The unfolding of iRFP713 in a crowded milieu

Plasmids, mutagenesis, protein expression and purification

iRFP713 and its variants in the holo- and apoforms were expressed and purified as described previously (Stepanenko et al., 2014) (Supplemental Methods 1). SDS/PAGE in a 12% polyacrylamide gels was used to confirm that the purity of the target proteins was at least 95% (Laemmli, 1970). The concentrated protein was stored in 20 mM Tris/HCl buffer, 150 mM NaCl, pH 8.0. The measurements were carried out at a low protein concentration (absorbance was kept less than 0.1, which corresponds to a protein concentration of 0.14 mg/ml) in 20 mM Tris/HCl buffer, pH 8.0, 1 mM tris(2-carboxyethyl)phosphine (TCEP).

GdnHCl, GTC, TCEP, and crowding agents PEG-8000, Dextran-40 and Dextran-70 were purchased from Sigma (St. Louis, MO, USA). The concentration of GdnHCl and GTC in stock solutions was calculated by the refraction coefficient measured by the Abbe refractometer (LOMO, St. Petersburg, Russia).

Spectrophotometric Experiments

The measurements of absorption spectra were conducted using a U-3900H spectrophotometer (Hitachi, Tokyo, Japan) with microcells 101.016-QS 5 × 5 mm (Hellma, Jena, Germany). All the measurements were carried out at room temperature.

Fluorescence Spectroscopy

The fluorescence experiments were performed using a Cary Eclipse spectrofluorometer (Agilent Technologies, Mulgrave, Australia) with 10 × 10 × 4 mm cells (Starna, Atascadero, CA, USA, USA). The temperature at measurements was maintained at 23° C using a thermostat. Selective excitation of the tryptophan residues of a protein was performed at the long-wave absorption spectrum edge (λ_ex = 295 nm) where the absorption of the tyrosine residues is negligible. The parameter A, which characterizes the position and form of the fluorescence spectra, was calculated using the equation: (1)${\displaystyle A=k{I}_{320}∕I_{365}}$ where I ₃₂₀ and I ₃₆₅ are the fluorescence intensities at the emission wavelengths of 320 and 365 nm, respectively, and k is the coefficient taking into account the instrument sensitivity (Turoverov & Kuznetsova, 2003). The specific fluorescence of the BV chromophore of iRFP713 in the holoform was recorded at 713 nm (λ_ex = 690 nm). The correction of the recorded fluorescence intensity for the primary inner filter effect was made according to the approach proposed in (Kuznetsova et al., 2012; Fonin et al., 2014). The anisotropy of tryptophan fluorescence was determined as follows: (2)${\displaystyle r=\frac{\left(I_V^V-G{I}_H^V\right)}{\left(I_V^V+2G{I}_H^V\right)}}$ where $I_V^V$ and $I_H^V$ are vertical and horizontal components of the fluorescence intensity excited by vertically polarized light, respectively, and $G=I_V^H∕I_H^H$ is the coefficient that determines the different instrument sensitivity for the vertical and horizontal components of the fluorescence light, λ_em = 365 nm (Turoverov et al., 1998).

The steady-state dependences of different fluorescent characteristics of iRFP713 and its variants on the denaturant concentration were recorded 24 h after manually mixing of the native protein (50 µL aliquot) with a buffer solution containing the desired concentration of a denaturant and a crowding agent (500 µL). Control experiments revealed no noticeable changes in the detected characteristics with an increase in the equilibration time.

Circular dichroism measurements

Circular dichroism (CD) spectra were recorded using Jasco-810 spectropolarimeter (Jasco, Japan). The cells with path length of 1 mm and 10 mm were used for measurements of the far-UV CD spectra (in the range of 250–190 nm) and the near-UV (in the 320–250 nm range) / visible CD spectra (in the range from 810 to 320 nm), respectively. For every probe, the average of three scans of CD spectra was obtained and the signal of buffer solution was subtracted. The experimental data, recorded in the far-UV and near-UV region of the spectrum, were converted into units of the molar ellipticity per amino acid residue and into units of the molar ellipticity, respectively. The visible CD spectra are represented in units of the ellipticity.

