FEBS Letters 582 (2008) 185–189
Correlation between Shiga toxin B-subunit stability and antigen crosspresentation: A mutational analysis David G. Pinaa,b,1, Bahne Stechmanna,b, Valery L. Shnyrovc, Lucien Cabanie´b, Nacilla Haicheurd,e, Eric Tartourd,e,2, Ludger Johannesa,b,*,2 b
a CNRS UMR144, France Institut Curie, Centre de Recherche, Laboratoire Trafic, Signalisation et Ciblage Intracellulaires, 75248 Paris Cedex 05, France c Departamento de Bioquı´mica y Biologı´a Molecular, Universidad de Salamanca, 37007 Salamanca, Spain d EA 4054, Universite´ Paris-Descartes-ENVA, 94704 Maisons Alfort, France e Hopital Europe´en Georges Pompidou, APHP, Paris, France
Received 22 October 2007; revised 28 November 2007; accepted 29 November 2007 Available online 10 December 2007 Edited by Felix Wieland
Abstract The homopentameric B-subunit of Shiga toxin (STxB) is used as a tool to deliver antigenic peptides and proteins to the cytosolic compartment of dendritic cells (DCs). In this study, a series of interface mutants of STxB has been constructed. All mutants retained their overall conformation, while a loss in thermal stability was observed. This effect was even more pronounced in trifluoroethanol solutions that mimic the membrane environment. Despite this, all mutants were equally efficient at delivering a model antigenic protein into the MHC class I-restricted antigen presentation pathway of mouse DCs, suggesting that the structural stability of STxB is not a key factor in the membrane translocation process. 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Shiga toxin B-subunit; Retrograde transport; Membrane translocation; Crosspresentation; Antigen presenting cell
1. Introduction Like verotoxins and cholera toxin, Shiga toxin belongs to the AB5 family of bacterial toxins [1]. Structurally, these toxins consist of two non-covalently associated subunits, a monomeric catalytic A-subunit, and a pentameric B-subunit that is responsible for receptor binding and intracellular trafficking of the holotoxins [2,3]. In the case of Shiga toxin, the A-subunit has ribosomal RNA N-glycosidase activity, leading to protein biosynthesis inhibition, while the B-subunit (STxB) binds specifically to the glycosphingolipid Gb3 [4]. The crystal structures of STxB and holotoxin have been solved [5–7]. STxB * Corresponding author. Address: Institut Curie, Centre de Recherche, Laboratoire Trafic, Signalisation et Ciblage Intracellulaires, 75248 Paris Cedex 05, France. Fax: +33 1 42346507. E-mail address:
[email protected] (L. Johannes). 1 Present address: Directorate-General for Research, European Commission, 1049 Brussels, Belgium. 2 ET and LJ are principal investigators.
Abbreviations: STxB, Shiga toxin B-subunit; CD, circular dichroism; DC, dendritic cell; ER, endoplasmic reticulum; TFE, trifluoroethanol
is a homopentamer composed of monomers of 69 residues (7.7 kDa). Each monomer has two three-stranded antiparallel b-sheets and an a-helix (see Fig. 1). The pentamer forms a ringlike structure with a central pore whose diameter is about ˚ , and that is delimited by the five a-helices and surrounded 11 A by b-sheets from pairs of adjacent monomers. Recent structural and mutational studies suggest that STxB has three Gb3 binding sites per monomer [8,9]. The toxin–receptor complex is internalized via clathrindependent and -independent endocytosis, and the toxin is then transported from the early endosome to the endoplasmic reticulum (ER), via the trans-Golgi network and Golgi apparatus, a pathway known as the retrograde route [10,11]. Transport to the ER is believed to be required for translocation of the A-subunit to the cytosol where its molecular target – ribosomal RNA – resides [12]. STxB has been used as a tool for the characterization of the retrograde route [13,14], and in biomedical research as a delivery tool for immunotherapy [3,15] and cancer targeting [16,17]. It was shown that STxB delivers exogenous peptides into the MHC class I and II pathways of human and mouse dendritic cells (DCs) and induces humoral and cell-mediated immune responses that protect mice from tumor growth [18–22]. Functional TAP and proteasome activities are required for STxB-induced MHC class I-restricted antigen presentation, strongly suggesting that STxB delivers exogenous antigenic proteins and peptides across membranes into the cytosolic compartment [19]. The mechanism and exact intracellular site of this membrane translocation event have not yet been identified, but intracellular trafficking studies on human monocyte-derived DCs suggested that the retrograde route might not be involved [23]. The interaction of STxB with membranes has been analyzed in some detail. The two-dimensional structure of STxB on Gb3-containing bilayer stacks overlaps perfectly with the projection map of its solution structure [24], and the binding of soluble Gb3 analogues does not induce structural changes at the atomic level of resolution [8]. In agreement with these data, no significant conformational changes occurred upon interaction of STxB with large unilamellar vesicles [25]. In the latter report, a fluorescence technique allowed us to monitor the conformational state of STxB due to the presence of tryptophan residues at the protein/membrane interface. However, when trifluoroethanol (TFE) was used to mimic a hydrophobic environment similar to that of membranes, a
0014-5793/$32.00 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.11.086
186
D.G. Pina et al. / FEBS Letters 582 (2008) 185–189
Fig. 1. Location of mutated residues in the three-dimensional structure of STxB.Structural elements shown in light grey or dark grey correspond to different monomers. Residues 39 and 41 are located in the a-helix, whereas residue 14 is in the b2 strand, interacting with the b6 strand of the adjacent monomer. Residue 33 is at the protein–membrane interface.
