Protein encapsulation via polyelectrolyte complex coacervation: Protection against protein denaturation

It has long been known that the electrostatic interaction of oppositely charged macromolecules in an aqueous environment leads to the formation of soluble complexes, or to phase separation including liquid-liquid (complex coacervation if talking about multiple types of macromolecules) and liquid-solid (precipitation) separation, depending on various factors such as the salt concentration, charge density, molecular weight, chirality of polyelectrolyte, and temperature.^1–4 When complex coacervation occurs, a dense polymer-rich phase (the coacervate phase) and a very dilute polymer-deficient phase (the supernatant phase), existing in equilibrium, are formed.⁵ Complex coacervation can be used as a route to compartmentalize small organic molecules,^6–8 inorganic nanoparticles,⁹ and biomacromolecules such as proteins^10–12 and DNA¹³ within microscale coacervate droplets. The generally low surface tension between the polymer-rich phase and polymer-deficient phase can possibly facilitate the transfer and uptake of solutes into the coacervate phase due to the low energetic barrier associated with crossing that interface.^14,15 It has been suggested that the drops initially formed during the coacervation process might have once been a primitive type of protocell, creating coacervate-mediated spatial compartmentalization and chemical enrichment in aqueous solution without a membrane.¹⁶ 

The importance of complex coacervate materials serving as microencapsulates for the uptake of a variety of solutes has applications in various fields. For example, gelatin-gum Arabic complex coacervates were developed for the encapsulation and thermally sensitive release of flavor compounds to improve the appeal of frozen foods upon heating.¹⁷ Chitosan based coacervates have been investigated for the delivery of proteins and vaccines, providing insight for the development of coacervates for delivery of protein biologics and vaccines.¹⁸ Dubin et al. developed a protein separation method via polyelectrolyte coacervation, providing a roadmap for further development of protein purification methods with high efficiency and selectivity.¹⁹ 

Polyelectrolyte complexes and related materials, including layer-by-layer films,^20,21 multilayer capsules,²² and bulk polyelectrolyte complexes,²³ have been shown to provide protection for proteins with an enhanced stability and preserved activity when exposed to harsh environmental conditions, such as high temperature and proteolysis. For example, Schwinté et al. reported that poly(styrenesylfonate) in contact with hen egg white lysozyme (HEL) could largely prevent the heat-induced aggregation of HEL.²¹ Encapsulation of enzymes in polyelectrolyte multilayer capsules provided a protective barrier for the enzyme against protease degradation, with enhanced stability and activity compared to the enzyme solubilized in aqueous solution.²² 

Polyelectrolyte complex coacervates provide a simple, rapid, and versatile alternative method to encapsulate and potentially protect proteins against harsh environmental conditions. Once encapsulated within the coacervates, proteins are concentrated within the droplets and protected from the surrounding environment, thereby increasing the local protein concentration within coacervate droplets as well as preserving their bioactivity.^10,24 Compared to other protein carriers such as hydrogels and microparticles, coacervates are formed simply by mixing oppositely charged macroions and taking advantage of the spontaneous phase separation and they do not require exposure to heat, organic solvents, modification of the proteins, or additional chemical reagents which might result in denaturation of proteins.^25,26 Additionally, the use of an aqueous system negates the need for separation steps to remove toxic organic solvents or chemical reagents prior to use in medical or food applications.²⁷ Another advantage of coacervates is the colloidal stability of charged complex coacervates induced by electrostatic repulsions, which maintains the high surface area of coacervates and therefore makes their interaction with the supernatant environment easier.^28,29

A number of factors can cause damage to proteins, including extremes of pH, high temperature, and addition of chemical reagents such as urea, guanidine HCl, or heavy metals.³⁰ Linderström-Lung first proposed that the stability of the protein was determined by electrostatic interactions to explain the lowered stability of proteins at extremes of pH.³¹ At acidic pH, the decreased stability would be the result of unfavorable electrostatic interactions introduced by the increase in positive charges on a protein. Similarly, the decreased stability at basic pH can be attributed to the repulsion of increased negative charges on the protein as well. It has long been know that incubation of protein solutions at high temperatures leads to protein denaturation, due to the disruption of non-covalent bonds.³² Urea may exert its effect on protein denaturation directly, by binding to the protein, or indirectly, by altering the solvent environment.^33–35 Heavy metals, with even a trace amount, can either interfere with the biological activity of native folded proteins through binding with functional side chain groups or displacing essential metal ions in metalloproteins,^36,37 or inhibit refolding of denatured proteins and cause aggregation of nascent proteins.^38,39 Because of the broad application of proteins for biosensors,⁴⁰ therapeutics,⁴¹ and enzyme catalysis,^24,42 it is of great importance to maintain protein stability and therefore functionality against extreme conditions.