Determination of volume fraction of crowders

The volume fraction, occupied by a crowding-agent, was calculated on the basis of the value of the partial specific volume. For dextran-40 and dextran-70, the partial specific volume of 0.623 and 0.56 ml/g, respectively, estimated in the works (Rotta et al., 2017; Senske et al., 2014) was used. The value of the partial specific volume for PEG-8000 of 0.84 ml/g was determined by approximating the density of the solution containing the crowder of different concentrations (Wandrey, Bartkowiak & Hunkeler, 1999); the data on the density of PEG-8000 was taken from the work (Gonzalez-Tello, Camacho & Blazquez, 1994) (Supplemental Methods 2).

The effect of crowding agents on the structure of iRFP713

We tested the effect of crowding agents Dextran-40 and Dextran-70 at a concentration of 240 mg/ml (Fig. 1) and PEG-8000 at a concentration of 80, 120 and 300 mg/ml (Fig. 2) on the structure of iRFP713 in the holoform (chromophore-bound) using absorption, fluorescence spectroscopy and circular dichroism.

Absorption spectra in the visible region, fluorescence spectra of the chromophore (Figs. 1A and 2A) and CD spectra in the visible region (Figs. 1B and 2B) recorded for iRFP713 in the holoform in solutions of Dextran-40, Dextran-70 and PEG-8000 practically coincided with the corresponding spectra of the holoprotein in the absence of a crowding agent. These data indicate that the microenvironment of the BV chromophore is preserved in the protein in the crowding conditions.

PEG-8000 at a concentration of 80–300 mg/ml and Dextran-40 and Dextran-70 at a concentration of 240 mg/ml did not affect CD spectra in the far-UV region of iRFP713 in the holoform (Figs. 1B and 2B). This testifies for the secondary structure of the protein being unchanged under these experimental conditions.

CD spectra in the near-UV region of iRFP713 in the holoform in solutions of crowding agents were almost undistinguishable from those of the protein in the absence of crowding agents (Figs. 1B and 2B). The shape of the intrinsic UV fluorescence spectra of iRFP713 in the holoform in the presence of Dextran-40, Dextran-70, PEG-8000 at a concentration of 80 and 120 mg/ml and in the absence of crowding agents coincided perfectly (Figs. 1A and 2A). The fluorescence spectrum of tryptophan residues of iRFP713 in the holoform in solutions of PEG-8000 at a concentration of 300 mg/ml deviated in the long-wavelength spectral region from the spectrum of the holoprotein in the absence of the crowder (Fig. 1A). In these experimental conditions, the value of parameter A of iRFP713 in the holoform was slightly higher than the value of parameter A of the protein in the absence of the crowding agent (Table 1). Since there observed no shift in the intrinsic UV fluorescence spectrum of iRFP713 in the holoform in the solution of PEG-8000 at high concentration relative to the position of the spectrum of the protein in the absence of the crowding agent, an increase in the parameter A of the protein under these conditions is hardly associated with any changes in the microenvironment of tryptophan residues. On the contrary, the small difference in the long-wavelength region of the tryptophan fluorescence spectra of iRFP713 in the holoform, recorded in the presence and in the absence of the crowding agent, is likely attributed to a high background contribution of PEG-8000 to the protein’s intrinsic fluorescence in this spectral region at increased crowder concentration. Thus, the tertiary structure of iRFP713 is preserved in crowded environment. The fact that the spatial structure of iRFP713 inherent to the native protein is unaltered under crowding conditions implies that the crowding agents used in this study do not interact with the protein. We previously demonstrated the lack of direct interaction between a number of proteins and several crowding agents, including PEG of different molecular mass, Dextran-70 and Ficoll-70 (Stepanenko et al., 2016b; Stepanenko et al., 2016c).

Unfolding of iRFP713 in the presence of PEG-8000

We studied the unfolding of iRFP713 in its holo- and apoform (BV-free) induced by GdnHCl in the presence of 80 and 120 mg/ml of PEG-8000 (Fig. 3) and induced by GTC in the presence of 300 mg/ml of PEG-8000 (Fig. 4).

