The Broadly Neutralizing Anti-Human Immunodeficiency Virus Type 1 4E10 Monoclonal Antibody Is Better Adapted to Membrane-Bound Epitope Recognition and Blocking than 2F5

Materials.The sequences AISpreTM, AISpreTM(9, 10)A, AISpreTM(17, 18), AIS, and PreTM displayed in Fig. 1A were synthesized as C-terminal carboxamides by solid-phase synthesis using Fmoc chemistry, and they were purified by high-performance liquid chromatography at the Proteomics Unit of the University Pompeu-Fabra (Barcelona, Spain). Peptide stock solutions were prepared in dimethyl sulfoxide (spectroscopy grade), and the concentrations were determined using a bicinchoninic acid microassay (Pierce, Rockford, IL). NAbs expressing hybridomas were originally generated by combined polyethylene glycol electrofusion of peripheral blood mononuclear cells from HIV-infected nonsymptomatic patients (14). The 2F5 and 4E10 MAbs used in this study were subsequently produced in recombinant CHO cells after the subclass switch to immunoglobulin G1 (IgG1) (33). 1-Palmitoyl-2-oleoylphosphatidylcholine (POPC) and cholesterol (Chol) were purchased from Avanti Polar Lipids (Birmingham, AL), while dodecylphosphocholine (DPC) was from Anatrace (Maumee, OH). The N-(7-nitro-benz-2-oxa-1,3-diazol-4-yl)phosphatidylethanolamine (N-NBD-PE), N-(lissamine rhodamine B sulfonyl)phosphatidylethanolamine (N-Rh-PE), N-(fluorescein-5-thiocarbamoyl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoeth-anolamine (f-DHPE), N-(5-dimethylaminonaphthalene-1-sulfonyl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine (d-DHPE), 8-aminonaphthalene-1,3,6-trisulfonic acid sodium salt (ANTS), and p-xylenebis(pyridinium)bromide (DPX) fluorescent probes were from Molecular Probes (Junction City, OR), while the PE-Cy5 mouse anti-human IgG was purchased from BD Biosciences.

Infrared spectroscopy.Infrared spectra were recorded at 25°C in a Bruker Tensor 27 spectrometer equipped with a mercury-cadmium-telluride detector using a Peltier-based temperature controller (TempCon; BioTools Inc., Wauconda, IL) with calcium fluoride cells (BioCell; BioTools Inc., Wauconda, IL). Typically, 370 scans were collected for each background and peptide-lipid sample prepared in D₂O-based buffer, and the spectra were obtained with a nominal resolution of 2 cm⁻¹. Data treatment and band decomposition of the original amide I have been described in detail elsewhere (4-6).

Production of vesicles.Large unilamellar vesicles (LUV) were prepared by following the extrusion method of Hope et al. (29). Phospholipids and Chol were mixed in chloroform and dried under an N₂ stream. Traces of organic solvent were removed by overnight vacuum pumping. Subsequently, the dried lipid films were dispersed in 5 mM HEPES-100 mM NaCl (pH 7.4) buffer and subjected to 10 freeze-thaw cycles prior to extrusion 10 times through two stacked polycarbonate membranes (Nuclepore, Inc., Pleasanton, CA). The size distributions of the vesicles were determined using a Zeta-Sizer Nano ZS instrument (Malvern Instruments, Malvern, United Kingdom). Extrusion through membranes with nominal pore sizes of 0.1 and 0.2 μm produced POPC-Chol (2:1, mol/mol) LUV with mean diameters (± standard deviations) of 90 ± 17 and 132 ± 36 nm, respectively. Lipid concentrations of liposome suspensions were determined by phosphate analysis.

