Control of Polymer Brush Morphology, Rheology, and Protein Repulsion by Hydrogen Bond Complexation

Abstract

Polymer brushes are widely used to alter the properties of interfaces. In particular, poly(ethylene glycol) (PEG) and similar polymers can make surfaces inert toward biomolecular adsorption. Neutral hydrophilic brushes are normally considered to have static properties at a given temperature. As an example, PEG is not responsive to pH or ionic strength. Here we show that, by simply introducing a polymeric acid such as poly(methacrylic acid) (PMAA), the highly hydrated brush barrier can change its properties entirely. This is caused by multivalent hydrogen bonds in an extremely pH-sensitive process. Remarkably, it is sufficient to reduce the pH to 5 for complexation to occur at the interface, which is two units higher than in the corresponding bulk systems. Below this critical pH, PMAA starts to bind to PEG in large amounts (comparable to the PEG amount), causing the brush to gradually compact and dehydrate. The brush also undergoes major rheology changes, from viscoelastic to rigid. Furthermore, the protein repelling ability of PEG is lost after reaching a threshold in the amount of PMAA bound. The changes in brush properties are tunable and become more pronounced when more PMAA is bound. The initial brush state is fully recovered when releasing PMAA by returning to physiological pH. Our findings are relevant for many applications involving functional interfaces, such as capture-release of biomolecules.

Figure 1

Analyzing intermolecular complex formation by SPR. (A) Kinetics of 20 kg/mol thiol-PEG grafting in 0.9 M Na₂SO₄ and subsequent 8 kg/mol PMAA binding (100 μg/mL) in PBS with pH lowered to 4.5. (B) Control shows only a bulk response when a high concentration (10 mg/mL) of PMAA is introduced at physiological pH. (C) Kinetics of PMAA (100 μg/mL introduced) binding to and dissociating from a PEG brush as the pH is changed. The arrows indicate dextran injections to probe brush height. (D) Exclusion heights of the PEG brush initially, after saturated PMAA binding, and after rinsing at high pH. Each line represents one experiment. (E) Example of exclusion height as a function of amount of PMAA bound at pH 4.5. (F) Angular spectra in dry state of clean Au, Au + PEG brush, and after saturated PMAA binding.

Figure 2

Probing rheology changes in the brush by QCMD. (A) Frequency and dissipation signals at different overtones upon PMAA (100 μg/mL) binding at pH 4.5 and release by increased pH. (B) Thickness changes upon PMAA binding based on Voight and Sauerbrey models. (C) Change in shear elasticity and viscosity upon complexation. After a pH increase the initial values are recovered.

Figure 3

Surface force apparatus experiments. (A) Simplified representation of the setup. The PEG brush is approached by a mica surface at a decreasing distance z and the force is measured. The separation between the surfaces is measured optically based on multiple beam interference fringes of equal chromatic order (FECO). (B) Semilogarithm plot of the force–distance curve profile during the approach of a mica surface to a PEG brush in pH 4.5 PBS (black markers), after PMAA bound to the PEG brush (red markers), and after increase of pH to 7.5 PBS (blue markers). The full lines are fits to eq 1, resulting in equilibrium brush heights (no compression) of 44.0, 42.0, and 43.4 nm. The inset shows the behavior of the polymer brush in the compressed regime (linear in the semilog plot).

Figure 4

Altering the protein repelling ability of PEG brushes. (A) PMAA binding followed by rinsing steps at pH 7.5 and injection of avidin, which gives no detectable binding. Subsequently, very brief injections of PMAA and new injections of avidin are performed. (B) Avidin injections for different amounts of PMAA already bound to the PEG brush (represented by SPR signals). Rinsing is performed as indicated by arrows. (C) Avidin signal as a function of PMAA signal. The detection limit is indicated. The dashed lines are guides to the eye, showing a threshold behavior.