Unusual effects of crowders on heme retention in myoglobin

FEBS Letters xxx (2015) xxx–xxx

journal homepage: www.FEBSLetters.org

Unusual effects of crowders on heme retention in myoglobin Jayanta Kundu, Uddipan Kar, Saurabh Gautam, Sandip Karmakar, Pramit K. Chowdhury ⇑ Department of Chemistry, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India

a r t i c l e

i n f o

Article history: Received 21 September 2015 Revised 5 November 2015 Accepted 10 November 2015 Available online xxxx Edited by Stuart Ferguson Keywords: Macromolecular crowder Glucose and sucrose Excluded volume effect Heme retention BSA Lysozyme Soft interaction

a b s t r a c t Myoglobin (Mb) undergoes pronounced heme loss under denaturing conditions wherein the proximal histidine gets protonated. Our data show that macromolecular crowding agents (both synthetic and protein based) can appreciably influence the extent of heme retention in Mb. Interestingly, glucose and sucrose, the monomeric constituents of dextran and ficoll-based crowders were much more effective in preventing heme dissociation of Mb, albeit, at much higher concentrations. The protein crowders BSA and lysozyme show very interesting results with BSA bringing about the maximum heme retention amongst all the crowding agents used while lysozyme induced heme dissociation even in the native state of Mb. The stark difference that these protein crowders exhibit when interacting with the heme protein is a testament to the varied interaction potentials that a test protein might be exposed to in the physiological (crowded) milieu. Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction The heme prosthetic group is one of the most diverse cofactors found in nature, catering to a wide range of essential biological functions viz. oxygen transport, storage, catalysis and electron transfer, etc. Of the different heme-based metalloproteins that exist, myoglobin (Mb), involved in oxygen storage, is one of the most extensively studied biomolecules and has long been considered a paradigm for understanding the structural and functional characteristics of such proteins [1–3]. One of the biggest obstacles/deterrents in the study of Mb has been the loss of heme under denaturing conditions, with the cofactor known to impart significant stability to the apo-protein [4–7]. Moreover, a majority of such studies have been carried out in dilute buffer medium [4–11]. However the physiological fluid is vastly different and is congested with a large concentration (50–400 g/L) of high molecular weight components namely DNA, RNA, enzymes, lipids and other proteins, that exhibit substantial intracellular volume occupancy [12–15]. These so-called macromolecular crowding agents are known to influence protein structure, stability and

dynamics primarily by the excluded volume effect, the latter arising from steric repulsions between the crowder molecules and proteins of interest [16–24]. In this communication, we report the effect of both synthetic [Ficoll 70 (F70), Dextran 70 (D70), Dextran 40 (D40) and Dextran 6 (D6), PEG8000 (P8) and PEG35000 (P35)] and protein based crowding agents (BSA and lysozyme) on the heme dissociation of Mb under denaturing conditions by monitoring primarily the Soret band absorbance changes. To check whether the effects so observed are macromolecular, the influence of the monomeric components of the crowding agents, namely, sucrose (F70), glucose (Dextran based crowders) and ethylene glycol (PEGs) were also examined. Our results reveal that the crowding agents have a significant effect on heme stability with bovine serum albumin (BSA) showing unusually high heme retention while lysozyme having the reverse influence, that is, exhibiting enhanced heme dissociation. Moreover glucose and sucrose were also quite effective in increasing heme retention, albeit, at much higher concentrations than the macromolecular crowders. 2. Materials and methods

Author contributions: JK and PKC conceived and designed the study; JK and UK carried out the bulk of the experiments while JK carried out the data analyses; SG helped with the aggregation studies and native page analyses along with SK. JK and PKC wrote the manuscript. ⇑ Corresponding author. Fax: +91 1126581102. E-mail address: [email protected] (P.K. Chowdhury).

Myoglobin (Mb) of equine skeletal muscle and bovine serum albumin (BSA) were purchased from Sigma Aldrich Private Ltd. (USA) while Lysozyme of Chicken Egg White was purchased from USB corporation (USA) and were used without further purification. Urea, guanidinium hydrochloride (GdmCl), Ficoll 70, Dextran

http://dx.doi.org/10.1016/j.febslet.2015.11.015 0014-5793/Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

