Building Polyelectrolyte Multilayers with Calmodulin: A Neutron and X

Publication Date (Web): April 5, 2017 ... There are no indications that ligand TFP can be involved in the multilayer buildup due to strong CaM-PAH int...
1 downloads 0 Views 1MB Size
Download PDF
Subscriber access provided by University of Colorado Boulder

Article

Building polyelectrolyte multilayers with calmodulin – a neutron and X-ray reflectivity study Süleyman Cinar, Simone Möbitz, Samy Al-Ayoubi, Beatrix-Kamelia Seidlhofer, and Claus Czeslik Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b00651 • Publication Date (Web): 05 Apr 2017 Downloaded from http://pubs.acs.org on April 8, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Building polyelectrolyte multilayers with calmodulin – a neutron and X-ray reflectivity study Süleyman Cinar,1 Simone Möbitz,1 Samy Al-Ayoubi,1 Beatrix-Kamelia Seidlhofer,2 Claus Czeslik*1 1

TU Dortmund University, Department of Chemistry and Chemical Biology, D-44221 Dortmund, Germany 2

Helmholtz-Zentrum Berlin, D-14109 Berlin, Germany

ABSTRACT We have studied the formation and functional properties of polyelectrolyte multilayers where calmodulin (CaM) is used as polyanion. CaM is known to populate distinct conformational states upon binding Ca2+ and small ligand molecules. Therefore, we have also probed the effects of Ca2+ ions and trifluoperazine (TFP) as ligand molecule on the interfacial structures. Multilayers with the maximum sequence PEI-(PSS-PAH)x-CaM-PAH-CaM-PAH have been deposited on silicon wafers and characterized by X-ray and neutron reflectometry. From the analysis of all data, several remarkable conclusions can be drawn. When CaM is deposited for the second time, a much thicker sub-layer is produced than in the first CaM deposition step. However, upon rinsing with PAH, very thin CaM-PAH sub-layers remain. There are no indications that the ligand TFP can be involved in the multilayer build-up due to strong CaMPAH interactions. However, there is a significant increase of the multilayer thickness upon removal of Ca2+ ions from holo-CaM and an equivalent decrease of the multilayer thickness upon subsequent saturation of apo-CaM with Ca2+ ions. Presumably, CaM can still be toggled between an apo- and holo-state, when it is embedded in polyelectrolyte multilayers, providing an approach to design bio-responsive interfaces.

1 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

INTRODUCTION The formation of polyelectrolyte multilayers (PEMs) provides an efficient way to cover material surfaces with a huge variety of different polyelectrolytes.1-6 It has soon been realized that proteins can be embedded in or adsorbed on such multilayers to design bio-functionalized interfaces.7-10 It could be demonstrated that the secondary structure of proteins embedded in PEMs can largely be retained.11,12 Furthermore, enzymes show catalytic behavior, when they are incorporated in PEMs, suggesting a native-like structure and some degree of internal dynamics.13-15 In some cases, diffusion of proteins and peptides within PEMs has also been recorded. For example, poly(L-lysine) and lysozyme can move in multilayers composed of poly(L-lysine) and hyaluronic acid.16,17 This diffusion can explain an exponential growth of PEMs, which is characterized by an increasing adsorbed mass with the number of deposition steps.17-19 Moreover, translational mobility and conformational flexibility has been inferred for insulin adsorbed on hyaluronic acid, where fibrillar structures could be observed.20 Thus, the conformation, biological activity and translational diffusion of proteins in PEMs have been characterized in some detail so far.

Figure 1. Calmodulin (CaM) populates distinct conformational states upon binding Ca2+ ions and ligand molecules, such as trifluoperazine (TFP). From left to right, apo-CaM, holo-CaM (Ca2+ shown in green), and holo-CaM with 4 bound TFP molecules (TFP shown in red). The images were prepared with PyMOL v1.3r1 using the PDB IDs 1DMO, 3CLN, and 1LIN.

Apparently, proteins that are adsorbed on or embedded in PEMs preserve some conformational freedom, because this is essential for enzymatic activity. However, there are strong interactions of the proteins with oppositely charged polyelectrolytes, which will limit this freedom to some degree. There is still a lack of knowledge about the extent of such conformational freedom of proteins adsorbed on or embedded in PEMs. To specifically address the conformational freedom of proteins in PEMs, we have used calmodulin (CaM) as polyanion in this study (Fig. 1). CaM is a messenger protein that is involved in many processes such as inflammation and metabolism. Its secondary structure is dominated by αhelices.21 It has a molar mass of about 16800 g mol-1 (148 amino acid residues) and a strong negative net charge of about -15 at neutral pH values with four Ca2+ ions.22 Ca2+ free (apo) CaM and Ca2+ saturated (holo) CaM have similar radii of gyration of 21.5 and 21.9 Å, respectively.23 The binding of Ca2+ ions leads to some conformational change and a concomitant increased affinity for ligands, such as proteins, peptides and small molecules.23-25 2 ACS Paragon Plus Environment

