DNA binding to zwitterionic model membranes
Supporting Information
DNA binding to zwitterionic model membranes Marie-Louise Ainalem, Nora Kristen, Karen J. Edler, Fredrik Höök, Emma Sparr, and Tommy Nylander Physical Chemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden; Chemistry Department, University of Bath, Claverton Down, Bath BA2 7AY, United Kingdom and Department of Applied Physics, Chalmers University of Technology, 41296 Gothenburg, Sweden
Experimental Section. Formation and characterization of surface-deposited model membranes. The formation of a DOPC bilayer using in situ null ellipsometry (a) as well as QCM-D (b) is shown in Figure S1. To deposit a bilayer using ellipsometry, the lipid-surfactant solution was injected into the cell to reach a total concentration of 0.114 mg mL-1. After reaching steady state, rinsing removed the sugar surfactant. In the second and third step, DOPC:DDM mixtures were repeatedly added but of lower total amount compared to the initial concentration, 10X and 100X dilution, respectively.1 Using QCM-D, the three adsorbing steps were performed using solutions prepared at the desired final concentration since all solutions added to the cell were subjected to circulation through the cell. The flow rate used for adsorption and rinsing was 50 µL min-1 and 100 µL min-1 respectively.
Figure S1. Adsorbed amount () as a function of time (t) for the deposition of a DOPC bilayer using (a) in situ null ellipsometry (denoted Γ) and (b) QCM-D (denoted ∆m). Also shown is the thickness (d) of the adsorbed layer and the dissipation ∆D (for on = 13) in (a) and (b), respectively (). The lipid-containing solution was added at t = 0 as well as when indicated using solid lines. Dashed lines correspond to the start of the rinsing cycles.
1
The d-DOPC bilayers that were used for NR were prepared as according to the bilayer deposition for QCM-D except that the total starting concentration of lipid and surfactant was 0.123 mg mL-1. Figure S2a shows the NR profiles in D2O of the bare surface and the complete bilayer after rinse. Figure S2b-c display the bilayer in the two additional aqueous solvents used; H2O and the D2O/H2O contrast that is matched to Si (cmSi). The fitting parameters for the bilayer components, headgroups (HGs) and acyl chains (ACs), are summarized in Table S1 using the model introduced in the Experimental section.
Figure S2. NR (R) as a function of momentum transfer (Q) for a d-DOPC bilayer. (a) The bare surface () and d-DOPC bilayer () in D2O. (b) and (c) display the reflectivity of the bilayer in cmSi () and H2O (), respectively. Error bars were omitted due to clarity but were all smaller or equal to the marker size. Lines correspond to calculated multilayer models of the reflectivity, see Table S1.
Table S1. Parameters obtained for fits to NR profiles (in D2O, H2O and cmSi) of the complete d-DOPC bilayers deposited on silica substrates. The membrane was modeled as a three-layer structure in which the bilayer was separated into an outer and inner headgroup (HG) region and an intermediate region containing the acyl chains (ACs)a
d (Å)
ξ (Å)
ρ×10-6 (Å-2)
φ (%)
A (Å2)
SiO2
15
3
3.41
16
----
HG
6
1
1.80
58
127
AC
31
1
6.20
49
125
a
The layer thickness is d and the interfacial roughness is described by ξ. The solvent content in the layer equals φ and ρ is the scattering length density of the component. A is the resulting mean molecular area. The gaussian roughness between the layer that is closest to the subphase (the outer HGs) and the subphase is 0 Å.
2
The mechanism in which DOPC is transported from the mixed micelles to the surface has been argued to be via diffusion within the stagnant layer followed by surface spreading.2 This theory was recently supported by QCMD studies on the change in dissipation during the bilayer formation process.3 This is further confirmed by the QCM-D results obtained in this study where ∆D shows sharp peaks for the two first lipid additions, which is indicative of extended structures of attached micelles at the interface. NR data has also been presented where the composition of the adsorbed film was evaluated after each sequential addition of lipid-surfactant mixture.2-4 The amount of solvent was found to increase upon bilayer formation, explained by the replacement of DDM by solvent rather than lipid, which resulted in a low surface coverage. However, both ellipsometry and QCM-D show higher degree of surface coverage compared to NR. This variation could be explained by the different flow profiles in the sample cells since NR uses the largest surface areas and requires the highest solvent volume to achieve complete rinsing. In QCM-D, the crystal substrates are also used repetitively and could become damaged, for example by scratching, whereas the silicon wafers for ellipsometry are only used once.
Results Effect of calcium on the lipid bilayer. Figure S3 shows QCM-D data for the addition of Ca2+ to a surfacedeposited DOPC bilayer. The total surface excess is observed to increase upon Ca2+ addition and we conclude that divalent salt associates with zwitterionic phospholipids which possibly turns them cationic.
Figure S3. Addition of Ca2+ to a surface-deposited DOPC bilayer using QCM-D. Adsorbed amount, ∆m, () is presented as a function of time, t. The solid line indicates the addition of Ca2+ and the dashed line indicates the start of rinsing using an
3
aqueous solution of 10 mM NaBr. At t = 0, the bilayer formation is completed.