2

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx

(70, 40 and 6), Poly Ethylene Glycol (35 and 8), ethylene glycol, sucrose, glucose, ANS (8-anilino-1-naphthalenesulphonic acid ammonium salt) and thioflavin T (ThT) were also purchased from the Sigma Aldrich Private Ltd. (USA) and used as received. Sodium phosphate dibasic (Na2HPO4) anhydrous, monobasic (NaH2PO4) dihydrate, sodium acetate (anhydrous) and acetic acid were purchased from Merck Specialties Private Limited (Mumbai, India). 2.1. Preparation of solutions Phosphate buffer solution (50 mM) of pH = 7 was prepared by dissolving weighed amounts of mono sodium phosphate and disodium phosphate while the acetate buffer solution (50 mM) of pH = 4 was made by dissolving weighed amounts of sodium acetate followed by the addition of desired amount of glacial acetic acid in Millipore water (Elix 3 UV; Millipore, Molsheim, France). Mb, ANS and ThT containing stock solutions were prepared in the aforesaid phosphate buffer (pH = 7), centrifuged (Centrifuge 5415R, Eppendorf) and diluted as required before carrying out the measurements. The concentrations of Mb, ANS and ThT were measured using an UV–Vis spectrophotometer (Model UV-2450, Shimadzu) in the range of 200–700 nm. The molar extinction coefficients used for Mb are as follows: 13 980 M
1 cm
1 at 280 nm and 188 000 M
1 cm
1 at 409 nm (the Soret band) [25] while those for ANS and ThT are 5000 M
1 cm
1 [26] and 26 620 M
1 cm
1 respectively [27]. For acid denaturation experiments of Mb, the urea solutions were prepared in acetate buffer (pH = 4) and for chemical denaturation experiments urea and GdmCl solutions were prepared in phosphate buffer (pH = 7). The concentrations (of chemical denaturants) were determined by measuring the corresponding refractive indices using a refractometer (KRUSS, A Kruss Optronic, Germany). pH of the solutions was maintained using a pH meter (Hanna HI3220) followed by the addition of NaOH, H3PO4 and acetic acid as per requirement. For the crowding experiments, different concentrations (100, 200, 300 and 400 g/l) of the macromolecular crowders were dissolved in phosphate and acetate buffers after weighing out the appropriate amounts using an analytical balance (Precisa XB 120A) to get the desired concentration. For the chemical denaturation experiments it was difficult to dissolve the crowders beyond 300 g/L while for PEG35 we were able to dissolve only up to 200 g/L of the crowder (further increase of crowder concentration resulted in turbidity). For the protein crowders, the maximum concentration for which reliable and reproducible data could be obtained was 200 g/L. Specific details of methods have been provided in the figure captions and the rest including experimental techniques are given in the Supporting information for further reference.

3. Results and discussion The absorption spectrum of native myoglobin in the visible region (Fig. 1A), is characterized by an intense peak at 409 nm (the Soret band or B-band) and two low-lying peaks at 506 and 637 nm (the Q-bands). The Soret band is very sensitive to the heme environment and hence has commonly been used to monitor heme pocket disruption in presence of external perturbations. Prior to subjecting the protein to denaturing environments, our initial goal was to probe whether the crowding agents were able to bring about any changes in the heme pocket of native Mb. F70, D70, D40, P35 and BSA gave rise to little or no change of the band in the Soret region; however in presence of D6, P8, and lysozyme, significant distortions and/or modulations could be seen (Fig. 1 and Supplementary information Fig. 1). For example, D6 (Fig. 1B) brought about a substantial red-shift in the Soret maximum (from 409 nm in absence of the crowder to 418 nm in 400 g/l of D6) while

in presence of P8 (Fig. 1C), the absorption intensity decreased to a large extent (without any wavelength shift) signifying a concomitant reduction in the probability of the electronic transition. One of the most notable effects on Mb was that of the protein lysozyme (Fig. 1D and E) wherein the spectrum not only became broad but showed a significant blue-shift akin to water exposure of heme that can either happen on its dissociation (as a result of Mb denaturation) or the heme being squeezed out of the pocket at such high concentrations of the crowder protein. On the other hand, the small molecules (glucose, sucrose and EG) had no visible effect on the Soret profile of Mb. Subsequently heme dissociation studies in presence and absence of the crowding agents were carried out by subjecting Mb to three different denaturing conditions: (i) pH = 4 with 1 M urea (ii) 1.9 M GdmCl and (iii) a mixture of 1.4 M GdmCl and 1.4 M urea. Before we proceed further, we would like to provide a brief rationalization with regards to why such denaturation conditions have been chosen. At low pH (pH < 4) Mb is known to undergo facile heme loss because of protonation of the proximal histidine [4]; however under such conditions crowding agents might not remain stable. Hence to mimic the low pH environment, at pH = 4, urea (1 M) was added. The other two denaturing conditions were chosen such that the extent of heme loss was very similar to that of the low pH, as determined by ratio of the absorbance of the bound to that of dissociated heme (explained below). Moreover, a recent simulation study has shown that use of mixed denaturants induces non-native collapsed states in proteins, with urea acting as a potential crowding agent [28]. The crowder concentrations were varied over an appreciable range, reaching as high as 400 g/l for the present study. For native myoglobin, the intact heme shows an absorption maximum at 409 nm (A409) while the broad absorption band with a maximum at 360 nm (A360) has often been attributed to that of dissociated heme remaining loosely attached (non-covalently and/or out of the heme pocket) to the protein matrix [10]. Decrease in pH or increase in the concentrations of GdmCl leads to significant changes in the Soret band. Under conditions wherein heme dissociation is minimal (in presence of urea), the Soret band shifts to the blue with concomitant broadening, signifying extensive water coordination as the heme pocket opens up (SI Fig. 2A). A convenient and informative means of monitoring the effect of crowding agents on heme stability/loss, is therefore to plot the A409/A360 ratio as a function of the crowder concentration. For example, for native myoglobin, at pH = 7, the ratio of bound to free heme (A409/A360) is 4.85, the same undergoing a considerable decrease as a function of chemical (SI Fig. 2A and B) and pH-induced denaturation (SI Fig. 2C). Shown in SI Fig. 3 are some representative spectra of the Soret band at pH = 4 (and 1 M urea) for the different crowding agents used in this study. As evident (Fig. 2A), the A409/A360 ratio increases for almost all the synthetic macromolecular crowders with P35 showing the maximum heme retention (amongst the synthetic crowders) by having the highest absorbance ratio of 4.5 at a concentration of 200 g/l. Here, it should be noted that this ratio is very close to that observed for native Mb (as mentioned above). P8 also showed a steep initial rise in the ratio but beyond 200 g/l this crowding agent adversely affected the heme stability as observed by the decrease in absorbance of the bound heme. On the other hand, the dextran based crowding agents showed a gradual increase in the heme retention as the crowder concentration was increased, with D6 showing the least among these while the profiles for D70 and D40 were quite similar. F70, having the same average molecular weight as that of D70, however was much less effective, reaching a final ratio of 2.80 as compared to that of 4.20 observed for D70. Fig. 2B (and SI Fig. 4) shows the guanidine hydrochloride induced denaturation of Mb where GdmCl concentration was kept at 1.9 M to reach the absorbance ratio