Page 2 of 19

Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Initially, holo-CaM is in an open, dumbbell-shaped conformation (Fig. 1). Bound to a ligand, holo-CaM has a closed, globular conformation (Fig. 1). For example, when Ca2+ saturated CaM binds trifluoperazine (TFP), the radius of gyration decreases from 21.9 to 17.6 Å, and the maximum diameter decreases from 61 to 48 Å.23 Thus, upon ligand binding, holo-CaM undergoes a major change in size. Therefore, it is highly interesting, if these conformational states can still be induced, when CaM is part of a PEM. We use X-ray and neutron reflectometry (XR and NR) to analyze the structure of PEMs incorporating CaM. XR is carried out with samples hydrated in a humidity chamber. The data are fitted with a structure model representing the electron density profile across the interface.26 In NR experiments, the sample is in contact with a D2O solution, and the corresponding data is fitted with a structure model showing the neutron scattering length density profile.26 In this way, the thickness, density and roughness of a PEM is determined by two independent methods. Furthermore, small-angle X-ray scattering (SAXS) is carried out to measure the radius of gyration and to obtain the pair distance distribution function of CaM in solution with and without ligands.27 These data yield the size and shape of CaM in aqueous environment without interfacial perturbations. Moreover, Fourier transform infrared (FTIR) spectroscopy in transmission and attenuated total reflection (ATR) mode has been applied to estimate potential changes in the secondary structure of CaM upon ligand binding and embedding in PEMs.28,29 As we will show in this study, the different conformational states of CaM, as found in solution, cannot be found in a PEM. However, removal and delivery of Ca2+ ions induce a considerable increase and decrease of the multilayer thickness suggesting a change between intrinsic and extrinsic charge compensation. In this way, a PEM containing CaM might be regarded as a bio-responsive interface, whose structure is controlled via protein functionality.

MATERIALS AND METHODS Sample preparation. Recombinant CaM from rat was produced using the pET-14b vector from GenScript (Piscataway, NJ, USA) and E. coli BL21-CodonPlus (DE3)-RIPL competent cells from Agilent Technologies (Santa Clara, CA, USA) as described in the literature.30 The purity of the protein was checked by SDS-PAGE (Fig. S1). Poly(ethyleneimine) (PEI), poly(allylamine hydrochloride) (PAH), poly(styrene sulfonate) (PSS), CaCl2, trifluoperazine (TFP), ethylene glycol-bis(β-aminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA), BisTris, and D2O were all purchased from Sigma-Aldrich. The pD-value of D2O solutions was determined by adding 0.4 to the pH-reading using a conventional pH-electrode.31 For XR, PEMs were built up on Si wafers of size 1.5 cm x 2.0 cm. The Si wafers were kindly donated by Siltronic (Burghausen, Germany). For NR, Si blocks of size 5.0 cm x 8.0 cm x 1.5 cm from Siliciumbearbeitung Andrea Holm (Tann, Germany) were used. They were cleaned in a 1:1:5 mixture of NH3(30%), H2O2(30%) and H2O at 70°C for 15 min before modification. PEMs were prepared by successive incubation with aqueous solutions of PEI (0.01 M monomer concentration), PSS (0.01 M monomer concentration with 1 M NaCl), 3 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

PAH (0.01 M monomer concentration with 1 M NaCl), and CaM (0.1 mg mL-1 with 10 mM CaCl2, 150 mM NaCl, 10 mM BisTris, pH = 6.5). Each layer deposition was carried out for 20 min followed by intensive rinsing with pure water. X-ray reflectometry (XR). XR was carried out with the PEM samples on Si wafers, which were equilibrated in a home-built humidity chamber (Fig. S2). This chamber is characterized by a sealed inner cell with a homogeneous temperature distribution that is maintained by an outer cell. 100 % relative humidity is achieved in this cell by a reservoir of pure water. Data were collected using the Seifert XRD 3000 TT reflectometer from GE Inspection Technologies (Ahrensburg, Germany), which was operated with the Mo-Kα wavelength (0.71 Å). Raw data were converted to reflectivity curves by normalization of the reflected X-ray intensity to the incident intensity. They are scaled as a function of wavevector transfer, Q = (4π/λ) sin θ, where λ is the wavelength and θ is the angle of incidence. Reflectivity curves were analyzed using the Motofit software32 by fitting simple layer models to the experimental data. Neutron reflectometry (NR). NR measurements were performed at the Helmholtz-Zentrum Berlin (Germany) using the instrument V6. The sample cell consists of the Si block with the PEM, which is covered by a Teflon trough filled with D2O. The neutron wavelength selected by a graphite monochromator was 4.66 Å. Neutron reflectivities were recorded with a fixed incident neutron beam in θ-2θ geometry using a 3He detector. Data have been collected at 23°C. θ ranges from 0° to 3.0°. Neutron reflectivities as a function of wavevector transfer, Q, were analyzed using the Motofit software.32 Small-angle X-ray scattering (SAXS). SAXS was carried out with aqueous solutions of CaM (10 mg mL-1 with 10 mM CaCl2, 150 mM NaCl, 10 mM BisTris, pH = 6.5) with or without TFP and PAH. Data were collected using the SAXSess mc2 instrument from Anton Paar (Graz, Austria), which was operated with Cu-Kα wavelength (1.54 Å), line collimation, Kratky optics, and an image plate as detector. The two-dimensional scattering data were integrated and analyzed using the SAXSquant 2D software from Anton Paar. Fourier transform infrared (FTIR) spectroscopy. FTIR spectra of CaM in D2O solution were collected using a Nicolet 6700 FTIR spectrometer from Thermo Fisher Scientific, which was operated with a liquid nitrogen-cooled MCT detector and was purged with dry air. CaM deposited on a PEI-PSS-PAH multilayer was analyzed in attenuated total reflection (ATR) mode. The ATR accessory from Pike Technologies (Madison, Wisconsin, USA) was placed in the sample compartment of the FTIR spectrometer. Constant temperature of the sample was maintained by a circulating water flow. Measurements were performed at a spectral resolution of 1 cm-1. Spectral analysis was carried out using the Grams/AI 8.1 software from Thermo Fisher Scientific. The amide I’ band (the prime indicates D2O as the solvent) is found between 1700 and 1600 cm-1 and is sensitive to the fractions of various secondary structure elements of proteins.

