Impact of Ligand Presentation on Protein Binding at Lipid Membrane Interfaces

Hyunsook Jung†, Tinglu Yang†, Mauricio D. Lasagna‡Jinjun Shi†,Gregory D. Reinhart‡, and Paul S. Cremer†

†Department of Chemistry, College Station, TexasA&MUniversity,

P. O. Box 30012, College Station, TX77843

‡Department of Biochemistry & Biophysics, TexasA&MUniversity, College Station, Texas77843

SUPPLEMENTARY MATERIALS

Figures S1-S6 areprovided below along with the corresponding figure legends.

FIGURE S1 Fluorescence anisotropy measurements. (A) B4F binding to

anti-biotin antibody (rF = 0.017 (± 0.001), rB = 0.200 (± 0.001)). (B) D5F binding to anti-DNP antibody (rF = 0.065 (± 0.001), rB = 0.190 (± 0.001)). The anisotropy values of free B4F and D5F were first measured in dilute solution (1 nM in PBS buffer, pH 7.2). Next, aliquots of antibody solution were added until no additional increase in the fluorescence anisotropy value could be observed. The bound anisotropy values are nearly the same for the two ligands as a function of IgG binding. This suggests that the microenvironment’s influence on the dynamic properties of the B4F/anti-biotin and D5F/anti-DNP complexes is relatively similar.

FIGURE S2 A solution binding measurement for DNP-PEG2000/anti-DNP by the fluorescence quenching method. The decrease in fluorescence intensity of anti-DNP, excited at 280 nm, was monitored through an emission filter (WG335, Schott) at varying DNP-PEG2000 concentrations. The intensity was corrected for dilution and collisional quenching. Each data point represents the average of three independent measurements. The value of KD is 3.5 ±0.6 nM. Detailed experimental procedures and data analysis are described elsewhere(1-2).

Reference

  1. Pisarchick, M. L., and N. L. Thompson. 1990. Binding of a monoclonal antibody and its Fab fragment to supported phospholipid monolayers measured by total internal reflection fluorescence microscopy. Biophys. J. 58:1235-1249.
  2. Anglister, J, T. Frey, and H. M. McConnell. 1984. Magnetic resonance of a monoclonal anti-spin-label antibody. Biochemistry. 23:1138-1142

FIGURE S3 Analytical reversed phase-HPLC chromatogram of nascently synthesized D5F. The gradient went from a 50/50 volume mixture of 0.05 % HCOOH in H2O and 0.05 % HCOOH in CH3OH to only the second component in a 45 min. time span. The flow rate was 2 mL/min and the column was a Resource RPC 3 mL packed with Source 15 RPC (Amersham Biosciences, Piscataway, NJ). Proton NMR studies were also performed. The results were as follows: 1H NMR (300 MHz, CD3OD) δ1.88-1.75 (m, 4H), 3.47-3.68 (m, 6H), 6.57-6.60 (dd, 2H, J = 8.7, 2.1 Hz), 6.70-6.74 (m, 2H), 7.19-7.22 (d, 1H, J = 9.3 Hz), 7.30-7.33 (d, 1H, J = 8.1 Hz), 8.14-8.17 (d, 1H, J = 7.5 Hz), 8.29-8.54 (m, 4H), 9.03-9.04 (d, 1H, J = 2.1 Hz).

FIGURE S4 (A) A representative TIRFM image of a nine microchannel device. The channels are 130 μm wide, 15 μm deep, and separated by 160 μm barriers. In this case Alexa Fluor 594-conjugated anti-biotin antibodies bound to lipid bilayers with 5 mol % biotin-cap-PE and 95 mol % POPC. (B) A fluorescence intensity line scan across the microchannel device (red-dotted line in (A)).


FIGURE S5 A TIRFM control experiment to demonstrate that NH2-PEG2000-PE does not bind with anti-DNP antibodies at 1 µM concentration in bulk solution. (1) ~1 mol % DNP-PEG2000-PE/99mol % POPC, (2) POPC, and (3) 1 mol % NH2-PEG2000-PE/99 mol % POPC.


FIGURE S6 MALDI-MS analysis of (A) SUVs and (B) SLBs. The vesicle composition (5 mol % biotin-cap-PE/95 mol % POPC) was analyzed by spotting 1 L of the SUV solution with 1 L of matrix (20 mg/mL, 2, 4, 6-trihydroxylacetophenone with 12 mM Na acetate in methanol). Following mixing, the sample was spotted onto the MALDI plate and analyzed. Two peaks corresponding POPC and biotin-DPPE were obtained in spectrum (A). Next, supported bilayers were formed by the vesicle fusion method from the same batch of vesicles. After the supported bilayers were formed, the sample was freeze-dried. The glass support, coated with the dried SLB layer, was directly attached to a MALDI plate and the same matrix was aerosol deposited on the sample. Again, the same two peaks corresponding to POPC and biotin-cap-PE were obtained in (B).

FIGURE S7 Comparison of the ratio of two dye lipids (R = INBD-PE / IPEG-PE) in SUVs and SLBs. (A) Intact vesicles consisting of 3 mol % NBD-PE (Avanti Polar Lipids, Alabaster, AL)/4 mol % dye-labeled PEG2000-PE/1 mol % biotin-cap-PE / 92 mol % POPC were attached to a lipid bilayer (1 mol % biotin-cap-PE/99 mol % POPC) via biotin-streptavidin (0.2 mg/mL) binding. NH2-PEG2000-PE was labeled with an Alexa Fluor 594 labeling kit. The intact state of the vesicles at the surface was verified by fluorescence recovery after photobleaching (FRAP) measurements. The data showed no evidence for fluorescence recovery (middle two micrographs). (B) The same vesicles as in (A) were fused to a clean glass support. The mobility of the SLB was confirmed by FRAP. The similar R values suggest that the lipid composition from the intact vesicles and in the SLBs were nearly identical within experimental error.

1