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

3

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx

0.8

(A)

Buffer 10 g/L 30 g/L 50 g/L 100 g/L 200 g/L

0.6 Absorbance

0.5 0.4 0.3 0.2

0.6 0.4 0.3 0.1

0.8

400 500 600 Wavelength (nm)

0.0

700

0.7

(C)

Buffer 50 g/L 100 g/L 150 g/L 200 g/L 250 g/L 300 g/L 350 g/L 400 g/L

0.7 0.6 0.5 0.4 0.3

0.3

0.1

0.1

6

700

(D)

0.4

0.2

400 500 600 Wavelength (nm)

500 600 Wavelength (nm)

0.5

0.2 0.0 300

400

0.6 Absorbance

0.0 300

Absorbance

0.5

Buffer 50 g/L 100 g/L 150 g/L 200 g/L 250 g/L 300 g/L 350 g/L 400 g/L

0.2

0.1

Absorbance Ratio (A409/A360)

(B)

0.7 Absorbance

0.7

0.0 300

400 500 600 Wavelength (nm)

700

Buffer 10 g/L 30 g/L 50 g/L 60 g/L 70 g/L 75 g/L 80 g/L 90 g/L 100 g/L 150 g/L 200 g/L

700

(E)

5 4 3 2

BSA Lysozyme P8 D6

1 0

100 200 300 [Crowder] (g/L)

400

Fig. 1. Absorption spectra of native myoglobin (10 lM) at pH = 7 (phosphate buffer) in presence of (A) BSA, (B) D6, (C) P8 and (D) lysozyme. In panel (E), the absorbance ratio (A409/A360) of native Mb in presence of BSA, lysozyme, D6 and P8 has been plotted. The dramatic effect of lysozyme on native Mb is clearly evident.

(A409/A360) of 1.25. Here, in presence of D40, Mb showed the maximum resistance towards heme dissociation followed by D6 (Fig. 2B). For both D70 and P8, the Soret band shows a bellshaped response with the ratio decreasing at higher concentrations of the crowding agents. F70 and P35 showed lower stabilizing effects as compared to the others. Similar studies in the denaturant mixture (GdmCl 1.4 M and urea 1.4 M) shown in Fig 2C (and SI Fig. 5), reveal that among the macromolecular crowding agents the PEG based ones impart the best heme retention environment for Mb, while at higher crowding concentrations D6, D40 and D70 predominate. Interestingly enough, the small molecule crowders (glucose, sucrose), while having no effect on the Soret band of native Mb, however, showed remarkable effects on heme retention. In particular, the extent of heme stability in presence of glucose and sucrose was the highest for the acid denaturation case giving rise to absorbance ratios as high as 4.88 (for glucose) and 4.70 (for sucrose). On the other hand, EG had a measurable stabilizing effect in the acidic pH while in case of the GdmCl based denaturation conditions, its effect was hardly visible. Taken together our data reveal differential crowder dependent heme retention, the latter also varying with the type of denaturation the protein was subjected to.