Fluorimetric assays.Peptide partitioning into membranes was evaluated by monitoring the change in the emitted Trp fluorescence. Corrected spectra were recorded using an SLM Aminco 8100 spectrofluorimeter (Spectronic Instruments, Rochester, NY) with excitation set at 280 nm and 5-nm slits. Partitioning curves were subsequently computed from the fractional changes in emitted Trp fluorescence when peptides were titrated with lipids at increasing concentrations. The signal was further corrected for dilution and inner filter effects (80) using NATA (N-acetyl-l-tryptophanamide), a soluble Trp analogue that does not partition into the membranes. The apparent mole fraction partition coefficients, K_x, were determined by fitting the experimental values to the hyperbolic function F/F₀ = 1 + {[(F_max/F₀) − 1][L]/(K + [L])}, where F and F₀, are the fluorescence intensities measured in the presence and absence of vesicles, respectively, [L] is the lipid concentration, and K is the lipid concentration at which the bound peptide fraction is 0.5. Therefore, K_x is given by [W]/K, where [W] is the molar concentration of water. Experimental free energies (ΔG_obs) of partitioning (Table 1) were subsequently inferred from the following expression (79): $$mathtex$$\[{\Delta}G_{\mathrm{obs}}{=}{-}RT\mathrm{ln}K_{x}\]$$mathtex$$(1) where R is the universal gas constant and T is temperature.

The release of vesicular contents into the medium was monitored by the ANTS/DPX assay (23). LUV containing 12.5 mM ANTS, 45 mM DPX, 20 mM NaCl, and 5 mM HEPES were obtained by separating the unencapsulated material by gel filtration on a Sephadex G-75 column eluted with 5 mM HEPES and 100 mM NaCl (pH 7.4). Osmolarities were adjusted to 200 mosM in a cryoscopic osmometer (Osmomat 030; Gonotec, Berlin, Germany). Fluorescence measurements were performed by setting ANTS emission at 520 nm and excitation at 355 nm. A cutoff filter (470 nm) was placed between the sample and the emission monochromator. Zero percent leakage corresponded to the fluorescence of the vesicles at time zero, whereas 100% leakage was the fluorescence value obtained after addition of Triton X-100 (0.5%, vol/vol).

Partitioning into the membrane interface as a function of time was determined by measuring energy transfer from the Trp peptide to the surface d-DHPE fluorescent probe as in reference 37. In brief, 6 mol% of the d-DHPE probe was included in the target vesicle composition, and its fluorescence was measured at an emission wavelength of 510 nm, while the excitation wavelength was that of the Trp residue (280 nm).

Membrane lipid mixing was monitored using the resonance energy transfer assay described by Struck et al. (69). The assay is based on the dilution of N-NBD-PE and N-Rh-PE, whereby dilution due to membrane mixing results in increased N-NBD-PE fluorescence. Vesicles containing 0.6 mol% of each probe were mixed with unlabeled vesicles at a ratio of 1:4 (final lipid concentration, 0.1 mM). The NBD emission was monitored at 530 nm, with the excitation wavelength set at 465 nm. A cutoff filter at 515 nm was used between the sample and the emission monochromator to avoid scattering interference. The fluorescence scale was calibrated such that the zero level corresponded to the initial residual fluorescence of the labeled vesicles and the value of 100% corresponded to the complete mixing of all the lipids in the system. The latter value was set by the fluorescence intensity of vesicles labeled with 0.12 mol% of each fluorophore at the same total lipid concentration as in the fusion assay.

Monolayer penetration.Surface pressure was determined in a fixed-area circular trough (μTrough S system; Kibron, Helsinki, Finland), and the measurements were taken at room temperature with constant stirring in an aqueous phase consisting of 1 ml 5 mM HEPES and 100 mM NaCl (pH 7.4). Lipid mixtures dissolved in chloroform were spread over the surface, and the desired initial surface pressure (π₀) was attained by changing the amount of lipid applied to the air-water interface. Peptide was injected into the subphase with a Hamilton microsyringe. At the concentrations used, the peptide alone induced only a negligible increase in surface pressure at the air-water interface.