4 ACS Paragon Plus Environment

Page 4 of 19

Page 5 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

RESULTS AND DISCUSSION Build-up of PEMs with CaM. In Fig. 2, typical XR curves are shown that track the stepwise build-up of the following PEM: Si-PEI-PSS-PAH-CaM-PAH-CaM-PAH- humid air The PEM is deposited on an Si wafer and kept under air. However, during measurements, it is equilibrated inside a sealed humidity chamber, which provides full hydration. Water-saturated air and pure liquid water have the same chemical potential and will provide full hydration of a PEM.4 However, odd-even effects have been observed, where the swelling behavior and the surface roughness of a PEM can depend on the outermost layer.33,34 For a determination of the multilayer structures, we have used a simple structure model that is characterized by the thickness, the X-ray scattering length density (electron density times classical electron radius), and the roughness of the multilayer, which are varied to fit the experimentally measured Xray reflectivity curves. The scattering length densities of the Si and humid air were fixed to 20.1 x 10-6 Å-2 and 0, respectively, during data fitting. The fits are shown as solid lines in Fig. 2, and the corresponding fit parameters are listed in Tab. 1.

Figure 2. Typical XR curves showing the stepwise deposition of a PEI-PSS-PAH-(CaM-PAH)2 multilayer on an Si wafer. The measurements were performed against air in a humidity chamber at 100 % relative humidity. Symbols reflect experimental data, solid lines are fits based on the layer model Si-multilayer-air.

For comparison, we have also carried out NR measurements of the structure of PEMs containing CaM. Typical data are plotted in Fig. 3. In this case, the humid air is replaced by D2O, which guarantees full hydration of the PEM. Thus, the NR data serve as control for the XR data in terms of hydration effects, but also as reproduction by an independent method. We use non-deuterated polyelectrolytes and protein, which generate a high contrast to D2O with little incoherent scattering. Kiessig interferences will reflect the total multilayer thickness without resolving the inner multilayer structure. The PEMs have been prepared with two additional layers of PSS and PAH: 5 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Si-PEI-(PSS-PAH)2-CaM-PAH-CaM-PAH-D2O In this way, the PEMs are somewhat thicker and show neutron reflectivity curves with a larger number of Kiessig interferences in the smaller Q-range, which is accessible in NR (the width of a Kiessig interference, ∆Q, is approximately related to the total thickness, d, of a multilayer by ∆Q ≈ 2π/d).26 The additional PSS-PAH layers as compared to the samples used for XR are not expected to have any effect on the conformation of CaM. The neutron reflectivity data have also been analyzed by fitting a simple layer model. Whereas the thickness, the neutron scattering length density, and the roughness of the PEM were varied in the fitting process, the neutron scattering length densities of Si and D2O were fixed to 2.07 x 10-6 Å-2 and 6.37 x 10-6 Å-2, respectively, using reference data provided by the National Institute of Standards and Technology (NIST).35 The fits are shown as solid lines in Fig. 3, and the corresponding fit parameters are listed in Tab. 2. It is noted that Ca2+ saturated CaM (holo-CaM) has always been used to build PEMs in this study.

Figure 3. Typical NR curves showing the stepwise deposition of a PEI-(PSS-PAH)2-(CaM-PAH)2 multilayer on an Si wafer. The measurements were performed against D2O. Symbols reflect experimental data, solid lines are fits based on a layer model.

The structural data listed in Tab. 1 and 2 reveal two remarkable features. When CaM is deposited for the first time, there is a smaller thickness increase than that observed, when CaM is deposited for the second time. Furthermore, upon deposition of PAH on a CaM layer, much of the CaM is desorbing, but a thin double layer of CaM-PAH remains. In order to verify these trends, we have studied a total of 26 different PEMs with XR and 3 PEMs with NR, which are listed in the supporting information. All PEMs follow the layer sequences given above. XR measurements are carried out in air at 100 % relative humidity, and NR measurements are carried out in D2O. As outlined in the introduction, the binding of four TFP ligands induces a transition of holo-CaM from an open to a closed conformation. Therefore, we have also prepared PEMs with holo-CaM binding TFP. This is denoted as CaM(TFP). The binding of four TFP molecules to one CaM molecules has been achieved by using solutions containing a molar ratio of 5 TFP to 1 CaM, as described in the literature.23 We will focus the 6 ACS Paragon Plus Environment

Page 6 of 19

Page 7 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

discussion of the results on the thickness of a film, because this parameter is largely determined by the Q-positions of the Kiessig fringes and cannot vary much (~1-2 Å) to obtain a good fit of an experimental reflectivity curve. The scattering length density and the roughness of the film are related to the amplitudes of the Kiessig fringes and show some larger uncertainties (~10-15 % and ~20-30 %, respectively). In Tab. 3, mean thickness increases are listed, as calculated from the X-ray and neutron reflectivity data of all 29 PEMs studied (see supporting information). The observed thicknesses of the initial PEI-PSS-PAH and PEI-(PSS-PAH)2 sequences are in good agreement with the literature,34,36 also suggesting a full hydration of the PEMs inside the humidity chamber used for XR. The thickness of the first CaM layer without and with bound TFP is 50 Å and 49 Å, respectively, as determined from XR data. Apparently, the initial conformation of CaM, open or closed, during the deposition step has no effect on the layer thickness, although CaM(TFP) has a smaller radius of gyration than CaM. In the case of NR, a somewhat smaller mean thickness increase of only 34 Å is observed for the first CaM or CaM(TFP) layer. However, this mean value is based on only three NR experiments and has a rather large experimental error. Furthermore, a small difference in the pH-value of the CaM solutions used for XR and NR can result in different amounts of deposited proteins. Thus, the difference between the XR and the NR CaM layer thickness appears not to be significant. For comparison, the maximum diameters of CaM and CaM(TFP) in solution are reported to be 61 Å and 48 Å, respectively.23 Therefore, about one monolayer of CaM or CaM(TFP) is formed on PAH.

