The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Léonetti, M.
Right arrow Articles by Ménez, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Léonetti, M.
Right arrow Articles by Ménez, A.
The Journal of Immunology, 1998, 160: 3820-3827.
Copyright © 1998 by The American Association of Immunologists

Increasing Immunogenicity of Antigens Fused to Ig-Binding Proteins by Cell Surface Targeting

Michel Léonetti1, Robert Thai, Joël Cotton, Sandrine Leroy, Pascal Drevet, Frédéric Ducancel, Jean Claude Boulain and André Ménez

Département d’Ingéniérie et d’Études des Protéines (DIEP) C. E. Saclay, Gif-Sur-Yvette, France


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Fusion of antigenic proteins to Ig-binding proteins such as protein A from Staphylococcus aureus and its derived ZZ fragment is known to increase immunogenicity of the fused Ag in vivo. To shed light on the origin of this effect, we used snake toxins as Ags and observed that 1) fusion of toxins to ZZ enhanced their presentation to a toxin-specific T cell hybridoma (T1B2), using A20 B lymphoma cells, splenocytes, or peritoneal exudate cells as APCs; 2) this enhancement further increased when the number of fused Ig-binding domains varied from two with ZZ to five with protein A; and 3) the phenomenon vanished when the fusion protein was preincubated with an excess of free ZZ or when P388D1 monocytes cells were used as APCs. Therefore, ZZ-fused toxins are likely to be targeted to surface Igs of APCs by their ZZ moiety. Furthermore, ZZ-{alpha} and toxin {alpha} stimulated similar profiles of toxin-specific T cells in BALB/c mice, suggesting a comparable processing and presentation in vivo for both toxin forms. To improve the targeting efficiency, ZZ-{alpha} was noncovalently complexed to various Igs directed to different cell surface components of APCs. The resulting complexes were up to 103-fold more potent than the free toxin at stimulating T1B2. Also, they elicited both a T cell and an Ab response in BALB/c mice, without the need of any adjuvant. This simple approach may find practical applications by increasing the immunogenicity of recombinant proteins without the use of adjuvant.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Improvement of immunogenicity of a protein Ag may be achieved by an appropriate manipulation of the molecular and/or cellular events that govern the immune response. In the case of a thymus-dependent Ag, the associated humoral response requires induction of both B and T helper cells (1). Stimulation of B cells results from interactions between the Ag and Ag-specific membrane Igs, and hence, it depends on both the affinity and local concentration of the two proteins. T cell stimulation follows more complex rules, depending on the efficiency of Ag capture by APC (2), efficiency of processing (3), affinity of processed peptides for the class II molecule (4), and stability of the peptide-class II complex (5).

Among the potentially interesting approaches that may increase the immunogenicity of a protein Ag is the possibility of fusing it with an appropriate moiety, such as Staphylococcus aureus protein A or some of its derived fragments. Protein A and its ZZ fragment, which can bind Igs, are often used as fusion moieties to facilitate production and purification of recombinant proteins produced in Escherichia coli (6, 7, 8). In addition, however, Ags fused to protein A or ZZ were observed to be substantially more immunogenic than the unfused Ags under in vivo conditions (9, 10, 11). Previous experiments performed with ZZ-fused peptides showed that this phenomenon is associated with an enhancement of T cell presentation and only suggested that it results from targeting the fused ZZ to Igs borne by APCs (12, 13). However, no experimental evidence has been brought to support the validity of this hypothesis. Such a proposal, however, is all the more interesting because targeting of a protein Ag to appropriate immune cells using anti-cell surface structure Abs was previously shown to dramatically increase the presentation efficiency of various Ags in vitro (14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24), and such an enhancement correlates well with an increase of the in vivo immune response (16, 25).

The aim of this work was twofold. First, we investigated whether the in vitro T cell presentation of an antigenic protein is increased by its fusion to ZZ or protein A, and if this is the case, whether this effect is related to a targeting via cell surface Igs. The antigenic proteins selected for this study are snake short-chain toxins of which the immunologic properties have been extensively studied (26). These toxins are small proteins of 60 to 62 residues with four disulfides, which block the nicotinic acetylcholine receptor with high affinity and great selectivity (27). Second, we explored the possibility of targeting Ags to APCs more efficiently, using a ZZ-toxin conjugate noncovalently complexed to Abs directed toward different cell surface components. The data presented in this study not only offer an explanation as to how protein A and its ZZ-derived fragments increase the immunogenicity of a fused Ag, but they also provide a simple and efficient approach to increasing immunogenicity of recombinant proteins without the need of any adjuvant.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Proteins and synthetic peptides

Erabutoxin a and toxin {alpha} were purified from Laticauda semifasciata and Naja nigricollis venom (Institut Pasteur, Paris), respectively, as previously described (28). Purity of the toxin was assessed by both reverse phase HPLC and isoelectric focusing. Biorex 70 and Bio-Gel P2 were obtained from Bio-Rad (Richmond, CA).

We synthesized five peptides encompassing the toxin {alpha} polypeptide chain with free cysteines, since we have previously shown that toxin {alpha} contains thiol-dependent epitopes (29, 30). The peptides 1–25SH, 15–30SH, 24–41SH, 32–49SH, and 39–62SH were synthesized using an Applied Biosystems (Foster City, CA) 430A peptide synthesizer. The Boc/benzyl strategy was followed using a p-methyl-benzydrylamine resin (0.77 meq/g). During synthesis, cysteines were protected with p-methyl-benzyl (pMeBzl) groups. Peptides were synthesized using 1-hydroxybenzotriozole (HOBT) esters. The peptides were cleaved from the resin and the side chain-protecting moieties were removed with anhydrous hydrofluoric acid (HF). Subsequently, the peptides were precipitated with ether, solubilized with 10% acetic acid, and lyophilized. After filtration through a Bio-Gel P2 column, the crude materials were purified to homogeneity by reverse phase HPLC on a C18 reverse phase column. The peptides were kept in a nitrogen atmosphere to avoid cysteine reoxidation.