Enzyme-linked immunosorbent assay (ELISA).Peptides were dissolved in phosphate-buffered saline and immobilized (100 μl/well) overnight in C96 Maxisorp microplate wells (Nunc, Denmark) at a concentration of 1 μM. Prior to incubation with the MAbs, the plates were blocked for 2 h with 3% (wt/vol) bovine serum albumin in phosphate-buffered saline. The binding of the MAbs was detected with an alkaline phosphatase-conjugated goat anti-human immunoglobulin (Pierce, Rockford, IL), which then catalyzed a color reaction with the p-nitrophenylphosphate substrate (Sigma, St. Louis, MO) that could be determined by measuring the absorbance at a wavelength of 405 nm in a Synergy HT microplate reader (Bio-TEK Instruments Inc., VT).

Electron microscopy.Experimental evidence for stable antibody-vesicle association was initially obtained by electron microscopy. Liposome-MAb samples were prepared by following a protocol analogous to that used for the cytometric analysis (see below). Samples were subsequently incubated on copper grids for 2 min, dried, and finally stained for 1 min with 2% uranyl acetate. Electron micrographs were obtained using a JEOL 1200EX II microscope operated at 100 kV and recorded on Kodak SO-163 film.

Flow cytometry.Antibody-vesicle association was also determined quantitatively using a FACSCalibur flow cytometer (Becton Dickinson Immunocytometry Systems, Mountain View, CA). Measuring conditions were initially optimized by attending to the following experimental parameters: (i) vesicle size (LUV extruded through polycarbonate filters with nominal sizes of 100, 200, 400 and 1,000 nm and multilayered vesicles were compared), (ii) lipid concentration in the samples, and (iii) percentage of fluorescent probe in the membrane tested with 1, 0.1, 0.01, and 0.001 mol of f-DHPE. The optimal conditions were set for LUV obtained through extrusion using filters with a nominal pore size of 200 nm, a lipid concentration of 0.25 mM, and 0.01 mol% fluorescent phospholipid probe. Thus, in a typical experiment fluorescently labeled vesicles doped with peptide epitopes (1:100 peptide/lipid molar ratio) were incubated for 10 min with increasing concentrations of the 2F5 or 4E10 MAbs and, subsequently, for 5 min with twice the concentration of anti-human IgG coupled to cyanine (IgG-PE-Cy5).

Partitioning of AISpreTM into membranes: structure and energetics.The aim of this study was to compare the abilities of the 2F5 and 4E10 MAbs to recognize complete membrane-inserted epitopes. Thus, the AISpreTM boundaries (residues 656 to 683) were defined as containing the full-length 2F5 and 4E10 epitopes, as well as both interfacial subdomains (Fig. 1A). The intrinsic capacity of the designed AISpreTM sequence to translocate into membranes and fold therein was first analyzed based on the sequence's amino acid composition (Table 1). Water-to-membrane interface transfer free energies for all amino acids have been determined by Wimley and White (79, 81). These values reflect real thermodynamic measurements compiling the energetic components that dictate the initial partitioning of unfolded peptide sequences into membranes. Furthermore, they are additive, which allows computation of the energetics of partitioning for arbitrary protein sequences. Computation of the overall free energies of partitioning for the unfolded AISpreTM and PreTM sequences rendered values on the order of −2.8 and −8.1 kcal/mol, respectively (Table 1). In accordance with equation 1 (Materials and Methods), these values corresponded to partitioning constants (K_x) on the order of 10² and 10⁶, which reflects the much higher solubility or reduced membrane affinity of the AISpreTM sequence in comparison to that of PreTM. Those K_x values actually predicted that AISpreTM partitioning would be insignificant at lipid concentrations that would otherwise permit substantial PreTM-membrane association (e.g., more than 60% of the added peptide at a lipid concentration of 100 μM).

However, C_α chain polarity may be reduced upon folding as an α-helix, which results in an increase of interfacial hydrophobicity (35, 78). Free energies of partitioning recalculated taking into account this conformational energy rendered K_x values higher than 10⁷ in both cases, consistent with the quantitative incorporation of PreTM and AISpreTM into membranes. Thus, from the sequence it can be inferred that decreasing the C_α chain polarity by adopting a defined secondary structure is a prerequisite for the efficient incorporation of AISpreTM into vesicles (35, 78, 79). In contrast, the free energy values computed for the AIS predicted that this gp41 stretch alone would not associate with the membrane interface even if it adopted an α-helix structure (Table 1). This implies that the embedding of the complete 2F5 epitope into the membrane interface would require its folding into an α-helix, as well as extending the sequence at its C terminus to include the aromatic PreTM residues.

Thus, we next set out to assess experimentally the structure adopted by AISpreTM in a membrane environment and to explore the conditions that permit the efficient incorporation of the peptide from water into lipid bilayers (see Fig. S1 and S2 and Table S1 in the supplemental material). Quantitative infrared spectroscopy determinations were indeed consistent with the helical structuring of the 2F5 and 4E10 epitopes in the context of membrane-bound AISpreTM. Decomposition of the amide I bands of PreTM and AISpreTM confirmed that these peptides adopted very similar secondary structures in DPC micelles (see Fig. S1 and Table S1 in the supplemental material). In both cases the α-helical conformation was predominant, although its contribution was slightly higher in the AISpreTM sequence. By contrast, the AIS peptide amide I band could not be fitted to a major component, consistent with the absence of a single main secondary structure. In this case, a strong contribution of unordered structures coexisted with a mixture of canonical structures. Notably, bands that could be assigned to β-turns and/or 3₁₀-helices contributed to the same extent as α-helices and β-sheet bands. Thus, these infrared spectroscopy results confirmed that the conformational plasticity previously reported for peptides representing the extended 2F5 epitope (7, 10) was maintained in the nonpolar environment provided by DPC micelles. Moreover, they also suggested that this flexible sequence was constrained into a main α-helical structure when anchored to the membrane interface as a part of AISpreTM.

We next analyzed experimentally AIS and AISpreTM membrane association (see Fig. S2 in the supplemental material). The monolayer insertion results reflected exclusion pressures (levels of lateral compression precluding insertion) of 43 and 28 mN m⁻¹ for AISpreTM and AIS, respectively, indicating that AISpreTM was capable of penetrating into comparatively more packed lipid bilayers (see Fig. S2A in the supplemental material). Moreover, while AISpreTM and PreTM clearly inserted at above the lateral pressures existing in biological membranes (π₀ ≥ 30 to 35 mN m⁻¹; Table 1), it was predicted that AIS was devoid of the capacity for inserting into lipid bilayers that are not expanded artificially (40). These observations were further confirmed by measurements of peptide insertion into LUV (see Fig. S2B in the supplemental material). LUV lipid bilayers are compressed to lateral pressures in the range of those existing in natural membranes, and, therefore, peptides showing exclusion pressures below 30 mN m⁻¹ are predicted not to insert into these vesicles. AISpreTM insertion into LUV bilayers could be deduced from the changes in the fluorescence Trp spectrum obtained upon titrating peptides with lipid vesicles (see Fig. S2B in the supplemental material). The polarity of the environment surrounding Trp residues decreased upon penetration of the peptide into the bilayer, which resulted in Trp fluorescence intensity enhancement and shifting of its maximum toward lower wavelengths in the presence of the vesicles (see Fig. S2B, left panel, in the supplemental material). The corresponding partitioning curve reflects the fraction of peptide bound to vesicles as the amount of lipid was augmented (see Fig. S2B, right panel, in the supplemental material). Fitting the experimental values to function (1) disclosed K_x values higher than 10⁷ for AISpreTM, in agreement with the free energy values calculated for a structured sequence (Table 1). By contrast no changes in AIS Trp emission were detected in the presence of vesicles (see Fig. S2 in the supplemental material). Thus, the MPER AIS stretch alone displayed no significant affinity for membranes, while extending the sequence to include the complete PreTM augmented the interfacial affinity so as to produce almost complete incorporation into vesicles at lipid concentrations above 100 μM (>95% of added peptide associated with membranes). Together, the results confirm that the AISpreTM sequence spontaneously inserts into electrically neutral vesicles, adopting an equilibrium α-helix structure therein.