Here, we present a study on the encapsulation of Bovine Serum Albumin (BSA) into poly(acrylic acid)-poly(allylamine hydrochloride) (PAA-PAH) coacervates. The influence of mixing sequence, total polyelectrolyte concentration, BSA overall concentration, and the molar ratio of PAA/PAH on the BSA encapsulation was studied, providing a useful roadmap for efficient encapsulation of proteins within polyelectrolyte complex coacervates. Preservation of the secondary structure of BSA during the complex coacervation process was confirmed using circular dichroism (CD) spectroscopy. Our study reveals that PAA-PAH coacervates can serve as a protective system against the denaturation of BSA when exposed to extremes of pH, high temperature, as well as in solution of urea. Additionally, it was found that by encapsulation of protein within coacervates via complex coacervation, the complexation between protein and heavy metal can be efficiently inhibited. Given that coacervates have been exploited as membrane-free artificial cellular models, our results provide new insights and methods for the design and construction of synthetic protocells capable of stabilizing protein structures under extreme conditions.

Bovine Serum Albumin (BSA, M_w ∼ 66 kDa, pI = 4.7) was purchased from Sigma-Aldrich. The positively charged polyelectrolyte, poly(allylamine hydrochloride) (PAH, M_w = 4 500 000), was purchased from Sigma. The negatively charged polyelectrolyte, poly(acrylic aid) (PAA, M_w = 50 000), was purchased from Polysciences, Inc. The negatively charged dye, 8-anilino-1-naphthalenesulfonic acid (ammonium salt, ANS), was purchased from Sigma-Aldrich. Tris(hydroxymethyl)aminomethane (Tris base) was purchased from Sigma-Aldrich. Urea was purchased from Sigma-Aldrich. CuCl₂·2H₂O was purchased from Alfa Aesar. All these materials were used as received without further purification. All water was dispensed from a Milli-Q water system at a resistivity of 18.2 MΩ·cm.

Stock solutions of PAH (20, 30, and 40 mM) and PAA (20, 30, and 40 mM) were prepared in 10 mM Tris buffer at pH 7.4. The concentration of PAH or PAA is with respect to the repeat unit. BSA stock solution (40 mg ml⁻¹) was prepared in 10 mM Tris buffer at pH 7.4. To study the encapsulation of BSA into PAA-PAH coacervates, the PAA-PAH coacervates with encapsulated BSA were prepared using three different mixing sequences (see Scheme 1 and Table I): (a) addition of BSA stock solution to a large excess of PAA stock solution to form BSA-PAA intermediate complexes under 1 h stirring, followed by the mixing of PAA-BSA mixture into PAH stock solution with a PAA:PAH molar ratio of 0.1, 0.2, 0.4, and 0.5, with another stirring for 3 h to prepare (BSA/PAA)-PAH coacervates; (b) addition of BSA stock solution to a large excess of PAH stock solution to form BSA-PAH intermediate complexes under 1 h stirring, followed by the mixing of PAA stock solution into BSA-PAH mixture with a PAA:PAH molar ratio of 0.1, 0.2, 0.4 and 0.5, with another stirring for 3 h to prepare (BSA/PAH)-PAA coacervates; (c) mixing of PAA stock solution into PAH stock solution with a PAA:PAH molar ratio of 0.1, 0.2, 0.4, and 0.5 to prepare the PAA-PAH coacervates under 1 h stirring, followed by the addition of BSA stock solution to prepare (PAA-PAH)/BSA coacervate. For the preparation of each coacervate sample, the concentration of PAA and PAH stock solution was selected to be the same to keep the total polyelectrolyte concentration constant regardless of the variation in mixing ratio. The detailed information on the mixing sequence, overall concentration of BSA, and the total polyelectrolyte concentration of the coacervate samples used in this study is listed in Table I.

Dynamic light scattering (DLS) was performed on a Zeta Phase Analysis Light Scattering (PALS) instrument (Brookhaven, USA) to study the formation of BSA-PAA or BSA-PAH intermediate complexes. Samples for DLS measurement were prepared as 20 mM PAA, 20 mM PAH, 0.25 mg ml⁻¹ BSA, mixture of 20 mM PAA and 0.25 mg ml⁻¹ BSA, and mixture of 20 mM PAH and 0.25 mg ml⁻¹ BSA in Tris buffer at pH 7.4.