Construction, expression, and purification of fusion proteins

The cDNA encoding erabutoxin a (Ea)2 from the sea snake L. semifasciata has been cloned and expressed as a fusion protein with either the complete protein A from S. aureus (6) or a double Ig-binding domain derived from protein A, called ZZ (7, 9). The experimental procedures followed to construct and express these two hybrids have been previously described in detail (11, 31). The cDNA encoding toxin {alpha} was fused to the gene that encodes ZZ, using a 195-bp synthetic gene built from 10 oligomers ranging in size from 26 to 61 nucleotides, according to standard procedures (32). Expression/secretion vectors were purchased from Pharmacia (Uppsala, Sweden). The fusion proteins were purified as described (11, 31). ZZ-{alpha} was homogeneously labeled with 14C by growing the transformed bacteria in a medium containing 14C-labeled glucose. The radioactive hybrid was purified as described in previous studies (11, 31).

Immunization in mice

To assess the in vivo T cell response to toxin {alpha} or ZZ-{alpha}, BALB/c mice (IFFA CREDO, Lyon, France) were injected at the base of the tail with 100 µl of a CFA emulsion containing 0.5 nmol of either toxin {alpha} or ZZ-{alpha}. Spleens were harvested 10 days after immunizing the mice. To assess the in vivo immune response to ZZ-{alpha}, BALB/c mice were similarly injected at the tail base with 100 µl of a 0.1 M phosphate buffer solution, pH 7.2, containing 0.1 nmol of ZZ-{alpha} or ZZ-{alpha} previously complexed with different Abs. Blood samples were taken 7, 14, and 26 days after the immunization. For T cell experiments, spleens were harvested 36 days after immunization.

T cell-stimulating assay

T cell experiments were performed in a medium containing no FCS; the synthetic medium used was DCCM1 (Biological Industries, Beit Haemek, Israel). For experiments made with T cell hybridoma, the different Ags were serially diluted in microculture wells, and 5.104 T1B2 hybridomas were added per well with 5.104 A20 or P388D1, 2.104 peritoneal exudate cells, or 5.105 splenocytes. Cells were cultured for 24 h at 37°C in a humidified 7% CO2 atmosphere. P388D1 cells were previously cultured for 2 days in the presence of IFN-{gamma} to induce their APC functions (33).

Bulk cultures of cells from immunized mice were made according to a procedure derived from a protocol described previously (34). Spleens were harvested and suspended in a proliferation medium. Cells (106/well) were cultured for 24 h at 37°C in a humidified 7% CO2 atmosphere with serial dilutions of the different Ags. The presence of IL-2 in culture supernatants from T cell hybridoma and bulk cultures was determined by measuring the proliferation of an IL-2-dependent cytotoxic T cell line (CTLL), using methyl [3H]thymidine ([3H]TdR; 5 Ci/mmol, Saclay, Commissariat à l’Energie Atomique, CEA France). The data are expressed in cpm.

Quantitative association of 14C-labeled ZZ-{alpha} with A20 cells

14C-labeled ZZ-{alpha} was diluted in DCCM1 medium (4.5 x 103 cpm/50 µl) and preincubated for 1 night at 4°C in the presence or absence of RAM-IgG-F(ab')2 (0.1 µM final concentration). The mixtures were added to wells containing 5 x 105 A20 cells and incubated for 2 h at 37°C. Cells were then centrifuged and washed two times with cold HBSS. The radioactivity associated with cells was counted.

Ab titration by ELISA

Microtiter ELISA plates were coated overnight with the native toxin (0.1 µg/well) in 0.05 M phosphate buffer, pH 7.4, at 4°C, then saturated with a 0.1 M phosphate buffer, pH 7.4, containing 0.3% BSA. Antisera were serially diluted in the same buffer containing 0.1% BSA and incubated overnight at 4°C. Binding of Abs was assessed using a goat anti-mouse peroxidase conjugate and 2,2'-azino-bis(3-ethylbenz-thiazoline-6-sulfonic acid) (ABTS) as described above. The titers were defined as the highest serum dilution giving an absorbance value of 0.6 above the negative control. For this control, we used pooled sera collected before immunization of the mice.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Presentation of ZZ-Ea and Ea to T1B2 hybridoma by different APCs

Protein A as well as its derived fragment ZZ (6, 7) are well known proteins that bind Igs. They are frequently used as fusion moieties to produce foreign proteins in the periplasm of E. coli (7, 8, 9, 10), including small proteins rich in disulfide bonds such as the curaremimetic toxin erabutoxin a (Ea) from the sea snake L. semifasciata (11). Also, the IgG-binding property of protein A or its derived fragments was exploited to purify recombinantly fused Ag on IgG affinity chromatography (11). Furthermore, the presence of an Ig- binding protein fused to an Ag was shown to increase its natural immunogenicity upon intradermal injection into rabbits (11, 31). As expected from previous studies made with peptides (12), we show here that a ZZ-fused protein Ag is more efficient than the unfused protein Ag at stimulating an Ag-specific T hybridoma. We used Ea as the Ag fused to ZZ and T1B2 as a toxin-specific murine T hybridoma. T1B2 (29) was previously prepared against toxin {alpha} from N. nigricollis, another curaremimetic toxin in which amino acid sequence differs from that of Ea by 17 amino acids (29). As shown in Figure 1Go, ZZ-Ea (Fig. 1Go, A and B) as well as ZZ-{alpha} (Fig. 1Go, C and D) are more potent than the corresponding unfused toxins at stimulating T1B2, using A20 cells or splenocytes as APCs. Toxin {alpha}, however, appeared to be slightly less efficient than Ea at stimulating T1B2, at least when A20 cells are used as APCs. Previously, it was reported that ZZ can enhance the capacity of small peptides to stimulate T cells (12, 13). Here, we bring evidence that a stimulating effect occurs also with small disulfide-rich proteins.



View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 1. Stimulation of T1B2 and T1C9 by ZZ-Ea, ZZ-{alpha}, Ea, and toxin {alpha} using A20 cells or splenocytes as APCs. Serial dilutions of Ag were incubated in wells with either T1B2 or T1C9 (5 x 104 hybridoma/well) and APCs for 24 h at 37°C. A, C, and E, 5 x 104 A20 cells/well were used as APCs B, D, and F, 5 x 105 splenocytes were added to the wells. IL-2 secretion was determined by CTLL assay.

