponding gate shown in A. A cytolytic assay was performed with 51Cr-labeled cells (Mel275and Mel290: HLA-A*0201ϩ, Melan-Aϩ; Na8: HLA-A*0201ϩ, Melan-AϪ) with increasing E:T cell ratios. f, Mel275; F, Mel290; E, Na8; ϩ, Na8 andMelan-A peptide).
peptide sequence. This step is of critical importance for introducing servation has been reported. To further assess the immunogenicity minimal, but efficient, structural modifications in the tumor antigenic of the nonnatural Melan-A peptides, we performed in vitro stim- sequence. As we previously showed for other human or murine tumor ulation of PBMC from healthy donors with the modified peptide antigenic peptides (19, 24), the Melan-A peptides are very sensitive to analogs. We used A2/Melan-A tetramer to quantify the number of exopeptidases in serum. Thus, this mechanism of peptide degradation Melan-A-specific T cells elicited after stimulation with the pro- involving both amino- and dipeptidyl-carboxy-peptidases seems to be tected peptides. This analysis shows that stimulation with doubly generic for antigenic peptide degradation in serum. In addition, be- substituted, fully protected analogs can give higher numbers of cause some doubly protected analogs were extremely stable in serum, Melan-A-specific T cells compared with those obtained by stim- we can deduce that the peptidic bonds left unprotected in the antigenic ulation with the parental peptide Melan-A sequence are not affected by endopeptidase activity. This finding is in sults thus indicate that protection against proteolysis of antigenic agreement with previous observations indicating that endopeptidases peptides could significantly enhance in vitro immunogenicity.
are not involved in the degradation of short peptides in serum (17, 37).
They further show the unique immunological properties of these Melan-A nonamer and decamers show marked quantitative and qual- itative differences in their degradation properties. However, the dif- ferences in the half-life of the Melan-A peptides could not be ex- of Melan-A-expressing tumors, the CTL elicited by stimulation plained solely by differences in peptide length, because the decamer A27L efficiently cross-recognize the endog- E26A (AAAGIGILTV) displayed a kinetic of degra- enously expressed Melan-A sequences (10, 11). We show in the , 15 min; data not shown) closer to that of Melan-A present study that CTL elicited by the nonnatural Melan-A analogs than that of the two Melan-A decamers. Thus, the nature of the first are also cross-reactive with the parental and endogenously ex- amino-terminal residues that greatly influence peptide susceptibility to pressed Melan-A peptides at the tumor cell surface. We thus val- amino-terminal degradation (37) seems to govern the kinetics of deg- idate and extend the tumor Ag cross-reactivity properties to the radation of the Melan-A peptides. Knowledge of the exact mechanism of peptide degradation was indeed of critical importance for rationally Antigenic peptide stability to protease degradation could have and efficiently designing nonnatural Melan-A analogs resistant to pro- important implications to elicit an efficient immune response. In tease degradation. To obtain fully protected peptides, the two pro- peptide-based cancer vaccines, after injection, the peptides have to tease-sensitive bonds (E1-L2 and L8-T9) of the Melan-A be loaded onto professional APCs such as DCs. Under activation, sequence needed to be modified. However, the doubly modified pep- DCs overexpress not only MHC, CD40, and CD80, but also CD13 tides did not show equal resistance to proteolysis. In particular, C- and CD26, molecules that are amino- and dipeptidyl-peptidases, terminal amidated analogs showed the weakest resistance and were respectively (40 – 42). Induction of such ectopeptidases on the sur- about twice as unstable as ␣-methylated ones. We may assume that face of activated DCs could increase the extracellular degradation this moderate protective effect is due to the capacity of a dipeptidyl- process of exogenous antigenic peptide and reduce the peptide carboxy-peptidase, such as angiotensin-converting enzyme, to cleave loading onto MHC molecules. If peptides are degraded before pre- C-terminal amidated substrates (38). Because studies have shown that sentation by DCs, repeated administration and/or high doses of peptide stability is similar in FCS and in the mouse (18, 37, 39), we antigenic peptide are then required for immunization (43, 44).
may reasonably postulate that the doubly protected analogs resistant Thus, the effective dose of a degradable peptide cannot be exactly to human serum proteases would also be protected against proteolysis estimated from the dose injected. Yet the concentration of peptide in vivo. However, in most antitumor vaccination protocols, the anti- used to elicit efficient antitumor-specific CTL is a critical param- genic peptide is delivered either by direct s.c. injection or after pulsing eter (45). Indeed, high-avidity CTL, which are highly efficient for onto dendritic cells (DCs). The biological significance of studying antitumor response, were only induced in vitro and in vivo with peptide degradation by serum proteases may then reasonably be ques- DCs pulsed with low concentrations of peptide (46 – 49). There- tioned. Interestingly, both the serum proteases that degrade tumor Ags fore, the quality of a peptide vaccine is probably determined not (Refs. 19 and 24 and this study), those expressed by human DCs (40) only by the density, but also by the duration, of the presence of able to degrade synthetic class I peptides (41), and those expressed by peptide at the injection site for loading onto activated DCs in vivo.
