Molecular Mechanisms of the Interactions of N-(2-Hydroxypropyl)methacrylamide Copolymers Designed for Cancer Therapy with Blood Plasma Proteins

The binding of plasma proteins to a drug carrier alters the circulation of nanoparticles (NPs) in the bloodstream, and, as a consequence, the anticancer efficiency of the entire nanoparticle drug delivery system. We investigate the possible interaction and the interaction mechanism of a polymeric drug delivery system based on N-(2-hydroxypropyl)methacrylamide (HPMA) copolymers (pHPMA) with the most abundant proteins in human blood plasma—namely, human serum albumin (HSA), immunoglobulin G (IgG), fibrinogen (Fbg), and apolipoprotein (Apo) E4 and A1—using a combination of small-angle X-ray scattering (SAXS), analytical ultracentrifugation (AUC), and nuclear magnetic resonance (NMR). Through rigorous investigation, we present evidence of weak interactions between proteins and polymeric nanomedicine. Such interactions do not result in the formation of the protein corona and do not affect the efficiency of the drug delivery.


Introduction
Currently, the design of anticancer therapeutics is focused on creating new systems with low toxicity, e.g., those based on nanocarriers, enabling targeted delivery and high efficiency of the treatment [1]. Usually, it is assumed that immediately after being introduced into the blood, nanoparticles (NPs) are surrounded by plasma proteins that form a biological coating around the individual NPs, known as the protein corona [1][2][3][4][5][6][7][8][9][10]. Various characteristics of NPs (their size and shape, surface properties, etc.) have an influence on a formation mechanism of the protein corona [5,7]. In this respect, two categories of proteins can be distinguished: opsonins, which can increase the recognition of the reticuloendothelial

SAXS Measurements
Synchrotron SAXS experiments were performed at the EMBL beamline P12 (DESY, Hamburg, Germany) using a pixel detector (PILATUS 2M). The X-ray scattering images were recorded for a sample-detector distance of 4.1 m, using a monochromatic incident X-ray beam with two different wavelengths (λ = 1.24 Å and λ = 0.92 Å) covering the range of momentum transfers 0.03 nm −1 < q < 5.1 nm −1 (q = 4π sin θ/λ, where 2θ is the scattering angle). A full description of the experimental details is available in the SupplementaryMaterials.

NMR Measurements
High-resolution 1 H NMR spectra were recorded with a Bruker Avance III 600 spectrometer operating at 600.2 MHz (Bruker BioSpin, Rheinstetten, Germany). The 1 H spin-spin relaxation times T2 of HDO were measured at 600.2 MHz using the CPMG18 pulse sequence with td = 5 ms. Every experiment was made with 16 scans, and the relaxation delay between scans was 100 s. The obtained T2 relaxation curves were monoexponential and the fitting process always made it possible to determine a single value of the relaxation time. The standard Bruker STD NMR pulse sequence STDDIFFESGP.3 with water suppression was used. An off-resonance at 20 ppm was used, and selective protein saturation was achieved by irradiating protein signals for 2 s with a spin-lock filter of 30 ms.

Analytical Ultracentrifugation
The sedimentation analysis was performed using a ProteomeLab XL-I analytical ultracentrifuge equipped with an An50Ti rotor (Beckman Coulter Life Sciences, Indianapolis, IN, USA) at a 0.5 or 40 mg mL −1 HSA and 1 or 18 mg mL −1 pHPMA-Chol NPs total loading concentration in 0.05 M sodium phosphate and 0.15 M NaCl buffer pH 7.4 (PBS), which was also used as a reference. A full description of the experimental details for analytical ultracentrifugation measurements is available in the Supplementary Materials.

SAXS Measurements
Synchrotron SAXS experiments were performed at the EMBL beamline P12 (DESY, Hamburg, Germany) using a pixel detector (PILATUS 2M). The X-ray scattering images were recorded for a sample-detector distance of 4.1 m, using a monochromatic incident X-ray beam with two different wavelengths (λ = 1.24 Å and λ = 0.92 Å) covering the range of momentum transfers 0.03 nm −1 < q < 5.1 nm −1 (q = 4π sin θ/λ, where 2θ is the scattering angle). A full description of the experimental details is available in the Supplementary Materials.