Since heme pocket disruption is almost always accompanied by changes at the secondary structure level as well as modulation in Tryptophan(Trp)-heme distance (thereby affecting the extent of energy transfer between the donor Trp and acceptor heme moieties), we also performed circular dichroism (CD) and steady state fluorescence experiments to monitor the same. The ellipticity ratios reveal a gain in secondary structure in most of the cases wherein there is also an increase in heme retention (SI Fig. 6). For example the CD profile in the acidic pH completely corroborates the effect on the Soret band even with regards to the bellshaped dependence for P8 (SI Fig. 7). However for the other two denaturing conditions, F70 brings about a decrease in the helical content though under identical conditions it increases heme retention (SI Figs. 6 and 7). Increase in helicity was also observed for glucose and sucrose under all conditions, with these having the maximum effect amongst the synthetic crowders in presence of the chemical denaturants. EG, in agreement with the changes in the absorbance ratio had little effect. In native Mb, the Trp emission intensity is severely quenched due to appreciable energy transfer to the heme moiety (SI Fig. 8A). Thus on unfolding, because of increase in the Trp-heme separation, a dramatic enhancement in the Trp fluorescence is observed. Hence it is expected that increase in

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx

(A) pH = 4 and 1M Urea

6 5 4 3 2 1 0

4.5 Absorbance Ratio (A409/A360)

Absorbance Ratio (A409/A360)

Absorbance Ratio (A409/A360)

4

100 200 300 [Crowder] (g/L)

400

(B) 1.9M GdmCl in pH = 7

4.0 3.5 3.0 2.5 2.0 1.5 1.0 0

100 200 300 [Crowder] (g/L)

400

(C) 1.4M GdmCl and 1.4M Urea

4.0 3.5 3.0 2.5 2.0 1.5 1.0

0

100 200 300 [Crowder] (g/L)

400

Fig. 2. Absorbance ratio (A409/A360) plots of Mb (10 lM) in presence of (A) pH = 4 + 1 M urea (acetate buffer), (B) 1.9 M GdmCl at pH = 7 (phosphate buffer) and (C) 1.4 M GdmCl and 1.4 M urea at pH = 7 (phosphate buffer) as a function of increasing crowder concentration: F70 ( ), D70 ( ), D40 ( ), D6 ( ), P8 ( ) P35 ( ), BSA ( ), lysozyme ( ), EG ( ), glucose ( ), sucrose ( ).

heme retention should lead to a concomitant decrease in Trp intensity in a crowder dependent manner. Our data (SI Fig. 9) reveal that the decrease in Trp fluorescence might not always be directly related to the increase in helicity and hence heme stability. For example, D6 for all the three cases show the maximum decrease in Trp emission (SI Fig. 9), in spite of the fact that P8, in particular, shows much higher heme retention (Fig. 2A and C). Moreover, while D70 and D40 show appreciable heme stability under all conditions, however, the change in Trp fluorescence for both is almost negligible. In other words, since heme plays an integral role in the stability of Mb it has been proposed that the denaturation profile of the holoprotein reports the heme affinity rather than the unfolding of the globin part of Mb [4]. Hence it was expected that the Trp emission would also be simultaneously quenched because of the Trp-heme proximity, provided that the secondary structure of the protein would retain its native-like packing in presence of the macromolecular crowders. On the other hand, based on our expectations, extensive quenching of Trp fluorescence almost reaching the value of that of native myoglobin was observed for the small molecule crowders, in particular glucose (SI Fig. 8B, E and G). Additionally, for the GdmCl based denaturation conditions, the emission maximum shifts red to  342 nm (SI Fig. 8E) suggesting significant solvent exposure of the Trp moieties as compared to a maximum at 330 nm as observed for native Mb. As seen (SI Fig. 8E and F) in presence of increasing concentrations of glucose and sucrose, the Trp emission maximum shifts to the blue from 342 nm to 330 nm in concurrence with the reduction in Trp intensity arising from progressive attainment of the native fold of Mb. However in case of acid denaturation at pH = 4, the emission maximum shifts only to 332 nm and is also accompanied by a large increase in Trp intensity. While addition of glucose again brings about a reduction in the Trp fluorescence, the fact that the shift in kmax is so small implies that the Trp environment is still predominantly hydrophobic. Thus the larger extent of heme retention and