 
The role of ZZ in the increased capacity of ZZ-Ea to stimulate T1B2

Stimulation of T cells, including T1B2, by toxin {alpha} was previously shown to require an appropriate processing of the toxin by APCs (29). More precisely, stimulation of T1B2 was associated with presentation of a T cell epitope localized in the toxin fragment 32–49. Ea possesses an identical region, which is likely to account for the ability of Ea to stimulate T1B2 nearly as efficiently as toxin {alpha} using splenocytes as APCs (Fig. 1Go, B and D). That stimulation of T1B2 by Ea results from a processing of the toxin is deduced from an experiment made with fixed A20 cells, which are known to retain their presentation capacity but not their processing ability. As shown in Figure 2GoA, the fragment 32–49 can stimulate T1B2, whereas Ea cannot. Similarly, the stimulating capacity of ZZ-Ea (Fig. 1Go, A and B) requires a processing of the fusion protein, since this capacity is lost when fixed A20 cells are used as APCs. Therefore, processing of Ea as well as ZZ-Ea seems to be a prerequisite for the stimulation of T1B2.



View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 2. Role of ZZ in the presentation of ZZ-Ea to T1B2. A, 2 x 105 fixed A20 cells were added to wells containing serial dilutions of the different Ags and 5 x 104 T1B2 cells. B, 5 x 104 A20 cells were preincubated for 1 h at 4°C with or without free ZZ (10 µM) and were then added to the wells. C, 5 x 105 splenocytes were preincubated for 1 h at 4°C with or without free ZZ (2 µM) and were then added to wells. D, 5 x 104 P388D1 previously cultured for 2 days in the presence of IFN-{gamma} were used as APCs. E, 2 x 104 peritoneal exudate cells (PEC) were preincubated for 1 h at 4°C with or without free ZZ (2 µM) and were then added to wells. IL-2 secretion was determined by the CTLL assay.

 
How can the presence of the ZZ moiety explain the increased stimulating activity of fused Ea? A first possible explanation is that ZZ may trigger an Ag-independent IL-2 release from APCs. To investigate this possibility, we used another hybridoma that is stimulated by toxin {alpha} but not by Ea (29). We reasoned that if the fused ZZ would caused an Ag-independent IL-2 release, an enhanced stimulating effect of T1C9 should be triggered by both ZZ-Ea and ZZ-{alpha}. As shown in Figure 1Go, E and F, ZZ-{alpha} did stimulate T1C9, whereas ZZ-Ea did not, irrespective of whether it was presented by A20 cells or splenocytes. Therefore, we can conclude that ZZ elicits an Ag-dependent IL-2 release from APCs. Moreover, it is not the presence of ZZ by itself in the medium that causes the observed phenomenon, because unfused Ea presented by living A20 cells (Fig. 2GoB) or living splenocytes (Fig. 2GoC), in the presence or absence of a large excess (10 µM) of free ZZ, had comparable stimulating capacity. Furthermore, that the presence of large amounts of ZZ did not cause any increase in the presentation of Ea indicates that the ZZ moiety had no mitogenic activity. Therefore, the observed increase of immunogenicity of fused Ea is possibly associated with the covalent coupling of the toxin to ZZ.

While the presence of an excess of free ZZ had no effect on the stimulating property of Ea, it dramatically reduced the stimulating effect of ZZ-Ea, irrespective of whether the APCs were A20s or splenocytes (Fig. 2Go, B and C). Furthermore, under these conditions, the stimulation potency of ZZ-Ea was virtually equal to that of the unfused Ea. These findings indicate that an excess of free ZZ inhibits the capacity of ZZ-Ea to exert its increased presentation capacity. It is well documented that ZZ can bind to Igs (35). Our data suggest, therefore, that membrane Igs could be responsible for the increased immunogenicity of ZZ-Ea and that the excess of free ZZ simply prevents the binding of the fusion protein to them. To confirm this hypothesis, we performed similar experiments using surface Ig-free monocyte cells (P388D1). As shown in Figure 2GoD, unfused Ea and ZZ-Ea displayed similar stimulating potencies when presented by P388D1, indicating that in the absence of surface Igs, the macrophages are inadequate APCs to enhance the stimulating effect of ZZ-Ea. Could this result be interpreted to indicate that macrophages are generally unable to enhance the stimulating effect of ZZ-Ea? We anticipated a negative answer to this question, provided the macrophages were appropriately loaded with Igs via their Fc receptors. To investigate this possibility, we used peritoneal exudate cells, which are predominantly populated with macrophages, loaded in vivo with Igs. Using such APCs, an enhanced stimulating property was observed (Fig. 2GoE), and this effect was specifically associated with fused ZZ because it vanished in the presence of an excess of free ZZ. Clearly, populations of B cells and macrophages, at least, can efficiently enhance the stimulating effect of ZZ-Ea. Therefore, ZZ not only targets the fused toxin to a cell surface component, but this component can be Igs expressed by B cells and Igs loaded on professional APCs such as macrophages.

We then investigated whether introduction of a higher number of Ig-binding domains in the Ea-fused protein would increase the targeting efficiency and hence the presentation potency. Knowing that protein A comprises five Ig-binding domains, we synthesized a protein A-Ea fusion protein and investigated its stimulating potency toward T1B2 using A20s as APCs. The results shown in Figure 3Go revealed that, in agreement with our expectations, the protein A-Ea conjugate was substantially (~13-fold) more potent at stimulating T1B2 than ZZ-Ea, which in turn was ~8-fold more potent than Ea. In other words, the protein A-Ea conjugate was ~100-fold more immunogenic than unfused Ea. Again, as in the case of ZZ-Ea (Fig. 2Go, B and C), the increase in the stimulating potency of protein A-Ea was inhibited by a large excess of free ZZ. Therefore, enhancement in immunogenicity of Ea fused to either protein A or ZZ is likely to be associated with the recognition of Igs at the surfaces of APCs.