In the presence of a moderate concentration of PEG-8000 (80 and 120 mg/ml), an increase in the intensity of tryptophan fluorescence of iRFP713 in the holoform was observed in the range of GdnHCl concentration from 0.7 to 1.2 M (Figs. 3A, 3B). Under these experimental conditions, the value of the parameter A of iRFP713 in holoform remained almost unchanged (Fig. 3E), and the fluorescence anisotropy of tryptophan residues of the holoprotein increased slightly (Fig. 3F). In the range of GdnHCl concentration from 0.7 to 1.5 M and from 0.5 to 1.8 M in solutions of PEG-8000 at a concentration of 80 mg/ml and 120 mg/ml, respectively, a noticeable decrease in the circular dichroism signal of iRFP713 in the holoform was detected (Fig. 3G). The effect is probably caused by protein aggregation. The aggregation of iRFP713 in the holoform in crowded environment mimicked by PEG-8000 at a concentration of 80 and 120 mg/ml at denaturant concentrations less than 2 M is confirmed by the shape of the recorded absorption spectra of the protein (Fig. 4). The dependencies of the absorbance at the maximum of the far-red absorption band and of the chromophore fluorescence intensity of iRFP713 in the holoform, recorded in solutions of PEG-8000 at a moderate concentration, coincided with the dependences of the protein in the absence of the crowding agent (Figs. 3C, 3D).

When examining the dependences of the tryptophan fluorescence intensity recorded at GdnHCl-induced unfolding of iRFP713 in the holoform in the absence of a crowding agent, we found two transitions at the pre-denaturing concentrations of the denaturant: from 0.6 to 1.0 M and from 1.6 to 2.8 M (Figs. 3A, 3B). Previously, we showed that the unfolding of iRFP713 in the apo- and holoform was accompanied by the formation of a compact monomeric state of the protein and an intermediate state of the protein, bearing the hydrophobic regions on its surface (Stepanenko et al., 2018). For careful characterization of iRFP713 in these states we analyzed three mutant variants of iRFP713 containing single tryptophan residue in different domain of the protein: iRFP713-W109 (iRFP713/W281F/W311F), iRFP713-W281 (iRFP713/W109F/W311F) and iRFP713-W311 (iRFP713/W109F/W281F).

The GdnHCl-induced unfolding of single-tryptophan mutant variants of iRFP713 in the holoforms were studied (Fig. 5). The dependences of the chromophore absorbance and fluorescence intensity, and ellipticity in the far-UV region of the spectrum of all single-tryptophan mutant variants of iRFP713 in the holoforms significantly overlapped with the corresponding dependences of the iRFP713 holoprotein (Figs. 5C, 5D). This evidences that amino acid substitutions introduced into the primary sequence of iRFP713 do not have marked effect on the stability of mutant proteins.

The change of tryptophan fluorescence intensity of iRFP713-W281 and iRFP713-W311 at GdnHCl-induced unfolding was identical to the change of the characteristic of the iRFP713 holoprotein (Figs. 5A, 5B). Indeed, in the range of pre-denaturing denaturant concentrations from 0.3–0.5 to 1.0 M and from 1.5 to 2.3–2.5 M an increase in the tryptophan fluorescence intensity of iRFP713-W281 and iRFP713-W311 was observed (Figs. 5A, 5B, green and blue curves). The rise in the tryptophan fluorescence intensity of iRFP713-W311 in the range of GdnHCl concentrations corresponding to the first transition was more pronounced compared to the change in the tryptophan fluorescence intensity of iRFP713-W281 and iRFP713 (Figs. 5A, 5B, green and blue curves). The tryptophan fluorescence intensity of iRFP713-W109 also increased in the range of GdnHCl concentrations from 0.5 to 1.0 M, then remained unchanged up to denaturant concentration of 2.5 M (Figs. 5A, 5B, red curves). In the range of GdnHCl concentrations from 0.2 to 1.0 M, the parameter A of the iRFP713-W311 in the holoform dropped from 1.2 to 0.8 (Fig. 5E, blue curve), indicating that the tryptophan residue becomes a more accessible to the solvent, most likely as a result of the protein dimer dissociation into monomers. The value of the parameter A of iRFP713-W109 and iRFP713-W281 in the range of pre-denaturing GdnHCl concentration descended slightly (Fig. 5E, red and green curves), pointing out some loosening of the structure around the tryptophan residues W109 and W281. The values of absorbance and fluorescence intensity of BV chromophore of all single-tryptophan mutants in the holoforms in the range of denaturant concentration from 0.5 to 1.0 M practically did not change (Figs. 5C, 5D). Under these experimental conditions, all single-tryptophan mutants of iRFP713 retained the values of ellipticity at 222 nm close to the values of the characteristic of the proteins, registered in the absence of a denaturant (Fig. 5G). Obviously, the structure of the chromophore-binding pocket of the GAF domain and the secondary structure of the protein are preserved in iRFP713 in the monomeric state. Taken together, these data are in line with the formation of the compact monomeric state of iRFP713 at the first denaturing transition in the range of low GdnHCl concentrations. The increased fluorescence quantum yield of the tryptophan residue W311 of iRFP713 in the compact monomeric state resulted from an increase in its accessibility to the solvent at the disruption of the native dimer iRFP713 (Fig. 5E, blue curves). The increased fluorescence quantum yield of the tryptophan residues W109 and W281 of iRFP713 in the compact monomeric state may be caused by a change in the distance and mutual orientation between these tryptophan residues and quenching groups in their vicinity due to a decrease in the density of their microenvironment.