lesser secondary structure distortion combined with the small fluorescence shift for Mb at pH = 4 point to the fact that the structural ensemble of Mb under these conditions is more native-like thereby facilitating attainment of near native conformations in presence of these small molecule crowders. This heme retention as observed here in presence of the polymeric synthetic crowders can be attributed to the excluded volume effect, the latter arising from the mutual impenetrability of the test species (the protein Mb) and the crowder molecules. The heme is held in its place by a single covalent linkage between the proximal histidine (from the globin fold) and the iron atom with the surrounding amino acids maintaining an environment such that facile ligand binding and release can take place from the heme. CD data reveal that the loss in secondary structure for Mb is higher in case of chemical denaturation (SI Fig. 6A) than that for the pH induced one. We believe this to be the reason behind an overall lesser heme loss (i.e. greater heme retention) for the pH induced denaturation as observed from the higher A409/A360 ratios obtained (SI Table 1). The fact that the extent of heme retention is crowder dependent also points to the intrinsic differences in the manner in which these affect Mb. If excluded volume had been the sole reason, then both D6 and P8, having the least molecular weights and hence the highest packing density (for a given crowder concentration) should have shown the maximum heme retention. While P8 fits this profile for the acid and mixed denaturation conditions, D6 only shows this effect at very high concentrations. Surprisingly, D40 and D70 show greater heme stabilizing effects than that expected from their excluded volume exertion. Further support to the fact that the observed variations are not due to changes in excluded volume only, comes from the manner in which glucose and sucrose affected the heme retention. Glucose and sucrose are well-known osmolytes that stabilize proteins by causing preferential hydration of the biomolecules [29]. Taken together these data reveal that besides excluded volume (expected based on the shape and size),

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

5

3.0

6 (A) 5 4 3 2 1

pH=4+1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl+1.4M Urea (pH=7)

0

50

100 150 [BSA] (g/L)

Absorbance Ratio (A409/A360)

Absorbance Ratio (A409/A360)

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx

2.4 2.1 1.8 1.5 (B) 1.2

200

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

2.7

0

100

200 300 [F70] (g/L)

400

3.5 3.0 2.5 2.0 (C)

Absorbance Ratio (A409/A360)

0

Absorbance Ratio (A409/A360)

200 300 [D70] (g/L)

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

3.0 2.5 2.0 1.5

(E) 100

200 [D6] (g/L)

300

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

2.5 2.0 1.5

(G) 0 5.0

100

200 [P8] (g/L)

300

3.0 2.5 2.0 1.5

400

(D) 0

5.0

100

200 300 [D40] (g/L)

400

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

4.5 4.0 3.5 3.0 2.5 2.0 1.5

(F) 0

4.0

3.0

3.5

400

3.5

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

4.0

400

3.5

0

Absorbance Ratio (A409/A360)

100

Absorbance Ratio (A409/A360)

1.5

Absorbance Ratio (A409/A360)

pH=4+1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

4.0

Absorbance Ratio (A409/A360)

Absorbance Ratio (A409/A360)

4.5

5.0

50

100 150 [P35] (g/L)

200

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

4.5 4.0 3.5 3.0 2.5 2.0 1.5

(H) 0

100 200 300 [Glucose] (g/L)

400

pH=4 + 1M Urea 1.9M GdmCl (pH=7) 1.4M GdmCl + 1.4M Urea (pH=7)

4.5 4.0 3.5 3.0 2.5 2.0 1.5

(I) 0

100 200 300 [Sucrose] (g/L)

400

Fig. 3. Fitted absorbance ratio (A409/A360) plots of Mb as a function of increasing concentration of (A) BSA, (B) F70, (C) D70, (D) D40, (E) D6, (F) P35, (G) P8, (H) glucose and (I) sucrose. The variation in the models used for the fits clearly reveal that the extent and manner of heme stabilization of Mb not only depends on the shape, size and concentration of the crowding agents but is also a function of the applied denaturant conditions.

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx

6

Intensity (10 )

other factors like internal architecture of the crowded medium and/or soft interactions with the protein can also play an important role as has been suggested previously [30–33]. In order to further analyse the crowder dependent effect on heme retention, the absorbance ratio trends were fit to some phenomenological models (Eqs. (1)–(7); see Annexure for details) in an effort to provide insights into the underlying differences in the manner in which the different crowders exert their respective influence. The range of equations used to analyse the observed trends reveal the degree of complexity associated in trying to provide a unified model based solely on excluded volume for these synthetic crowding agents (Fig. 3). Even more intriguing is the fact that for the same crowding agent, depending on the type of denaturation, the model used to describe the observed heme retention profile changes (SI Table 4). For example, at low pH, the trend in presence of D40, EG and sucrose showed a linear variation with respect to the crowder concentration while for GdmCl only and the mixed denaturation, sigmoidal and exponential fits were necessary for describing the response of Mb. On the other hand in presence of D6, the profile was described by a biphasic dose response curve (Eq. (6)) for all the three types of denaturing environments revealing that D6 behaved in a similar fashion throughout, an aspect that points towards this crowder exerting its influence primarily via the excluded volume effect. Similarly the profiles in presence of P8 were all modeled based on a bellshaped response (Eq. (7)) further confirming that beyond a certain concentration, P8 brings about a reduction in the heme retention. The most interesting and surprising effects have been obtained with the protein based crowding agents, BSA and lysozyme. For all the three denaturing conditions, in presence of BSA, Mb experienced the maximum prevention in heme loss (amongst all the crowders used here) while for lysozyme the effect is just the opposite, with the latter inducing heme dissociation. Our data show that the absorbance ratio (A409/A360) increases dramatically with increase in BSA for all the three denaturing conditions. Moreover, in case of the chemical denaturants, the increase in heme retention in presence of BSA shows almost a biphasic dependence on the crowder (BSA) concentration with the steepest response observed in the 0–30 g/l (Figs. 2 and 3A) where the excluded volume effects are presumed to be very low. Furthermore, the increase in absorbance ratio for Mb is much higher in case of acid denaturation than chemical denaturation (SI Fig. 10), signifying that the resistance towards the heme dissociation of myoglobin is higher in the former case, an aspect that is common with that of the synthetic crowders. Additionally, the final ratio in presence of 200 g/l BSA is 5.95 (acid denaturation) which is even higher than that observed for