Table 1. Structure of a growing PEM containing holo-CaM in air in a humidity chamber, as determined by XR.a

Multilayer PEI-PSS-PAH PEI-PSS-PAH-CaM PEI-PSS-PAH-CaM-PAH PEI-PSS-PAH-CaM-PAH-CaM PEI-PSS-PAH-CaM-PAH-CaM-PAH

d/Å 79 116 89 173 96

ρe / 10-6Å-2

σ/Å

8.7 10.9 9.4 12.0 10.1

12 20 15 29 21

a

For each step, the multilayer has been modeled by a single layer of thickness d, X-ray scattering length density ρe and roughness σ.

Table 2. Structure of a growing PEM containing holo-CaM in contact with D2O, as determined by NR.

Multilayer PEI-(PSS-PAH)2 a PEI-(PSS-PAH)2-CaM b PEI-(PSS-PAH)2-CaM-PAH a PEI-(PSS-PAH)2-CaM-PAH-CaM b PEI-(PSS-PAH)2-CaM-PAH-CaM-PAH a

d/Å 123 97/88 130 105/117 138

ρn / 10-6Å-2

σ/Å

4.0 4.8/5.2 4.2 4.6/5.4 4.0

20 20/21 22 20/19 20

7 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The multilayer has been modeled by a single layer of thickness d, neutron scattering length density ρn and roughness σ.

a

b

The multilayer has been modeled by two layers.

Remarkably, when CaM or CaM(TFP) is deposited on PAH for the second time, a much thicker layer is generated. Values of 87 Å and 74 Å are found with XR and even 112 Å with NR (Tab. 3). Such effect has already been reported in the literature for other systems and denoted as exponential growth. For example, PEMs composed of hyaluronic acid / poly(Llysine) or poly(L-glutamic acid) / PAH show exponential growth.18,19 More mass is deposited in each deposition step than in the preceding step. It has been found that at least one of the two components can diffuse within the homogeneous structure of the PEM. Thus, a CaM / PAH multilayer might also grow according to an exponential growth mechanism. It is noted that no effect of TFP can be detected in terms of the thickness of the second CaM layer within the experimental error (Tab. 3), as found for the first layer (see above). However, when a CaM or CaM(TFP) terminated PEM is rinsed with a PAH solution, there is no further increase of the PEM thickness but a decrease (Tab. 1 and 2). This can only be explained by a strong interaction of PAH with CaM or CaM(TFP) leading to a complexation and a partial desorption of CaM or CaM(TFP). Though, a thin double layer CaM-PAH or CaM(TFP)-PAH of only 14 Å thickness (XR) and 10 Å thickness (NR) remains (Tab. 3). Again, no effect of TFP on the double layer thickness can be detected. Moreover, CaM must be in the open conformation and oriented parallel to the interface within this double layer to allow for such small thickness. The open conformation is consistent with the idea that adsorbing CaM(TFP) is losing TFP inside the PEM. It is noted that CaM is not totally desorbing upon rinsing with a PAH solution, but is still remaining in the PEM. Furthermore, it can still be detected by ATR-FTIR spectroscopy (see below). The data in Tab. 1-3 are in agreement with complementary ATR-FTIR data shown in Fig. S3. The amide I’ band area, which is proportional to the amount of CaM at the interface, has been recorded during the buildup of a PEI-PSS-PAH-(CaM-PAH)x multilayer. The data confirm that much more CaM is deposited in the second rinsing step than in the first one. Furthermore, the high amount of deposited CaM is not washed away by buffer solution. There is a small but steady growth of the multilayer. However, after a few double layers of CaMPAH, the multilayer eventually becomes instable and some fraction of CaM and PAH can desorb. Therefore, unfortunately, the data cannot provide sufficient evidence for an exponential growth mechanism. Complexation and partial desorption of oppositely charged polyelectrolytes has been explained by a medium charge density of the polyelectrolytes, and the pH-value of the solution has been found to be important, where the components have a pH-dependent charge.37 Both conditions apply to CaM and PAH used in this study.

8 ACS Paragon Plus Environment

Page 8 of 19

Page 9 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Table 3. Mean thickness increases of PEMs after various deposition steps as determined by XR (in watersaturated air) and NR (in D2O).a

∆d / Å by XR +62 ± 14 (23)

Deposition unit PEI-PSS-PAH PEI-(PSS-PAH)2 1st CaM 1st CaM(TFP) 2nd CaM 2nd CaM(TFP) CaM-PAH

∆d / Å by NR +125 ± 4 (3) + 34 ± 25 (3) b

+50 ± 21 (7) +49 ± 18 (11) +87 ± 15 (3) +74 ± 8 (3) +14 ± 7 (17) c

+112 ± 28 (2) b + 10 ± 3 (5) c

a

numbers in round brackets indicate the numbers of independent experiments (see supporting information). Errors are given as one standard deviation. b

averaged over CaM and CaM(TFP) deposition steps.

c

averaged over CaM, CaM(TFP), 1st and 2nd deposition steps.

Interaction of CaM with TFP and PAH. To shed some light on the interaction of CaM with TFP and PAH, we have carried out SAXS measurements in solution (Fig. 4). Four samples are compared: holo-CaM without ligand, with TFP, with PAH, and with TFP + PAH. To bind TFP to CaM, the molar concentration of TFP is five times higher than that of CaM.23 To bind PAH to CaM, a 1:1 molar ratio has been chosen. For each scattering curve, I(Q), the pair distance distribution function defined as ∞

r2 sin Qr p( r ) = I (Q) ⋅ 4πQ 2 dQ 3 ∫ (2π ) 0 Qr

(1)

has been deduced (Fig. 5).27 The fits in Fig. 4 represent the Fourier transformations of the p(r) curves in Fig. 5. p(r)dr is proportional to the number of distances in a molecule between r and r+dr. Whereas the maximum of p(r) locates approximately the radius of the molecule, the point of intersection on the r-axis corresponds to the maximum diameter of the molecule. In addition, the radius of gyration, Rg, has been calculated from p(r) according to ∞

1 Rg = ∫ r 2 p (r )dr 20

(2)

where the area under p(r) is normalized to one.