The transition in the range of denaturant concentrations from 1.5 to 2.3–2.5 M, revealed by the increase in the tryptophan fluorescence intensity of iRFP713-W281 and iRFP713-W311 in the holoforms (Figs. 5A, 5B, green and blue curves), is associated with the formation of an intermediate state of the protein. At these denaturant concentrations, the values of absorbance and fluorescence intensity of the BV chromophore of all single-tryptophan variants of iRFP713 decreased to zero (Figs. 5C, 5D). This indicates that BV chromophore is no longer embedded in the GAF domain pocket of the iRFP713 in its intermediate state, meaning the disruption of the GAF domain structure at the formation of the intermediate state of iRFP713. A decrease in the ellipticity in the far-UV region of the spectrum of all single-tryptophan variants of iRFP713 in the range of GdnHCl concentrations from 1.5 to 2.3–2.5 M testifies the loss of the secondary structure of iRFP713 in its intermediate state (Fig. 5G). The increased fluorescence quantum yield of the tryptophan residues W281 and W311 of iRFP713 at the intermediate state is explained by the absence of the conditions for the effective nonradiative energy transfer from tryptophan residues of the protein to the chromophore (Stepanenko et al., 2014; Stepanenko et al., 2017). The fact that the formation of the intermediate state was not accompanied by any increase in fluorescence intensity of iRFP713-W109 may be connected with less efficient nonradiative energy transfer from this residue to the chromophore compared to the other tryptophan residues, as W109 is the most distant tryptophan residue from the pocket of the GAF domain of iRFP713 (Stepanenko et al., 2017). An increase in the parameter A of iRFP713-W311 in the holoform, which was observed in the range of GdnHCl concentrations from 1.5 to 2.3–2.5 M, is obviously related with the re-shielding of this tryptophan residue in the intermediate state of iRFP713 (Fig. 5E, blue curves). The latter implies the oligomerization of iRFP713 in its intermediate state.

Thus, an increase in the intensity of tryptophan fluorescence of iRFP713 in the holoform in the range of GdnHCl concentrations from 0.6 to 1.0 M in the absence of crowding agents and from 0.7 to 1.2 M in solutions of PEG-8000 at a concentration of 80 and 120 mg/ml is associated with the formation of the compact monomeric state of the protein. PEG-8000 at a moderate concentration induced no shift of the dependences of the tryptophan fluorescence intensity of iRFP713 in the holoform in the region of this transition (Figs. 3A, 3B). This indicates that at crowded environment created by PEG-8000 at a concentration of 80 and 120 mg/ml, the structure of the native dimer of iRFP713 in the holoform is not stabilized or destabilized.

In the region of the denaturation transition, the values of all recorded characteristics of iRFP713 in the holoform in the solutions with the moderate concentration of PEG-8000 were indistinguishable from the values of the corresponding characteristics of the protein in the absence of the crowder (Figs. 3A–3G). This evidences that PEG-8000 at a concentration of 80 and 120 mg/ml does not have a noticeable stabilizing or destabilizing effect on iRFP713 in the compact monomeric or intermediate state.