Zeta potential measurements were recorded at room temperature on PAA-PAH coacervates using a Zeta PALS instrument (Brookhaven, USA) to study the influence of urea on the PAA-PAH coacervates. Samples for zeta potential measurement were prepared by mixing 20 mM PAA stock solution into 20 mM PAH stock solution with a PAA/PAH molar ratio of 0.5, supplemented with 0–10M urea.

PAA-PAH/BSA coacervate samples were centrifuged for 3 h at 9000 rpm (Allegra X-30R Centrifuge, Beckman Coulter). After centrifugation, the supernatant was removed using a micropipette and the coacervate phase was left in the bottom of centrifuge tubes. Ultraviolet-visible (UV-vis) measurement (Agilent 8453 spectrophotometer) was used to determine the BSA content of the supernatant. The encapsulation of BSA into the PAA-PAH coacervate phase was calculated as per Eq. (1). The partition coefficient (K) of BSA into the PAA-PAH coacervate phase was calculated as per Eq. (2),

To determine the secondary structure of free versus encapsulated BSA, CD spectroscopy measurements were performed at room temperature on a J-1500 circular dichroism spectrophotometer (Jasco, Inc., Japan) using a quartz cuvette of 1 mm path length. (BSA_0.25/PAH)-PAA₂₀ coacervate samples with various PAA/PAH ratios (0.1, 0.2, and 0.5) as well as 0.25 mg ml⁻¹ BSA solution samples were chosen. The coacervate samples were stirred for 3 h at room temperature before the CD measurement. To study the influence of urea on the denaturation of free versus encapsulated BSA, CD spectra of (BSA_0.25/PAH)-PAA₂₀ coacervate samples with a fixed PAA:PAH ratio of 0.5 as well as 0.25 mg ml⁻¹ BSA solution samples, supplemented with 0–10M urea, were measured. To examine how the coacervates can protect BSA against extremes of pH, CD spectra of (BSA_0.25/PAH)-PAA₂₀ coacervate samples with a fixed PAA:PAH ratio of 0.5 as well as 0.25 mg ml⁻¹ BSA solution, with pH adjusted to 3.0, 7.4, and 12.0, respectively, were measured. After pH adjustment, both the coacervate and BSA solution samples were stirred at room temperature for 3 h before the CD measurements. To ensure that the pH of coacervate samples and BSA solution do not drift during the measurements, the pH was measured before and after the CD measurements. To study the protection of BSA against high temperature using coacervates, CD spectra of (BSA_0.25/PAH)-PAA₂₀ coacervate samples with a fixed PAA:PAH ratio of 0.5 as well as 0.25 mg ml⁻¹ BSA solution samples, kept either at room temperature or heated at 70 °C for 90 min and cooled to room temperature, were measured. Measurements were performed from 280 nm to 200 nm.

The mean residue molar ellipticity (MRE) was calculated according to the equation

where $θ$ is the measured ellipticity, c_r is the mean residue molar concentration, and l is the path length. The fraction of unfolded proteins (Σ) was calculated according to the equation

where $MRE222 nm(native)$ and $MRE222 nm(unfolded)$ correspond to the values obtained for the native proteins (c_urea = 0M) and fully unfolded proteins (c_urea = 10M), respectively.

Steady state emission spectra of 0.25 mg ml⁻¹ BSA in aqueous solution (free BSA) as well as in PAA-PAH coacervate (encapsulated BSA) suspension with the addition of 0 to 0.5 mM Cu²⁺ were recorded within a range of 290 to 500 nm upon excitation at a wavelength of 280 nm, using a Horiba FluoroMax 4 spectrofluorometer. The (BSA_0.25/PAH)-PAA₄₀ coacervate samples were prepared with a PAA/PAH stoichiometry of 0.2.

To study the influence of urea on the coacervation of PAA and PAH, turbidity was used to qualitatively measure the extent of coacervate formation as a function of urea concentration. Turbidity measurements were performed using a 2 cm path length fiber-optic probe colorimeter (Brinkmann PC 950) at 420 nm. Turbidity was reported as 100%–T%, where T corresponds to the transmittance. Samples for turbidity measurement were prepared by mixing 20 mM PAA stock solution into 20 mM PAH stock solution with a PAA/PAH molar ratio of 0.5, supplemented with 0–10M urea. Turbidity was also used to study the phase behavior of aqueous solutions of PAA and PAH as functions of PAA/PAH stoichiometry at different total polyelectrolyte concentration. The PAA and PAH solutions were prepared in 10 mM Tris buffer at pH 7.4. Turbidity measurements for the titration of PAA into PAH solution with matching polyelectrolyte concentration were performed using a 2 cm path length fiber-optic colorimeter (Brinkmann PC950) at a wavelength of 420 nm. The transmittance was recorded at 60 s after each PAA addition.

To better understand and gain insight into the encapsulation of BSA into PAA-PAH coacervates, the phase behavior of the PAA-PAH system on its own as well as the influence of BSA on this phase behavior were both studied (Fig. S1 of the supplementary material). The total polyelectrolyte concentration has an impact on the phase behavior of the PAA-PAH system. An increase in total polyelectrolyte concentration promotes coacervation and precipitation, which both occur at a lower PAA/PAH mixing ratio as the total polyelectrolyte concentration increases. Previous work on the phase behavior of aqueous solutions of poly(diallyldimethylammonium chloride) and poly(sodium 4-styrene sulfonate) shows a similar trend.⁴³ Mixtures with higher polymer concentrations have a higher density of oppositely charged sites available to interact, leading to an earlier coacervation and precipitation at a smaller PAA/PAH molar ratio. Another factor is the higher concentration of the relatively hydrophobic backbones, which promote phase separation as well. The impact of BSA on the PAA-PAH system was studied at a fixed total polyelectrolyte concentration of 20 mM. It was observed that as the BSA concentration increases, the precipitation occurs earlier at a smaller PAA/PAH mixing ratio, suggesting that the presence of BSA favors precipitation.

The encapsulation of BSA into PAA-PAH coacervates was systematically studied as a function of different mixing sequences, overall polyelectrolyte concentration, molar ratio of PAA to PAH, and overall BSA concentration to optimize the encapsulation of BSA. The ionization degree of the weak polyelectrolytes PAA and PAH was determined using potentiometric titration as a function of pH (Fig. S2 of the supplementary material). BSA, with an isoelectric point (pI) of 4.7,⁴⁴ has an overall negative charge at the pH of 7.4 used throughout this work. However, both positive and negative charges are present on the BSA macromolecule even though its overall charge is negative, making electrostatic interactions between BSA and either PAA or PAH chains possible.

Multimodal size distribution of the BSA-PAA or BSA-PAH mixture prepared at pH 7.4 (Fig. 1), obtained using dynamic light scattering, both show three separate peaks that can be attributed to free BSA, free PAA or PAH, and BSA-PAA or BSA-PAH primary complexes, respectively, which suggests the formation of the primary complex of BSA with PAA or PAH as well as macromolecules being present in mixture that do not associate with other macromolecules. Previous studies on protein-polyelectrolyte complexation showed similar multiple modes in DLS results, which were usually attributed to the unbound protein, unbound polyelectrolyte chains, intrapolymer complexes, and even interpolymer complexes.^45–48 Different mixing sequences of the three components were then studied in order to optimize the encapsulation of BSA into PAA-PAH coacervates: (a) mixing BSA with PAA before coacervation with PAH to prepare (BSA/PAA)-PAH sample; (b) mixing BSA with PAH before coacervation with PAA to prepare (BSA/PAH)-PAA sample; (c) mixing BSA with the preformed PAA-PAH coacervates to prepare (PAA-PAH)/BSA sample. Figure 2(a) shows the importance of the mixing order for the BSA encapsulation efficiency: mixing of BSA having a negative net charge with the polycation PAH before coacervation leads to more efficient encapsulation than both mixing BSA with the polyanion PAA before coacervation and mixing BSA with the preformed PAA-PAH coacervates, presumably due to stronger interactions between BSA and PAH over PAA. Due to the overall negative charge of BSA at pH 7.4, BSA tends to form a larger number of soluble BSA-PAH complexes with positively charged PAH than with negatively charged PAA, leading to a higher BSA encapsulation using mixing sequence (b) by complex coacervation of PAA with the BSA-bounded PAH, which can also complex with the free PAH. Mixing order has been shown to be important in the formation of polyelectrolyte complexes and coacervates as kinetic trapping of the polyelectrolytes in a given configuration is possible due to the strength of the ionic interactions.^28,49,50

The influence of total polyelectrolyte concentration on the encapsulation of BSA was studied using the (BSA/PAH)-PAA sample prepared by mixing BSA with the PAH before coacervation with a fixed overall BSA concentration of 0.25 mg·ml⁻¹, as shown in Fig. 2(b). An increase in the total polyelectrolyte concentration leads to an increase in the encapsulation of BSA into PAA-PAH coacervate [Fig. 2(b)], probably due to the increasing coacervate yield with overall polyelectrolyte concentration. Increasing yield of coacervate with polyelectrolyte concentration is consistent with the reported literature.^5,51 The coacervate yield was defined using Eq. (5). As shown in Fig. S3 of the supplementary material, the coacervate yield increases with the increase in both the overall polyelectrolyte concentration as well as the molar ratio of PAA/PAH,

A study on the encapsulation of BSA as a function of overall concentration of BSA reveals that an increase in the overall concentration of BSA leads to an increase in the BSA encapsulation [Fig. 2(c)]. For the (BSA/PAH)-PAA coacervate samples prepared in this study, the encapsulation of BSA increases with the increase in the PAA/PAH molar ratio [Fig. 2(c)], which may be a consequence of the increasing coacervate yield with the PAA/PAH molar ratio. Another possible explanation for this increased encapsulation of BSA into the PAA-PAH coacervate with the PAA/PAH molar ratio is that the excess PAH in the supernatant at a low PAA/PAH ratio can compete with the PAA-PAH coacervates for the protein, as previously reported in a study of protein uptake performance of chitosan/tripolyphosphate micro- and nanogels.⁵² However, for the (BSA/PAH)-PAA coacervate samples, the partition coefficient of BSA into PAA-PAH coacervates exhibits an opposite trend with respect to the PAA/PAH molar ratio compared to the encapsulation of BSA [Figs. 2(c) and 2(d)]. The partition coefficient of BSA decreases as the PAA/PAH molar ratio increases, showing that the affinity with PAH is drawing the BSA into the coacervate.

Both the total polyelectrolyte concentration and the overall BSA concentration in the coacervate suspension system have an impact on the encapsulation of BSA into PAA-PAH coacervate. To further study the overall influence of the concentration of both polyelectrolyte and BSA on the encapsulation of BSA, the ratio of encapsulated BSA (mg) to overall polyelectrolyte (mM) (BSA_encap/PE_overall) was plotted as a function of the ratio of overall BSA (mg) to overall polyelectrolyte (mM) (BSA_overall/PE_overall), as shown in Fig. 3. For all the PAA/PAH molar ratio studied here (PAA/PAH = 0.1, 0.2 and 0.5), at low BSA_overall/PE_overall ratios, the BSA_encap/PE_overall ratio increases linearly with the BSA_overall/PE_overall ratio, followed by a deviation from the linearity to a lower value, which indicates the saturation of BSA in the PAA-PAH coacervate at high BSA_overall/PE_overall ratio. Another phenomenon observed here is that up until the saturation of BSA in the PAA-PAH coacervates, the ratio of encapsulated BSA to overall polyelectrolyte scales linearly with the ratio of overall BSA to overall polyelectrolyte for a fixed PAA/PAH molar ratio. To explain this, a possible mechanism might be that the main driving force of the uptake of BSA is the coacervation of PAA with the BSA bounded PAH, which may also complex with free PAH. As the ratio of overall BSA to the overall polyelectrolyte increases, the number of soluble BSA-PAH complexes increases, leading to a higher amount of encapsulated BSA. However, as the overall amount of BSA continues to increase, PAH will be saturated by the bound BSA, leading to the deviation of BSA uptake from the linear scale to a plateau.

To study the permeability of PAA-PAH coacervates toward other solutes, BSA and the negatively charged dye ANS were selected as models for protein and small organic molecule, respectively. PAA-PAH coacervates were prepared by mixing PAA with PAH, followed by the addition of BSA or ANS into the pre-prepared coacervate suspension. The encapsulation of BSA or ANS into the PAA-PAH coacervates was measured as a function of time, as shown in Fig. S4 of the supplementary material. The dashed lines in Fig. S4 of the supplementary material represent the encapsulation of BSA or ANS into PAA-PAH coacervates prepared via mixing of negatively charged BSA or ANS with the PAH before coacervation. The encapsulation of ANS into the PAA-PAH coacervate reaches a steady state at 9 h, while the encapsulation of BSA does not reach the steady state until 24 h. Additionally, the encapsulation of ANS after reaching the steady state is independent of the mixing sequence, while the encapsulation of BSA into the pre-prepared coacervates even after reaching the steady state is much lower than the encapsulation of BSA by mixing BSA with PAH before the coacervation with PAA. The longer time for reaching the steady state and reduced encapsulation of BSA indicate that the permeation of the larger protein into PAA-PAH coacervates is lesser that the permeation of small molecules.

Preservation of the secondary structure is an important criterion for protein encapsulation, as the structure directly relates to the activity and function of the protein. Circular dichroism (CD) spectroscopy is an excellent tool for rapid determination of the secondary structure of proteins. The CD spectrum obtained for native free BSA consisted of two minima at 222 and 208 nm, which are characteristic of an α-helical secondary structure (Fig. 4).¹¹ A similar α-helical secondary structure was observed for encapsulated BSA in PAA-PAH coacervates as well (Fig. 4), which confirms the retention of the protein’s secondary structure during the complex coacervation process. The observed decrease in the intensity of encapsulated BSA is likely the result of loss of signal due to scattering from coacervate droplets.

Various attempts have been made to stabilize proteins in extreme environments. For example, it has been reported that proteins can be stabilized against extremes of pH in the presence of small molecules serving as cosolvents such as glucose through the strengthening of hydrophobic interactions between the non-polar side chains of protein due to the polar environment produced by sugar molecules and the steric exclusion effect of the cosolvent molecules.⁵³ In our study, instead of using small molecules as cosolvents, polyelectrolyte coacervates were used as the carrier and protector for the protein which shows ability of protein encapsulation and stabilization. While sugars have been shown to have some benefit in stabilizing proteins, it is at much higher concentrations than the polyelectrolytes used in this work.^53,54 Encapsulation of BSA within PAA-PAH coacervates was found to be capable to efficiently stabilize the BSA against extremes of pH (Fig. 5). It has been long been known that the stability of proteins is lowered at extremes of pH, probably due to the unfavorable electrostatic repulsion introduced by the increase in positive charges at low pH and negative charges at high pH on the protein. BSA unfolding at extremes of pH (3.0 and 12.0) was studied using CD spectroscopy [Fig. 5(a)]. The unfolded fraction of encapsulated BSA within PAA-PAH coacervates at pH 3.0 and 12.0 is approximately 0.1, while the unfolded fraction of free BSA at these harsh pH values is approximately 0.5 [Fig. 5(b)], indicating that coacervates serve as efficient protection for BSA under extreme pH conditions. Several reasons for this can be postulated. For one, hydronium or other ions diffusing into the coacervate will first encounter BSA molecules closest to the phase boundary between coacervate and water rich solution, and those protein molecules may act to shield the BSA proteins further within the coacervate. That is, the crowded environment experienced by the encapsulated BSA inhibits the unfolding at extreme pH conditions. However, it is likely more important that the weak polyelectrolyte environment can act as a sort of buffer in this situation, extending the pH range over which BSA can be stable. It has been previously shown that polyelectrolyte multilayer films consisting of linear polyethylenimine and polyacrylic acid maintain similar fractions of charged and uncharged carboxylic acid groups over an exposure to aqueous solution from pH 3–10, something that is not true for polyacrylic acid on its own.⁵⁵ This buffering effect, while expected to be not as strong in the coacervate due to a less dense ionic network compared to multilayer films, may be creating the demonstrated stability toward pH. In addition to the buffering effect of the polyelectrolytes, the protein itself may also be an active contributor to the buffering through a charge regulation mechanism, meaning the protein’s ability to regulate its charge due to the shift in acid-base equilibrium.^56–58 At extremely low pH, the protein within the coacervate experiences a positively charged environment, which increases the ionization degree of carboxylic groups and decreases the ionization degree of amine groups of the protein, while at extremely high pH values, the protein within the coacervate experiences a negatively charged environment, which decreases the ionization degree of carboxylic groups and increases the ionization degree of amine groups.

The impact of urea on the properties of coacervates was investigated using turbidity and zeta potential measurements. In general, there was only a small decrease in both the turbidity (Fig. S5a of the supplementary material) and positive zeta potential values (Fig. S5b of the supplementary material) of the samples prepared with increasing concentration of urea. The positive zeta potential values were attributed to the excess of positively charged PAH in the PAA-PAH coacervates prepared with a PAA/PAH molar ratio of 0.5 at pH 7.4. BSA unfolding in the presence of increasing amounts of urea was confirmed using CD spectroscopy (Fig. S6 of the supplementary material). The peak intensities and MRE were progressively reduced as the urea concentration was increased up to 10M. The transition from the folded to unfolded state of BSA was observed for both the aqueous solution (control) and coacervates, using the normalization of MRE_{222 nm} with corresponding values obtained for the native BSA (c_urea = 0M) and fully unfolded BSA (c_urea = 10M) to calculate the unfolded fraction (Fig. S6c of the supplementary material). The unfolded fraction curve of encapsulated BSA in PAA-PAH coacervates is slightly lower than the unfolded fraction curve of the free BSA in aqueous solution, indicating that the PAA-PAH coacervates stabilized the folded state of BSA. It has been shown that complexation of proteins with polyelectrolyte undergoing urea-induced unfolding tends to destabilize the protein conformers as a result of competition between polyelectrolyte-protein intermolecular interactions and intra-protein interactions.^59,60 The enhanced stability of encapsulated BSA within coacervates can be attributed to the crowded environment within PAA-PAH coacervates experienced by the BSA, which probably disfavors protein unfolding.^10,61

Contributions to protein folding arise from hydrophobic interactions, hydrogen bonding, van der Waal’s forces, electrostatic forces, and local peptide interactions.³² Incubation of protein solutions at high temperatures leads to protein denaturation, due to the disruption of the non-covalent bonds. BSA unfolding upon heating was studied using CD spectroscopy (Fig. S7a of the supplementary material). Free BSA in aqueous solution and encapsulated BSA within PAA-PAH coacervates were incubated at either room temperature or 70 °C for 90 min, followed by a cooling down procedure to room temperature for another 30 min before taking CD measurements. The unfolded fraction of encapsulated BSA within PAA-PAH coacervates after the heating-cooling down process was approximately 0.45, while the unfolded fraction of free BSA in aqueous solution after the heating-cooling down process was about 0.55 (Fig. S7b of the supplementary material), indicating that the coacervates are capable of protecting the protein against high temperature to a small degree. The enhanced stability of BSA within PAA-PAH coacervates compared to the free BSA in aqueous solution may be attributed to the crowded environment experienced by the sequestered protein, which disfavors protein unfolding due to the excluded volume effect.⁶¹ It has been suggested that thermal stability of proteins can be improved by wrapping the protein in a soft, hydrophilic, water-soluble, flexible polymer environment such as polyethylene glycol (PEG) or polyacrylic acid (PAA).^62–64 Improvement in stability against high temperature can be due to either the increase in intrinsic thermal stability of the protein via encapsulation in the polymer matrix or inhibition of the aggregation of the denatured peptide which may facilitate refolding back to the native state.⁶³ The structural change of BSA was reversible in the temperature range below 45 °C, while the structural change of BSA was only partially reversible upon cooling to room temperature subsequent to heating at higher temperatures. Another possible explanation for the reduced unfolded fraction of encapsulated BSA within coacervates compared to free BSA after the heating-cooling down process is that the coacervates may help with the refolding or recovery of heat-induced unfolded BSA. Martin et al. investigated that the accumulation of protein into coacervate microdroplets facilitates protein refolding and gives increased levels of protein renaturation.¹⁰ 

Heavy metals can either interfere with the biological activity of native folded proteins through binding with functional side chain groups, or displacing essential metal ions in metalloproteins, or inhibit refolding of denatured proteins and cause aggregation of nascent proteins.⁶⁵ The inhibition of the binding of proteins with heavy metal ions in order to retain the biological activity of native proteins or correct refolding of denatured protein is therefore urgent and important.

Our study reveals that the encapsulation of protein into coacervates via complex coacervation is capable of protecting the protein against heavy metal contamination by inhibiting the complexation of the protein with heavy metal ions. Copper ion (Cu²⁺) and BSA were selected as representative of the heavy metal and protein, respectively, to show how the encapsulation of protein via complex coacervation can significantly reduce the complexation between the protein and heavy metal. The BSA macromolecule contains two tryptophan residues, one located on the protein surface (Trp 134) and the other located closer to the internal region of the protein (Trp 212). A decrease in the emission intensity of the fluorescence of tryptophan residues in BSA was observed upon addition of Cu²⁺ into free BSA aqueous solution [Fig. 6(a)], as a result of the non-fluorescent complex formation in the ground state between the fluorophore groups of BSA and the heavy metal Cu²⁺. However, by encapsulating BSA into PAA-PAH coacervate, the fluorescence quenching was readily reduced as compared to the quenching of free BSA fluorescence induced by Cu²⁺ [Fig. 6(b)], indicating the efficient protection of BSA from Cu²⁺ simply by complex coacervation with PAA and PAH.

It has been reported that both PAA and PAH can act as polymeric ligands for a variety of heavy metal ions via coordination interactions with the carboxylic acid or amine groups, including Cu²⁺, and therefore compete with BSA to complex with Cu²⁺, realizing the protection of BSA through efficient encapsulation.^66,67 The quenching of BSA fluorescence upon Cu²⁺ binding was assumed to be static quenching in this study. Static quenching of fluorescence can be described by the Stern-Volmer equation [see Eq. (6)],⁶⁸ 

where F₀ and F are the fluorescence intensities in the absence and presence of the quencher, respectively, K_S-V is the Stern-Volmer constant, which can be assumed to be equal to the association constant K, and [Q] is the concentration of the quencher.

The Stern-Volmer plots of the BSA fluorescence quenching using Cu²⁺ for free and encapsulated BSA are shown in Fig. 7. For quantitative evaluation of association of Cu²⁺ with free BSA and encapsulated BSA, values of the Stern-Volmer constants were calculated, as shown in Table II. The Stern-Volmer constant of association for BSA with Cu²⁺ was significantly reduced from 5705 for free BSA to 352 for the encapsulated BSA within PAA-PAH coacervates, suggesting that encapsulation of proteins within coacervates can efficiently inhibit the complexation of proteins with heavy metal ions.

Presented here is a study of the encapsulation of protein into synthetic polyelectrolyte coacervates via electrostatic interaction. PAA and PAH were selected as representative weak polyelectrolytes to form complex coacervates. A model protein, BSA, was encapsulated with a tunable encapsulation ranging from approximately 14% to 50% corresponding to the variation in mixing sequence, total polyelectrolyte concentration, BSA overall concentration, and the molar ratio of PAA/PAH, which provides insights into regulation and control of protein loading efficiency within coacervates. The ability to tune and control the protein uptake efficiency is particularly important for the use of protein-based biosensors and protein therapeutics.^69,70 It was confirmed via CD spectroscopy that the encapsulation process did not affect the secondary structure of the model protein BSA, which is often an important condition for protein activity. Maintaining the stability and functionality of proteins against extreme conditions such as extremes of pH, high temperature, and the presence of chemicals including urea, guanidine HCl, and heavy metals³⁰ is a particular relevant issue today in academia and research and because of its broad application in fields of biosensors,⁴⁰ therapeutics,⁴¹ and enzyme catalysis.^24,42 Polyelectrolyte related materials have been shown to be capable of stabilizing proteins against harsh conditions. A previous study on the uptake of globular proteins in poly(diallyldimethylammonium)/poly(acrylate) coacervate droplets shows similar stabilization property of BSA against urea-induced denaturation.¹⁰ It was also reported that embedding proteins into polyelectrolyte films did not cause considerable further changes in their secondary structure and provided some protection against the denaturation upon increasing the temperature.^21,71 It is observed here that the PAA-PAH coacervate serves to strongly protect against the denaturation of BSA when exposed to extremes of pH, due to its ability to act as a sort of buffer and aid the protein in its ability to regulate charge, as well as to provide some protection from high temperature and the presence of urea. Additionally, it was found that coacervate encapsulation can inhibit the complexation between protein and heavy metal. These experiments may shed some light into the kinetics of encapsulating protein into macromolecular environments. Additionally, by choosing the acid base chemistry of the coacervate pair more carefully, it is possible that a stronger buffering environment could be designed. Given that coacervates have been thought of as membrane-free artificial cellular models, our results provide new insights and methods to design and construction of synthetic protocells capable of stabilizing protein structures under extreme conditions.

See supplementary material for phase behavior of PAA-PAH system, titrations of polyelectrolytes, turbidity and other characterizations of the coacervate, and circular dichroism data for BSA in solution and in coacervate under exposure to temperature and the presence of urea.

The authors would like to thank Professor Xiong Gong at the University of Akron (Department of Polymer Engineering) for use of his UV-vis spectroscopy, Professor Jie Zheng and Baiping Ren at the University of Akron (Department of Chemical Engineering) for help with circular dichroism measurement, as well as Professor David Modarelli and Dr. Mahesh Dawadi also at the University of Akron (Department of Chemistry) for help with fluorescence emission measurements. Additionally, the authors acknowledge funding from NSF CBET No. 1744459.