View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 3. Stimulation of T1B2 by ProtA-Ea, ZZ-Ea, and Ea using A20 cells as APCs. Serial dilutions of Ag were incubated in wells with T1B2 hybridoma and A20 cells for 24 h at 37°C. IL-2 secretion was determined by the CTLL assay.

 
The T cell response elicited in vivo by toxin {alpha} fused or unfused to ZZ

Although the ZZ-toxin conjugates were more potent than the unfused toxins at stimulating T cells in vitro, it was unclear as to whether or not the T epitopes that are generated by these distinct entities were the same in vivo. To address this question, we immunized BALB/c mice with either toxin {alpha} alone or ZZ-{alpha}, both mixed with Freund’s adjuvant. We then collected immunized spleens 10 days later and investigated the abilities of five toxin peptides to stimulate a T cell response. These peptides overlap each other and encompass the whole toxin sequence (peptides 1–25, 15–30, 24–41, 32–49, and 39–62). They were all maintained in a thiol-free form to avoid a lack of response to thiol-dependent epitopes, as observed previously (29, 30, 36, 37, 38). As shown in Figure 4Go (right) and in agreement with previous data (34), immunization with unfused toxin revealed that fragment 24–41 was even more potent at stimulating T cells than the toxin itself, whereas the four other fragments displayed substantially less or no stimulating capacity. Using ZZ-{alpha} as an immunogen, a similar pattern of T cell response was observed (Fig. 4Go, left). Therefore, in BALB/c mice, the immunodominant region of toxin {alpha} and ZZ-{alpha} is located between cysteines 24 and 41, with two subdominant regions located between residues 32 and 49 and residues 15 and 30. One may notice that the T cell response to the subdominant epitopes was not strictly identical in each case, the response to peptide 32–49 being somewhat higher when mice were immunized with the ZZ-{alpha}. Nevertheless, these data show that the presence of fused ZZ did not cause a major alteration in the profile of peptides that stimulated T cells, suggesting that in vivo the processing of the fused and unfused toxins are qualitatively comparable.



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 4. T cell response to ZZ-{alpha} and toxin {alpha}. BALB/c mice were injected with 0.5 nmol of either toxin {alpha} or ZZ-{alpha} emulsified with CFA. Ten days after the immunization, spleen cells were challenged with serial dilutions of native toxin, free ZZ, or the five overlapping peptides. After a culture period of 24 h, each supernatant was assayed for its capacity to stimulate incorporation of [3H]thymidine in the IL-2-dependent CTLL.

 
Further increase of T cell presentation by noncovalent association of a ZZ-fused toxin to Abs directed to cell surface components

The data presented above showed that the fusion of ZZ to a protein Ag 1) enhanced the T cell response of the fused protein in vitro; 2) did not virtually change the profile of epitopes that are generated during APC processing in vivo; and 3) targeted the Ag to a ZZ receptor, which are presumably Igs expressed on B cell surface and Igs bound to RFc of macrophages. To further enhance the immunogenicity of the fused toxin, we investigated the possibility of targeting it even more efficiently to APCs. We reasoned that complexing the genetic conjugate with an Ig that specifically recognizes a cell surface structure on the surface of APCs may selectively target the ZZ-fused toxin to APCs carrying such molecules. We therefore tentatively complexed the ZZ-{alpha} with the mouse IgG2a mAb 14-4-4S, which specifically recognizes the murine class II MHC I-Ed (39), a rabbit polyclonal anti-IgM Ab (RAMµ), or a rabbit polyclonal anti-IgG-F(ab')2 Ab (RAM-IgG-F(ab')2). Then, we investigated the capacity of these different complexes to stimulate T cell presentation under in vitro conditions. As shown in Figure 5Go, a remarkable enhancement of the T cell presentation was observed with these complexes. More precisely, using either splenocytes or A20s as APCs, 14-4-4S/ZZ-{alpha}, RAMµ/ZZ-{alpha}, and RAM-IgG-F(ab')2/ZZ-{alpha} were at least 10- and 103-fold more potent at stimulating T1B2 as compared with the fused ZZ-{alpha} and unfused toxin {alpha}, respectively (Fig. 5Go, A and B). A series of control experiments showed that the enhancing effect is provided by a targeting to the cell surface structure recognized by the Ab involved in the complex. In particular, no increase was observed with RAMµ/ZZ-{alpha} when A20 cells, which do not not carry surface IgM (40), were used as APCs (Fig. 5GoB), whereas a marked enhancement was seen using splenocytes (Fig. 5GoA). This simple comparison indicates that the Ab used in the targeting can discriminate between various APCs. As an additional control, the boosting effect was not provided by a nonspecific rabbit polyclonal IgG or an unrelated mouse IgG2a mAb (Fig. 5Go, A and B). Finally, no enhancement was observed when the unfused toxin was incubated with both free ZZ and the different Abs (Fig. 5Go, C and D). Altogether, these findings indicate that complexing the ZZ-{alpha} with anti-cell surface structure Abs is an efficient way to enhance the T cell presentation of toxin {alpha} in vitro.



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 5. Stimulation of T1B2 by fusion protein ZZ-{alpha}, ZZ + toxin {alpha}, and toxin {alpha} in the presence or absence of different Abs. Serial dilutions of the different proteins were preincubated in wells for 1 night at 4°C with or without mAb 14-4-4S, RAMµ, RAM-IgG-F(ab')2, rabbit IgG, or mouse IgG2a mAb (0.05 µM final concentration for each). T1B2 and APCs were then added. A and C, 5 x 105 splenocytes were used as APCs; B and D, 5 x 104 A20 cells. IL-2 secretion was determined by the CTLL assay.