Lipid vesicles. DNA adsorption from NaBr. Figure S4 shows the scattering from linearized plasmid DNA in 10 mM NaBr using SANS. The scattering data were modeled using a power-law model in which a flat background (B) was included as according to:
I (Q ) = AQ − m + B
(1)
where the intensity at Q = 1 is A, considered independent of contrast, equal to 2.37×10-7, and decaying with power –m, Table S2. The scattering from mixed samples containing DNA and vesicles were compared with the scattering from free DNA.
Figure S4. SANS spectra showing the intensity as a function of momentum transfer (Q) for DNA in the presence of 10 mM NaBr. The three contrasts displayed are; D2O (), H2O () and 83 % D2O (). Error bars were omitted due to clarity but were all smaller or equal to the marker size. The scattering data were fitted using a power-law model, see Table S2.
Table S2. The power –m with which the intensity of free DNA decays in H2O, D2O and cmSi.
m
H2O
D2O
83 % D2O
2.96
2.84
2.63
4
Figure S5 shows SANS data of DMPC and d54-DMPC vesicles in the presence of DNA and 10 mM NaBr. The scattering curves were fitted using the model for unilamellar vesicles (ULVs) that also was applied to vesicles alone (in solutions containing 10 mM NaBr).5
Figure S5. SANS spectra showing the intensity as a function of momentum transfer (Q) for extruded DMPC and d54-DMPC vesicles in the presence of DNA and 10 mM NaBr. The three contrasts displayed are DMPC in D2O (), DMPC in 83 % D2O () and d54-DMPC in D2O (). The best fit was obtained using the same model as for vesicles in the absence of DNA (Table 4). Error bars were omitted due to clarity but were all smaller or equal to the marker size.
References (1)
Vandoolaeghe, P.; Rennie, A. R.; Campbell, R. A.; Thomas, R. K.; Hook, F.; Fragneto, G.; Tiberg, F.;
Nylander, T. Soft Matter 2008, 4, 2267-2277. (2)
Vacklin, H. P.; Tiberg, F.; Thomas, R. K. Biochim. Biophys. Acta, Biomembr. 2005, 1668, 17-24.
(3)
Vandoolaeghe, P.; Rennie, A. R.; Campbell, R. A.; Nylander, T. Langmuir 2009, 25, 4009-4020.
(4)
Vacklin, H. P.; Tiberg, F.; Fragneto, G.; Thomas, R. K. Langmuir 2005, 21, 2827-2837.
(5)
Bartlett, P.; Ottewill, R. H. J. Chem. Phys. 1992, 96, 3306-3318.
5
6
DNA binding to zwitterionic model membranes Marie-Louise Ainalem, Nora Kristen, Karen J. Edler, Fredrik Höök, Emma Sparr, and Tommy Nylander Physical Chemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden; Chemistry Department, University of Bath, Claverton Down, Bath BA2 7AY, United Kingdom and Department of Applied Physics, Chalmers University of Technology, 41296 Gothenburg, Sweden
Experimental Section. Formation and characterization of surface-deposited model membranes. The formation of a DOPC bilayer using in situ null ellipsometry (a) as well as QCM-D (b) is shown in Figure S1. To deposit a bilayer using ellipsometry, the lipid-surfactant solution was injected into the cell to reach a total concentration of 0.114 mg mL-1. After reaching steady state, rinsing removed the sugar surfactant. In the second and third step, DOPC:DDM mixtures were repeatedly added but of lower total amount compared to the initial concentration, 10X and 100X dilution, respectively.1 Using QCM-D, the three adsorbing steps were performed using solutions prepared at the desired final concentration since all solutions added to the cell were subjected to circulation through the cell. The flow rate used for adsorption and rinsing was 50 µL min-1 and 100 µL min-1 respectively.
Figure S1. Adsorbed amount () as a function of time (t) for the deposition of a DOPC bilayer using (a) in situ null ellipsometry (denoted Γ) and (b) QCM-D (denoted ∆m). Also shown is the thickness (d) of the adsorbed layer and the dissipation ∆D (for on = 13) in (a) and (b), respectively (). The lipid-containing solution was added at t = 0 as well as when indicated using solid lines. Dashed lines correspond to the start of the rinsing cycles.
1
The d-DOPC bilayers that were used for NR were prepared as according to the bilayer deposition for QCM-D except that the total starting concentration of lipid and surfactant was 0.123 mg mL-1. Figure S2a shows the NR profiles in D2O of the bare surface and the complete bilayer after rinse. Figure S2b-c display the bilayer in the two additional aqueous solvents used; H2O and the D2O/H2O contrast that is matched to Si (cmSi). The fitting parameters for the bilayer components, headgroups (HGs) and acyl chains (ACs), are summarized in Table S1 using the model introduced in the Experimental section.
Figure S2. NR (R) as a function of momentum transfer (Q) for a d-DOPC bilayer. (a) The bare surface () and d-DOPC bilayer () in D2O. (b) and (c) display the reflectivity of the bilayer in cmSi () and H2O (), respectively. Error bars were omitted due to clarity but were all smaller or equal to the marker size. Lines correspond to calculated multilayer models of the reflectivity, see Table S1.