T cells present in the skin and the afferent lymph (42), such as CD13 Thus, peptide protection against protease degradation appears to be and CD26, display amino- and dipeptidyl-peptidase activities. We an attractive and effective way to control the concentration and may thus assume that modified peptides that resist serum proteases local persistence of an antigenic peptide after injection. In this probably resist peptidases expressed by DCs and remain stable after context, the use of peptides protected against degradation should s.c. injection. Nevertheless, direct determination of the fate and sta- then allow the optimization of immunization protocols for cancer bility of antigenic peptides in the interstitial space of the dermis and/or vaccine. However, concerns may be raised about possible negative effects that injection of nondegradable peptides may have. In par- Modifications of the peptide structure leading to resistance to ticular, it has been reported that due to persistence, diffusion of proteolysis may alter both the affinity of peptide-MHC binding and such peptides into the systemic circulation could induce peripheral the efficiency of CTL recognition. The impact on peptide-MHC immune tolerance (50). A recent study has also shown pharmaco- binding and CTL recognition clearly differed among the structural kinetic differences between a T cell-tolerizing and a T cell-acti- modifications. As previously described, we show that peptide bond vating peptide (51). Indeed, knowledge of the pharmacokinetics of alteration is deleterious for peptide-MHC binding (20, 24, 27). The peptides is a prerequisite before their possible use in peptide-based results presented here indicate that amino acid modifications such immunotherapy. Pharmacokinetic studies of the peptidase-resis- as ␣-methylation, N-hydroxylation, or ␤-amino acids are among tant Melan-A peptides are thus currently underway in our labora- the most efficient to protect against proteolysis and yet cause min- tories. In addition, the demonstration of increased in vitro immu- imal reduction of the peptide antigenicity. Interestingly, despite an nogenicity of the protected peptide should also be extended in vivo overall decrease in MHC presentation, some of the doubly pro- and validated. In particular, the correlation between high protease tected analogs sensitized target cells to lysis by the Melan-A-spe- resistance and increased immunogenicity remains to be addressed cific CTL with similar or even higher efficacy than the parental in vivo. Immunization of transgenic HLA-A2/Kb mice with dif- peptide. To our knowledge, this is the first time that such an ob- ferent kinds of adjuvant or several doses of peptide could allow an A NEW GENERATION OF MELAN-A/MART-1 Ag DERIVATIVES evaluation of the benefit of the increased resistance to proteolysis high performance liquid chromatography/electrospray ionization mass spectrom- on immunogenicity of antigenic peptides. These experiments are etry: application to the design of peptidase-resistant analogs. Rapid Commun.
Mass Spectrom. 12:557. currently being performed in our laboratories.
20. Brinckerhoff, L. H., V. V. Kalashnikov, L. W. Thompson, G. V. Yamshchikov, In conclusion, the results shown in the present study demon- R. A. Pierce, H. S. Galavotti, V. H. Engelhard, and C. L. Slingluff, Jr. 1999.
Terminal modifications inhibit proteolytic degradation of an immunogenic strate that protease-resistant, nonnatural tumor Ag derivatives can MART-1(27–35) peptide: implications for peptide vaccines. Int. J. Cancer 83: be highly immunogenic and potent activators of melanoma-spe- cific CTL. They may represent promising new tools for molecular 21. Valmori, D., J. F. Romero, Y. Men, J. L. Maryanski, P. Romero, and G. Corradin.
1994. Induction of a cytotoxic T cell response by co-injection of a T helper peptide and a cytotoxic T lymphocyte peptide in incomplete Freund’s adjuvant(IFA): further enhancement by pre-injection of IFA alone. Eur. J. Immunol. 24: Acknowledgments
22. Powell, M. F., T. Stewart, L. J. Otvos, L. Urge, F. Gaeta, A. Sette, T. Arrhenius, We thank Immanuel Luescher for invaluable advice and support with the D. Thomson, K. Soda, and S. M. Colon. 1993. Peptide stability in drug devel- use of tetramers and Pascal Batard and Sophie Millot for technical opment. II. Effect of single amino acid substitution and glycosylation on peptide reactivity in human serum. Pharm. Res. 10:1268. 23. Mayer, U., K. Zimmermann, K. Mann, D. Reinhardt, R. Timpl, and R. Nischt.
References
1995. Binding properties and protease stability of recombinant human nidogen.
Eur. J. Biochem. 227:681. 1. Boon, T., J. C. Cerottini, B. Van den Eynde, P. van der Bruggen, and A. Van Pel.
24. Ayyoub, M., H. Mazarguil, B. Monsarrat, B. Van den Eynde, and J. E. Gairin.
1994. Tumor antigens recognized by T lymphocytes. Annu. Rev. Immunol. 12: 1999. A structure-based approach to designing non-natural peptides that can ac- tivate anti-melanoma cytotoxic T cells. J. Biol. Chem. 274:10227. 2. Kawakami, Y., and S. A. Rosenberg. 1997. Human tumor antigens recognized by 25. Stemmer, C., A. Quesnel, A. Prevost-Blondel, C. Zimmermann, S. Muller, J. P. Briand, and H. Pircher. 1999. Protection against lymphocytic choriomenin- 3. Hofbauer, G. F., J. Kamarashev, R. Geertsen, R. Boni, and R. Dummer. 1998.
gitis virus infection induced by a reduced peptide bond analogue of the H-2Db- Melan A/MART-1 immunoreactivity in formalin-fixed paraffin-embedded pri- restricted CD8ϩ T cell epitope GP33. J. Biol. Chem. 274:5550. mary and metastatic melanoma: frequency and distribution. Melanoma Res. 26. Maillere, B., G. Mourier, M. Herve, and A. Menez. 1995. Fine chemical modi- fications at N- and C-termini enhance peptide presentation to T cells by increas- 4. Kawakami, Y., S. Eliyahu, C. H. Delgado, P. F. Robbins, K. Sakaguchi, ing the lifespan of both free and MHC-complexed peptides. Mol. Immunol. 32: E. Appella, J. R. Yannelli, G. J. Adema, T. Miki, and S. A. Rosenberg. 1994.
Identification of a human melanoma antigen recognized by tumor-infiltrating 27. Calbo, S., G. Guichard, P. Bousso, S. Muller, P. Kourilsky, J. P. Briand, and lymphocytes associated with in vivo tumor rejection. Proc. Natl. Acad. Sci. USA J. P. Abastado. 1999. Role of peptide backbone in T cell recognition. J. Immunol. 5. Pittet, M. J., D. Valmori, P. R. Dunbar, D. E. Speiser, D. Lienard, F. Lejeune, 28. Hudrisier, D., and J. E. Gairin. 1998. Peptide-major histocompatibility complex K. Fleischhauer, V. Cerundolo, J. C. Cerottini, and P. Romero. 1999. High fre- class I complex: from the structural and molecular basis to pharmacological prin- quencies of naive Melan-A/MART-1-specific CD8ϩ T cells in a large proportion ciples and therapeutic applications. Curr. Top. Microbiol. Immunol. 232:75. of human histocompatibility leukocyte antigen (HLA)-A2 individuals. J. Exp. 29. Garcia, K. C., L. Teyton, and I. A. Wilson. 1999. Structural basis of T cell recognition. Annu. Rev. Immunol. 17:369. 6. Romero, P., N. Gervois, J. Schneider, P. Escobar, D. Valmori, C. Pannetier, 30. Salter, R. D., D. N. Howell, and P. Cresswell. 1985. Genes regulating HLA class A. Steinle, T. Wolfel, D. Lienard, V. Brichard, et al. 1997. Cytolytic T lympho- I antigen expression in T-B lymphoblast hybrids. Immunogenetics 21:235. cyte recognition of the immunodominant HLA-A*0201-restricted Melan-A/MART-1 antigenic peptide in melanoma. J. Immunol. 159:2366. 31. Fehrentz, J. A., and D. Castro. 1983. An efficient synthesis of optically active 7. Dunbar, P. R., C. L. Smith, D. Chao, M. Salio, D. Shepherd, F. Mirza, M. Lipp, -(t-butyloxycarbonylamino)-aldehydes from ␣-amino acids. Synthesis 8:676.
A. Lanzavecchia, F. Sallusto, A. Evans, et al. 2000. A shift in the phenotype of 32. Bianco, A., C. Zabel, P. Walden, and G. Jung. 1998. N-hydroxy-amide analogues Melan-A-specific CTL identifies melanoma patients with an active tumor-specific of MHC-class I peptide ligands with nanomolar binding affinities. J. Pept. Sci. immune response. J. Immunol. 165:6644. 8. Romero, P., P. R. Dunbar, D. Valmori, M. Pittet, G. S. Ogg, D. Rimoldi, 33. Du¨rr, H., M. Goodman, and J. Gu¨nther. 1992. Retro-inverso amide bonds be- J. L. Chen, D. Lienard, J. C. Cerottini, and V. Cerundolo. 1998. Ex vivo staining tween trifunctional amino acids. Angew. Chem. Int. Ed. Engl. 31:785. of metastatic lymph nodes by class I major histocompatibility complex tetramers 34. Guichard, G., F. Connan, R. Graff, M. Ostankovitch, S. Muller, J. G. Guillet, reveals high numbers of antigen-experienced tumor-specific cytolytic T lympho- J. Choppin, and J. P. Briand. 1996. Partially modified retro-inverso pseudopep- tides as non-natural ligands for the human class I histocompatibility molecule 9. Rivoltini, L., P. Squarcina, D. J. Loftus, C. Castelli, P. Tarsini, A. Mazzocchi, F. Rini, V. Viggiano, F. Belli, and G. Parmiani. 1999. A superagonist variant of 35. Guichard, G., A. Zerbib, F. A. Le Gal, J. Hoebeke, F. Connan, J. Choppin, peptide MART1/Melan A27–35 elicits anti-melanoma CD8ϩ T cells with en- J. P. Briand, and J. G. Guillet. 2000. Melanoma peptide MART-1(27–35) ana- hanced functional characteristics: implication for more effective immunotherapy.
logues with enhanced binding capacity to the human class I histocompatibility molecule HLA-A2 by introduction of a ␤-amino acid residue: implications for 10. Valmori, D., J. F. Fonteneau, C. M. Lizana, N. Gervois, D. Lienard, D. Rimoldi, recognition by tumor-infiltrating lymphocytes. J. Med. Chem. 43:3803. V. Jongeneel, F. Jotereau, J. C. Cerottini, and P. Romero. 1998. Enhanced gen- 36. Parkhurst, M. R., M. L. Salgaller, S. Southwood, P. F. Robbins, A. Sette, eration of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1 S. A. Rosenberg, and Y. Kawakami. 1996. Improved induction of melanoma- immunodominant peptide analogues. J. Immunol. 160:1750. reactive CTL with peptides from the melanoma antigen glycoprotein 100 mod- 11. Men, Y., I. Miconnet, D. Valmori, D. Rimoldi, J. C. Cerottini, and P. Romero.
ified at HLA-A*0201-binding residues. J. Immunol. 157:2539. 1999. Assessment of immunogenicity of human Melan-A peptide analogues in 37. Powell, M. F., H. Grey, F. Gaeta, A. Sette, and S. Colon. 1992. Peptide stability HLA-A*0201/Kb transgenic mice. J. Immunol. 162:3566. in drug development: a comparison of peptide reactivity in different biological 12. Valmori, D., M. J. Pittet, D. Rimoldi, D. Lienard, R. Dunbar, V. Cerundolo, F. Lejeune, J. C. Cerottini, and P. Romero. 1999. An antigen-targeted approach 38. Araujo, M. C., R. L. Melo, M. H. Cesari, M. A. Juliano, L. Juliano, and to adoptive transfer therapy of cancer. Cancer Res. 59:2167. A. K. Carmona. 2000. Peptidase specificity characterization of C- and N-terminal 13. Kugler, A., G. Stuhler, P. Walden, G. Zoller, A. Zobywalski, P. Brossart, catalytic sites of angiotensin I-converting enzyme. Biochemistry 39:8519. U. Trefzer, S. Ullrich, C. A. Muller, V. Becker, et al. 2000. Regression of human 39. Calbo, S., G. Guichard, S. Muller, P. Kourilsky, J. P. Briand, and J. P. Abastado.
metastatic renal cell carcinoma after vaccination with tumor cell-dendritic cell 2000. Antitumor vaccination using a major histocompatibility complex (MHC) class I-restricted pseudopeptide with reduced peptide bond. J. Immunother. 23: 14. Jaeger, E., H. Bernhard, P. Romero, M. Ringhoffer, M. Arand, J. Karbach, C. Ilsemann, M. Hagedorn, and A. Knuth. 1996. Generation of cytotoxic T-cell 40. Hart, D. N. 1997. Dendritic cells: unique leukocyte populations which control the responses with synthetic melanoma-associated peptides in vivo: implications for primary immune response. Blood 90:3245. tumor vaccines with melanoma-associated antigens. Int. J. Cancer 66:162. 41. Amoscato, A. A., D. A. Prenovitz, and M. T. Lotze. 1998. Rapid extracellular 15. Offringa, R., S. H. van der Burg, F. Ossendorp, R. E. Toes, and C. J. Melief. 2000.
degradation of synthetic class I peptides by human dendritic cells. J. Immunol. Design and evaluation of antigen-specific vaccination strategies against cancer.
42. Yawalkar, N., R. E. Hunger, W. J. Pichler, L. R. Braathen, and C. U. Brand. 2000.
16. Melief, C. J., R. Offringa, R. E. Toes, and W. M. Kast. 1996. Peptide-based Human afferent lymph from normal skin contains an increased number of mainly cancer vaccines. Curr. Opin. Immunol. 8:651. memory/effector CD4ϩ cells expressing activation, adhesion and co-stimulatory 17. Falo, L. D. J., L. J. Colarusso, B. Benacerraf, and K. L. Rock. 1992. Serum molecules. Eur. J. Immunol. 30:491. proteases alter the antigenicity of peptides presented by class I major histocom- 43. Salgaller, M. L., F. M. Marincola, J. N. Cormier, and S. A. Rosenberg. 1996.
patibility complex molecules. Proc. Natl. Acad. Sci. USA 89:8347. Immunization against epitopes in the human melanoma antigen gp100 following 18. Widmann, C., J. L. Maryanski, P. Romero, and G. Corradin. 1991. Differential patient immunization with synthetic peptides. Cancer Res. 56:4749. stability of antigenic MHC class I-restricted synthetic peptides. J. Immunol. 147: 44. Cormier, J. N., M. L. Salgaller, T. Prevette, K. C. Barracchini, L. Rivoltini, N. P. Restifo, S. A. Rosenberg, and F. M. Marincola. 1997. Enhancement of 19. Ayyoub, M., B. Monsarrat, H. Mazarguil, and J. E. Gairin. 1998. Analysis of the cellular immunity in melanoma patients immunized with a peptide from MART- degradation mechanisms of MHC class I-presented tumor antigenic peptides by 1/Melan A. Cancer J. Sci. Am. 3:37. 45. Toes, R. E., R. J. Blom, R. Offringa, W. M. Kast, and C. J. Melief. 1996. En- 48. Alexander-Miller, M. A., G. R. Leggatt, and J. A. Berzofsky. 1996. Selective hanced tumor outgrowth after peptide vaccination: functional deletion of tumor- expansion of high- or low-avidity cytotoxic T lymphocytes and efficacy for adop- specific CTL induced by peptide vaccination can lead to the inability to reject tive immunotherapy. Proc. Natl. Acad. Sci. USA 93:4102. 49. Bullock, T. N., T. A. Colella, and V. H. Engelhard. 2000. The density of peptides 46. Zeh, H. J., 3rd, D. Perry-Lalley, M. E. Dudley, S. A. Rosenberg, and J. C. Yang.
displayed by dendritic cells affects immune responses to human tyrosinase and 1999. High avidity CTLs for two self-antigens demonstrate superior in vitro and glycoprotein 100 in HLA-A2 transgenic mice. J. Immunol. 164:2354. in vivo antitumor efficacy. J. Immunol. 162:989.
50. Aichele, P., K. Brduscha-Riem, R. M. Zinkernagel, H. Hengartner, and 47. Gervois, N., N. Labarriere, S. Le Guiner, M. C. Pandolfino, J. F. Fonteneau, H. Pircher. 1995. T cell priming versus T cell tolerance induced by synthetic Y. Guilloux, E. Diez, B. Dreno, and F. Jotereau. 2000. High avidity melanoma- reactive cytotoxic T lymphocytes are efficiently induced from peripheral blood 51. Weijzen, S., S. C. Meredith, M. P. Velders, A. G. Elmishad, H. Schreiber, and lymphocytes on stimulation by peptide-pulsed melanoma cells. Clin. Cancer Res. W. M. Kast. 2001. Pharmacokinetic differences between a T cell-tolerizing and a T cell-activating peptide. J. Immunol. 166:7151.

Source: http://www.blanchet.wyenet.co.uk/Blanchet_JI2001.pdf