NMR Measurements
High-resolution 1 H NMR spectra were recorded with a Bruker Avance III 600 spectrometer operating at 600.2 MHz (Bruker BioSpin, Rheinstetten, Germany). The 1 H spin-spin relaxation times T 2 of HDO were measured at 600.2 MHz using the CPMG18 pulse sequence with t d = 5 ms. Every experiment was made with 16 scans, and the relaxation delay between scans was 100 s. The obtained T 2 relaxation curves were monoexponential and the fitting process always made it possible to determine a single value of the relaxation time. The standard Bruker STD NMR pulse sequence STDDIFFESGP.3 with water suppression was used. An off-resonance at 20 ppm was used, and selective protein saturation was achieved by irradiating protein signals for 2 s with a spin-lock filter of 30ms.

Analytical Ultracentrifugation
The sedimentation analysis was performed using a ProteomeLab XL-I analytical ultracentrifuge equipped with an An50Ti rotor (Beckman Coulter Life Sciences, Indianapolis, IN, USA) at a 0.5 or 40 mg mL −1 HSA and 1 or 18 mg mL −1 pHPMA-Chol NPs total loading concentration in 0.05 M sodium phosphate and 0.15 M NaCl buffer pH 7.4 (PBS), which was also used as a reference. A full description of the experimental details for analytical ultracentrifugation measurements is available in the Supplementary Materials.

Results
Three techniques were utilized to study pHPMA-Chol copolymer NPs, the plasma proteins (HSA, IgG, Fbg, Apolipoprotein E4, and A1), the blood plasma itself, and the polymer/protein mixtures. Firstly, the individual components, i.e., NPs and protein solutions, were separately analyzed by SAXS, and the examples of the scattering curves from different protein samples are shown in Figure 2 (see details in the Supplementary Materials).