2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0

(A)

350

B 5 μM BM 5 μM B 10 μM BM 10 μM B 20 μM BM 20 μM B 40 μM BM 40 μM B 60 μM BM 60 μM B 80 μM BM 80 μM B 100 μM BM 100 μM B 120 μM BM 120 μM B 140 μM BM 140 μM B 160 μM BM 160 μM

400 450 500 Wavelength (nm)

550

native Mb (ratio is 4.85) in spite of the fact that the protein crowder hardly had any effect on native Mb. Since proteins are known to self-associate at high concentrations [34,35], we further probed the same using ThT assay (SI Fig. 11), time resolved fluorescence studies (SI Fig. 12 and SI Table 3) and native gel analyses (SI Fig. 13 and SI Table 2). These studies clearly reveal appreciable reversible oligomer formation for both the protein based crowders as their respective concentrations are increased, implying that protein oligomers are also able to exhibit macromolecular crowding. CD studies on the protein crowders show that BSA undergoes distinct changes at the secondary structure level, with the extent of such distortion being the least for the pH = 4 solution (SI Fig. 6B). These data taken together with the fact that Mb shows unusually high heme retention in presence of BSA at pH = 4 arises from the nearly intact secondary structure of BSA and lesser extent of helix loss for Mb at this pH. On the other hand lysozyme hardly shows any changes under the denaturing conditions studied here (SI Fig. 6C); however the fact that it affects the heme stability significantly provides an insight into how different the interactions are for this protein with Mb as compared to that of BSA. Moreover, since BSA (pI = 4.7) shows almost no effect on native Mb (pI = 7.1) while lysozyme (pI = 11.2) brings about the maximum heme loss under similar conditions (SI Fig. 10), we propose that the interaction between the serum and the heme proteins is primarily hydrophobic while for lysozyme electrostatics predominates. Since in the native state, there is a predominance of charged or polar amino acid residues on the surface of the water soluble protein Mb, perturbation of its native structure to such an extent can only occur under charge-charge interactions that are destabilizing. Further proof of this hypothesis is available from the heme retention plots in presence of pH and chemical denaturation (SI Fig. 10). Had BSA exhibited non-specific electrostatic interactions with Mb, then it was expected that in presence of GdmCl (a charged denaturant), or at low pH, the same would decrease (since both BSA and Mb would be predominantly positively charged) because of efficient electrostatic screening or like-charge repulsion. However the fact that the heme retention was still quite high proves that hydrophobic forces predominate. Comparison of the hydropathy index (SI Fig. 14) [36] also shows that both Mb and BSA are on average more hydrophobic than lysozyme. To further probe the presence of soft interactions between the proteinbased crowders and Mb, we also examined the extent of energy transfer, FRET, occurring between the Trp residues (as donor) of BSA with the heme cofactor (as acceptor) of Mb (Fig. 4). Since in the native state, the fluorescence of the Trp residues of Mb is highly quenched due to energy transfer to the heme, hence the Mb-Trp

20 Transfer Efficiency (%)

6

(B)

15 10 5 0

0

40

80 120 [BSA] ( M)

160

Fig. 4. (A) Tryptophan intensity, with increasing concentration of BSA in presence of myoglobin at pH = 7 (phosphate buffer) where the Mb concentration was kept at 10 lM and BSA concentration was varied from 5 to 160 lM (B represents only BSA and BM represents BSA with myoglobin). We were unable to increase the BSA concentration beyond 160 lM because the absorbance of Tryptophans of BSA at 295 nm was quite high and hence would have had severe interference from the inner filter effect. (To reduce the inner filter effect, the fluorescence spectra were acquired in a 3 mm path length fluorescence cuvette.) (B) Variation of the efficiency of energy transfer between the Trp residues of BSA and heme (myoglobin) as a function of increasing concentration of BSA.