9 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. SAXS curves of holo-CaM in aqueous solution (10 mg mL-1 with 10 mM CaCl2, 150 mM NaCl, 10 mM BisTris, pH = 6.5). Symbols reflect experimental data, solid lines are fits representing the Fourier transformations of the pair distance distribution functions, p(r), as shown in Fig. 5. Curve 1, CaM without ligand; curve 2, CaM with TFP (1:5 molar ratio); curve 3, CaM with PAH (1:1 molar ratio); curve 4, CaM with TFP and PAH (1:5:1 molar ratio).

As can be seen in Fig. 5, holo-CaM without ligand has an open conformation with a maximum diameter of about 70 Å and a radius of gyration of 22.8 Å. The shoulder of p(r) at higher r-values results from the dumbbell shape of the protein. In the presence of 5 mol TFP per mol of CaM, 4 TFP molecules bind to CaM and induce the closed conformation. The corresponding p(r) curve is symmetric now (Fig. 5), as expected for a spherical shape.27 The maximum diameter and the radius of gyration are reduced to about 45 Å and 16.5 Å, respectively. These data are in excellent agreement with an earlier study.23 When PAH is added to CaM in 1:1 ratio of polymer chains and protein molecules, the p(r) curve is asymmetric (Fig. 5). This means CaM is not forming the closed, globular conformation as observed with TFP, but has an elongated shape resembling the open conformation. However, it is much larger now. A maximum diameter of about 130 Å and a radius of gyration of 40.9 Å have been determined. Finally, we have also added PAH to CaM(TFP) to see, if PAH can replace the bound TFP molecules. Indeed, as seen in Fig. 5, the p(r) curve is asymmetric as found in the absence of TFP. Similar values of about 130 Å and 37.4 Å can be determined for the maximum diameter and radius of gyration (Fig. 5). It must be noted that CaM and PAH form a white precipitate which has been dissolved by sonication. Clearly, these findings indicate that CaM(TFP), which is initially in the closed conformation, interacts with PAH. A strong interaction between holo-CaM(TFP) and PAH will probably lead to a replacement of TFP by PAH. When TFP is lost, holo-CaM will lose its closed conformation as shown in Fig. 1. Therefore, the deposition of holo-CaM without and with TFP will end up in the same open conformation of holo-CaM. The thin double layers of CaM-PAH (~10-14 Å, Tab. 3) are consistent with holo-CaM in the open conformation, when the protein is oriented parallel to 10 ACS Paragon Plus Environment

Page 10 of 19

Page 11 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

the interface, because the diameter of the lobes of open holo-CaM can be estimated to range over only 10 – 15 Å.

Figure 5. Pair distance distribution functions, p(r), of holo-CaM in aqueous solution, as derived from the SAXS curves in Fig. 4. Curve 1, CaM without ligand; curve 2, CaM with TFP (1:5 molar ratio); curve 3, CaM with PAH (1:1 molar ratio); curve 4, CaM with TFP and PAH (1:5:1 molar ratio). The arrows indicate the maximum diameters.

The observed SAXS results are in favorable agreement with FTIR experiments. The amide I’ band of holo-CaM has its maximum located at 1644±1 cm-1. The binding of 4 TFP molecules shifts this maximum to 1648±1 cm-1 (Fig. 6A). The shape and position of the amide I’ band of a protein reflects the distribution of the secondary structure elements.28 Thus, the small shift of the maximum wavenumber indicates a small change of the secondary structure of CaM upon binding TFP. Using ATR-FTIR spectroscopy, we have also measured the amide I’ band, when CaM is deposited on a PEI-PSS-PAH layer (Fig. 6B). Interestingly, the presence or absence of TFP does not have a noticeable effect on the shape and position of the amide I’ band of CaM confirming the idea that TFP is not binding to CaM in a PEM. The maximum wavenumber can be found at 1645±1 cm-1, which is between the wavenumbers of 11 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

CaM and CaM(TFP). Finally, as found by XR and NR, a CaM-PAH double layer is very thin (Tab. 3). A major fraction of CaM is desorbing, when a PAH layer is deposited on a CaM layer. However, as can be seen in Fig. 6C, CaM is not totally desorbing upon rinsing with a PAH solution. The amide I’ band of the PEI-PSS-PAH-CaM-PAH multilayer is clearly detectable.

Figure 6. A: Normalized FTIR solution spectra of holo-CaM without ligand and with 4 bound TFP molecules (1:5 molar ratio of CaM and TFP). B: Normalized ATR-FTIR spectra of holo-CaM (two different samples) deposited on PEI-PSS-PAH in the absence and the presence of TFP (1:5 molar ratio of CaM and TFP). C: Nonnormalized ATR-FTIR spectra of PEI-PSS-PAH-CaM and PEI-PSS-PAH-CaM-PAH. These spectra were recorded using the same sample to show the relative amount of immobilized CaM. The latter spectrum has been enhanced by a factor of 10.