monomeric intermediate was observed [26], which could not be detected when STxB was analyzed in the absence of solvent [27]. Here, a series of single-residue mutants at the interface between STxB monomers was constructed. We demonstrate that STxB pentamer stability is compromised to variable extents in these mutants, and in some cases, this effect is particularly pronounced in membrane-mimicking conditions. However, translocation to the cytosol does not appear to be favored, as assessed by MHC class I-restricted antigen presentation.
2. Materials and methods 2.1. Cells and materials HeLa cells were grown in DulbeccoÕs modified EagleÕs medium containing 4.5 g/l glucose (Life Technologies Inc.) supplemented with 10% fetal calf serum (Life Technologies Inc.), 0.01% penicillin/streptomycin, 4 mM glutamine, and 5 mM pyruvate in a 5% CO2 incubator. D1 cells were cultured in IscoveÕs Modified DulbeccoÕs Medium (Sigma), supplemented with 10% heat-inactivated fetal calf serum, 2 mM L -glutamine, 5 mM sodium pyruvate (all Invitrogen), and 30% conditioned medium from GM-CSF-producing NIH-3T3 cells, as previously described [28]. 2.2. Plasmid construction The pSU108 plasmid was used for the construction of STxB mutants, as described [29]. Primers encoding the desired point mutations and specific primers Shiga-AtpE and Shiga-fd were used to produce fragments that were cloned into the SphI and SalI restriction sites of pSU108. For the construction of OVA257–264 fusion proteins, the same strategy was used, starting with a plasmid encoding wild-type STxB fused to OVA257–264 peptide including three amino-terminal flanking amino acids (QLESIINFEKL). Sequences derived by polymerase chain reaction were verified by dideoxy sequencing. 2.3. Expression, purification, and size-exclusion chromatography of recombinant STxB and mutants Purification of recombinant STxB was essentially done as described elsewhere [30]. After preparation of periplasmic extracts, these were
loaded on a QFF column (Pharmacia) and eluted by a linear NaCl gradient in 20 mM Tris/HCl, pH 7.5. Depending on the construction, recombinant STxB eluted between 120 and 400 mM. STxB-containing fractions were dialyzed against 20 mM Tris/HCl, pH 7.5, reloaded on a Mono Q column (Pharmacia), and eluted as before. The resulting proteins, estimated to be 95% pure by SDS–PAGE, were stored at 80 C until use. Protein concentration was estimated by spectrophotometry using an extinction coefficient of 9100 M 1 for STxB wild-type and R33A, 8700 for Y14A, and 14 500 for L39W and L41W. For sizeexclusion chromatography experiments, samples were loaded on a 30 ml Sephadex 75-column (GE Healthcare). 2.4. Circular dichroism spectroscopy Far-UV (190–250 nm) circular dichroism (CD) spectra were recorded with a Jasco J-715 spectropolarimeter coupled to a Peltier thermal control unit. Spectra are the average of five accumulations that all were background corrected and smoothed using the noise reduction software of the instrument. Data were converted to molar ellipticity per mean residue: [H] = MresHobsl 1p 1, where Mres = 111.5 is the mean residue molar mass, Hobs the measured ellipticity in millidegrees, l the optical path length of the cell in millimetres, and p is the protein concentration (mg/ml). Experiments were performed in a 1-mm pathlength cuvette and sample concentration of about 0.2 mg/ml. Thermal stability curves were collected at the fixed wavelength of 225 nm, and between 40 and 90 C, with a constant heating rate of 1 C/min. 2.5. Immunofluorescence Immunofluorescence was performed as previously described [11]. Briefly, cells were fixed at room temperature for 10 min in 3% paraformaldehyde, quenched with ammonium chloride, permeabilized with 0.05% saponin, incubated with the indicated primary or secondary antibodies, mounted, and viewed by confocal microscopy (Leica Microsystems, Mannheim, Germany). 2.6. Scatchard analysis 1 · 105 cells were placed on ice and washed once with ice-cold PBS. The indicated concentrations of recombinant iodinated wild-type or mutant STxB (5.000 cpm/ng; concentrations from 0.05 to 2 lM) were added in DulbeccoÕs modified EagleÕs medium supplemented with 10% fetal calf serum. After 90 min incubation on ice in presence of 10 mM HEPES, pH 7.2, the cells were washed three times with ice-cold PBS. Cells were lysed in 0.1 M KOH, cell-associated radioactivity and radioactivity in the culture medium and wash solutions were counted, and binding data were obtained as published [29].