 
Why is ZZ-{alpha} less efficient than RAM-IgG-F(ab')2/ZZ-{alpha}? There is no definitive explanation; however, one may anticipate that the proportion of both surface-bound and incorporated ZZ-{alpha} should be higher with RAM-IgG-F(ab')2/ZZ-{alpha}. To investigate this aspect, we incubated A20 cells with a constant amount of 14C-labeled ZZ-{alpha} incubated in the absence or presence of RAM-IgG-F(ab')2, and counted the radioactivity associated with the cells. The results of these experiments indicate that RAM-IgG-F(ab')2/14C-labeled ZZ-{alpha} caused a higher cell labeling than free 14C-labeled ZZ-{alpha}. Indeed, the radioactivity associated with 5 x 105 cells was equal to 375 ± 24 cpm for the complex and 138 ± 41 cpm for free 14C-labeled ZZ-{alpha} (respectively, 8 and 3% of the total radioactivity). Whether this observation reflects differential affinities of the constructs for their targets now remains to be elucidated.

Immune response elicited in vivo by ZZ-{alpha} noncovalently complexed to Abs

We investigated whether the noncovalent complexes could enhance the immune response to toxin {alpha} under in vivo conditions. The different Abs used in the above in vitro experiments were respectively incubated with ZZ-{alpha} and injected into BALB/c mice. Immunizations were performed in the absence of adjuvant as suggested by previous targeting experiments made in vivo with Ags chemically coupled to MHC-specific Abs (41). When RAMµ, RAM-IgG-F(ab')2, and 14-4-4S Ab were used, an anti-toxin Ab response was observed 1 week after immunization, and this response increased progressively during at least the following 3 weeks (Fig. 6Go). This finding is all the more interesting because 1) such an adjuvant-free immune response is not seen with ZZ-{alpha} that is free or that has been previously complexed with the nonspecific polyclonal rabbit IgGs or a nonspecific mouse IgG2a mAb; and 2) the amount of injected fused toxin could be lowered to 0.1 nmol per mouse, a value fivefold lower than the minimal dose that raises an anti-toxin Ab response in the presence of adjuvant. It should be noted that at this dose and in the absence of adjuvant, the free toxin is still lethal and therefore cannot be used as a reference for the present experiment.



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 6. Anti-toxin Ab response of BALB/c mice immunized with either free ZZ-{alpha} or ZZ-{alpha} complexed to different Abs. The fusion protein was preincubated for 1 night at 4°C with or without the different Abs used in the in vitro experiments. Groups of five mice were then immunized s.c. at the base of the tail without adjuvant. Mice were bled at days 7, 15, and 28 after the injection, and the presence of anti-toxin Abs in individual sera was assessed by enzyme immunoassay described in Materials and Methods.

 
We also assessed whether the noncovalent complexes are able to elicit an anti-toxin T cell response in vivo. We collected spleens from immunized mice 36 days after the injection and investigated the ability of the free toxin {alpha}, peptide 24–41, and HEL to restimulate a T cell response in vitro. As shown in Figure 7Go, toxin-specific T cells were elicited when ZZ-{alpha} was noncovalently complexed with 14-4-4S, RAMµ, and RAM-IgG-F(ab')2 Ab. Conversely, T cells were poorly stimulated when ZZ-{alpha} was injected in a free form or complexed with the nonspecific rabbit IgGs or mouse IgG2a mAb. Furthermore, T cells from toxin responder mice were efficiently restimulated by the peptide 24–41, which, as shown above and previously (34), is the main T cell immunodominant region when BALB/c mice are immunized with the free toxin emulsified with Freund’s adjuvant.



View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 7. T cell response of BALB/c mice immunized with either free ZZ-{alpha} or ZZ-{alpha} complexed to different Abs. BALB/c mice were immunized as described in the legend of Figure 6Go. Thirty-six days after the injection, three spleens were collected per immunization, and pooled spleen cells were challenged with native toxin (1 µM final concentration), peptide 24–41 (1.5 µM final), or hen egg lysozyme (1 µM final). After a culture period of 24 h, each supernatant was assayed for its capacity to stimulate incorporation of [3H]thymidine in the IL-2-dependent CTLL.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Previous studies have shown that immunogenicity of recombinant Ags is enhanced upon fusion to either protein A or its derived fragment ZZ (9, 10, 11). It was proposed that this phenomenon could be mediated by a receptor uptake via binding of the hybrids to membrane Igs of APCs (12, 13). The data reported in this paper strongly support such a scenario. Clearly, T cell presentation of two snake toxins was enhanced when they were fused to ZZ, and this effect vanished in the presence of a large excess of free ZZ, showing that presentation of ZZ-toxin conjugates is associated with binding to a ZZ-specific cell surface structure. That the ZZ-toxin conjugates are primarily targeted to surface Igs is supported by the fact that ZZ is known to bind to Igs and by the observation that T cell presentations are enhanced only if APCs possess surface Igs. Therefore, fusion of an Ag to ZZ is an appropriate way to target a recombinant toxin to surface Igs and to enhance its in vitro T cell presentation.

The number of Ig-binding domains fused to the Ag clearly influences the efficiency of presentation, since the monomeric protein A, which contains five Ig-binding domains, called EDABC (42), is substantially more efficient than ZZ, which only contains two Ig-binding domains (7). This effect, however, can have multiple origins including the possibility that it results from fine differences in specificity and/or affinity of the different Ig-binding domains for surface Igs. In effect, protein A binds to Igs via the Fc domain but also contains an accessory site for binding via the Fab region (43). Furthermore, the number of accessible binding sites of protein A and ZZ for human polyclonal IgG, IgM, and F(ab')2 was recently identified (44). Thus, when an arbitrary number of binding sites of 100 is given for intact protein A, the relative number of binding sites for ZZ is 123, 3, and 20 for IgG, IgM, and F(ab')2, respectively. Therefore, if protein A is clearly more potent than ZZ, there is no direct proportional relationship between these properties and the number of Ig binding sites.