Response of PEM structure to solution conditions. After having characterized the build-up of PEMs containing CaM, we have also performed some rinsing experiments to probe the response of these PEMs to the solution conditions. The idea is to induce the same conformational transitions of CaM inside a PEM as observed free in solution (Fig. 1). Using an EGTA solution, the Ca2+ ions can be removed from holo-CaM, because EGTA has a specific binding affinity for Ca2+ ions. In this way, the affinity of CaM for ligands, such as TFP, decreases and CaM forms an open conformation, when it is free in solution. Afterwards, using a Ca2+/TFP solution, apo-CaM binds 4 Ca2+ ions and 4 TFP molecules and undoes a transition to the closed conformation, when it is free in solution. To probe these conformational changes of CaM in PEMs, we have chosen the PEI-PSS-PAH, PEI-PSS-PEICaM and PEI-PSS-PAH-CaM-PAH multilayers as samples. The thicknesses of these multilayers before and after rinsing experiments can be found in the supporting information and are used to calculate mean thickness changes. In the following, mean values from at least 3 independent experiments are reported. The PEI-PSS-PAH multilayer serves as reference. Upon exposure to an EGTA solution (2 mM with 10 mM BisTris and ca. 2 mM NaCl, pH = 6.5), we observe a thickness increase of this multilayer of +9 Å (mean of 5 experiments, samples #4, 14, 17, 24, 25). Since the PEM does not contain any Ca2+ ions, there will be no ion exchange. The thickness increase can simply be explained by a monolayer of adsorbed EGTA molecules, which are negatively charged and are attracted by the positively charged PAH layer. Indeed, upon subsequent rinsing with a CaCl2 solution (10 mM with 10 mM BisTris and 150 mM NaCl, pH 12 ACS Paragon Plus Environment

Page 12 of 19

Page 13 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

= 6.5), EGTA is washed away, and the PEM thickness decreases by -12 Å (mean of 3 experiments, samples #17, 24, 25). The PEI-PSS-PAH-CaM multilayer, prepared with CaM or CaM(TFP), represents a simple PEM to probe conformational changes of immobilized CaM. CaM is bound on top of a PAH layer and expected to show maximum conformational flexibility. Indeed, as reported recently, CaM that is bound to a self-assembled monolayer (SAM) still undergoes all conformational transitions as observed in solution.38 When a PEI-PSS-PAH-CaM multilayer is rinsed with an EGTA solution, a small thickness increase of +6 Å (mean of 7 experiments, samples #19, 19, 20, 20, 22, 22, 23) is observed, which does not suggest a major conformational change of CaM. This would be consistent with the similar apo- and holo forms (Fig. 1). However, we have also observed that the thickness of the multilayer decreases by -25 Å (mean of 3 experiments, samples # 19, 20, 22) upon rinsing with a CaCl2 solution indicating a substantial desorption of CaM, considering an initial CaM layer thickness of 50 Å (Tab. 3). Thus, unfortunately, conformational changes of CaM in this PEM cannot be tracked by simple thickness measurements using XR. In the PEI-PSS-PAH-CaM-PAH multilayer, CaM is formally sandwiched between two PAH layers. In this way, it is firmly immobilized but expected to be restricted in its conformational freedom. However, when a PEI-PSS-PAH-CaM-PAH multilayer (with or without TFP) is rinsed with an EGTA solution, we observe a relatively large thickness increase of +25 Å by XR and +28 Å by NR (mean of 11 XR and 1 NR experiments, samples #3, 5, 5, 9, 9, 10, 10, 11, 12, 18, 18, 27). This swelling of the PEM can be explained by a simple scenario (Fig. 7). Initially, there are probably 4 Ca2+ ions bound to CaM. When these bound ions are removed by EGTA, they will be replaced by 8 free monovalent cations, such as Na+ or H+, to restore the charge balance. Thus, there will be an increase of the osmotic pressure, an uptake of water and a swelling of the PEM. Moreover, apo-CaM has a smaller conformational stability than holo-CaM.39,40 Thus, a potential partial unfolding or distortion of CaM upon swelling is facilitated in the absence of Ca2+ ions.

Figure 7. Simple scenario explaining the observed shrinking and swelling of a PEM containing CaM. Initially, the PEM is built with Ca2+ saturated CaM. When the Ca2+ ions are removed from CaM by EGTA, Na+ or H+ ions must enter the PEM for extrinsic charge compensation. This will increase the osmotic pressure and leads to an uptake of water. Subsequent rinsing with a Ca2+ solution restores the intrinsic charge compensation with a reduced osmotic pressure.

Subsequently, the PEI-PSS-PAH-CaM-PAH multilayer has also been rinsed with a CaCl2/TFP solution (10 mM and 24 µM, respectively), which leads to reduction in thickness of -21 Å by XR and -10 Å by NR (5 XR and 1 NR experiments, samples # 5, 9, 10, 11, 12, 27). CaM is saturated with Ca2+ ions again, and the monovalent counterions are washed out of 13 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the PEM. The reduced osmotic pressure due to the significant lower number of free ions in the PEM will cause the observed contraction of the PEM (Fig. 7). In contrast, TFP is not binding to CaM, because the formation of the closed, globular conformation of CaM(TFP) is not consistent with the shrinking of the PEM upon rinsing with the CaCl2/TFP solution and the formation of very thin CaM-PAH double layers (Tab. 3). The latter requires the open conformation of CaM oriented parallel to the interface (Fig. 7). Apparently, the experiments of this study cannot find any evidence that the closed, globular conformations of CaM can be triggered inside a PEM. However, the presence of CaM inside a PEM provides a mean to control the osmotic pressure via removal and binding of Ca2+ ions. This process is reversible and can be repeated several times, as shown in the SI and in Tab. 4.

Table 4. Switching of a CaM containing PEM upon rinsing with EGTA and Ca2+ solutions.a

multilayer ∆d / Å PEI-PSS-PAH-CaM-PAH 90 +EGTA +17 -17 +Ca2+ +EGTA +14 +Ca2+ -8 +EGTA +14 +Ca2+ -16 +EGTA +22 +Ca2+ -6 a

EGTA solution contains 2 mM EGTA, 10 mM BisTris, ca. 2 mM NaCl, pH = 6.5; Ca2+ solution contains 10 mM CaCl2, 10 mM BisTris, 150 mM NaCl, pH = 6.5. The data correspond to sample #26 in the SI.

CONCLUSIONS PEMs are well characterized surface coatings that can easily be used to bio-functionalize interfaces with proteins. In many studies, it has been shown that proteins largely maintain their native conformation, and enzymes are still active, when they are embedded in or adsorbed on a PEM. Although the latter requires some preserved dynamics of enzymes, larger distinct conformational changes of proteins have not been reported so far inside a PEM. In this study, we have addressed this issue using CaM that can be triggered between an open and a closed conformation upon ligand binding, when it is free in solution. Apparently, the results of this study do not suggest that the closed conformation can be induced by ligand binding inside a PEM. Rather, TFP binding is suppressed by the counter-polyelectrolyte PAH, which binds firmly to CaM. This represents some limitation of using PEMs as a native host environment for immobilized proteins. However, CaM is a Ca2+ binding protein. Utilizing this property, swelling and shrinking of a PEM containing CaM could be demonstrated by removing and adding Ca2+ ions. Although this effect is likely to be related to osmosis, the involvement of CaM as Ca2+ binding partner provides a PEM with bio-responsive properties. 14 ACS Paragon Plus Environment

Page 14 of 19

Page 15 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

For example, the release of small drug molecules from a CaM-PEM might be triggered by an EGTA / Na+ solution, when the CaM-PEM is swelling. In contrast, addition of Ca2+ ions could induce a shrinking and possibly a concomitant trapping of dissolved small molecules inside the CaM-PEM. Whereas conformational transitions of CaM are suppressed in a PEM, the native conformation of CaM in terms of secondary structure and Ca2+ binding property seems to be preserved.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: SDS-PAGE of calmodulin. Drawing of humidity sample chamber for XR. Complementary ATR-FTIR data of multilayer build-up. List of all XR and NR experiments.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS Financial support from the Deutsche Forschungsgemeinschaft (DFG Forschergruppe 1979) is gratefully acknowledged.

REFERENCES (1) Moehwald, H.; Brezesinski, G. From Langmuir Monolayers to Multilayer Films. Langmuir 2016, 32, 10445-10458. (2) Boudou, T.; Crouzier, T.; Ren, K.; Blin, G.; Picart, C. Multiple Functionalities of Polyelectrolyte Multilayer Films: New Biomedical Applications. Adv. Mater. 2010, 22, 441467. (3) Schlenoff, J. B. Retrospective on the Future of Polyelectrolyte Multilayers. Langmuir 2009, 25, 14007-14010. 15 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(4) von Klitzing, R. Internal structure of polyelectrolyte multilayer assemblies. Phys. Chem. Chem. Phys. 2006, 8, 5012-5033. (5) Schönhoff, M. Self-assembled polyelectrolyte multilayers. Curr. Opin. Colloid Interface Sci. 2003, 8, 86-95. (6) Decher, G. Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites. Science 1997, 277, 1232-1237. (7) Erol, M.; Du, H.; Sukhishvili, S. Control of Specific Attachment of Proteins by Adsorption of Polymer Layers. Langmuir 2006, 22, 11329-11336. (8) Ngankam, A. P.; Mao, G.; Van Tassel, P. R. Fibronectin Adsorption onto Polyelectrolyte Multilayer Films. Langmuir 2004, 20, 3362-3370. (9) Jeworrek, C.; Hollmann, O.; Steitz, R.; Winter, R.; Czeslik, C. Interaction of IAPP and insulin with model interfaces studied using neutron reflectometry. Biophys. J. 2009, 96, 11151123. (10) Jackler, G.; Czeslik, C.; Steitz, R.; Royer, C. A. Spatial distribution of protein molecules adsorbed at a polyelectrolyte multilayer. Phys. Rev. E 2005, 71, 041912-1 – 041912-7. (11) Schwinté, P.; Ball, V.; Szalontai, B.; Haikel, Y.; Voegel, J.-C.; Schaaf, P. Secondary Structure of Proteins Adsorbed onto or Embedded in Polyelectrolyte Multilayers. Biomacromolecules 2002, 3, 1135-1143. (12) Schwinté, P.; Voegel, J.-C.; Picart, C.; Haikel, Y.; Schaaf, P.; Szalontai, B. Stabilizing Effects of Various Polyelectrolyte Multilayer Films on the Structure of Adsorbed/Embedded Fibrinogen Molecules: An ATR-FTIR Study. J. Phys. Chem. B 2001, 105, 11906-11916. (13) Caruso, F.; Schüler, C. Enzyme Multilayers on Colloid Particles: Assembly, Stability, and Enzyme Activity. Langmuir 2000, 16, 9595-9603. (14) Disawal, S.; Qiu, J.; Elmore, B. B.; Lvov, Y. M. Two-step sequential reaction catalyzed by layer-by-layer assembled urease and arginase multilayers. Colloids Surf. B: Biointerfaces 2003, 32, 145-156. (15) Schuabb, V.; Cinar, S.; Czeslik, C. Effect of interfacial properties on the activation volume of adsorbed enzymes. Colloids Surf. B: Biointerfaces 2016, 140, 497-504. (16) Velk, N.; Uhlig, K.; Vikulina, A.; Duschl, C.; Volodkin, D. Mobility of lysozyme in poly(L-lysine)/hyaluronic acid multilayer. Colloids Surf. B: Biointerfaces 2016, 147, 343-350. (17) Garza, J. M.; Schaaf, P.; Muller, S.; Ball, V.; Stoltz, J.-F.; Voegel, J.-C.; Lavalle, P. Multicompartment Films Made of Alternate Polyelectrolyte Multilayers of Exponential and Linear Growth. Langmuir 2004, 20, 7298-7302. (18) Hübsch, E.; Ball, V.; Senger, B.; Decher, G.; Voegel, J.-C.; Schaaf, P. Controlling the Growth Regime of Polyelectrolyte Multilayer Films: Changing from Exponential to Linear 16 ACS Paragon Plus Environment

Page 16 of 19

Page 17 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Growth by Adjusting the Composition of Polyelectrolyte Mixtures. Langmuir 2004, 20, 19801985. (19) Porcel, C.; Lavalle, P.; Ball, V.; Decher, G.; Senger, B.; Voegel, J.-C.; Schaaf, P. From Exponential to Linear Growth in Polyelectrolyte Multilayers. Langmuir 2006, 22, 4376-4383. (20) Koo, J.; Czeslik, C. Probing aggregation and fibril formation of insulin in polyelectrolyte multilayers. Colloids Surf. B: Biointerfaces 2012, 94, 80-88. (21) Fabian, H.; Yuan, T.; Vogel, H. J.; Mantsch, H. H. Comparative analysis of the aminoand carboxy-terminal domains of calmodulin by Fourier transform infrared spectroscopy. Eur. Biophys. J. 1996, 24, 195-201. (22) André, I.; Kesvatera, T.; Jönsson, B.; Åkerfeldt, K. S.; Linse, S. The Role of Electrostatic Interactions in Calmodulin-Peptide Complex Formation. Biophys. J. 2004, 87, 1929-1938. (23) Matsushima, N.; Hayashi, N.; Jinbo, Y.; Izumi, Y. Ca2+-bound calmodulin forms a compact globular structure on binding four trifluoperazine molecules in solution. Biochem. J. 2000, 347, 211-215. (24) Erwin, N.; Patra, S.; Winter, R. Probing conformational and functional substates of calmodulin by high pressure FTIR spectroscopy: influence of Ca2+ binding and the hypervariable region of K-Ras4B. Phys. Chem. Chem. Phys. 2016, 18, 30020-30028. (25) Kataoka, M.; Head, J. F.; Vorherr, T.; Krebs, J.; Carafoli, E. Small-Angle X-ray Scattering Study of Calmodulin Bound to Two Peptides Corresponding to Parts of the Calmodulin-Binding Domain of the Plasma Membrane Ca2+ Pump. Biochemistry 1991, 30, 6247-6251. (26) Russell, T. P. X-ray and neutron reflectivity for the investigation of polymers. Mater. Sci. Rep. 1990, 5, 171-271. (27) Blanchet, C. E.; Svergun, D. I. Small-Angle X-Ray Scattering on Biological Macromolecules and Nanocomposites in Solution. Annu. Rev. Phys. Chem. 2013, 64, 37-54. (28) Byler, D. M.; Susi, H. Examination of the secondary structure of proteins by deconvolved FTIR spectra. Biopolymers 1986, 25, 469-187. (29) Goormaghtigh, E.; Raussens, V.; Ruysschaert, J.-M. Attenuated total reflection infrared spectroscopy of proteins and lipids in biological membranes. Biochim. Biophys. Acta 1999, 1422, 105-185. (30) Hayashi, N.; Matsubara, M.; Takasaki, A.; Titani, K.; Taniguchi, H. An Expression System of Rat Calmodulin Using T7 Phage Promoter in Escherichia coli. Protein Expression Purification 1998, 12, 25-28. (31) Covington, A. K.; Paabo, M.; Robinson, R. A.; Bates, R. G. Use of the Glass Electrode in Deuterium Oxide and the Relation between the Standardized pD (paD) Scale and the Operational pH in Heavy Water. Anal. Chem. 1968, 40, 700–706. 17 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(32) Nelson, A. Co-refinement of multiple-contrast neutron/X-ray reflectivity data using MOTOFIT. J. Appl. Cryst. 2006, 39, 273-276. (33) Zerball, M.; Laschewsky, A.; von Klitzing, R. Swelling of Polyelectrolyte Multilayers: The Relation Between Surface and Bulk Characteristics. J. Phys. Chem. B 2015, 119, 1187911886. (34) Nestler, P.; Block, S.; Helm, C. A. Temperature-Induced Transition from Odd-Even to Even-Odd Effect in Polyelectrolyte Multilayers Due to Interpolyelectrolyte Interactions. J. Phys. Chem. B 2012, 116, 1234-1243. (35) https://www.ncnr.nist.gov/resources/n-lengths/list.html. (36) Schwarz, B.; Schönhoff, M. Surface Potential Driven Swelling of Polyelectrolyte Multilayers. Langmuir 2002, 18, 2964-2966. (37) Hoogeveen, N. G.; Cohen Stuart, M. A.; Fleer, G. J.; Böhmer, M. R. Formation and Stability of Multilayers of Polyelectrolytes. Langmuir 1996, 12, 3675-3681. (38) Noguchi, H.; Adachi, T.; Nakatomi, A.; Yazawa, M.; Uosaki, K. Biofunctionality of Calmodulin Immobilized on Gold Surface Studied by Surface-Enhanced Infrared Absorption Spectroscopy: Ca2+-Induced Conformational Change and Binding to a Target Peptide. J. Phys. Chem. C 2016, 120, 16035-16041. (39) Masino, L.; Martin, S. R.; Bayley, P. M. Ligand binding and thermodynamic stability of a multidomain protein, calmodulin. Prot. Science 2000, 9, 1519-1529. (40) Gibrat, G.; Hoa, G. H. B.; Craescu, C. T.; Assairi, L.; Blouquit, Y.; Annighöfer, B.; May, R. P.; Bellissent-Funel, M.-C. High-pressure SANS and fluorescence unfolding study of calmodulin. Biochim. Biophys. Acta 2014, 1844, 1560-1568.

18 ACS Paragon Plus Environment

Page 18 of 19

Page 19 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

TOC Graphic

19 ACS Paragon Plus Environment