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

7

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx Table 1 Conversion of crowder concentration from g/L to moles/L. (A) For protein crowders g/L moles/L

BSA Lysozyme

10

30

50

100

150

200

0.00015 0.00070

0.00045 0.0021

0.00075 0.0035

0.0015 0.0070

0.0023 0.0105

0.0030 0.0140

(B) For synthetic crowders g/L moles/L

F70 D70 D40 D6 P8 P35 EG Glucose Sucrose

50

100

150

200

250

300

350

400

0.00071 0.00071 0.0013 0.0083 0.0063 0.0014 0.80 0.28 0.15

0.0014 0.0014 0.0025 0.017 0.012 0.0029 1.61 0.55 0.29

0.0021 0.0021 0.0038 0.025 0.019 0.0043 2.42 0.83 0.44

0.0029 0.0029 0.0050 0.033 0.025 0.0057 3.22 1.11 0.58

0.0036 0.0036 0.0063 0.042 0.030 – 4.03 1.39 0.73

0.0043 0.0043 0.0075 0.050 0.037 – 4.83 1.67 0.88

0.0050 0.0050 0.0088 0.058 0.044 – 5.64 1.94 1.02

0.0057 0.0057 0.010 0.067 0.050 – 6.40 2.22 1.17

signal (which was quite low) was subtracted from the Trp emission of BSA to calculate the efficiency (E) of transfer of energy from Trp132 and Trp-214 of BSA to the heme moiety of Mb. Our data show that at the lowest concentration of BSA used, the FRET efficiency was the highest at 0.18 leading to a distance of 37.1 Å of the crowder protein from Mb. Increase of the BSA concentration however brought about a decrease in the average FRET efficiency (Fig. 4 and SI Table 6). These data suggest that at higher concentrations, the BSA themselves formed a cluster (supporting the reversible oligomer formation for BSA) around Mb thereby increasing the apparent distance between the individual protein crowder and test protein molecules, and hence providing us a glimpse of the how the protein crowders might be arranged around Mb. 4. Conclusions In conclusion, our study reveals a significant and remarkable effect of crowders (both small and macromolecular) on the heme stability of myoglobin. That the Soret band of Mb is affected in the native state too by some of the crowding agents presents an interesting phenomenon. Previous studies have ascribed redshifts (arising from hybrid orbital deformation effect) in the heme visible bands (as observed for D6) to the ruffling (non-planarity) of this cofactor brought about by changes of protein structure in its immediate proximity [37–39]. This implies that D6 based on its high packing density perturbs native Mb in a manner that the heme pocket is distinctly altered, an aspect that can have several consequences on the manner in which Mb functions. On the other hand, P8 brings about a reduction in the Soret band absorbance only which again conforms to heme pocket modulation. Moreover, for P8, the sudden drop in Soret absorbance beyond 200 g/L corresponds to an appreciable decrease in the helical signal (CD) of Mb at 222 nm. Similar effects of PEG on DNA stability has been reported wherein it was proposed that the reversal of free energy change at high PEG concentrations is related to the changing extent of entanglement of the polymer coils [40]. That the effects need not be fully macromolecular were evident based on the significant influence that both glucose and sucrose had on the heme retention and Trp fluorescence. Moreover, glucose showed much more stabilization than sucrose, a finding that is at odds with a recent report wherein the free energy of unfolding was augmented as the number of monomer units in the polysaccharide increased, the latter having been attributed to increased steric repulsion [41]. This disagreement further shows that excluded volume effect is not the sole contributor and that the nature of interactions involved is also very much dependent on the specific biomacromolecule in question. Again, a recent size dependent

analysis of crowding agents shows that small molecules are much better crowders (based purely on the hard sphere model) [42] which is in accordance with our observations. Thus it is obvious from our results that the molar concentration of the osmolytes, glucose and sucrose, wherein these are effective, are much higher than that of the macromolecular crowders (Table 1) in agreement with a couple of recent studies [43,44]. BSA and lysozyme exhibit significant soft interactions with Mb as evident from the changes in the Soret band absorbance at low concentrations of these crowder proteins. A few recent studies have emphasized upon the presence of such weak non-specific interactions for such protein based crowders [45,46]. It has been proposed that the cell instead of being considered as a uniformly crowded medium should rather be thought of having regions of supercrowding interspersed with domains that behave like dilute buffer solutions [47,48]. In this regard the oligomers formed by BSA and lysozyme can be thought of as such superstructures which bring about enhanced effects of volume exclusion along with associated soft potentials. Furthermore the absence of correlation between the extent of heme retention (through absorption studies) and fluorescence (and CD) data implies the appearance of different non-native states as induced by the macromolecular crowding agents under the denaturing conditions employed here. In other words, high A409/A360 ratio does not necessarily guarantee that the protein has been able to retain its native-like ensemble in presence of the crowding agents. Finally, it would be worthwhile to investigate how the different biological functions of Mb (e.g. binding of gaseous ligands, redox properties) are affected under such crowder-induced distortions and non-native conformational ensembles. For example, extensive studies on ligand (CO) migration in Mb have shown the presence of distinct channels along with xenon pockets where the ligands can temporarily lodge themselves and also a hierarchy of substates giving rise to a distribution of ligand binding pathways [49]. It would thus be an exciting proposition to investigate how these aforementioned phenomena are affected for native Mb in the physiological milieu, in particular with proteins as crowders having been shown (in this work) to exhibit significant effects on the Mb heme pocket, thereby bringing us closer to the physiological interior. As an ending note, it should be mentioned that studying the effects of protein based crowders can be a challenge since these have their own huge contributions in CD and fluorescence signals. Hence the presence of an alternate window of detection, like cofactor absorption (here the Soret band), that can reliably report structural changes of the test protein (here Mb), can be of great use and also provide a simpler approach for monitoring perturbations in cell-mimicking environments.