Although the fusion of ZZ to the Ag clearly enhances the T cell response in vitro, no unique explanation can a priori account for the consecutive increase of the in vivo humoral immune response (10, 11). Thus, among the possible and nonexclusive explanations are the possibilities that 1) the half-life of the fused Ags is increased; 2) the ZZ moiety acts as a carrier; and 3) the ZZ domain targets the fused Ags to the appropriate immune cells. A carrier effect is unlikely to occur, since splenocytes from mice immunized with the fusion protein are not restimulated in vitro by free ZZ, indicating that ZZ-specific T cells are induced weakly, if at all, during the immunization process. Therefore, we suggest that under in vivo conditions, the conjugate has targeted the toxin by its ZZ moiety to Igs expressed on the surface of B cells or bound to RFc of macrophages and/or has increased the half-life of the hybrid in the mouse. Previous reports have shown that, under in vivo conditions, the activity and immunodominant characteristics of T cell determinants are not only an intrinsic property of the epitope but also depend on the molecular context (45, 46, 47). To examine whether the presence of the ZZ moiety had modified the selection of the toxin-specific T cell epitopes by APCs, we investigated the capacity of peptides encompassing the whole toxin sequence to stimulate splenocytes from BALB/c mice primed with either free toxin {alpha} or ZZ-{alpha}. Clearly, similar patterns of stimulating toxin fragments were observed, with a major T cell immunostimulating region being located between cysteines 24 and 41 and less potent stimulating regions between residues 32 and 49, and 15 and 30. Therefore, the fusion of ZZ to the N-terminal part of the toxin seems to have no major influence on the profile of T cell epitopes of the toxin. We noted, however, that the T cell response to peptide 32–49 was quantitatively higher in the fusion protein when compared with the free toxin. It is unclear, however, whether this increase is related to the targeting effect toward immune cells.

A number of previous studies have shown that the presence of a T cell epitope on an antigenic protein covalently coupled to a targeting Ab (16, 17, 18, 19, 20, 21, 22, 23), or within the sequence of the Ab (14, 15, 24), increases the efficiency of T cell presentation in vitro. However, the efficiency varies considerably according to the system used. Thus, using a chemical coupling between an Ag and an Ab, the enhancement of presentation ranges from 10- to 104-fold as compared with the free Ag. Insertion of the epitope directly into the Ab seems more powerful, as the efficiency ranges from 103- to >105-fold. Possibly, the lower efficiency of the first approach is related to the difficulty in achieving an appropriate chemical coupling between the protein Ag and the Ab. In contrast, a ZZ-fusion protein offers the considerable advantage of binding specifically to unique areas of an Ab, mainly its Fc part, leaving the paratopic region of the Ab freely accessible for binding to its specific Ag. In principle, therefore, an anti-MHC class II Ab or an anti-Ig Ab complexed to a ZZ-Ag is anticipated to target the Ag to cells bearing MHC class II or Ig molecules and hence to increase its T cell presentation. However, a major uncertainty with such an approach was associated with the noncovalent nature of the ZZ-Ag/Ab complex. We did not know the affinity of the ZZ moiety for mAb 14-4-4S and for rabbit IgG; however, it was previously shown that the affinity constants of ZZ for polyclonal IgG is approximately 4.5 108 M-1 (44). Although we did not know its stability, the complex was capable of targeting the hybrid in vitro and hence of increasing the T cell presentation efficacy of the Ag. Therefore, a ZZ-toxin conjugate can be readily associated with either an I-Ed MHC-specific Ab or an Ig-specific Ab, and such construction increases the T cell presentation efficiency up to 103-fold, irrespective of the type of APCs that are used. Moreover, the weak T cell response was observed with a RAM-IgM Ab when non-IgM-bearing A20 cells were used as APCs. This result shows that it is possible to discriminate between various populations of APCs according to the specificity of the Ab involved in the complex.

Targeting of Ags to surface components of appropriate immune cells has previously been shown to increase immunogenicity of an Ag in vivo with no adjuvant. Adjuvant-free immune responses were elicited in mice against various Ags using 1) avidin bound to a biotinylated anti-class II MHC Ab (21, 41); 2) HEL complexed with bispecific Abs made of an anti-HEL Ab cross-linked to Abs specific for a particular cell surface component (25); 3) OVA covalently linked to an anti-Fc{epsilon}RII Ab (22); and 4) HEL and porcine pancreatic elastase chemically coupled to {alpha}2-macroglobulin (23). In all these approaches, however, a chemical coupling, usually not easy to achieve in a regioselective manner, was required. The ZZ-Ag-Ab complex described in the present paper shows that such a chemical step can be avoided, since the noncovalent complex is stable enough to be used under in vivo conditions. Thus, we found that in the absence of adjuvant, injection of rabbit anti-IgM, rabbit anti-IgG-F(ab')2, or mAb 14-4-4S noncovalently complexed to ZZ-{alpha} elicited a toxin-specific immune response. This observation is all the more interesting because the Ag dose used here was fivefold lower than the minimum dose at which the free Ag, injected in the presence of Freund’s adjuvant, elicited an immune response. Clearly, this simple approach may find applications by increasing the immunogenicity of any recombinant protein genetically fused to ZZ.


   Footnotes
 
1 Address correspondence and reprint requests to Dr Michel Léonetti, Département d’Ingéniérie et d’Études des Protéines (DIEP) CEA/Saclay, 91191 Gif-Sur-Yvette Cedex, France. E-mail address: Back

2 Abbreviations used in this paper: Ea, erabutoxin a; ZZ-Ea, ZZ fused to erabutoxin a; ProtA-Ea, protein A fused to erabutoxin a; ZZ-{alpha}, ZZ fused to toxin {alpha}; CTLL, cytotoxic T cell line; RAM, rabbit anti-mouse; RAMµ, rabbit polyclonal anti-IgM Ab. Back