Table S1. Parameters obtained for fits to NR profiles (in D2O, H2O and cmSi) of the complete d-DOPC bilayers deposited on silica substrates. The membrane was modeled as a three-layer structure in which the bilayer was separated into an outer and inner headgroup (HG) region and an intermediate region containing the acyl chains (ACs)a
d (Å)
ξ (Å)
ρ×10-6 (Å-2)
φ (%)
A (Å2)
SiO2
15
3
3.41
16
----
HG
6
1
1.80
58
127
AC
31
1
6.20
49
125
a
The layer thickness is d and the interfacial roughness is described by ξ. The solvent content in the layer equals φ and ρ is the scattering length density of the component. A is the resulting mean molecular area. The gaussian roughness between the layer that is closest to the subphase (the outer HGs) and the subphase is 0 Å.
2
The mechanism in which DOPC is transported from the mixed micelles to the surface has been argued to be via diffusion within the stagnant layer followed by surface spreading.2 This theory was recently supported by QCMD studies on the change in dissipation during the bilayer formation process.3 This is further confirmed by the QCM-D results obtained in this study where ∆D shows sharp peaks for the two first lipid additions, which is indicative of extended structures of attached micelles at the interface. NR data has also been presented where the composition of the adsorbed film was evaluated after each sequential addition of lipid-surfactant mixture.2-4 The amount of solvent was found to increase upon bilayer formation, explained by the replacement of DDM by solvent rather than lipid, which resulted in a low surface coverage. However, both ellipsometry and QCM-D show higher degree of surface coverage compared to NR. This variation could be explained by the different flow profiles in the sample cells since NR uses the largest surface areas and requires the highest solvent volume to achieve complete rinsing. In QCM-D, the crystal substrates are also used repetitively and could become damaged, for example by scratching, whereas the silicon wafers for ellipsometry are only used once.
Results Effect of calcium on the lipid bilayer. Figure S3 shows QCM-D data for the addition of Ca2+ to a surfacedeposited DOPC bilayer. The total surface excess is observed to increase upon Ca2+ addition and we conclude that divalent salt associates with zwitterionic phospholipids which possibly turns them cationic.
Figure S3. Addition of Ca2+ to a surface-deposited DOPC bilayer using QCM-D. Adsorbed amount, ∆m, () is presented as a function of time, t. The solid line indicates the addition of Ca2+ and the dashed line indicates the start of rinsing using an
3
aqueous solution of 10 mM NaBr. At t = 0, the bilayer formation is completed.
Lipid vesicles. DNA adsorption from NaBr. Figure S4 shows the scattering from linearized plasmid DNA in 10 mM NaBr using SANS. The scattering data were modeled using a power-law model in which a flat background (B) was included as according to:
I (Q ) = AQ − m + B
(1)
where the intensity at Q = 1 is A, considered independent of contrast, equal to 2.37×10-7, and decaying with power –m, Table S2. The scattering from mixed samples containing DNA and vesicles were compared with the scattering from free DNA.
Figure S4. SANS spectra showing the intensity as a function of momentum transfer (Q) for DNA in the presence of 10 mM NaBr. The three contrasts displayed are; D2O (), H2O () and 83 % D2O (). Error bars were omitted due to clarity but were all smaller or equal to the marker size. The scattering data were fitted using a power-law model, see Table S2.
Table S2. The power –m with which the intensity of free DNA decays in H2O, D2O and cmSi.
m
H2O
D2O
83 % D2O
2.96
2.84
2.63
4
Figure S5 shows SANS data of DMPC and d54-DMPC vesicles in the presence of DNA and 10 mM NaBr. The scattering curves were fitted using the model for unilamellar vesicles (ULVs) that also was applied to vesicles alone (in solutions containing 10 mM NaBr).5
Figure S5. SANS spectra showing the intensity as a function of momentum transfer (Q) for extruded DMPC and d54-DMPC vesicles in the presence of DNA and 10 mM NaBr. The three contrasts displayed are DMPC in D2O (), DMPC in 83 % D2O () and d54-DMPC in D2O (). The best fit was obtained using the same model as for vesicles in the absence of DNA (Table 4). Error bars were omitted due to clarity but were all smaller or equal to the marker size.
References (1)
Vandoolaeghe, P.; Rennie, A. R.; Campbell, R. A.; Thomas, R. K.; Hook, F.; Fragneto, G.; Tiberg, F.;
Nylander, T. Soft Matter 2008, 4, 2267-2277. (2)
Vacklin, H. P.; Tiberg, F.; Thomas, R. K. Biochim. Biophys. Acta, Biomembr. 2005, 1668, 17-24.
(3)
Vandoolaeghe, P.; Rennie, A. R.; Campbell, R. A.; Nylander, T. Langmuir 2009, 25, 4009-4020.
(4)
Vacklin, H. P.; Tiberg, F.; Fragneto, G.; Thomas, R. K. Langmuir 2005, 21, 2827-2837.
(5)
Bartlett, P.; Ottewill, R. H. J. Chem. Phys. 1992, 96, 3306-3318.
5
6