Results
Three techniques were utilized to study pHPMA-Chol copolymer NPs, the plasma proteins (HSA, IgG, Fbg, Apolipoprotein E4, and A1), the blood plasma itself, and the polymer/protein mixtures. Firstly, the individual components, i.e., NPs and protein solutions, were separately analyzed by SAXS, and the examples of the scattering curves from different protein samples are shown in Figure 2   The SAXS data were used for ab initio shape reconstruction of the free polymer NPs and for comparison with computed scattering from the available high-resolution crystal structures of proteins, using the programs DAMMIN and CRYSOL, respectively [26]. For the proteins displayed in Figure 2, the results confirmed the monomeric state in solution. Next, SAXS experiments were performed on mixtures of proteins and polymers to check for possible interactions. In the absence of interactions between the polymers and proteins, the scattering from their mixture can be represented as a linear combination of the scattering curves from the two components with appropriate volume fractions; if complexes are present, such a representation would not fit the experimental data. For all analyzed samples, the scattering patterns were computed from the best-fitting mixtures using the program OLIGOMER, which yielded strong agreement with the experimental data ( Figure 3) from the mixtures of individual proteins and free NPs (see details in the Supplementary Materials) [31,32]. This finding clearly pointed to the absence of significant interactions between the investigated proteins and pHPMA-Chol NPs. A similar result was also obtained for the native blood plasma and pHPMA-Chol NPs, indicating that other proteins present in the plasma do not interact with the NPs either. The SAXS data were used for ab initio shape reconstruction of the free polymer NPs and for comparison with computed scattering from the available high-resolution crystal structures of proteins, using the programs DAMMIN and CRYSOL, respectively [26]. For the proteins displayed in Figure 2, the results confirmed the monomeric state in solution. Next, SAXS experiments were performed on mixtures of proteins and polymers to check for possible interactions. In the absence of interactions between the polymers and proteins, the scattering from their mixture can be represented as a linear combination of the scattering curves from the two components with appropriate volume fractions; if complexes are present, such a representation would not fit the experimental data. For all analyzed samples, the scattering patterns were computed from the best-fitting mixtures using the program OLIGOMER, which yielded strong agreement with the experimental data ( Figure 3) from the mixtures of individual proteins and free NPs (see details in the Supplementary Materials) [31,32]. This finding clearly pointed to the absence of significant interactions between the investigated proteins and pHPMA-Chol NPs. A similar result was also obtained for the native blood plasma and pHPMA-Chol NPs, indicating that other proteins present in the plasma do not interact with the NPs either.
Absorbance and interference optical detection systems were used for AUC measurements, allowing for accurate monitoring of the sedimentation in real time and for the determination of the possible interactions between pHPMA-Chol NPs and human blood plasma proteins ( Figure 4) [33]. Analyses of the mixed solutions of proteins and NPs did not show additional peaks in the case of HSA and IgG, even at a HSA concentration as high as 40 mg·mL -1 , which represents its physiological level in human blood. Absorbance and interference optical detection systems were used for AUC measurements, allowing for accurate monitoring of the sedimentation in real time and for the determination of the possible interactions between pHPMA-Chol NPs and human blood plasma proteins ( Figure 4) [33]. Analyses of the mixed solutions of proteins and NPs did not show additional peaks in the case of HSA and IgG, even at a HSA concentration as high as 40 mg·mL -1 , which represents its physiological level in human blood. Conformational changes and internal motions are just as important for the function of biomolecules as their chemical structures. NMR is an experimental technique that provides a clear insight into the behavior of these systems on an atomic level.
Saturation Transfer Difference (STD NMR) is one of the handiest NMR methods for the detection of temporal ligand-protein interactions in solution through monitoring the signals of a ligand (with spectroscopic properties suitable for high-resolution studies) regardless of the protein size and structure [34,35]. In our case, a part of the copolymer (located on the surface of the NPs) and cholesterol (located not deeply in the structure of NPs) possesses the necessary spectroscopic properties and allows us to explore the nature of the interaction of NPs with the protein. The  Absorbance and interference optical detection systems were used for AUC measurements, allowing for accurate monitoring of the sedimentation in real time and for the determination of the possible interactions between pHPMA-Chol NPs and human blood plasma proteins ( Figure 4) [33]. Analyses of the mixed solutions of proteins and NPs did not show additional peaks in the case of HSA and IgG, even at a HSA concentration as high as 40 mg·mL -1 , which represents its physiological level in human blood. Conformational changes and internal motions are just as important for the function of biomolecules as their chemical structures. NMR is an experimental technique that provides a clear insight into the behavior of these systems on an atomic level.
Saturation Transfer Difference (STD NMR) is one of the handiest NMR methods for the detection of temporal ligand-protein interactions in solution through monitoring the signals of a ligand (with spectroscopic properties suitable for high-resolution studies) regardless of the protein size and structure [34,35]. In our case, a part of the copolymer (located on the surface of the NPs) and cholesterol (located not deeply in the structure of NPs) possesses the necessary spectroscopic properties and allows us to explore the nature of the interaction of NPs with the protein. The Conformational changes and internal motions are just as important for the function of biomolecules as their chemical structures. NMR is an experimental technique that provides a clear insight into the behavior of these systems on an atomic level.
Saturation Transfer Difference (STD NMR) is one of the handiest NMR methods for the detection of temporal ligand-protein interactions in solution through monitoring the signals of a ligand (with spectroscopic properties suitable for high-resolution studies) regardless of the protein size and structure [34,35]. In our case, a part of the copolymer (located on the surface of the NPs) and cholesterol (located not deeply in the structure of NPs) possesses the necessary spectroscopic properties and allows us to explore the nature of the interaction of NPs with the protein. The presence of both a polymer and cholesterol signal on the STD spectrum (red spectrum, Figure 5) confirms the presence of a notable interaction between the HSA and the NP. Comparison of samples with different ratios of protein and NP concentrations suggests that enhancement of STD can only be explained by a saturation transfer between the protein and the NP. The STD enchantment for the NP in the presence of other proteins (Fbg, Apo A) can be neglected due to a weak signal comparable to the signal-to-noise ratio.
presence of both a polymer and cholesterol signal on the STD spectrum (red spectrum, Figure 5) confirms the presence of a notable interaction between the HSA and the NP. Comparison of samples with different ratios of protein and NP concentrations suggests that enhancement of STD can only be explained by a saturation transfer between the protein and the NP. The STD enchantment for the NP in the presence of other proteins (Fbg, Apo A) can be neglected due to a weak signal comparable to the signal-to-noise ratio.

Discussion
Two possible mechanisms of the interaction were proposed: (i) HSA binds to the NPs' cholesterol groups, or (ii) HSA is restricted in the meshes formed by pHPMA chains [21]. The finding from SAXS measurements and fitting clearly pointed to the absence of significant interactions between the investigated proteins and NPs. A similar result was also obtained for the native blood plasma and NPs, indicating that other proteins present in the plasma do not have a strong interaction with the NPs either. Some weak interactions between NPs with Fbg and Apo E4 were represented by a slight shift of the sedimentation coefficient distribution and the presence of an additional peak, respectively ( Figure 4). However, for Apo A1, the whole distribution changes upon mixing it with NPs and resembles the shape of the distribution of free NPs, confirming the presence of weak cholesterol-protein interactions. Spin-spin relaxation times (T2) measured by NMR provide the dynamic picture of the segment's movement on a picoseconds-nanoseconds time scale. This allows for characterization of the motions of flexible polymer segments and solvent separately [36,37]. The spin relaxation time T2 is sensitive to both the local molecular dynamics and the local density of protons in the medium [37]. The effect of the component in the mixture can be evaluated using the T2 ratio from neat NPs or protein in comparison to their mixture. The ratio, approximately equal to 1, characterizes the absence of changes in the behavior of the components in the mixture in comparison with the pure sample. Increasing the value of the ratio T2 HDO (HSA)/T2 HDO (HSA+NP) assumes the presence of some kind of interaction that affects the mobility of water in the solution. Moreover, we followed the changes in the pHPMA methyl group (which remains mobile on the surface of the nanoparticle), and no restriction in mobility due to the pHPMA and HSA interaction was observed (Table 1). This result suggests that the interaction of the NP with HSA does not go through the HPMA loop and can be an argument in favor of the second hypothesis that the interaction occurs between HSA and cholesterol.

Discussion
Two possible mechanisms of the interaction were proposed: (i) HSA binds to the NPs' cholesterol groups, or (ii) HSA is restricted in the meshes formed by pHPMA chains [21]. The finding from SAXS measurements and fitting clearly pointed to the absence of significant interactions between the investigated proteins and NPs. A similar result was also obtained for the native blood plasma and NPs, indicating that other proteins present in the plasma do not have a strong interaction with the NPs either. Some weak interactions between NPs with Fbg and Apo E4 were represented by a slight shift of the sedimentation coefficient distribution and the presence of an additional peak, respectively ( Figure 4). However, for Apo A1, the whole distribution changes upon mixing it with NPs and resembles the shape of the distribution of free NPs, confirming the presence of weak cholesterol-protein interactions. Spin-spin relaxation times (T 2 ) measured by NMR provide the dynamic picture of the segment's movement on a picoseconds-nanoseconds time scale. This allows for characterization of the motions of flexible polymer segments and solvent separately [36,37]. The spin relaxation time T 2 is sensitive to both the local molecular dynamics and the local density of protons in the medium [37]. The effect of the component in the mixture can be evaluated using the T 2 ratio from neat NPs or protein in comparison to their mixture. The ratio, approximately equal to 1, characterizes the absence of changes in the behavior of the components in the mixture in comparison with the pure sample. Increasing the value of the ratio T 2 HDO (HSA)/T 2 HDO (HSA+NP) assumes the presence of some kind of interaction that affects the mobility of water in the solution. Moreover, we followed the changes in the pHPMA methyl group (which remains mobile on the surface of the nanoparticle), and no restriction in mobility due to the pHPMA and HSA interaction was observed (Table 1). This result suggests that the interaction of the NP with HSA does not go through the HPMA loop and can be an argument in favor of the second hypothesis that the interaction occurs between HSA and cholesterol. The measurement of relaxation times does not confirm that HSA is restricted in meshes formed by pHPMA chains, but it also does not provide information about the interaction of protein with NPs via cholesterol. However, by changing the saturation time in STD NMR experiments, we can evaluate the influence of the polymer and cholesterol signals on the STD amplitude separately. STD NMR profiles ( Figure 6) for pHPMA-Chol copolymer NPs in the presence of HSA demonstrate a more intense effect on the cholesterol in comparison with the STD enhancement on the pHPMA signals. This fact proves the mechanism of the interaction between protein and NPs through cholesterol.
The measurement of relaxation times does not confirm that HSA is restricted in meshes formed by pHPMA chains, but it also does not provide information about the interaction of protein with NPs via cholesterol. However, by changing the saturation time in STD NMR experiments, we can evaluate the influence of the polymer and cholesterol signals on the STD amplitude separately.
STD NMR profiles ( Figure 6) for pHPMA-Chol copolymer NPs in the presence of HSA demonstrate a more intense effect on the cholesterol in comparison with the STD enhancement on the pHPMA signals. This fact proves the mechanism of the interaction between protein and NPs through cholesterol.

Conclusions
The presence of strong or weak interactions between nanoparticles and proteins determines the possibility of occurrence and the mechanism of corona formation. The presence or absence of a corona is an important factor in determining the effectiveness of a nanoparticle as a drug carrier. The complementary methods SAXS and AUC witness that no thick, hard, or soft protein corona from HSA, IgG, Fbg, Apo E4, Apo A1, or plasma itself is formed around pHPMA-Chol copolymer NPs. This result proves the perfect "stealth" properties of pHPMA without any absorption of protein on the nanocarriers designed for the cancer therapy. Meanwhile, the results from a combination of SAXS, AUC, and NMR demonstrate the existence of weak interactions between proteins (HSA and Apo A1) and cholesterol groups of NPs. However, these interactions do not hamper the drug delivery potency of the studied pHPMA-Chol copolymer NPs.
Supplementary Materials: The following are available online at xxx. Figure S1: The dependence of the scattered intensity I(s) on the momentum transfer s for HSA, Fbg, IgG, and plasma. Figure S2: Analysis of pHPMA NPs mixed with HSA at different NPs/HSA ratios. Figure

Conclusions
The presence of strong or weak interactions between nanoparticles and proteins determines the possibility of occurrence and the mechanism of corona formation. The presence or absence of a corona is an important factor in determining the effectiveness of a nanoparticle as a drug carrier. The complementary methods SAXS and AUC witness that no thick, hard, or soft protein corona from HSA, IgG, Fbg, Apo E4, Apo A1, or plasma itself is formed around pHPMA-Chol copolymer NPs. This result proves the perfect "stealth" properties of pHPMA without any absorption of protein on the nanocarriers designed for the cancer therapy. Meanwhile, the results from a combination of SAXS, AUC, and NMR demonstrate the existence of weak interactions between proteins (HSA and Apo A1) and cholesterol groups of NPs. However, these interactions do not hamper the drug delivery potency of the studied pHPMA-Chol copolymer NPs.
Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4923/12/2/106/s1. Figure S1: The dependence of the scattered intensity I(s) on the momentum transfer s for HSA, Fbg, IgG, and plasma. Figure S2: Analysis of pHPMA NPs mixed with HSA at different NPs/HSA ratios. Figure