Received for publication May 29, 1997. Accepted for publication December 12, 1997.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Mitchinson, N. A.. 1971. The carrier effect in the secondary response to hapten-protein conjugates. Eur. J. Immunol. 1:18.[Medline]
  2. Lanzavecchia, A.. 1990. Receptor-mediated antigen uptake and its effect on antigen presentation to class II-restricted T lymphocytes. Annu. Rev. Immunol. 8:773.[Medline]
  3. Ziegler, H. K., E. Unanue. 1982. Decrease in macrophage antigen catabolism caused by ammonia and chloroquine with inhibition of antigen presentation to T cells. Proc. Natl. Acad. Sci. USA 79:175.[Abstract/Free Full Text]
  4. Adorini, L., E. Appella, G. Doria, Z. A. Nagy. 1988. Mechanisms influencing the immunodominance of T cell determinants. J. Exp. Med. 168:2091.[Abstract/Free Full Text]
  5. Nelson, C. A., S. J. Petzold, E. R. Unanue. 1994. Peptides determine the lifespan of MHC class II molecules in the antigen-presenting cell. Nature 371:250.[Medline]
  6. Nilsson, B., E. Holmgren, S. Josephson, S. Gatenbeck, L. Philipson, M. Uhlén. 1985. Efficient secretion and purification of human insulin-like growth factor I with a gene fusion vector in staphylococci. Nucleic Acids Res. 13:1151.[Abstract/Free Full Text]
  7. Nilsson, B., T. Moks, B. Jansson, L. Abrahmsén, A. Emblad, E. Holmgren, C. Henrickson, T. Alwiyn Jones, M. Uhlén. 1987. A synthetic IgG-binding domain based on staphylococcal protein A. Protein Eng. 1:107.[Abstract/Free Full Text]
  8. Löwenadler, B., B. Nilsson, L. Abrahmsén, T. Moks, L. Ljungqvist, E. Holmgren, S. Paleus, S. Josephson, L. Philipson, M. Uhlén. 1986. Production of specific antibodies against protein A fusion proteins. EMBO J. 5:2393.[Medline]
  9. Löwenadler, B., B. Jansson, S. Paleus, E. Holmgren, B. Nilsson, T. Moks, G. Palm, S. Josephson, L. Philipson, M. Uhlén. 1987. A gene fusion system for generating antibodies against short peptides. Gene 58:87.[Medline]
  10. Löwenadler, B., A.-M. Svernnerholm, M. Gidlund, E. Holmgren, K. Krook, C. Svanholm, S. Ulff, S. Josephson. 1990. Enhanced immunogenicity of recombinant peptide fusions containing multiple copies of a heterologous T helper epitope. Eur. J. Immunol. 20:1541.[Medline]
  11. Ducancel, F., J. C. Boulain, O. Trémeau, A. Ménez. 1989. Direct expression in E. coli of a functionally active protein A-snake toxin fusion protein. Protein Eng. 3:139.[Abstract/Free Full Text]
  12. Löwenadler, B., N. Licke, C. Svanholm, A.-M. Svernnerholm, K. Krook, M. Gidlund. 1992. T and B cell responses to chimeric proteins containing heterologous T helper epitopes inserted at different positions. Mol. Immunol. 29:1185.[Medline]
  13. Löwenadler, B., N. Lycke. 1994. Fusion proteins with heterologous T helper epitopes: Recombinant E. coli heat stable enterotoxin proteins. Int. Rev. Immunol. 11:103.[Medline]
  14. Chesnut, R., S. Colon, H. M. Grey. 1981. Studies on the capacity of B cells to serve as antigen-presenting cells. J. Immunol. 126:1075.[Medline]
  15. Tony, H. P., D. C. Parker. 1985. Major histocompatibility complex-restricted, polyclonal B cell responses resulting from helper T cell recognition of anti-immunoglobulin presented by small B lymphocytes. J. Exp. Med. 161:223.[Abstract/Free Full Text]
  16. Kawamura, H., J. A. Berzofsky. 1986. Enhancement of antigenic potency in vitro and immunogenicity in vivo by coupling the antigen to anti-immunoglobulin. J. Immunol. 136:58.[Abstract]
  17. Snider, D. P., D. Segal. 1987. Targeted antigen presentation using cross-linked antibody heteroaggregates. J. Immunol. 139:1609.[Abstract]
  18. Casten, L. A., S. Pierce. 1988. Receptor-mediated B cell antigen processing: increased antigenicity of a globular protein covalently coupled to antibodies specific for B cell surface structures. J. Immunol. 140:404.[Abstract]
  19. Snider, D. P., D. Segal. 1989. Efficiency of antigen presentation after antigen targeting to surface IgD, IgM, MHC, Fc{gamma}RII, and B220 molecules on murine splenic B cells. J. Immunol. 143:59.[Abstract]
  20. Gosselin, E. J., K. Wardwell, D. R. Gosselin, N. Alter, J. L. Fisher, P. M. Guyre. 1992. Enhanced antigen presentation using human Fc{gamma} receptor (monocyte/macrophage)-specific immunogens. J. Immunol. 149:3477.[Abstract]
  21. Skea, D. L., B. H. Barber. 1993. Studies of the adjuvant-independent antibody response to immunotargeting: target structure dependence, isotype distribution, and induction of long term memory. J. Immunol. 151:3557.[Abstract]
  22. Squire, C. M., E. J. Studer, A. Lees, F. D. Finkelman, D. H. Conrad. 1994. Antigen presentation is enhanced by targeting antigen to Fc{epsilon}RII by antigen-anti-Fc{epsilon}RII conjugates. J. Immunol. 152:4388.[Abstract]
  23. Berg, S. F., S. Mjaaland, S. Fosssum. 1994. Comparing macrophages and dendritic leukocytes as antigen-presenting cells for humoral responses in vivo by antigen targeting. Eur. J. Immunol. 24:1262.[Medline]
  24. Lanzavecchia, A., S. Abrignani, D. Scheidegger, R. Obrist, B. Dôrken, G. Moldenhauer. 1988. Antibodies as antigens: the use of mouse monoclonal antibodies to focus human T cells against selected targets. J. Exp. Med. 167:345.[Abstract/Free Full Text]
  25. Snider, D. P., A. Kaubisch, D. M. Segal. 1990. Enhanced antigen immunogenicity induced by bispecific antibodies. J. Exp. Med. 171:1957.[Abstract/Free Full Text]
  26. Ménez, A., L. Pillet, M. Léonetti, F. Bontems, B. Maillère. 1992. Snake toxins as antigens. M. H. V. Van Regenmortel, ed. In Structure of Antigens Vol. 1:293. CRC, Boca Raton, FL.