As we mentioned before the apparent aggregation of the iRFP713 holoprotein was observed in the presence of PEG-8000 at a moderate concentration in the pre-denaturing range of GdnHCl concentrations (Fig. 3G). The shape of the recorded absorption spectra of iRFP713 under these experimental conditions implies that the protein aggregates most strongly at 1.2 M GdnHCl (Fig. 4). We showed previously that at the concentration of GdnH⁺ of about 1.0 M the surface charge of iRFP713 in the intermediate state is neutralized, leading to the protein aggregation (Stepanenko et al., 2018). PEG-8000 at a moderate concentration did not affect the unfolding of iRFP713 in the holoform. The GdnHCl-induced unfolding of iRFP713 in the holoform was highly cooperative (Fig. 3D). At the same time, a monotonic decrease in the fluorescence intensity of the iRFP713 chromophore was registered in the range of GdnHCl concentrations from 0 to 2.3 M (Fig. 3D) indicative of the accumulation of a small amount of the holoprotein in its intermediate state at the treatment with small denaturant concentrations. We assume that this small fraction of iRFP713 in the intermediate state is responsible for observed protein aggregation in the presence of the crowder in the range of GdnH⁺ concentrations that are optimal for the neutralization of the charge of the protein surface. Crowded milieu is known to promote protein aggregation by the excluded volume effect (Minton, 2000; Ellis & Minton, 2006).

In the presence of PEG-8000 at a high concentration (300 mg/ml) the solubility of GdnHCl decreases that makes impossible to prepare a solution of GdnHCl with a concentration above 3M. We used a stronger denaturing agent, GTC, to unfold iRFP713 in the apo- and holoform in concentrated solutions of PEG-8000.

In the presence of PEG-8000 at a high concentration, the intensity of tryptophan fluorescence of iRFP713 in the apo- and holoform increased in the range of GTC concentrations from 0.3 to 0.5 M and from 0.35 M to 0.6 M, respectively, as a result of the accumulation of the compact monomeric state of the protein (Figs. 6A, 6B). We demonstrated previously that iRFP713 in the intermediate state in the presence of GTC was capable of forming large aggregates that precipitated, which was manifested as local minima on the dependences of tryptophan fluorescence intensity of apo- and hololoprotein at a denaturant concentration of 0.4 and 1.0 M, respectively. We did not find such local minima on the dependences of the intensity of tryptophan fluorescence in the case of GTC-induced unfolding of iRFP713 in the apoform in the presence of a high concentration of PEG-8000 (Figs. 6A, 6B). The local minimum, arising from the protein aggregation, on the dependences of the intensity of tryptophan fluorescence in the case of GTC-induced unfolding of the iRFP713 holoprotein under these experimental conditions was shifted to about 1.2–1.4 M GTC and was much less pronounced than the local minimum found on the dependences of tryptophan fluorescence intensity of the holoprotein in the absence of a crowder (Figs. 6A, 6B). The aggregation of iRFP713 at GTC-induced unfolding in concentrated solutions of PEG-8000 is proved by the shape of the absorption spectra of iRFP713 in the holoform recorded in the presence of the crowding agent in the range of GTC concentrations from 0.4 to 1.8 M, with the strongest protein aggregation being observed at 1.2 M GTC. The absence of pronounced minima on the dependences of the tryptophan fluorescence intensity of iRFP713 in the apo- and holoform in the presence of PEG-8000 at a high concentration can be explained by two reasons. First, the high viscosity of concentrated solutions of PEG-8000 can slow down the sedimentation of the protein aggregates. Secondly, these experimental conditions might be less favorable for aggregation of the protein compared to its unfolding in the absence of a crowder.

In the range of GTC concentrations from 1.0 to 1.3 M in the presence of PEG-8000 at a high concentration, the absorbance and fluorescence intensity of the BV chromophore of iRFP713 in the holoform dropped to zero. The shift of the denaturation transition in concentrated solutions of PEG-8000 to higher denaturant concentrations testifies to the stabilization of the compact protein monomeric state (Fig. 5C, 5D). Stabilization of iRFP713 in the monomeric compact state in concentration solutions of PEG-8000, which hydrodynamic dimensions are comparable to the size of the iRFP713 monomer, is consistent with the theory of excluded volume (Kuznetsova, Turoverov & Uversky, 2014; Tokuriki et al., 2004). As a result of the stabilization of the compact monomeric state of iRFP713 in concentrated solutions of PEG-8000, the formation of the aggregation-prone intermediate state of the protein occurred at higher concentrations of the denaturant. It further confirmed by the shift of the local minimum on dependences of the tryptophan fluorescence intensity of the iRFP713 holoprotein, recorded in the presence of the crowder at a concentration of 300 mg/ml, to higher GTC concentration (1.2–1.4 M) compared to that of the protein, unfolded in the absence of a crowder (Fig. 5A, 5B).

In the presence of PEG-8000 at a high concentration, the unfolding of iRFP713 in the holo- and apoform, accompanied by a decrease in the parameter A and the fluorescence anisotropy, occurred at higher GTC concentrations than in the absence of a crowder (Figs. 5E, 5F). This indicates the stabilization of the structure of iRFP713 in its intermediate state under these conditions.

Unfolding of iRFP713 in the presence of Dextran-40 and Dextran-70

In the presence of Dextran-70 at a concentration of 240 mg/ml, the cooperativity of the denaturing transition followed by the change in absorbance at the maximum of the far-red absorption band and the chromophore fluorescence intensity of iRFP713 in the holoform decreased (Figs. 7C, 7D). In the presence of Dextran-40 at a concentration of 240 mg/ml, this effect was even more pronounced (Figs. 8C, 8D). These data show that Dextran-70 and, particularly, Dextran-40 destabilize the structure of iRFP713 in the compact monomeric state.

In the presence of Dextran-70, the intensity of tryptophan fluorescence of iRFP713 in the holoform increased by about 40% in the range of GdnHCl concentrations from 2.0 to 2.8 M compared to the fluorescence intensity of the protein at zero denaturant concentration (Figs. 7A, 7B). Small changes in the tryptophan fluorescence intensity of iRFP713 in the holoform, recorded at the protein unfolding in the presence of Dextran-40, make their interpretation unfeasible (Figs. 8A, 8B). In the range of denaturant concentrations above 2.0 M, the absorbance and the fluorescence intensity of the BV chromophore of iRFP713 in the holoform dropped to zero in the presence of both crowding agents (Figs. 7C, 7D and 8C, 8D). Based on the above data, the denaturing transition at the GdnHCl concentrations from 2.0 to 2.8 M can be associated with the formation of the intermediate state of iRFP713. The change in the intensity of tryptophan fluorescence of the iRFP713 holoprotein occurring in the range of GdnHCl concentrations above 3.8 M in the presence of Dextran-70 is caused by the unfolding of the protein (Figs. 7A, 7B). In this range of GdnHCl concentrations, a significant decrease in the parameter A and the fluorescence anisotropy of iRFP713 in the holoform was detected in the presence of both crowders (Figs. 7E, 7F and 8E, 8F). The dependences of parameter A and fluorescence anisotropy, recorded at unfolding of iRFP713 in the apoform, were also shifted to higher concentrations of GdnHCl in the presence of both crowders compared to that of apoprotein in the crowder’s absence (Fig. 7E, 7F and 8E, 8F). This argues for the stabilization of the intermediate state of iRFP713 in the presence of Dextran-70 and Dextran-40.

In the range of pre-denaturing concentrations of GdnHCl, a noticeable decrease in molar ellipticity in the far-UV CD spectra of the iRFP713 holoform in solutions of both crowding agents was detected (Figs. 7G and 8G). Examination of the recorded absorption spectra of the iRFP713 in the holoform under these experimental conditions revealed that protein aggregated in a wide range of denaturant concentrations. Two ranges of GdnHCl concentrations were identified in which the iRFP713 in the holoform strongly aggregated in the presence of Dextran-70 and Dextran-40: about 1.0 M and about 2.6 M (Fig. 9). Probably, the aggregation of the iRFP713 holoprotein in the presence of Dextran-40 contributes to the character of the dependences of the tryptophan fluorescence intensity, recorded at these conditions (Figs. 8A, 8B). In the range of GdnHCl concentrations of about 1.0 M, the aggregation of iRFP713 in the holoform in crowded environment is facilitated by the conditions that are suitable for the neutralization of the surface charge of the protein in the intermediate state. In the range of GdnHCl concentrations of about 2.6 M, the aggregation of the iRFP713 holoprotein is related to an increase in the portion of the protein in the intermediate state. The unfolding of iRFP713 in the apoform in concentrated solutions of Dextran-70 and Dextran-40 resulted in even stronger aggregation of the apoprotein with respect to the holoprotein (Figs. 7G and 8G). Thus, the CD spectrum in the far UV region of the iRFP713 apoprotein in the presence of 1.3 M GdnHCl and of 240 mg/ml of Dextran-40 can not be detected (Fig. 8G). The enhanced aggregation of the apoprotein compared to the holoprotein in crowded environment is obviously determined by its lower stability.