D.G. Pina et al. / FEBS Letters 582 (2008) 185–189
187
2.7. Antigen presentation assays The presence of OVA257–264/Kb complexes at the cell surface was determined by performing a functional assay, as described previously [20]. Briefly, D1 DCs (105 cells/well) were first pulsed with antigen for 5 hours and washed twice before being co-cultured with the B3Z hybridoma (2 · 105 cells/well). A colorimetric assay with O-nitrophenyl b-D -galactopyranoside (ONPG, Sigma) as a substrate was used to detect b-galactosidase activity in B3Z lysates.
3. Results and discussion 3.1. Design of STxB mutants A series of single-residue mutants of STxB were constructed by site-directed mutagenesis. The four mutation sites are situated in interacting regions between monomers. The Y14A mutation is located in the b-sheet monomer–monomer interface, whereas R33A is located proximal to the region where binding of Gb3 takes place. Two other mutations, L39W and L41W, are located in the helical region. In the wild-type protein, L39 of one monomer is localized in close proximity of L41 of the next monomer (Fig. 1). Introducing a Trp in these positions is expected to perturb monomer packing, if one considers the increased size of the Trp side chain. 3.2. STxB mutants are destabilized but do not exhibit major structural changes Fig. 2A shows the far-UV CD spectra of wild-type STxB and mutants at 0.2 mg/ml concentration in 20 mM phosphate buffer, pH 7.0 at 25 C. Superimposition of the spectra suggests that the secondary structure of the mutants is not perturbed. Furthermore, the size-exclusion chromatography profiles of the mutants were similar to wild-type STxB (Fig. 2B, inset), strongly suggesting that the oligomeric state of the mutants is not altered. The thermal denaturation of the mutants was monitored by changes in ellipticity at 225 nm (Fig. 2B), and unfolding curves were successfully fitted to the two-state unfolding model coupled to dissociation (with equations detailed in [26]). Table 1 summarizes the melting temperatures observed for each mutant protein. In all cases, a destabilization was observed, with unfolding temperature differences ranging from 6 to 11 C. The C-terminal fusion of STxB with the OVA257–264 peptide does not significantly affect the CD spectra nor the thermal denaturation curves (data not shown), strongly suggesting that STxB structure was not altered. This result is in good agreement with previous 2D crystallization studies in which at 8.5A resolution, the structures of a C-terminal fusion variant of STxB and of wild-type STxB were found to be perfectly superposable [24]. We also performed thermal unfolding experiments on STxB mutants in presence of 10% TFE (v/v) to evaluate their affinity to the more hydrophobic environment provided by the TFE/ water mixture. In a previous study, analytical ultracentrifugation experiments showed that high concentrations of TFE – mimicking the membrane environment – induce dissociation of wild-type STxB into monomers with high a-helical contents [26]. However, at concentrations below 20% TFE (v/v), the protein retains its pentameric conformation (Fig. 2A, inset). Under our experimental conditions – 10% TFE (v/v) concentration – Y14A and R33A mutants were much more destabilized than wild-type STxB or L39W and L41W mutants (see Fig. 2), suggesting these mutants have a higher propensity to dissociate into monomers.
Fig. 2. (A) Far-CD spectra of native wild-type STxB (continuous line), mutants (L39W: thick dotted line, Y14A: thin dashed line, L41W: thick dashed line, R33A: thin dotted line) and thermally-denatured STxB (thin continuous line). The inset shows the effect on wild-type STxB of increasing concentrations of TFE at 208 nm (solid symbols) and 260 nm (open symbols) at 25 C and pH 7.0. Lines are the best fit using the two-state unfolding coupled to dissociation model (see details in [26]). (B) Fraction of unfolded protein, monitored by changes in ellipticity at 225 nm, as a function of temperature, for wild-type (d) and mutants of STxB (m: L39W, n: Y14A, s: L41W, n: R33A). The inset shows the normalized Absorbance at 280 nm as a function of the retention volume for wild-type STxB (thick continuous line), R33A (dotted line) and L41W mutants (thin continuous line). STxB and mutants elute from the column at the level of the main peaks. Table 1 Melting temperatures for the unfolding of STxB wild-type and mutants monitored by CD spectroscopy at pH 7.0, in aqueous conditions and in presence of 10% TFE (v/v), DTm between these two values, and binding constants on HeLa cells obtained by Scatchard analysis Protein
Tm (C) aqueous buffer
Tm (C) 10% TFE (v/v)
DTm
KD (nM)
Wild-type L39W L41W R33A Y14A
79.3 ± 0.3 73.4 ± 0.2 73.4 ± 0.3 68.4 ± 0.1 68.3 ± 0.1
65.6 ± 0.2 59.2 ± 0.2 58.3 ± 0.1 49.3 ± 0.1 50.5 ± 0.2
13.7 ± 0.3 14.2 ± 0.2 15.1 ± 0.2 19.1 ± 0.1 17.8 ± 0.2
136 ± 17 201 ± 21 87 ± 11 838 ± 75 477 ± 42
CD experiments were performed at a fixed wavelength of 225 nm and 0.2 mg/ml protein concentration. For Scatchard analysis, see Section 2.
3.3. STxB mutants retain their retrograde transport properties and affinity for Gb3 Scatchard analysis showed that the dissociation constants of the L39W and L41W mutants are similar to that of wild-
188
D.G. Pina et al. / FEBS Letters 582 (2008) 185–189
Fig. 4. Comparison of the respective efficiency of STxB and mutants to deliver exogenous peptides into the MHC class I pathway. D1 dendritic cells (105) were pulsed for 5 h with wild-type (s) or mutant (R33A, h; L41W, e) STxB coupled to OVA257–264, an immunodominant Kb restricted OVA derived peptide. After washing, the cells were cocultured with the specific anti-OVA257–264 B3Z hybridoma (2 · 105) carrying a LacZ construct driven by NF-AT elements of the IL-2 promotor. A colorimetric assay with O-nitrophenyl b-D -galactopyranoside (ONPG, Sigma) as a substrate was used to detect b-galactosidase activity in B3Z lysates.
3.4. STxB mutants are efficient at inducing MHC class Irestricted antigen presentation As described previously for wild-type STxB [19], we fused the OVA257–264 peptide to the C-terminus of the R33A and L41W mutants, and analyzed Kb-restricted antigen presentation using the B3Z hybridoma. The L41W mutant was similarly efficient as wild-type STxB at introducing the OVA257–264 peptide into the MHC class I pathway of D1 cells, whereas the R33A mutant was slightly less efficient (Fig. 4). This reduced efficiency is likely due to the apparent loss in affinity for cells of R33A (see Table 1).
Fig. 3. Retrograde trafficking analysis on wild-type STxB and the R33A and L41W mutants. The indicated proteins (2 lM) were bound to (A) Hela cells or (B) D1 cells on ice. The cells were washed, shifted to 37 C for 45 min, fixed, and labeled for STxB (or mutants) and the trans-Golgi/TGN marker Rab6. Right images show overlay. STxB: red; Rab6: green.
type STxB, whereas those of the Y14A and R33A mutants were 3- to 6-fold increased (Table 1). Despite these differences, all mutants were efficiently internalized into both HeLa and D1 cells, and their intracellular trafficking was qualitatively comparable to that of wild-type STxB (Fig. 3). These results demonstrate that the mutated residues are not essential for Gb3 binding and that retrograde transport remains efficient for all proteins.
3.5. Conclusion In the present study, we observed that STxB mutants whose pentamer stability in membrane mimicking environments is strongly reduced, are not more efficient than wild-type STxB at introducing the Kb-restricted immunodominant OVA257– 264 peptide of chicken ovalbumin into the MHC class Irestricted antigen presentation pathway of DCs. This finding suggests that a monomeric molten globule-like conformation that has been observed for STxB [26] may not be the membrane translocation competent intermediate of the protein. Other mechanisms such as inverted micelle formation or endosomal membrane destabilization need to be considered in future studies. Acknowledgements: This work was supported by Association pour la Recherche sur le Cancer (n 3105), Cance´ropoˆle Ile-de-France, the Cancerimmunotherapy program (7th PCRD), and PIC Vectorization from Institut Curie. We thank Drs. J. Me´ne´trey and A. Houdusse for valuable technical assistance. D.G.P. was recipient of a postdoctoral fellowship from Fundac¸a˜o para a Cieˆncia e a Tecnologia, Portugal (Ref. SFRH/BPD/14568/2003).
D.G. Pina et al. / FEBS Letters 582 (2008) 185–189
References [1] Merritt, E.A. and Hol, W.G. (1995) AB5 toxins. Curr. Opin. Struct. Biol. 5, 165–171. [2] Falnes, P.O. and Sandvig, K. (2000) Penetration of protein toxins into cells. Curr. Opin. Cell Biol. 12, 407–413. [3] Johannes, L. and Decaudin, D. (2005) Protein toxins: intracellular trafficking for targeted therapy. Gene Ther. 16, 1360–1368. [4] Smith, D.C., Lord, J.M., Roberts, L.M. and Johannes, L. (2004) Glycosphingolipids as toxin receptors. Semin. Cell Dev. Biol. 15, 397–408. [5] Stein, P.E., Boodhoo, A., Tyrrell, G.J., Brunton, J.L. and Read, R.J. (1992) Crystal structure of the cell-binding B oligomer of verotoxin-1 from E. coli. Nature 355, 748–750. [6] Fraser, M.E., Chernaia, M.M., Kozlov, Y.V. and James, M.N. (1994) Crystal structure of the holotoxin from Shigella dysenteriae at 2.5 A resolution. Nat. Struct. Biol. 1, 59–64. [7] Richardson, J.M., Evans, P.D., Homans, S.W. and DonohueRolfe, A. (1997) Solution structure of the carbohydrate-binding B-subunit homopentamer of verotoxin VT-1 from E. coli. Nat. Struct. Biol. 4, 190–193. [8] Ling, H., Boodhoo, A., Hazes, B., Cummings, M.D., Armstrong, G.D., Brunton, J.L. and Read, R.J. (1998) Structure of Shiga-like toxin I B-pentamer complexed with an analogue of its receptor Gb3. Biochemistry 37, 1777–1788. [9] Bast, D.J., Banerjee, L., Clark, C., Read, R.J. and Brunton, J.L. (1999) The identification of three biologically relevant globotriaosyl ceramide receptor binding sites on the verotoxin 1 B subunit. Mol. Microbiol. 32, 953–960. [10] Sandvig, K., Garred, O., Prydz, K., Kozlov, J.V., Hansen, S.H. and van Deurs, B. (1992) Retrograde transport of endocytosed Shiga toxin to the endoplasmic reticulum. Nature 358, 510–512. [11] Mallard, F., Tenza, D., Antony, C., Salamero, J., Goud, B. and Johannes, L. (1998) Direct pathway from early/recycling endosomes to the Golgi apparatus revealed through the study of Shiga toxin B-fragment transport. J. Cell Biol. 143, 973–990. [12] Lord, J.M., Roberts, L.M. and Lencer, W.I. (2005) Entry of protein toxins into mammalian cells by crossing the endoplasmic reticulum membrane: co-opting basic mechanisms of endoplasmic reticulum-associated degradation. Curr. Top. Microbiol. Immunol. 300, 149–168. [13] Amessou, M., Popoff, V., Yelamos, B., Saint-Pol, A. and Johannes, L. (2006) Measuring retrograde transport to the trans-Golgi network. Current Protocols in Cell Biology (eds. Juan S. Bonifacino et al.), John Wiley & Sons Inc. [14] Lamaze, C. and Johannes, L. (2006) Intracellular trafficking of bacterial and plant protein toxins in: Comprehensive Sourcebook of Bacterial Protein Toxins (Alouf, Joseph E. and Popoff, Michel R., Eds.), third ed, Academic Press, London. [15] Smith, D.C., Lord, J.M., Roberts, L.M., Tartour, E. and Johannes, L. (2002) 1st class ticket to class I: protein toxin as pathfinders for antigen presentation. Traffic 3, 697–704. [16] Janssen, K.-P. et al. (2006) In vivo tumor targeting using a novel intestinal pathogen-based delivery approach. Cancer Res. 66, 7230–7236.
189 [17] El Alaoui, A., Schmidt, F., Amessou, M., Sarr, M., Decaudin, D., Florent, J.C. and Johannes, L. (2007) Shiga toxin-mediated retrograde delivery of a topoisomerase I inhibitor prodrug. Angew. Chem., Int. Ed. Engl. 46, 6469–6472. [18] Lee, R.-S., Tartour, E., van der Bruggen, P., Vantomme, V., Goud, B., Fridman, W.-H. and Johannes, L. (1998) Major histocompatibility complex class I presentation of exogenous soluble tumor antigen fused to the B-fragment of Shiga toxin. Eur. J. Immunol. 28, 2726–2737. [19] Haicheur, N. et al. (2000) The B-subunit of Shiga toxin fused to a tumor antigen elicits CTL and targets dendritic cells to allow MHC class I restricted presentation of peptides derived from exogenous antigens. J. Immunol. 165, 3301–3308. [20] Haicheur, N. et al. (2003) The B-subunit of Shiga toxin coupled to full-size protein elicits humoral and cellular immune responses associated with a TH1 dominant polarization. Int. Immunol. 15, 1161–1171. [21] Vingert, B. et al. (2006) The Shiga toxin B-subunit targets antigen in vivo to dendritic cells and elicits anti-tumor immunity. Eur. J. Immunol. 36, 1124–1135. [22] Adotevi, O. et al. (2007) B-subunit of Shiga toxin based vaccines synergize with alpha-galactosylceramide to break tolerance against self antigen and elicit anti-viral immunity. J. Immunol. 179, 3371–3379. [23] Falguie`res, T., Mallard, F., Baron, C.L., Hanau, D., Lingwood, C., Goud, B., Salamero, J. and Johannes, L. (2001) Targeting of Shiga toxin B-subunit to retrograde transport route in association with detergent resistant membranes. Mol. Biol. Cell 12, 2453– 2468. [24] Hegnerelle, X., Plisson, C., Lambert, O., Marco, S., Rigaud, J.L., Johannes, L. and Levy, D. (2002) Two-dimensional structures of the Shiga toxin B-subunit and of a chimera bound to the glycolipid receptor Gb3. J. Struct. Biol. 139, 113–121. [25] Pina, D.G., Johannes, L. and Castanho, M.A. (2007) Shiga toxin B-subunit sequential binding to its natural receptor in lipid membranes. Biochim. Biophys. Acta 1768, 628–636. [26] Pina, D.G., Go´mez, J., England, P., Craescu, C.T., Johannes, L. and Shnyrov, V.L. (2006) Characterization of the non-native trifluoroethanol-induced intermediate conformational state of the Shiga toxin B-subunit. Biochimie 88, 1199–1207. [27] Pina, D.G., Gomez, J., Villar, E., Johannes, L. and Shnyrov, V.L. (2003) Thermodynamic analysis of the structural stability of Shiga toxin B-subunit. Biochemistry 42, 9498–9506. [28] Winzler, C. et al. (1997) Maturation stages of mouse dendritic cells in growth factor-dependent long-term cultures. J. Exp. Med. 185, 317–328. [29] Johannes, L., Tenza, D., Antony, C. and Goud, B. (1997) Retrograde transport of KDEL-bearing B-fragment of Shiga toxin. J. Biol. Chem. 272, 19554–19561. [30] Mallard, F. and Johannes, L. (2003) Shiga toxin B-subunit as a tool to study retrograde transport, in: Methods Mol. Med. Shiga Toxin Methods and Protocols (Philpott, D. and Ebel, F., Eds.), vol. 73, Chapter 17, pp. 209–220.