The blocking of the AISpreTM-induced membrane restructuring by MAbs 2F5 and 4E10.In a previous paper we showed that Chol sustained the lytic and fusogenic activities of the PreTM peptide, as measured in a vesicular system (59). Since the membrane activity of PreTM is preserved within the longer sequence, AISpreTM also altered the integrity of Chol-containing vesicles (Fig. 2A). Indeed, there was extensive leakage of the aqueous contents from the vesicle in the presence of 1 μM peptide (>60%), while comparable levels of membrane-lipid mixing required a dose four times higher. In these assays AISpreTM incorporation into vesicles as a function of time was fluorescently monitored using the membrane surface-anchored d-DHPE probe. Close contact with inserting peptide residues enables the energy transfer process that results in d-DPHE emission upon excitation of Trps at 280 nm. In both cases, kinetic traces of d-DHPE fluorescence increase demonstrated that the process of peptide incorporation into vesicles occurred within a few seconds (Fig. 2A). By contrast and in accordance with the rearrangement of peptide monomers at the membrane surface (59), the kinetics of leakage and fusion were slower and both processes required several minutes to stabilize (Fig. 2A).

• Open in new tab
• Download powerpoint

FIG. 2.

AISpreTM-induced membrane-restructuring in POPC-Chol (2:1, mol/mol) vesicles and its inhibition by the 2F5 and 4E10 MAbs. (A, left and center) Continuous lines correspond to the kinetic traces of leakage (left) or fusion (center) upon AISpreTM injection into stirred solutions of LUV (arrows) at a final peptide concentration of 1 or 4 μM, respectively. The dotted traces follow the incorporation of AISpreTM into the vesicles monitored through energy transfer from peptide tryptophans to membrane-residing d-DHPE. (Right) Percentages of AISpreTM-induced leakage (circles) and lipid mixing (squares) after a 30-min incubation with vesicles plotted as a function of the peptide concentration. The L B arrow indicates the condition selected to measure MAb-induced blocking of leakage and MAb binding to vesicles, and the F arrow indicates that selected to measure MAb-induced blocking of fusion. The final lipid concentration was 100 μM in all cases. (B) Inhibition of ANTS leakage (top) and mixing of vesicle lipids (bottom). LUV suspensions (100 μM lipid) were treated with 1 or 4 μM AISPreTM, respectively, and subsequently they were supplemented with 10 μg/ml of 2F5 or 4E10 MAbs (addition times indicated by 1 and 2, respectively). Dotted traces correspond to the controls in the absence of antibody. (C) Dose dependency of AISpreTM-induced ANTS leakage and fusion inhibition by the 4E10 MAb (circles) or 2F5 MAb (squares). The initial rates were determined as the increase in fluorescence during the 20 s that followed MAb addition. Inhibition percentages were plotted as a function of the MAb concentration, and the data are represented as the means ± standard deviations of three (leakage) and four (fusion) independent experiments.

The fast incorporation and slower membrane restructuring induced by AISpreTM allowed the capacities of MAbs 2F5 and 4E10 to block MPER membrane activity to be compared by assessing the effect of adding the MAbs on AISpreTM-induced leakage or fusion (Fig. 2B, top and bottom, respectively). Accordingly, the peptide was added to a stirred solution of lipid vesicles (1) and the MAb was injected into the mixture after the peptide had incorporated into the vesicles according to the d-DHPE assay (2). Consistent with the recognition of membrane-bound species under these conditions, the 4E10 MAb efficiently interfered with both leakage and fusion (right), whereas injection of similar doses of the 2F5 MAb had almost no effect on these processes (left).

The inhibition of membrane restructuring was dose dependent (Fig. 2C), and the percentage of inhibition could be inferred from the sudden reductions in the kinetic slopes. For leakage (left), 50% inhibitory concentrations on the order of 30 and 200 nM were observed for the 4E10 and 2F5 MAbs, respectively, and thus, the 4E10 MAb was approximately 10 times more potent than the 2F5 MAb in inhibiting AISpreTM-induced leakage. The 50% inhibitory concentration for the inhibition of fusion was on the same order in the case of the 4E10 MAb, while the 2F5 MAb was unable to completely inhibit fusion within the concentration range tested. Thus, the difference in the capacity to inhibit AISpreTM-induced fusion was even higher.

Stable AISpreTM-mediated MAb-bilayer association.In order to confirm the in-membrane recognition of the inserted AISpreTM, we evaluated MAb-liposome association under comparable experimental conditions (i.e., stirred, diluted liposome samples). The existence of a stable association of the antibody with the surface of vesicles was first assessed by electron microscopy (Fig. 3). Samples of control untreated POPC-Chol vesicles showed heavily stained membrane edges that clearly contrasted with the surrounding medium (Fig. 3A), and a similar morphology was displayed by vesicles incubated with AISpreTM alone at concentrations that induced leakage from individual vesicles but not fusion (Fig. 3B). However, incubation of these peptide-containing vesicles with the 4E10 (Fig. 3D and E) or 2F5 (Fig. 3G and H) MAb produced vesicles with electron-dense particles and punctate edges, which were consistent with antibody molecules protruding from the bilayer surface. These antibody particles could not be detected when POPC-Chol vesicles devoid of the peptide epitope were exposed to the MAbs (Fig. 3C and F). Thus, the morphological data support a physical interaction of the MAbs with the lipid bilayer surrounding the vesicles in the presence of AISpreTM.

• Open in new tab
• Download powerpoint

FIG. 3.

Physical MAb-bilayer association as seen by electron microscopy in POPC-Chol (2:1) LUV. (A and B) Electron micrographs of control vesicles devoid of peptide and AISpreTM-containing vesicles (peptide-to-lipid mole ratio, 1:100), respectively. (D, E, G, and H) AISpreTM-containing vesicles incubated with 4E10 (D and E) or 2F5 (G and H) MAbs. These samples consist of vesicles (1 mM) incubated for 5 min with AISpreTM and then supplemented with 50 μg/ml MAb before staining with uranyl acetate. (C and F) Control samples devoid of the AISpreTM peptide. Bars = 100 nm.

This association of the MAbs with the vesicles was next monitored by fluorescence-activated cell sorting (FACS) set up to simultaneously detect liposome and antibody fluorescent signals (Fig. 4; see Materials and Methods). As judged from the appearance of a main antibody-stained vesicle population (Fig. 4A), both Mabs efficiently associated with POPC-Chol vesicles that contained AISpreTM (>90% of the vesicles were labeled with fluorescent antibody upon incubation with 50 μg/ml MAb). In contrast, when the MAbs were incubated under comparable conditions with POPC-Chol liposomes devoid of peptides, the antibody-stained vesicles constituted a minimal fraction (<5% of the vesicle population).

• Open in new tab
• Download powerpoint

FIG. 4.

Stable MAb-vesicle association as determined by flow cytometry. (A) The 4E10 and 2F5 MAbs (50 μg/ml) were incubated with a stirred solution of f-DHPE-labeled LUV (250 μM) under experimental conditions that were otherwise similar to those for leakage blocking (Fig. 2). After 10 min, the fluorescently labeled secondary antibody was added and the resulting mixtures were incubated for 5 min before being analyzed by flow cytometry. The incubation of AISpreTM-containing POPC-Chol vesicles with the 4E10 and 2F5 MAbs rendered double-labeled vesicles (center and top right panels, respectively). This pattern could not be observed in the absence of the MAbs (left panels) or when the MAbs were incubated with bare POPC-Chol vesicles (center and bottom right panels). (B) Differential association of the 4E10 and 2F5 MAbs with AISpreTM-containing POPC-Chol vesicles (1:100 peptide-to-lipid molar ratio) as determined by flow cytometry. The 4E10 (blue traces) and 2F5 (red traces) MAbs were incubated at the final concentrations (μM) indicated with AISpreTM-containing LUV under conditions that were otherwise similar to those in the panel A. Black traces correspond to control samples not incubated with MAb.

Hence, this flow cytometry analysis confirmed that AISpreTM mediated the stable association of MAbs with the lipid bilayer. As a consequence, we performed further cytometric measurements to quantitatively compare the capacities of both MAbs to stably associate with vesicles doped with AISpreTM (Fig. 4B). These measurements served to ascertain whether the weaker inhibition induced by the 2F5 MAb was due to reduced binding or whether the binding of this MAb to AISpreTM did not effectively block membrane activity. Titration data confirmed that the 4E10 MAb associated with vesicles at lower concentrations (Fig. 4B). Moreover, in both cases a population of MAb-stained vesicles could be recovered at the MAb concentrations that inhibited leakage. Together, these results support the notion that stable association of the 2F5 and 4E10 MAbs with peptide-containing POPC-Chol vesicles correlates with their capacity to inhibit AISpreTM-induced membrane restructuring.

Specificity of membrane-bound AISpreTM recognition and blocking.Our previous results are consistent with the in situ recognition and blocking of membrane-inserted AISpreTM, and, in addition, they suggest that the 4E10 MAb is more potent in mediating both effects. In order to demonstrate the involvement of epitope recognition in these processes, we analyzed sequences in which Ala substituted for Asp9 and Lys10 [AISpreTM(9,10)A] or Trp17 and Phe18 [AISpreTM(17,18)A] to control for the specificity of the interaction of the 2F5 and 4E10 Mabs, respectively (Fig. 5). These substitutions were inspired by alanine-scanning mutagenesis, which pinpointed these residues as crucial for the neutralizing activity of the 2F5 and 4E10 Mabs (87). The AISpreTM(9,10)A and AISpreTM(17,18)A sequences were therefore designed to produce the strongest effect on recognition by only one of the two MAbs, while reducing the effect on the membrane restructuring efficiency of the peptide variants as much as possible.

• Open in new tab
• Download powerpoint

FIG. 5.

Dependence of MAb blockage of induced membrane restructuring and MAb-vesicle association on specific epitope recognition. (A) Binding of the 2F5 (red) and 4E10 (blue) MAbs to 1 μM AISpreTM(9,10)A (left) or AISpreTM(17,18)A (right) immobilized on ELISA plaques. (B) Blocking of leakage induced by the addition of 2 μM AISpreTM(9,10)A (left) or 1 μM AISpreTM(17,18)A (right) to POPC-Chol vesicles by the 2F5 (red) and 4E10 (blue) MAbs. The black trace corresponds to the control without MAb. The MAbs were added at a concentration of 50 μg/ml. The insets display the blocking induced by the same concentration of the 2F5 (left) or 4E10 (right) MAb on the leakage induced by parental AISpreTM. (C) Flow cytometry determination of the association of the 2F5 (red) and 4E10 (blue) MAbs with POPC-Chol vesicles bearing AISpreTM(9,10)A (left) or AISpreTM(17,18)A (right). MAbs were incubated with the vesicles at a concentration of 30 μg/ml.

ELISA studies confirmed the failure of the 4E10 MAb to recognize AISpreTM(17,18)A while it efficiently bound to plaques coated with the AISpreTM(9,10)A variant (Fig. 5A). Likewise, the 2F5 MAb bound to plates coated with AISpreTM(17,18)A but not to plates coated with AISpreTM(9,10)A. We then evaluated the capacity of these MAbs to block the vesicle permeabilization induced by these variants in phosphatidylcholine (POPC)-Chol vesicles (Fig. 5B). Both MAbs were added at doses that almost completely arrested the process induced by the wild-type AISpreTM (Fig. 5B, insets). In accordance with specific recognition in membranes, the 4E10 MAb blocked AISpreTM(9,10)A-induced but not AISpreTM(17,18)A-induced leakage. Conversely, the 2F5 MAb induced the opposite effect, and, specifically, it arrested AISpreTM(17,18)A-induced leakage while it exerted almost no effect on the process induced by the AISpreTM(9,10)A variant. The results of a FACS analysis (Fig. 5C) confirmed that the membrane-blocking activity correlated with MAb-vesicle association. Thus, vesicles labeled with fluorescent antibodies were recovered from a mixture of AISpreTM(17,18)A and MAb 2F5 or AISpreTM(9,10)A and MAb 4E10 but not from samples in which AISpreTM(17,18)A and MAb 4E10 or AISpreTM(9,10)A and MAb 2F5 were combined.