  27. Ménez, A.. 1995. Les venins et toxines de serpents. M. Goyffon, and J. Heurthault, eds. La Fonction Venimeuse 200. Masson, Paris.
  28. Fryklund, L., D. Eaker. 1975. The complete covalent structure of a cardiotoxin from the venom of Naja nigricollis toxin {alpha}. Biochemistry 14:2865.[Medline]
  29. Maillère, B., J. Cotton, G. Mourier, M. Léonetti, S. Leroy, A. Ménez. 1993. Role of thiols in the presentation of a snake toxin to murine T cells. J. Immunol. 150:5270.[Abstract]
  30. Mourier, G., B. Maillère, J. Cotton, M. Hervé, S. Leroy, M. Léonetti, A. Ménez. 1994. Boc-Cys(Npys)-OH (BCNP): an appropriate reagent for the identification of T cell epitopes in cystine and/or cysteine-containing proteins. J. Immunol. Methods 171:65.[Medline]
  31. Pillet, L., O. Trémeau, F. Ducancel, P. Drevet, S. Zinn-Justin, S. Pinkasfeld, J.-C. Boulain, A. Ménez. 1993. Genetic engineering of snake toxins: role of invariant residues in the structural and functional properties of a curaremimetic toxin, as probed by site-directed mutagenesis. J. Biol. Chem. 268:909.[Abstract/Free Full Text]
  32. J. Sambrook, and E. F. Fritsch, and T. Maniatis, eds. Molecular Cloning: A Laboratory Manual 1989 Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
  33. Birmingham, J. R., R. W. Chesnut, J. W. Kappler, P. Marrack, R. Kubo, H. M. Grey. 1982. Antigen presentation to T cell hybridomas by a macrophage cell line: an inducible function. J. Immunol. 128:1491.[Abstract]
  34. Léonetti, M., L. Pillet, B. Maillère, H. Lam Thanh, P. Frachon, J. Couderc, A. Ménez. 1990. Immunization with a peptide having both T and conformationally restricted B cell epitopes elicits neutralizing antisera against a snake neurotoxin. J. Immunol. 145:4214.[Abstract]
  35. Langone, J. J.. 1982. Protein A of Staphylococcus aureus and related immunoglobulin receptors produced by streptococci and pneumococci. Adv. Immunol. 32:157.[Medline]
  36. Jensen, E. P.. 1991. Reduction of disulfide bonds during antigen processing: evidence from a thiol dependent insulin determinant. J. Exp. Med. 174:1121.[Abstract/Free Full Text]
  37. Hampl, J., G. Gradehandt, D. Plachov, H. G. Gattner, H. Kalbacher, W. Voelter, M. Meyer-Delius, E. Rüde. 1991. Presentation of insulin A chain peptides to mouse T cells: involvement of cysteines residues. Mol. Immunol. 28:479.[Medline]
  38. Hampl, J., G. Gradehandt, H. Kalbacher, E. Rüde. 1992. In vitro processing of insulin for recognition by murine T cells results in the generation of a chains with free CysSH. J. Immunol. 148:2664.[Abstract]
  39. Ozato, K., N. Mayer, D. H. Sachs. 1980. Hybridoma cell lines secreting monoclonal antibodies to mouse H-2 and Ia antigens. J. Immunol. 124:533.[Abstract]
  40. Kim, K. J., C. Kanellopoulos-Langevin, R. M. Merwin, D. H. Sachs, R. Asofsky. 1979. Establishment and characterization of BALB/c lymphoma lines with B cell properties. J. Immunol. 122:549.[Abstract/Free Full Text]
  41. Carayanniotis, G., B. H. Barber. 1987. Adjuvant-free IgG responses induced with antigen coupled to antibodies against class II MHC. Nature 327:59.[Medline]
  42. Uhlén, M., B. Guss, B. Nilsson, S. Gatenbeck, L. Philipson, M. Lindberg. 1984. Complete sequence of the staphylococcal gene encoding protein A: a gene evolved through multiple duplications. J. Biol. Chem. 259:1695.[Abstract/Free Full Text]
  43. Myhre, E. B.. 1990. Interaction of bacterial immunoglobulin receptors with sites in the Fab region. M. D. P. Boyle, ed. In Bacterial Immunoglobulin Binding Proteins Vol. 1:243. Academic, New York, NY.
  44. Ljundberg, U. K., B. Jansson, U. Niss, R. Nilsson, B. E. Sandberg, B. Nilsson. 1993. The interaction between different domains of staphylococcal protein A and human polyclonal IgG, IgA, IgM, and F(ab')2: separation of affinity from specificity. Mol. Immunol. 30:1279.[Medline]
  45. Perkins, D. L., G. Berriz, T. Kamradt, J. A. Smith, M. L. Gefter. 1991. Immunodominance: intramolecular competition between T cell epitopes. J. Immunol. 146:2137.[Abstract]
  46. Leclerc, C., P. Martineau, A. Charbit, R. Lo-Man, E. Dériaud, M. Hofnung. 1993. Immunodominance of a recombinant T cell epitope depends on its molecular environment. Mol. Immunol. 30:1561.[Medline]
  47. Lo-Man, R., P. Martineau, J.-M. Betton, M. Hofnung, C. Leclerc. 1994. Molecular context of a viral T cell determinant within a chimeric bacterial protein alters the diversity of its T cell recognition. J. Immunol. 152:5660.[Abstract]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
L. Agren, E. Sverremark, L. Ekman, K. Schon, B. Lowenadler, C. Fernandez, and N. Lycke
The ADP-Ribosylating CTA1-DD Adjuvant Enhances T Cell-Dependent and Independent Responses by Direct Action on B Cells Involving Anti-Apoptotic Bcl-2- and Germinal Center-Promoting Effects
J. Immunol., June 15, 2000; 164(12): 6276 - 6286.
[Abstract] [Full Text] [PDF]

Home page
 J. Exp. Med.Home page
M. Leonetti, J. Galon, R. Thai, C. Sautes-Fridman, G. Moine, and A. Menez
Presentation of Antigen in Immune Complexes Is Boosted by Soluble Bacterial Immunoglobulin Binding Proteins
J. Exp. Med., April 19, 1999; 189(8): 1217 - 1228.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Léonetti, M.
Right arrow Articles by Ménez, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Léonetti, M.
Right arrow Articles by Ménez, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS