Pharmaceutical Perspective on Opalescence and Liquid–Liquid

Review pubs.acs.org/molecularpharmaceutics

Pharmaceutical Perspective on Opalescence and Liquid−Liquid Phase Separation in Protein Solutions Ashlesha S. Raut* and Devendra S. Kalonia* Department of Pharmaceutical Sciences, University of Connecticut, Storrs, Connecticut 06269, United States ABSTRACT: Opalescence in protein solutions reduces aesthetic appeal of a formulation and can be an indicator of the presence of aggregates or precursor to phase separation in solution signifying reduced product stability. Liquid−liquid phase separation of a protein solution into a protein-rich and a protein-poor phase has been well-documented for globular proteins and recently observed for monoclonal antibody solutions, resulting in physical instability of the formulation. The present review discusses opalescence and liquid−liquid phase separation (LLPS) for therapeutic protein formulations. A brief discussion on theoretical concepts based on thermodynamics, kinetics, and light scattering is presented. This review also discusses theoretical concepts behind intense light scattering in the vicinity of the critical point termed as “critical opalescence”. Both opalescence and LLPS are affected by the formulation factors including pH, ionic strength, protein concentration, temperature, and excipients. Literature reports for the effect of these formulation factors on attractive protein−protein interactions in solution as assessed by the second virial coefficient (B2) and the cloud-point temperature (Tcloud) measurements are also presented. The review also highlights pharmaceutical implications of LLPS in protein solutions. KEYWORDS: opalescence, liquid−liquid phase separation, aggregation, light scattering, thermodynamics, phase diagram, formulation, stability, protein−protein interactions, crystallization, B2, Tcloud product and cause immunogenicity concerns.16,17 Solid−liquid phase separation results in the formation of protein crystals or amorphous precipitate, indicating reduced solubility of the proteins in the solution. Liquid−liquid phase separation into a protein-rich and protein-poor phase occurs at lower temperature and may be observed for protein formulations stored at refrigerated conditions. Dictionary definition of opalescence is “reflection of iridescent light (dichroism) similar to opal”. The solution exhibits different colors depending on the angle between the incident light and the transmitted light. However, the terms “turbidity” or “opacity” of the solution, i.e., cloudy-whitish appearance and “opalescence”, are used interchangeably in the pharmaceutical literature. Opalescent/turbid appearance of the solution is due to the increased scattering of light. This scattering can be attributed to the particles in solution indicating aggregation or can be due to fluctuations of the density and concentration resulting in liquid−liquid phase separation. The graphic in the abstract shows opalescence in a monoclonal antibody solution followed by liquid−liquid phase separation after the solution was stored at refrigerated condition (adapted from Raut et al.15). For a system exhibiting

1. INTRODUCTION Monoclonal antibodies (mAbs) have emerged as successful biotherapeutics in the treatment of several terminal diseases.1 Continual efforts to enhance the safety and efficacy of these molecules against numerous indications have led to a growing interest in the second generation of antibody-based therapeutics such as ADCs, bispecific antibodies, engineered antibodies, and antibody fragments or domains.2,3 These antibodies and antibody-like molecules are generally formulated at high concentrations (>100 mg/mL) because of their low potency and low volume restriction ( 0), i.e., the energy of interactions between solute−solute and solvent−solvent is greater than the solute−solvent interactions, ΔGmix shows a strong temperature dependence. As observed in Figure 2, at some temperature T1 ≫Tc, entropic contribution (−TΔS) balances the positive enthalpic contributions resulting in negative ΔGmix and hence promotes mixing. As temperature T2 approaches and decreases below the critical temperature (T → Tc), the entropic contribution decreases and does not compensate for the positive mixing enthalpy resulting in phase separation. At an extremely low temperature, T3 ≪ Tc, ΔGmix is positive, and the system spontaneously phase separates. On undergoing LLPS, protein separates into a protein-rich phase with a higher enthalpy (attractive protein−protein interactions) and lower entropy, while the protein-poor phase has lower enthalpy and higher entropy. This enthalpy−entropy effect balances the free energy of the system, hence stabilizing the two phases.15,64 2.2. Kinetics of Phase Separation. Attaining thermodynamic equilibrium forms the basis of separation of a binary mixture; however, kinetics controls the ultimate phase separation process. This section discusses the kinetic mechanism by which binary system undergoes phase separation as the system transitions from a stable region to metastable or unstable region (on lowering the temperature or increasing the concentration). Phase separation of a binary mixture into a protein-rich and protein-poor phase occurs in the metastable region by nucleation and growth process, while the kinetics of phase separation in spinodal or unstable region is by spinodal decomposition.50,63 Table 2 outlines the fundamental difference between the two kinetic mechanisms for phase separation. 2.2.1. Nucleation and Growth Process. In the metastable region, the formation of a protein-rich phase is initiated either by homogeneous or heterogeneous nucleation. The phase separation occurs down the concentration gradient as the

total energy = 4πr 2σ +

spinodal decomposition

formation of a stable nucleus of a critical size is essential for growth of the phases downhill diffusion, down the concentration gradient sharp interface formation

no thermodynamic barrier, results in formation of domains

explained by Ostwald ripening

(6)

Smaller sized clusters formed on smaller fluctuations revert back to one-phase; however, as the size of the nucleus reaches a critical size termed as critical radius (represented by eq 7), these clusters become stable and grow further, resulting in the formation of new phase droplets. R=

2σ ε

(7)

The growth of the nucleus or the nucleation rate is as follows, J = Ae−ω / kT

(8)

where A is a factor that depends on many parameters such as the size, surface area, and total number of droplets.65,66 Nucleation rate increases as the system goes further down the coexistence curve, i.e., as the temperature of the system is lowered. Due to the formation of spherical droplets on nucleation, the two phases so formed have a sharp interface between them.66 2.2.2. Spinodal Decomposition. The phase separation process inside the unstable region occurs by spinodal decomposition mechanism. Since there is no kinetic barrier, small fluctuations in density or concentration grow in time and space resulting in an uphill diffusion (against the concentration gradient) with a negative diffusion coefficient obtained by the Cahn−Hilliard approximation.66 While nucleation and growth is a local phenomenon, systems exhibiting spinodal decomposition have long-range order. This results in the formation of domains in the unstable region instead of droplets with no sharp interface between the two phases at the initial stage.44,63,66 2.3. Light Scattering from Solution. Opalescence in solution is due to scattering of light, which can either be due to the presence of particles or fluctuations of the thermodynamic quantity in the solution. Theoretical concepts behind both are briefly discussed below, and readers are directed to general references on the topics focusing on light scattering from protein/colloid solutions for further details.67−73 2.3.1. Scattering due to Particles in Solution. The intensity of scattering from solution increases with the increasing size of the particles. Scattering in the presence of small, isotropic particles (size < λ/10) is described by Rayleigh theory as,

Table 2. Compare and Contrast for Kinetics of Phase Separation by Nucleation and Growth Mechanism and by Spinodal Decomposition nucleation and growth

4 3 πr ε 3

uphill diffusion, against the concentration gradient no sharp boundaries between the phases due to long-range fluctuations explained by Cahn−Hillard approximation

Is = I0

2 2π 2 n pNAα (1 + cos2 θ ) r 2λ 4 V

(9)

where Is is the intensity of scattered light, I0 is the intensity of incident light, r is the distance from the scattering center, λ is D

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics the wavelength of the light, np is the number of particles, α is the polarizability of molecule, V is the molar volume, and θ is the scattering angle. Scattering intensity can also be expressed in terms of turbidity (τ), which is the total relative amount of light scattered by unit volume in all directions. For particles scattering light in accordance with the Rayleigh theory, the scattering volume geometry changes with the position (angle) of the detector. Maximum scattering intensity is at θ = 0° and 180°, while minimum scattering intensity is at θ = 90°.73 For larger particles, more than one site acts as scattering center. At θ = 0°, path lengths are identical, and the absence of destructive interference results in maximum scattering intensity; while, as scattering angle increases, the scattering intensity decreases due to destructive interference. Larger particles exhibit an angular asymmetry; the scattering intensity in the forward direction is greater than the intensity for back scattering. Several theories including Mie, Rayleigh-Gans, and Debye provide a general correction for scattering intensity in solution for larger particles by considering a form factor.70 Table 3 briefly summarizes the size and wavelength dependence of scattering from the solution.69 According to

Is = I0

diameter (d)a

small (x ≪ 1)b (Rayleigh theory)

Is ∝ d6

large (x ∼ 1) (Mie Scattering, Rayleigh-Debye-Gans theory) very large (x ≫ 1) (Fraunhofer diffraction)

Is ∝ d4 Is ∝ d2

turbidity (τ) turbidity (τ) depends on d3 τ increases linearly with d τ decreases with increase in d

τ=

32π 2η2⟨Δη⟩2 ΔV 3λ 4

(12)

Density fluctuations in pure liquid and concentration fluctuations in binary mixtures are related to the turbidity of the solution and given by Einstein−Smoluchowski equation: τ=

2 ⎧ ⎫ ⎪ ⎛ dn ⎞2 4π 2 ⎪⎛ dn ⎞ c ⎜n ⎟ kTV ⎨ ⎬ kT ρ χ + ⎜ ⎟ ⎝ dc ⎠ ( −dπosm/dc) ⎪ λ4 ⎪ ⎩⎝ dρ ⎠ ⎭

(13)

τ is the turbidity, λ is the wavelength of the light, ρ is the solvent density, χ is the isothermal compressibility, η is the refractive index, πosm is the osmotic pressure, V is the molar volume, and c is the concentration. Isothermal compressibility (χ) is related to the density fluctuations in pure liquids and is defined as the change in volume as a response to change in pressure of the system.18 Osmotic pressure is related to the concentration fluctuations, and as solute−solute interactions increase, osmotic pressure decreases. Therefore, a protein system with significant intermolecular interactions results in opalescent appearance. Every protein system has some inherent opalescence associated with it (for most cases which cannot be detected by visual observation) and may become evident at higher concentrations. 2.3.3. Critical Opalescence. An infinite increase in fluctuations close to the critical point results in increased scattering from the solution termed as critical opalescence.18,40,74−76 Scattering of light due to fluctuations in solution is given by Einstein eq (eq 13); however, Einstein assumed that the fluctuations in solution were independent or there are no correlations between these fluctuations. In that case, in the vicinity of critical point, fluctuations would be infinite (1/χ and dπosm/dc are zero from pressure−volume and temperature−concentrations phase diagram, respectively), and hence the intensity of scattered light would be infinite. But in real solutions, the intensity of scattering is finite. Ornstein− Zernike accounted for finite correlation between the fluctuations in the volume element ΔV characterized by a correlation length (ξ).44 By definition, correlation length is the spatial extent of correlations between the fluctuations and is given by

wavelength (λ) λ−4 λ−2 independent of λ

a

d: diameter of the particle. bx: dimensionless optical size parameter, x = πd/λ.

Rayleigh theory, the intensity of scattered light is inversely proportional to the fourth power of wavelength (Is ∝ λ−4), and hence, shorter wavelengths scatter more light, resulting in bluish appearance of the solution. As the size of the particles increases above that defined by the Rayleigh theory, but smaller than the wavelength of light, scattering intensity exhibits an inverse dependence on the second power (Is ∝ λ−2) of the wavelength of the incident light. As the size of the particles increases and becomes larger than the wavelength of the light, intensity is independent of the wavelength; hence, solution appears white (turbid). 2.3.2. Scattering due to Fluctuations in Solution. There are constant density and concentration fluctuations in solution due to thermal (Brownian) motions. These density (Δρ) and concentration fluctuations (Δc) result in fluctuations of refractive index ⟨Δη⟩2 (represented as eq 10).69

⎛ T − Tc ⎞−ν ξ∝⎜ ⎟ ⎝ Tc ⎠

(14)

where T is any temperature, Tc is the critical temperature, and ν is the critical exponent. Figure 3 represents the schematic for correlation length of fluctuations in solution as temperature approaches critical temperature. At some temperature T ≫ Tc, fluctuations are correlated in small volume elements as observed in Figure 3a. As T approaches Tc, correlation between fluctuations increases (Figure 3b), and at Tc, ξ diverges; i.e., fluctuations become correlated over the entire volume element (Figure 3c). This correlation length is related to the intensity of scattered light by a correlation factor, and eq 11 is modified as

2 ⎛ ∂η ⎛ ∂η ⎞2 ⎛ ∂η ⎞2 ∂η ⎞ ⟨Δη⟩2 = ⎜ Δρ + Δc ⎟ = ⎜ ⎟ ⟨Δρ⟩2 + ⎜ ⎟ ⎝ ∂c ⎠ ∂c ⎠ ⎝ ∂ρ ⎝ ∂ρ ⎠

⟨Δc⟩2

(11)

where η is the refractive index of the medium and is related to polarizability (α). Equation 10 can also be expressed in terms of turbidity (τ) of the solution as

Table 3. Effect of Particle Size and Wavelength Dependence for Light Scattering from Solution size

4π 2ΔVη2⟨Δη⟩2 (1 + cos2 θ ) r 2λ 4

(10)

Scattering of light in solution occurs due to fluctuations of refractive index ⟨Δη⟩2 in the volume element, ΔV, and is represented as follows (modified eq 9, Rayleigh theory): E

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics

correlation length between the density and concentration fluctuations in the solution (Section 3.3). 3.2. Phase SeparationConstructing a Phase Diagram. Two techniques are generally employed to study phase separation in solution: (1) temperature quenching and centrifugation method and (2) cloud-point measurement.55,79 In temperature quenching method, samples are placed in tubes and then centrifuged at certain speed and time. If there is a phase separation in the sample, concentration is determined in protein-rich and protein-poor phase by UV-spectrophotometer. Sample tubes are then gently inverted and centrifuged again at lower temperatures. The phase diagram is constructed by plotting temperatures against the concentrations measured in the two phases. In cloud-point measurement, the sample is placed in a temperature-controlled water bath/chamber and turbidity of the sample is measured as temperature is lowered (usually by 1 °C step-size). The onset of the liquid−liquid phase separation is characterized by a dramatic increase in solution turbidity; this temperature is marked as Topacity. The temperature is then increased stepwise (0.1 °C) until solution becomes clear, and this temperature is marked as Tclear. Average temperature between Topacity and Tclear is the Tcloud temperature.80 Cloudpoint measurements can also be performed using microscope fitted with a temperature controller. On lowering the temperature, the onset of phase separation can be visually assessed by darkening of the field.78 As described previously (Section 2.1.1), spinodal curve marks the separation of metastable and unstable system on the phase diagram. The temperatures on the spinodal curve (Tsp) are determined by measuring the scattering intensity (Is) at 90° angle for varying concentrations and then extrapolating the inverse of this intensity to zero on the plot of 1/Is vs Tsp.79,80 The coexistence curve and spinodal curve are then fitted to eq 1 to determine the critical point. 3.3. Characterizing Critical Opalescence. Several studies are reported in the literature where LLPS for globular proteins is related to critical opalescence in the solution.79−81 Though there are speculations for increased solution opalescence for protein as critical opalescence,10 there are very few studies where divergence of the thermodynamic properties close to the critical point are actually quantified.82 Earliest studies on liquid−liquid phase separation in lysozyme−salt solution using light scattering methods were performed by Ishimoto and Tanaka, where they demonstrated the existence of a critical point for protein in binary solutions. They characterized the asymptomatic behavior of fluctuations close to the critical point, confirming critical opalescence for lysozyme solution. Correlation length (ξ) of the fluctuations in solution was determined using mean-field approximation.80 In another study, authors observed that, for γ-crystallins, the scattering at coexistence temperatures and spinodal temperatures are similar (and converge) in the vicinity of the critical point.79 Detailed experimentation and relationships to determine osmotic isothermal compressibility (χ) and correlation length is described in the literature.81

Figure 3. Correlations between fluctuations of thermodynamic quantity in a system (density fluctuations in pure liquids or concentration fluctuations in binary systems) in terms of correlation length ξ at (a) temperature above critical temperature, (b) temperature approaching critical temperature, and (c) at critical temperature.44 2 ⎞ Is 4π 2 ⎛ dn ⎞ ⎛ kTc 1 = 4 2 ⎜η ⎟ ⎜ ⎟ I0 λ r ⎝ dc ⎠ ⎝ ( −dπosm/dc) ⎠ 1 + ξ 4π sin θ λ 2

(

× (1 + cos2 θ )

2

)

(15)

Near critical point, scattering intensity due to concentration fluctuations is much larger than density fluctuations.77 As T → Tc, scattering intensity depends on λ−2 as compared to λ‑4, in simple/regular solutions; hence solution appears opalescent in the vicinity of the critical point.76 Solution exhibits intense turbidity close to the critical point as the two phases with different concentration and density coexist in solution.

3. CHARACTERIZING OPALESCENCE AND PHASE SEPARATION IN PROTEIN SOLUTIONS An overview of the commonly used techniques/methods to measure opalescence and characterize liquid−liquid phase separation and critical opalescence for protein solutions is discussed in this section. 3.1. Opalescence Measurements. Opalescence in solution depends on the size and concentration of the particles, and it is routinely measured using turbidity meter, nephelometer, or spectrophotometer. Nephelometric turbidity units (NTUs), Formazin turbidity units (FTUs), absorbance (optical density), or percent transmittance are the commonly used units representing opalescence in solution. Opalescence or turbidity of the solution is measured at a fixed wavelength, usually between 350 and 650 nm, where protein is nonabsorbing34 and it can be expressed in standard NTU’s by using appropriate calibration standards.10,13 Opalescence and LLPS can also be visualized by commonly used phase contrast or polarized light microscope.13,56,78 LLPS and critical opalescence are further characterized by measuring opalescence as a function of temperature (to determine Tc/Tcloud) and determining the

4. PROTEIN−PROTEIN INTERACTIONS AND LLPS As discussed previously, both enthalpy (solute−solute interactions) and entropy (lower temperatures) promote liquid− liquid phase separation in a binary system. From a pharmaceutical perspective, attractive protein−protein interactions (PPI) between the protein molecules result in phase F

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics

proteins; however, recent publication reports indicate inefficacies of these models to explain complex phase behaviors of the protein systems (as restated by Prausnitz).98,99A patchy colloid system is a better model for mAbs as it closely mimics anisotropic protein surfaces (as opposed to the point charges).100−102 Though interactions in the protein solutions is a well-researched area;4,70,73,103,104 with the advent of newer and more complex molecules, an in-depth understanding of the contributions of the protein−protein interactions to the solution phase behavior remains a challenge. 4.2. Second Virial Coefficient (B2). Second virial coefficient (B2) is routinely used to characterize protein− protein interactions in formulation development. B2 can be measured using various techniques including membrane osmometry, static light scattering, self-interaction chromatography, and so forth. Virial expansion for osmotic pressure in protein solution is presented as,

separation as the temperature of the system is lowered. Intermolecular interactions in protein solution as assessed by B2 (second virial coefficient) and Tcloud/Tc (cloud temperature/ critical temperature) and the challenges associated with it are discussed in this section. 4.1. Nature of Attractive PPI in Protein Solution. Protein−protein interactions play a significant role in opalescence and phase separation in solutions and determining the nature of these interactions holds the key for the selection of appropriate solution conditions (pH, ionic strength, salt type, etc.) and excipients to formulate a physically stable product. Molecular properties and solution conditions that strongly influence protein−protein interactions are discussed below. 4.1.1. Charges. Though the presence of charges on proteins results in repulsive intermolecular interactions, as the interseparation distance decreases and protein molecules approach each other, charges on one protein molecule may interact with oppositely charged dipole/induced-dipole on another protein molecule resulting in attractive interactions in solution. 4.1.2. Dipolar Interactions. At high protein concentration, van der Waals interaction including dipole−dipole, dipole− induced dipole, and induced dipole−induced dipole, are the dominant forces in solution. Dipole-mediated interactions in proteins are further enhanced for asymmetrical and larger molecules like mAb (and mAb-like molecules) due to their geometry and orientation/asymmetric charge distribution on the protein surface.83 Dipolar interactions are dominant near the pI of the molecule, while charge-dipole interactions dominate at pH away from the pI (where protein carries net charge). The mean potential force due to attractive charge and dipole interactions decreases on increasing the ionic strength of the solution.84,85 4.1.3. Hydrophobic Interactions. The presence of hydrophobic amino acids/patches on protein surface results in attractive hydrophobic interactions in solution which are shortrange in nature (i.e, proximity energy increases as protein molecules approach each other at high concentrations). Low ionic strengths have no effect on these interactions; however, high ionic strength increases hydrophobic interactions. Any excipient that interacts with the surface hydrophobic amino acids reduces these attractive interactions in solution.86−89 4.1.4. Specific Interactions. The presence of certain amino acids on the protein surface results in specific interactions between the protein molecules.90 These attractive interactions decrease on addition of amino acids and its derivatives or other excipients that bind specifically with amino acids on protein surface.89,91−93 Proteins are flexible molecules and exhibit anisotropy (asymmetric charge distribution) with respect to surface charges and surface roughness. This inherent nature of the protein impacts the nature of intermolecular interactions in the solution. Depending on the solution conditions and dynamics in the solution (spatial configuration and range of interactions) protein systems exhibit complex phase behavior and can exist as crystals, dense phases, equilibrium clusters/arrested metastable state, or nonequilibrium gels.94−97 For understanding the phase behavior of protein and colloidal system, it is important to understand that the effective attractive potential is short-ranged compared to the size of the particle.95 Intermolecular interaction studied using colloid theory (DLVO, Hard-sphere model) have been successfully applied to many globular

⎛ 1 ⎞ π = RTc ⎜ + 2B2 c + 3B3c 2 + 4B4 c 3 + ...⎟ ⎝ Mw ⎠

(16)

where π is osmotic pressure, R is the universal gas constant, T is the absolute temperature, c is the solute concentration, and Mw is the average molecular weight. At infinite dilution, higher order terms are neglected, and eq 16 reduces to the van’t Hoff equation for ideal solution. B2 is the first measure of deviation from ideality in the solution and characterizes interactions in solutions. The value of B2 reflects the magnitude of deviation from the ideality, while its sign reflects the nature of this deviation. A positive value corresponds to net repulsive interactions between the solute molecules wherein the osmotic pressure increases above that for an ideal solution whereas a negative value corresponds to net attractive interactions between the solute molecules with a consequent decrease in solution osmotic pressure below that for an ideal solution.105 Good correlation has been established between B2 and protein solubility,106,107 crystallization,51,108,109 and protein precipitation,110 all of which characterize phase separation in solution.107,111 Crystallization and its relation to protein−protein interactions in solution has been thoroughly studied and reported in the literature (as reviewed by Wilson et al.).112 LLPS is metastable with respect to crystal formation and hence, similar to solid−liquid phase transition, attractive interactions result in liquid−liquid phase separation in solution. Several studies in the literature refer to a “crystallization slot”, i.e., a range of B2 values which allows for the formation and growth of crystals in solution. Crystallization occurs only when B2 values are between −1 and −8 × 10−4 mol·mL/g2; high B2 indicating strong attractive interactions would result in the formation of amorphous precipitate, while low B2 indicating weak interactions would prevent crystal formation.113 These B2 values are in good agreement with crystal formation for globular proteins; however, as the complexity of the molecule increases as observed for mAb, the crystallization slot may not be an ideal predictor of crystallization or LLPS in solution. There are several examples in the literature for crystallization in solution, where B2 values are outside the crystallization slot.114−116 On a similar note, Rakel et al. have reported inconsistencies in predicting the phase behavior for mAbs with B2 measurements as the parameter (B2 indicates interactions only) does not take into account kinetic and concentration dependence of the protein phase behavior.117 G

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics 4.3. Critical Temperature Tc, Tcloud. Critical temperature is the temperature at the critical concentration, i.e., concentration indicating maximum opalescence, below which the system is unstable, while Tcloud marks the onset of liquid− liquid phase separation in solution at any concentration on the coexistence curve. For a system exhibiting attractive interactions (positive enthalpy), the change in temperature modulates the entropy and hence the free energy of the system. On similar lines, the change in temperature of onset of LLPS indicates increased or decreased attractive interactions in solution. B2 or the second virial coefficient is measured under dilute solution conditions; however, LLPS occurs at relatively higher concentrations, and hence Tcloud or Tc, which can be measured at high concentrations, can be a better indicator of physical instabilities due to attractions at higher concentrations. Tcloud can be determined by any of the opalescence measurement techniques mentioned earlier (Section 3). Taratuta and co-workers established the coexistence curve for lysozyme solution at various pH conditions and demonstrated that the coexistence curve shifts parallel with temperature as a function of solution conditions. Shape and width of the coexistence curve as well as critical concentration Cc (230 ± 10 mg/mL) remains fairly constant. As a result, they optimized Tcloud as a parameter to asses LLPS in lysozyme solution with change in pH, ionic strength and salt type, details of which are provided in the next section. The shift in Tcloud with solution conditions were modeled to thermodynamic Gibbs free energy with contribution from attractive interactions between the hard spheres. Authors suggested that Cc is similar for similar sized molecules while Tc is a physical property that changes with the solution conditions.55 Similar observations were reported by Broide et al., where concentration (Cc ∼ 289 ± 20 mg/mL) and width of the coexistence curve (A ∼ 2.6 ± 0.1) remains same for different γ-crystallins, while Tc changes.56 In another study, authors investigated Tcloud for lysozyme as a function of different salt ions and correlated it to the strength of the protein interactions using DLVO theory model for colloids.118 Similar studies are reported in the literature for globular proteins and monoclonal antibodies, where critical temperature or Tcloud is measured as a function of solution conditions to assess attractive interactions in solution (elaborated in the next section).

disrupting hydrophobic interactions between the protein molecules. It was also concluded from their studies that the ionic strength rather than the specific ions from Hofmeister series influences the opalescence in solution.11 On the contrary, Woods et al. concluded that at low salt concentration different salts affect opalescence differently; citrate and succinate buffers reduce opalescence, while acetate increases opalescence. They rank-ordered the effect of different buffer species on reducing opalescence by their ability to interact with hydrophobic regions and preventing mAb self-association.30 Though LLPS was not observed in any of the above studies, opalescence in solution increased on lowering the temperature (Wang and coworkers did not study the temperature effect) and was reversible with changing solution conditions. 5.2. Effect of Excipients on LLPS in Protein Solution. 5.2.1. pH and Ionic Strength. Maintaining a constant pH and ionic strength is necessary not only from physiological point of view, but for proper storage of the formulation and maintaining its stability and solubility throughout the shelf life. Most studies reported in the literature indicate increased tendency of a protein to undergo LLPS close to the pI of the molecule. Taratuta and co-workers observed Tcloud shifts to higher temperature for lysozyme solution as pH was adjusted close to the pI (∼11.2) of the protein.55 Similar observation was reported for a certain mAb (IgG2), where solution opalescence was maximum close to the pI (pH ∼ 7.1) and decreased on increasing the ionic strength, while at pH away from pI, opalescence increased at high ionic strength.13 This change in opalescence with changing pH and ionic strength correlated well with change in critical temperature at a constant critical concentration (Cc ∼ 90 mg/mL).119 Nishi et al. reported increased opalescence, due to LLPS, close to the pI (pH ∼ 6.5) of the molecule for another mAb (IgG1) at low ionic strength conditions, which reversed on increasing the ionic strength of the solution.12 This pH dependence of opalescence and phase separation is similar to the solubility of the protein, where the solubility increases at pH conditions away from pI and is minimal at the pI. At low ionic strength, strong attractive interactions are present close to the pI of the molecule and can be attributed to dipoles and multipoles on the molecule, which are shielded on increasing the ionic strength of the solution. At pH conditions away from the pI, molecules carry a net charge and hence repel each other resulting in increased solubility (salting-in) and lower tendency to phase separate. However, as ionic strength is increased, charges on the molecules are shielded, and there may be attractions between the protein molecules due to preferential exclusion of salts resulting in reduced solubility (salting-out). Broide et al. correlated the solubility of lysozyme with Tcloud measurements as a function of ionic strength and observed that point of minimal solubility correlated to the maximum Tcloud in the solution.118 Though, pH and ionic strength result in increased or decreased tendency of protein to undergo LLPS, this behavior cannot be generalized as it also depends on the nature/identity of the salt-ion. For lysozyme protein, Taratuta and co-workers observed that the effect of anionic species is more prominent than cationic species as protein carries a net positive charge at conditions evaluated in the study.55 While in another study, it was determined that both cationic and anionic salt species have a varying effect on Tcloud of lysozyme depending on the pH and ionic strength.120 Broide et al. reported the effect of two anions, Cl− and Br−, on Tcloud, which reversed with ionic strength;

5. FACTORS AFFECTING OPALESCENCE AND LLPS In this section, the effect of formulation factors including pH, ionic strength, salt types, and other excipients on LLPS in protein solutions are discussed with relevant examples from the literature. These factors modulate the PPI in solution, thereby increasing or decreasing opalescence and tendency to undergo LLPS. 5.1. Effect of Excipients on Opalescence in mAb Solutions. Formation of irreversible aggregates and LLPS in solution results in increased opalescence; however, there are few studies where mAb solution is opalescent and neither aggregation nor LLPS is observed in the solution. Salinas et al. reported increased opalescence at high ionic strength (due to attractive interactions as determined from B2) which decreased on reducing the ionic strength of the solution.10 Similar observations were reported by Wang et al., where solution was clear at low ionic strength and opalescence increased on addition of salt. They attributed an increased opalescence to hydrophobic interactions, which becomes dominant on charge shielding; consequently, Tween 80 reduces opalescence by H

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics below 1 M ionic strength, Tcloud value for Br− is larger than Tcloud for Cl−, while between ionic strength of 1 and 2 M, Tcloud for Cl− is larger than that for Br−.118 Mason and co-workers performed extensive studies on the effect of different salts from Hofmeister series on LLPS in mAb solution. It was observed that the Tc decreases with increasing ionic strength and anions have a nonmonotonic (no particular trend observed) influence on LLPS in solution at pH conditions away from the pI.119 Salts and buffer species exert their effects mostly by electrostatic interactions which can be explained by DLVO theory. This is supported by several studies in the literature that discuss the differential hydration of the protein molecules in the presence of different salt ions as the cause for change in Tcloud.118,120−122 The exact mechanism by which salt-species affect salting-in vs salting-out behavior is still under investigation, and the discussion is beyond the scope of this review. 5.2.2. Polyethylene Glycol (PEG). Polyethylene glycol is a known macromolecular crowder and a precipitating agent for proteins.123,124 It is frequently used in purification of proteins and production of protein crystals in the solution. PEG precipitation has also been reported as a method for solubility screening of proteins in solution.125−127 PEG exerts its effect by depletion force or excluded volume effect decreasing the solvent accessibility for proteins, thereby increasing protein− protein interactions in solution. This results in the increased tendency of proteins to precipitate out of the solution either as solid phase or liquid phase. Liquid−liquid coexistence curve shifts to higher or lower temperatures with change in pH or ionic strength of the solution; similarly, increasing concentration as well as molecular weight of PEG raises Tc to higher temperatures. Extensive work has been carried out for the determination of colloidal stability by assessing shifts in the coexistence curve of protein solutions using PEG-induced LLPS.128−132 PEG raises the phase separation temperature of the protein and hence is extensively used to study the solubility of proteins as a function of solution conditions. Historically, PEG is known to exert the preferential exclusion effect by steric mechanism; however, it is reported in the literature that, in addition to the steric effects (entropic effects), PEGs also show weak (nonspecific) binding to the proteins (enthalpic effects).133−137 In a recent study, Samanta et al. investigated hydration dynamics around the HSA protein at different PEG concentrations. It was observed that at low PEG concentration steric effect dominates, while at higher PEG concentration interactions between protein and PEG hydration become evident.138 Though, PEG has been used as protein−precipitant for a long time, with the unfolding of new research the exact mechanism by which it exerts its effect in protein solution remains questionable. 5.2.3. Other Proteins. Wang et al. studied LLPS for a monoclonal antibody in the presence of human serum albumin (HSA) at physiological pH 7.4 to mimic blood serum conditions where HSA forms a major component. At pH 7.4, the two molecules carry opposite charges resulting in interaction of HSA (pI ∼ 5.7) with mAb (pI ∼ 8.8) and hence partitioning into protein-rich phase. Favorable interactions of mAb with HSA reduce attractive interactions/selfassociation between the mAb molecules and shift the coexistence curve to lower temperatures.139 5.2.4. Polyols. Though polyols are commonly used for reducing thermal aggregation in solution, its effect on phase separation are not well understood. There is only one study in

the literature, where increasing concentration of glycerol has been reported to decrease the phase separation temperature for lysozyme, which was hypothesized to be due to specific binding of glycerol with protein.78 5.2.5. Excipients with Specific Binding. From our recent studies, it was observed that sucrose and hydroxyl-propyl βcyclodextrin decreased phase separation for DVD-Ig by possible binding with specific amino acids on the protein surface. A significant decrease in both, Tcloud (indicating phase separation temperature) and kD (indicating protein−protein interactions) was observed in the presence of these excipients.29 5.3. Effect of Molecular Properties on LLPS. Other than the formulation factors, the inherent nature of the protein molecule including its size, shape, surface charge heterogeneity, amino acid sequence, hydrophobicity, and so forth influences interparticle interactions and hence its phase behavior in the solution. Broide et al. compared LLPS for four different γcrystallins which differed in their amino acid sequence. It was observed that the coexistence curve for different γ-crystallins can be divided into low Tc (5 °C) and high Tc (38 °C) groups; γ-crystallins with high Tc resulted in cataract at body temperature.56 In another study on LLPS in protein solutions,139 it was reported that globular proteins, crystallins (MW ∼ 20 kDa), and lysozyme (MW ∼ 14 kDa) have narrow and symmetrical coexistence curves, similar to that for a hard sphere, while mAb (MW ∼ 150 kDa) exhibits a wider and asymmetrical coexistence curve due to its nonspherical shape and increased flexibility. Critical concentrations for crystallins and lysozyme are 240 ± 10 mg/mL and 230 ± 10 mg/mL,55 respectively, while for the mAb the critical concentration is a range of 50−100 mg/mL.10,13,15,139 A recent publication on LLPS in bispecific antibody solution showed that depending on the solution conditions LLPS occurs in the concentration range of 10−100 mg/mL, indicating that, as the size and the complexity of the protein molecules increases, it shows a higher tendency to phase separate even at lower protein concentrations.28,29 These literature findings further confirm that, in addition to the solution conditions, inherent molecular properties also impact phase separation.

6. IMPLICATIONS OF LIQUID−LIQUID PHASE SEPARATION Liquid−liquid phase separation of globular proteins under physiological conditions results in certain diseases, and its physiological perspective have been reported in the literature. Some examples include cold cataracts,32,140 sickle-cell disease,141−143 neurodegenerative and amyloidogenic diseases,144 and cryoimmunoglobinemia.145 This section will briefly discuss the implications of liquid−liquid phase separation in biopharmaceutical formulations. 6.1. Aesthetic Appeal. Marketed protein formulations are generally stored at refrigerated conditions where they may exhibit slight turbidity or cloudiness in the solution. Although the cloudiness/haziness is reversible as the formulation is brought to the room temperature and may not be a serious issue for its physical stability, an opalescent appearance of the solution reduces its aesthetic appeal. It can also raise concerns about the quality of the product leading to patient incompliance. Mahler et al. have compiled the description/ definition of opalescence in solution as per European Pharmacopeia limits.34 Any solution above 3 FTUs is termed as opalescent and from a formulation point of view should be I

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics

also discussed with relevant literature examples. Finally, the review discusses the implications of LLPS in biopharmaceutical formulations. In summary, many factors contribute to opalescence and LLPS in protein solutions resulting in a constant challenge to develop a successful biopharmaceutical formulation, but at the same time providing a wider scope for further research.

investigated for possibility of the presence of aggregates or phase separation. 6.2. Physical Instability. Phase separation, both solid− liquid and liquid−liquid, indicates reduced physical stability of a protein solution. Though native structure of the protein is retained on phase separation, the overall integrity of the product is compromised. Also, the formation of solid phase or solid−liquid phase separation in solutions is associated with reduced solubility of proteins. The high concentration phase formed on LLPS can also promote formation of irreversible aggregates in solution which is of serious concern in formulation development. On phase separation, the chemical potential of the protein is same in both phases even though their concentrations are different.42,48 To account for the Donnan effect, salts in the solution would also partition in two phases according to the concentration gradient and maintain the chemical potential across the solution.146 This can result in pH shift/ionic strength difference in two phases and may further enhance physical instability.147 6.3. Concentrating Proteins in Solution. Phase separation is a concern for formulation development; however, under controlled conditions LLPS can be used to concentrate proteins as the formation of a protein-rich phase in the solution is spontaneous.59,148,149 High concentrations of monoclonal antibody formulation can be easily achieved on a large scale compared to other such techniques as ultrafiltration, drying, chromatography, and dialysis,6 which are time-consuming and increase production cost. Nishi and co-workers studied the properties of concentrated phase obtained on phase separation and reported no change in physical or chemical properties and binding ability of the antibody before phase separation.12 6.4. Crystallization in Solution. Obtaining high-quality crystals of proteins is one of the major challenges for crystallographers, as crystal formation occurs in very narrow range of conditions.150 Also, crystallization is a tedious process; nucleation and crystal growth may normally take weeks to occur and grow. LLPS is metastable with respect to protein crystallization, and hence for systems that exhibit LLPS, crystallization follows a two-step nucleation mechanism where formation of dense phase is followed by nucleation within the phase.149,151 The process of nucleation can be hastened in the dense phase reducing the time and energy for crystal growth. For any protein solutions, formulation conditions where crystal growth can occur can be easily optimized by studying LLPS for that system.152,153 Measuring B2 and Tcloud, that exhibits a good correlation with LLPS, can also be easily optimized for high throughput techniques resulting in better and efficient formulation development process.

■

AUTHOR INFORMATION

Corresponding Authors

*A.S.R.: 69 North Eagleville Rd., Unit 3092, Storrs, CT 06269. Phone: 347-331-8669. E-mail: [email protected]. *D.S.K.: 69 North Eagleville Rd., Unit 3092, Storrs, CT 06269. Phone: 860-486-3655. Fax: 860-486-2076. E-mail: kalonia@ uconn.edu. Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS Cc, critical concentration; Tc, critical temperature; Tcloud, cloudpoint temperature; Φ, order parameter; A, width of the coexistence curve (eq 1) and also represents factor that controls nucleation in solution as presented in eq 8; β, υ, critical exponent; μp, chemical potential of the protein; np, total number of protein molecules; nx, total number of solute (other components) molecules; Gmix, Gibbs free energy of mixing of the system; Smix, entropy of mixing; Hmix, enthalpy of mixing; σ, interfacial tension; ε, bulk energy; R, critical radius (eq 7); J, nucleation rate; Is, intensity of scattered light; I0, intensity of incident light; λ, wavelength of the light; α, polarizability of molecule; V, molar volume; θ, scattering angle; η, refractive index of the medium; τ, turbidity of solution; ξ, correlation length (between the fluctuations); χ, isothermal compressibility; πosm, osmotic pressure; B2, second virial coefficient

■

REFERENCES

(1) Elbakri, A.; Nelson, P. N.; Abu Odeh, R. O. The state of antibody therapy. Hum. Immunol. 2010, 71 (12), 1243−50. (2) Reichert, J. M.; Beck, A.; Lugovskoy, A. A.; Wurch, T.; Coats, S.; Brezski, R. J. 9th annual European Antibody Congress, November 11− 13, 2013, Geneva, Switzerland. mAbs 2014, 6 (2), 309−26. (3) Evans, J. B.; Syed, B. A. From the analyst’s couch: Nextgeneration antibodies. Nat. Rev. Drug Discovery 2014, 13 (6), 413−4. (4) Saluja, A.; Kalonia, D. S. Nature and consequences of proteinprotein interactions in high protein concentration solutions. Int. J. Pharm. 2008, 358 (1−2), 1−15. (5) Hall, D.; Minton, A. P. Macromolecular crowding: qualitative and semiquantitative successes, quantitative challenges. Biochim. Biophys. Acta, Proteins Proteomics 2003, 1649 (2), 127−39. (6) Shire, S. J.; Shahrokh, Z.; Liu, J. Challenges in the development of high protein concentration formulations. J. Pharm. Sci. 2004, 93 (6), 1390−402. (7) Daugherty, A. L.; Mrsny, R. J. Formulation and delivery issues for monoclonal antibody therapeutics. Adv. Drug Delivery Rev. 2006, 58 (5−6), 686−706. (8) Harris, R. J.; Shire, S. J.; Winter, C. Commercial Manufacturing Scale Formulation and Analytical Characterization of Therapeutic Recombinant Antibodies. Drug Dev. Res. 2004, 61, 137−154. (9) Sukumar, M.; Doyle, B. L.; Combs, J. L.; Pekar, A. H. Opalescent appearance of an IgG1 antibody at high concentrations and its relationship to noncovalent association. Pharm. Res. 2004, 21 (7), 1087−93. (10) Salinas, B. A.; Sathish, H. A.; Bishop, S. M.; Harn, N.; Carpenter, J. F.; Randolph, T. W. Understanding and modulating opalescence and

7. SUMMARY A thorough understanding of the opalescence and liquid−liquid phase separation in solution is important for formulation development of therapeutic proteins. This review addresses the basic theoretical difference between scattering of light due to aggregates in solution and due to liquid−liquid phase separation, an understanding of which may help in better optimization of solution conditions to reduce physical instabilities. The thermodynamic and kinetic basis for phase separation is different from that of aggregation and is briefly discussed in this review. The nature of protein−protein interactions and parameters to measure the same, effect of formulation factors and molecular properties affecting opalescence and liquid−liquid phase separation in solutions is J

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics viscosity in a monoclonal antibody formulation. J. Pharm. Sci. 2010, 99 (1), 82−93. (11) Wang, N. H. B.; Ionescu, R.; Mach, H.; Sweeney, J.; Hamm, C.; Kirchmeier, M. J.; Meyer, B. K. Opalescence of an IgG1 monoclonal antibody formulation is mediated by ionic strength and excipients. BioPharm. Int. 2009, 22 (4), 36−47. (12) Nishi, H.; Miyajima, M.; Nakagami, H.; Noda, M.; Uchiyama, S.; Fukui, K. Phase separation of an IgG1 antibody solution under a low ionic strength condition. Pharm. Res. 2010, 27 (7), 1348−60. (13) Mason, B. D.; Zhang, L.; Remmele, R. L., Jr.; Zhang, J. Opalescence of an IgG2 monoclonal antibody solution as it relates to liquid-liquid phase separation. J. Pharm. Sci. 2011, 100 (11), 4587−96. (14) Sahin, E.; Weiss, W. F. t.; Kroetsch, A. M.; King, K. R.; Kessler, R. K.; Das, T. K.; Roberts, C. J. Aggregation and pH-temperature phase behavior for aggregates of an IgG2 antibody. J. Pharm. Sci. 2012, 101 (5), 1678−87. (15) Raut, A. S.; Kalonia, D. S. Opalescence in monoclonal antibody solutions and its correlation with intermolecular interactions in dilute and concentrated solutions. J. Pharm. Sci. 2015, 104 (4), 1263−1274. (16) Wang, W.; Singh, S.; Zeng, D. L.; King, K.; Nema, S. Antibody structure, instability, and formulation. J. Pharm. Sci. 2007, 96 (1), 1− 26. (17) Saluja, A.; Badkar, A. V.; Zeng, D. L.; Nema, S.; Kalonia, D. S. Application of high-frequency rheology measurements for analyzing protein-protein interactions in high protein concentration solutions using a model monoclonal antibody (IgG2). J. Pharm. Sci. 2006, 95 (9), 1967−83. (18) Gopal, E. S. R. Critical opalescence. Resonance 2000, 5 (4), 37− 45. (19) Pearlman, R.; Bewley, T. Stability and Characterization of Human Growth Hormone. In Stability and Characterization of Protein and Peptide Drugs; Wang, Y. J., Pearlman, R., Eds.; Springer: New York, 1993; Vol. 5, pp 1−58. (20) Eckhardt, B. M.; Oeswein, J. Q.; Yeung, D. A.; Milby, T. D.; Bewley, T. A. A turbidimetric method to determine visual appearance of protein solutions. PDA J. Pharmaceut. Sci. Technol. 1994, 48 (2), 64−70. (21) Hsu, C. C.; Nguyen, H. M.; Yeung, D. A.; Brooks, D. A.; Koe, G. S.; Bewley, T. A.; Pearlman, R. Surface denaturation at solid-void interface–a possible pathway by which opalescent particulates form during the storage of lyophilized tissue-type plasminogen activator at high temperatures. Pharm. Res. 1995, 12 (1), 69−77. (22) Eberlein, G. A.; Stratton, P. R.; Wang, Y. J. Stability of rhbFGF as determined by UV spectroscopic measurements of turbidity. PDA J. Pharmaceut. Sci. Technol. 1994, 48 (5), 224−30. (23) Yu, C. M.; Chin, C. Y.; Franses, E. I.; Wang, N. H. In situ probing of insulin aggregation in chromatography effluents with spectroturbidimetry. J. Colloid Interface Sci. 2006, 299 (2), 733−9. (24) Mahler, H. C.; Muller, R.; Friess, W.; Delille, A.; Matheus, S. Induction and analysis of aggregates in a liquid IgG1-antibody formulation. Eur. J. Pharm. Biopharm. 2005, 59 (3), 407−17. (25) Kiese, S.; Papppenberger, A.; Friess, W.; Mahler, H. C. Shaken, not stirred: mechanical stress testing of an IgG1 antibody. J. Pharm. Sci. 2008, 97 (10), 4347−66. (26) Sharma, D. K.; Oma, P.; Pollo, M. J.; Sukumar, M. Quantification and characterization of subvisible proteinaceous particles in opalescent mAb formulations using micro-flow imaging. J. Pharm. Sci. 2010, 99 (6), 2628−42. (27) Kiese, S.; Pappenberger, A.; Friess, W.; Mahler, H.-C. Equilibrium studies of protein aggregates and homogeneous nucleation in protein formulation. J. Pharm. Sci. 2010, 99 (2), 632− 644. (28) Raut, A. S.; Kalonia, D. S. Liquid−Liquid Phase Separation in a Dual Variable Domain Immunoglobulin Protein Solution: Effect of Formulation Factors and Protein−Protein Interactions. Mol. Pharmaceutics 2015, 12 (9), 3261−3271. (29) Raut, A. S.; Kalonia, D. S. Effect of Excipients on Liquid-Liquid Phase Separation and Aggregation in Dual Variable Domain Immunoglobulin Protein Solutions. Mol. Pharmaceutics 2016, 13, 774.

(30) Woods, J. M.; Nesta, D. Formulation effects on opalescence of a high-concentration MAb. BioProcess Int. 2010, 8, 48−59. (31) Asherie, N. Protein crystallization and phase diagrams. Methods 2004, 34 (3), 266−72. (32) Pande, A.; Pande, J.; Asherie, N.; Lomakin, A.; Ogun, O.; King, J.; Benedek, G. B. Crystal cataracts: human genetic cataract caused by protein crystallization. Proc. Natl. Acad. Sci. U. S. A. 2001, 98 (11), 6116−20. (33) Roberts, C. J. Protein aggregation and its impact on product quality. Curr. Opin. Biotechnol. 2014, 30 (0), 211−217. (34) Mahler, H. C.; Friess, W.; Grauschopf, U.; Kiese, S. Protein aggregation: pathways, induction factors and analysis. J. Pharm. Sci. 2009, 98 (9), 2909−34. (35) Weiss, W. F.; Young, T. M.; Roberts, C. J. Principles, approaches, and challenges for predicting protein aggregation rates and shelf life. J. Pharm. Sci. 2009, 98 (4), 1246−1277. (36) Li, Y.; Roberts, C. J. Protein Aggregation Pathways, Kinetics, and Thermodynamics. In Aggregation of Therapeutic Proteins; John Wiley & Sons, Inc., 2010; pp 63−102. (37) Sharma, V. K.; Kalonia, D. S. Experimental Detection and Characterization of Protein Aggregates. In Aggregation of Therapeutic Proteins; John Wiley & Sons, Inc., 2010; pp 205−256. (38) Sanchez-Ruiz, J. M. Protein kinetic stability. Biophys. Chem. 2010, 148 (1−3), 1−15. (39) den Engelsman, J.; Garidel, P.; Smulders, R.; Koll, H.; Smith, B.; Bassarab, S.; Seidl, A.; Hainzl, O.; Jiskoot, W. Strategies for the assessment of protein aggregates in pharmaceutical biotech product development. Pharm. Res. 2011, 28 (4), 920−33. (40) Furth, R.; Williams, C. L. Proc. R. Soc. London, Ser. A 1954, 224, 104−119. (41) Rowlinson, J. Liquids and liquid mixtures; Butterworth Scientific, 1969; p 1. (42) Van Dijk, M. A.; Wakker, A. Concepts in Polymer Thermodynamics; CRC Press, 1998; Vol. 2. (43) Stanley, H. E. Introduction to Phase Transition and Critical Phenomena; Oxford University Press: New York, 1971. (44) Beysens, D.; Straub, J.; Turner, D. J. Phase Transition and Near Critical Phenomena; Springer: Berlin, 1987. (45) Utracki, L. A. Thermodynamics and Kinetics of Phase Separation. In Interpenetrating Polymer Networks; American Chemical Society, 1994; Vol. 239, pp 77−123. (46) Strobl, G. R.; Strobl, G. R. The physics of polymers; Springer, 1997; Vol. 2. (47) Young, R. J.; Lovell, P. A. Introduction to polymers; Chapman & Hall: London, 1991; Vol. 2. (48) Atkins, P.; De Paula, J. Atkins’ physical chemistry; Oxford University Press, 2014. (49) Zhang, W. Phase behavior and phase separation kinetics in polymer solutions under high pressure. Virginia Polytechnic Institute and State University, 2005. (50) Lipatov, I. S. Colloid chemistry of polymers; Elsevier, 1988. (51) Rosenbaum, D. F.; Zukoski, C. F. Protein interactions and crystallization. J. Cryst. Growth 1996, 169 (4), 752−758. (52) Piazza, R. Interactions and phase transitions in protein solutions. Curr. Opin. Colloid Interface Sci. 2000, 5 (1−2), 38−43. (53) Tisza, L. The The Theory of Critical Points. J. Phys. Colloid Chem. 1950, 54 (9), 1317−1323. (54) Thomson, J. A.; Schurtenberger, P.; Thurston, G. M.; Benedek, G. B. Binary liquid phase separation and critical phenomena in a protein/water solution. Proc. Natl. Acad. Sci. U. S. A. 1987, 84 (20), 7079−83. (55) Taratuta, V. G.; Holschbach, A.; Thurston, G. M.; Blankschtein, D.; Benedek, G. B. Liquid−liquid phase separation of aqueou lysozyme solutions: Effects of pH and salt identity. J. Phys. Chem. 1990, 94, 2140−2144. (56) Broide, M. L.; Berland, C. R.; Pande, J.; Ogun, O. O.; Benedek, G. B. Binary-liquid phase separation of lens protein solutions. Proc. Natl. Acad. Sci. U. S. A. 1991, 88 (13), 5660−4. K

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics

Effect of salt and role of concentration fluctuations. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 2003, 68 (1), 011904. (83) Yadav, S.; Liu, J.; Shire, S. J.; Kalonia, D. S. Specific interactions in high concentration antibody solutions resulting in high viscosity. J. Pharm. Sci. 2010, 99 (3), 1152−68. (84) Chari, R.; Jerath, K.; Badkar, A. V.; Kalonia, D. S. Long- and short-range electrostatic interactions affect the rheology of highly concentrated antibody solutions. Pharm. Res. 2009, 26 (12), 2607−18. (85) Lockhart, D. J.; Kim, P. S. Electrostatic screening of charge and dipole interactions with the helix backbone. Science 1993, 260 (5105), 198−202. (86) Du, W.; Klibanov, A. M. Hydrophobic salts markedly diminish viscosity of concentrated protein solutions. Biotechnol. Bioeng. 2011, 108 (3), 632−6. (87) Guo, Z.; Chen, A.; Nassar, R. A.; Helk, B.; Mueller, C.; Tang, Y.; Gupta, K.; Klibanov, A. M. Structure-activity relationship for hydrophobic salts as viscosity-lowering excipients for concentrated solutions of monoclonal antibodies. Pharm. Res. 2012, 29 (11), 3102− 9. (88) Kamerzell, T. J.; Pace, A. L.; Li, M.; Danilenko, D. M.; McDowell, M.; Gokarn, Y. R.; Wang, Y. J. Polar solvents decrease the viscosity of high concentration IgG1 solutions through hydrophobic solvation and interaction: Formulation and biocompatibility considerations. J. Pharm. Sci. 2013, 102 (4), 1182−1193. (89) He, F.; Woods, C. E.; Trilisky, E.; Bower, K. M.; Litowski, J. R.; Kerwin, B. A.; Becker, G. W.; Narhi, L. O.; Razinkov, V. I. Screening of monoclonal antibody formulations based on high-throughput thermostability and viscosity measurements: Design of experiment and statistical analysis. J. Pharm. Sci. 2011, 100 (4), 1330−1340. (90) Mahadevi, A. S.; Sastry, G. N. Cation-pi interaction: its role and relevance in chemistry, biology, and material science. Chem. Rev. 2013, 113 (3), 2100−38. (91) Falconer, R. J.; Chan, C.; Hughes, K.; Munro, T. P. Stabilization of a monoclonal antibody during purification and formulation by addition of basic amino acid excipients. J. Chem. Technol. Biotechnol. 2011, 86 (7), 942−948. (92) Bye, J.; Platts, L.; Falconer, R. Biopharmaceutical liquid formulation: a review of the science of protein stability and solubility in aqueous environments. Biotechnol. Lett. 2014, 36 (5), 869−875. (93) Inoue, N.; Takai, E.; Arakawa, T.; Shiraki, K. Specific Decrease in Solution Viscosity of Antibodies by Arginine for Therapeutic Formulations. Mol. Pharmaceutics 2014, 11 (6), 1889−1896. (94) Stradner, A.; Sedgwick, H.; Cardinaux, F.; Poon, W. C. K.; Egelhaaf, S. U.; Schurtenberger, P. Equilibrium cluster formation in concentrated protein solutions and colloids. Nature 2004, 432 (7016), 492−495. (95) Wolf, M.; Roosen-Runge, F.; Zhang, F.; Roth, R.; Skoda, M. W. A.; Jacobs, R. M. J.; Sztucki, M.; Schreiber, F. Effective interactions in protein−salt solutions approaching liquid−liquid phase separation. J. Mol. Liq. 2014, 200 (Part A), 20−27. (96) Zhang, F.; Roosen-Runge, F.; Sauter, A.; Roth, R.; Skoda, M. W. A.; Jacobs, R. M. J.; Sztucki, M.; Schreiber, F. The role of cluster formation and metastable liquid-liquid phase separation in protein crystallization. Faraday Discuss. 2012, 159 (0), 313−325. (97) Lewus, R. A.; Levy, N. E.; Lenhoff, A. M.; Sandler, S. I. A comparative study of monoclonal antibodies. 1. Phase behavior and protein-protein interactions. Biotechnology progress 2015, 31 (1), 268− 76. (98) Sarangapani, P. S.; Hudson, S. D.; Jones, R. L.; Douglas, J. F.; Pathak, J. A. Critical Examination of the Colloidal Particle Model of Globular Proteins. Biophys. J. 2015, 108 (3), 724−737. (99) Prausnitz, J. The Fallacy of Misplaced Concreteness. Biophys. J. 2015, 108 (3), 453−454. (100) Bianchi, E.; Blaak, R.; Likos, C. N. Patchy colloids: state of the art and perspectives. Phys. Chem. Chem. Phys. 2011, 13 (14), 6397− 6410. (101) Quang, L. J.; Sandler, S. I.; Lenhoff, A. M. Anisotropic Contributions to Protein−Protein Interactions. J. Chem. Theory Comput. 2014, 10 (2), 835−845.

(57) Manno, M.; Xiao, C.; Bulone, D.; Martorana, V.; San Biagio, P. L. Thermodynamic instability in supersaturated lysozyme solutions: effect of salt and role of concentration fluctuations. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 2003, 68 (1), 011904. (58) Wentzel, N.; Gunton, J. D. Liquid-liquid coexistence surface for lysozyme: role of salt type and salt concentration. J. Phys. Chem. B 2007, 111 (6), 1478−81. (59) Dumetz, A. C.; Chockla, A. M.; Kaler, E. W.; Lenhoff, A. M. Protein phase behavior in aqueous solutions: crystallization, liquidliquid phase separation, gels, and aggregates. Biophys. J. 2008, 94 (2), 570−83. (60) Vekilov, P. G.; Feeling-Taylor, A. R.; Petsev, D. N.; Galkin, O.; Nagel, R. L.; Hirsch, R. E. Intermolecular interactions, nucleation, and thermodynamics of crystallization of hemoglobin C. Biophys. J. 2002, 83 (2), 1147−56. (61) Galkin, O.; Chen, K.; Nagel, R. L.; Hirsch, R. E.; Vekilov, P. G. Liquid-liquid separation in solutions of normal and sickle cell hemoglobin. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (13), 8479−83. (62) Holyst, R. Solubility and mixing in fluids. arXiv preprint condmat/9603062, 1996. (63) Favvas, E.; Mitropoulos, A. C. What is spinodal decomposition. J. Eng. Sci. Technol. Rev. 2008, 1, 25−27. (64) Drenth, J. The Nucleation of Lysozyme from a Fluctuation Point of View. Cryst. Growth Des. 2005, 5 (3), 1125−1127. (65) De Yoreo, J. J.; Vekilov, P. G. Principles of crystal nucleation and growth; Mineral Society of America: Washington, DC, 2003. (66) Gunton, J. D. Homogeneous Nucleation. J. Stat. Phys. 1999, 95 (5−6), 903−923. (67) Debye, P. Light Scattering in Solutions. J. Appl. Phys. 1944, 15 (4), 338−342. (68) Morel, A. Optical properties of pure water and pure sea water. Optical aspects of oceanography; 1974, 1. (69) Oster, G. The Scattering of Light and its Applications to Chemistry. Chem. Rev. 1948, 43 (2), 319−365. (70) Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 3rd ed.; Marcel Dekker, Inc.: New York, 1997. (71) Berne, B. J.; Pecora, R. Dynamic light scattering: with applications to chemistry, biology, and physics; Courier Dover Publications, 2000. (72) Brown, W. Dynamic light scattering: The method and some applications; Oxford University Press: New York, 1993. (73) Hunter, R. J. Foundations of Colloid Science; Oxford University Press, 2001. (74) Debye, P.; Chu, B.; Kaufmann, H. Critical Opalescence of Binary Liquid Mixtures: MethanolCyclohexane; AnilineCyclohexane. J. Chem. Phys. 1962, 36 (12), 3378−3381. (75) Zimm, B. H. The Opalescence of a Two-Component Liquid System near the Critical Mixing Point. J. Phys. Colloid Chem. 1950, 54 (9), 1306−1317. (76) Stenland, C.; Pettitt, B. M. Binary-solution critical opalescence: mole fraction versus temperature phase diagram. J. Chem. Educ. 1995, 72 (6), 560. (77) Chu, B. Critical Opalescence of a Binary Liquid Mixture, nDecaneβ,β′-Dichloroethyl Ether. I. Light Scattering. J. Chem. Phys. 1964, 41 (1), 226−234. (78) Galkin, O.; Vekilov, P. G. Control of protein crystal nucleation around the metastable liquid−liquid phase boundary. Proc. Natl. Acad. Sci. U. S. A. 2000, 97 (12), 6277−6281. (79) Thomson, J. A.; Schurtenberger, P.; Thurston, G. M.; Benedek, G. B. Binary liquid phase separation and critical phenomena in a protein/water solution. Proc. Natl. Acad. Sci. U. S. A. 1987, 84 (20), 7079−7083. (80) Ishimoto, C.; Tanaka, T. Critical Behavior of a Binary Mixture of Protein and Salt Water. Phys. Rev. Lett. 1977, 39 (8), 474−477. (81) Schurtenberger, P.; Chamberlin, R. A.; Thurston, G. M.; Thomson, J. A.; Benedek, G. B. Observation of critical phenomena in a protein-water solution. Phys. Rev. Lett. 1989, 63, 2064. (82) Manno, M.; Xiao, C.; Bulone, D.; Martorana, V.; San Biagio, P. L. Thermodynamic instability in supersaturated lysozyme solutions: L

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics (102) Roberts, C. J.; Blanco, M. A. Role of Anisotropic Interactions for Proteins and Patchy Nanoparticles. J. Phys. Chem. B 2014, 118 (44), 12599−12611. (103) Laue, T. Proximity energies: a framework for understanding concentrated solutions. J. Mol. Recognit. 2012, 25 (3), 165−73. (104) Israelachvili, J. N. Intermolecular and Surface Forces, 3rd ed.; Academic Press: San Diego, 2011. (105) Neal, B. L.; Asthagiri, D.; Lenhoff, A. M. Molecular origins of osmotic second virial coefficients of proteins. Biophys. J. 1998, 75, 2469−2477. (106) Demoruelle, K.; Guo, B.; Kao, S.; McDonald, H. M.; Nikic, D. B.; Holman, S. C.; Wilson, W. W. Correlation between the osmotic second virial coefficient and solubility for equine serum albumin and ovalbumin. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2002, 58, 1544− 1548. (107) Valente, J. J.; Payne, R. W.; Manning, M. C.; Wilson, W. W.; Henry, C. S. Colloidal behavior of proteins: Effects of the second virial coefficient on solubility, crystallization and aggregation of proteins in aqueous solution. Curr. Pharm. Biotechnol. 2005, 6, 427−436. (108) George, A.; Chiang, Y.; Guo, B.; Arabshahi, A.; Cai, Z.; Wilson, W. W. Second virial coefficient as predictor in protein crystal growth. Methods Enzymol. 1997, 276, 100−110. (109) Neal, B. L.; Asthagiri, D.; Velev, O. D.; Lenhoff, A. M.; Kaler, E. W. Why is the osmotic second virial coefficient related to protein crystallization? J. Cryst. Growth 1999, 196, 377−387. (110) Curtis, R. A.; Prausnitz, J. M.; Blanch, H. W. Protein-protein and protein-salt interactions in aqueous protein solutions containing concentrated electrolytes. Biotechnol. Bioeng. 1998, 57 (1), 11−21. (111) Dumetz, A. C. Protein interactions and phase behavior in aqueous solutions: Effects of salt, polymer, and organic additives; ProQuest: 2007. (112) Wilson, W. W.; DeLucas, L. J. Applications of the second virial coefficient: protein crystallization and solubility. Acta Crystallogr., Sect. F: Struct. Biol. Commun. 2014, 70 (5), 543−554. (113) George, A.; Wilson, W. W. Predicting protein crystallization from a dilute solution property. Acta Crystallogr., Sect. D: Biol. Crystallogr. 1994, 50 (Pt 4), 361−365. (114) Bonnete, F.; Vivares, D. Interest of the normalized second virial coefficient and interaction potentials for crystallizing large macromolecules. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2002, 58 (10), 1571−1575. (115) Lewus, R. A.; Darcy, P. A.; Lenhoff, A. M.; Sandler, S. I. Interactions and phase behavior of a monoclonal antibody. Biotechnology progress 2011, 27 (1), 280−9. (116) Rakel, N.; Bauer, K. C.; Galm, L.; Hubbuch, J. From osmotic second virial coefficient (B22) to phase behavior of a monoclonal antibody. Biotechnology progress 2015, 31 (2), 438−451. (117) Rakel, N.; Galm, L.; Bauer, K. C.; Hubbuch, J. Influence of macromolecular precipitants on phase behavior of monoclonal antibodies. Biotechnology progress 2015, 31 (1), 145−153. (118) Broide, M.; Tominc, T.; Saxowsky, M. Using phase transitions to investigate the effect of salts on protein interactions. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1996, 53 (6), 6325− 6335. (119) Mason, B. D.; Zhang-van Enk, J.; Zhang, L.; Remmele, R. L., Jr.; Zhang, J. Liquid-liquid phase separation of a monoclonal antibody and nonmonotonic influence of Hofmeister anions. Biophys. J. 2010, 99 (11), 3792−800. (120) Grigsby, J. J.; Blanch, H. W.; Prausnitz, J. M. Cloud-point temperatures for lysozyme in electrolyte solutions: effect of salt type, salt concentration and pH. Biophys. Chem. 2001, 91 (3), 231−43. (121) Zhang, Y.; Cremer, P. S. The inverse and direct Hofmeister series for lysozyme. Proc. Natl. Acad. Sci. U. S. A. 2009, 106 (36), 15249−15253. (122) Dumetz, A. C.; Chockla, A. M.; Kaler, E. W.; Lenhoff, A. M. Effects of pH on protein−protein interactions and implications for protein phase behavior. Biochim. Biophys. Acta, Proteins Proteomics 2008, 1784 (4), 600−610.

(123) Ingham, K. C. Precipitation of proteins with polyethylene glycol. In Methods in Enzymology; Murray, P. D., Ed.; Academic Press, 1990; Vol. 182, pp 301−306. (124) Atha, D. H.; Ingham, K. C. Mechanism of precipitation of proteins by polyethylene glycols. Analysis in terms of excluded volume. J. Biol. Chem. 1981, 256 (23), 12108−17. (125) Gibson, T. J.; McCarty, K.; McFadyen, I. J.; Cash, E.; Dalmonte, P.; Hinds, K. D.; Dinerman, A. A.; Alvarez, J. C.; Volkin, D. B. Application of a high-throughput screening procedure with PEGinduced precipitation to compare relative protein solubility during formulation development with IgG1 monoclonal antibodies. J. Pharm. Sci. 2011, 100 (3), 1009−21. (126) Stevenson, C. L.; Hageman, M. J. Estimation of recombinant bovine somatotropin solubility by excluded-volume interaction with polyethylene glycols. Pharm. Res. 1995, 12 (11), 1671−6. (127) Li, L.; Kantor, A.; Warne, N. Application of a PEG precipitation method for solubility screening: a tool for developing high protein concentration formulations. Protein science: a publication of the Protein Society 2013, 22 (8), 1118−23. (128) Annunziata, O.; Asherie, N.; Lomakin, A.; Pande, J.; Ogun, O.; Benedek, G. B. Effect of polyethylene glycol on the liquid-liquid phase transition in aqueous protein solutions. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (22), 14165−70. (129) Wang, Y.; Annunziata, O. Comparison between proteinpolyethylene glycol (PEG) interactions and the effect of PEG on protein-protein interactions using the liquid-liquid phase transition. J. Phys. Chem. B 2007, 111 (5), 1222−30. (130) Wang, Y.; Lomakin, A.; Latypov, R. F.; Laubach, J. P.; Hideshima, T.; Richardson, P. G.; Munshi, N. C.; Anderson, K. C.; Benedek, G. B. Phase transitions in human IgG solutions. J. Chem. Phys. 2013, 139 (12), 121904. (131) Wang, Y.; Latypov, R. F.; Lomakin, A.; Meyer, J. A.; Kerwin, B. A.; Vunnum, S.; Benedek, G. B. Quantitative evaluation of colloidal stability of antibody solutions using PEG-induced liquid-liquid phase separation. Mol. Pharmaceutics 2014, 11 (5), 1391−402. (132) Jion, A. I.; Goh, L. T.; Oh, S. K. Crystallization of IgG1 by mapping its liquid-liquid phase separation curves. Biotechnol. Bioeng. 2006, 95 (5), 911−8. (133) Bloustine, J.; Virmani, T.; Thurston, G. M.; Fraden, S. Light Scattering and Phase Behavior of Lysozyme-Poly(Ethylene Glycol) Mixtures. Phys. Rev. Lett. 2006, 96 (8), 087803. (134) Knowles, D. B.; LaCroix, A. S.; Deines, N. F.; Shkel, I.; Record, M. T., Jr. Separation of preferential interaction and excluded volume effects on DNA duplex and hairpin stability. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (31), 12699−704. (135) Mukherjee, S. K.; Gautam, S.; Biswas, S.; Kundu, J.; Chowdhury, P. K. Do Macromolecular Crowding Agents Exert Only an Excluded Volume Effect? A Protein Solvation Study. J. Phys. Chem. B 2015, 119 (44), 14145−14156. (136) Sapir, L.; Harries, D. Is the depletion force entropic? Molecular crowding beyond steric interactions. Curr. Opin. Colloid Interface Sci. 2015, 20 (1), 3−10. (137) Shkel, I. A.; Knowles, D. B.; Record, M. T. Separating chemical and excluded volume interactions of polyethylene glycols with native proteins: Comparison with PEG effects on DNA helix formation. Biopolymers 2015, 103 (9), 517−527. (138) Samanta, N.; Luong, T. Q.; Das Mahanta, D.; Mitra, R. K.; Havenith, M. Effect of Short Chain Poly(ethylene glycol)s on the Hydration Structure and Dynamics around Human Serum Albumin. Langmuir 2016, 32 (3), 831−837. (139) Wang, Y.; Lomakin, A.; Latypov, R. F.; Benedek, G. B. Phase separation in solutions of monoclonal antibodies and the effect of human serum albumin. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (40), 16606−11. (140) Siezen, R. J.; Fisch, M. R.; Slingsby, C.; Benedek, G. B. Opacification of gamma-Crystallin solutions from calf lens in relation to cold cataract formation. Proc. Natl. Acad. Sci. U. S. A. 1985, 82 (6), 1701−5. M

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Review

Molecular Pharmaceutics (141) Eaton, W. A.; Hofrichter, J. Sickle cell hemoglobin polymerization. Adv. Protein Chem. 1990, 40, 63−279. (142) Chen, Q.; Vekilov, P. G.; Nagel, R. L.; Hirsch, R. E. Liquidliquid phase separation in hemoglobins: distinct aggregation mechanisms of the beta6 mutants. Biophys. J. 2004, 86 (3), 1702−12. (143) Galkin, O.; Chen, K.; Nagel, R. L.; Hirsch, R. E.; Vekilov, P. G. Liquid−liquid separation in solutions of normal and sickle cell hemoglobin. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (13), 8479−8483. (144) Koo, E. H.; Lansbury, P. T., Jr.; Kelly, J. W. Amyloid diseases: abnormal protein aggregation in neurodegeneration. Proc. Natl. Acad. Sci. U. S. A. 1999, 96 (18), 9989−90. (145) Yagi, H.; Takahashi, N.; Yamaguchi, Y.; Kato, K. Temperaturedependent isologous Fab-Fab interaction that mediates cryocrystallization of a monoclonal immunoglobulin G. Mol. Immunol. 2004, 41 (12), 1211−5. (146) Vis, M.; Peters, V. F.; Tromp, R. H.; Erne, B. H. Donnan potentials in aqueous phase-separated polymer mixtures. Langmuir 2014, 30 (20), 5755−62. (147) Bolton, G. R.; Boesch, A. W.; Basha, J.; Lacasse, D. P.; Kelley, B. D.; Acharya, H. Effect of protein and solution properties on the Donnan effect during the ultrafiltration of proteins. Biotechnology progress 2011, 27 (1), 140−52. (148) Johnson, H. R.; Lenhoff, A. M. Characterization and suitability of therapeutic antibody dense phases for subcutaneous delivery. Mol. Pharmaceutics 2013, 10 (10), 3582−91. (149) Trilisky, E.; Gillespie, R.; Osslund, T. D.; Vunnum, S. Crystallization and liquid-liquid phase separation of monoclonal antibodies and fc-fusion proteins: screening results. Biotechnology progress 2011, 27 (4), 1054−67. (150) Navarro, A.; Wu, H.-S.; Wang, S. S. Engineering problems in protein crystallization. Sep. Purif. Technol. 2009, 68 (2), 129−137. (151) Kuznetsov, Y. G.; Malkin, A. J.; McPherson, A. The liquid protein phase in crystallization: a case studyintact immunoglobulins. J. Cryst. Growth 2001, 232 (1−4), 30−39. (152) Tanaka, S.; Yamamoto, M.; Ito, K.; Hayakawa, R.; Ataka, M. Relation between the phase separation and the crystallization in protein solutions. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1997, 56 (1), R67−R69. (153) Astier, J.-P.; Veesler, S. Using Temperature To Crystallize Proteins: A Mini-Review†. Cryst. Growth Des. 2008, 8 (12), 4215− 4219.

N

DOI: 10.1021/acs.molpharmaceut.5b00937 Mol. Pharmaceutics XXXX, XXX, XXX−XXX