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

8

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx

Acknowledgements

Appendix A. Supplementary data

JK and SG gratefully acknowledge CSIR while SK acknowledges IIT Delhi for providing fellowships. PKC gratefully acknowledges Indian Institute of Technology (IIT) Delhi for providing start up funding and Department of Science and Technology (DST), New Delhi, India for providing financial support (SR/FT/CS-007/2010).

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2015.11. 015.

Appendix Annexure:. Equations used to fit the absorbance ratio trends for the various crowders under different denaturation conditions (as mentioned below the respective model). (1) y ¼ mx þ c

ð1Þ

[D70, glucose (1.9 M GdmCl) and D40, EG and sucrose (pH = 4 + 1 M urea)] (2) y ¼ y0 þ A1  ð1
exp ð
x=c1 ÞÞ

ð2Þ

[BSA, P35 and glucose (pH = 4 + 1 M urea) and D40 and P35 (1.4 M GdmCl + 1.4 M urea)] (3) y ¼ y0 þ A1  ð1
expð
x=c1 ÞÞ þ A2  ð1
expðx=c2 ÞÞ

ð3Þ

[BSA (1.9 M Gdmcl and 1.4 M Gdmcl + 1.4 M urea)] (4) y ¼ A2 þ ðA1
A2 Þ=ð1 þ expðx
x0 Þ=dxÞ

ð4Þ

[F70 and D70 (pH = 4 + 1 M urea) and D40, P35 and sucrose (1.9 M GdmCl) and D70, glucose and sucrose (1.4 M GdmCl + 1.4 M urea)] Boltzmann sigmoid model: where A1 and A2 are the initial and final absorbance ratio values respectively, x0 is the concentration at which the increase in absorbance ratio is halfway between initial and final value and dx is the slope of the curve. (5) y ¼ A0 þ A1  x þ A2  x2 þ A3  x3 þ A4  x4 þ A5  x5

ð5Þ

[F70 (1.9 M GdmCl and 1.4 M GdmCl + 1.4 M Urea)] (6) y ¼ A þ ðA
A Þ½p=ð1 þ 10ððLOGx01
xÞh1 Þ Þ 1 2 1

þ ð1
pÞ=ð1 þ 10ððLOGx02
xÞh2 Þ Þ

ð6Þ

[D6 in presence of all three denaturating conditions] Biphasic dose–response model: where A1 and A2 are the initial and final absorbance ratio value respectively, LOGx01 and LOGx02 are the concentrations at which the increase in absorbance ratio is halfway between initial value and final value, h1 and h2 are the slopes and p is the proportion of maximal response due to the more potent phase. (7) y ¼ A þ ðA
A Þ=ð1 þ 10ððLOGx01
xÞh1 Þ Þ 0 1 0

þ ðA2
A0 Þ=ð1 þ 10ððx
LOGx02Þh2 Þ Þ

ð7Þ

[P8 in presence of all three denaturating conditions] Bell shaped dose response model: where A1, A2 and A0 are the initial, final, and peak absorbance ratio values respectively, LOGx01 and LOGx02 are the concentrations at which the increase in absorbance ratio is halfway between initial value and final value, h1 and h2 are the slopes All the fit parameters have been summarised in SI Table 5.

References [1] Irace, G., Bismuto, E., Savy, F. and Colonna, G. (1986) Unfolding pathway of myoglobin: molecular properties of intermediate forms. Arch. Biochem. Biophys. 244, 459. [2] Bismuto, E., Colonna, G. and Irace, G. (1983) Unfolding pathway of myoglobin. Evidence for a multistate process. Biochemistry 22, 4165. [3] Pace, C.N. and Vanderburg, K.E. (1979) Determining globular protein stability: guanidine hydrochloride denaturation of myoglobin. Biochemistry 18, 288. [4] Hargrove, M.S., Wilkinson, A.J. and Olson, J.S. (1996) Structural factors governing hemin dissociation from metmyoglobin. Biochemistry 35, 11300. [5] Hargrove, M.S., Krzywda, S., Wilkinson, A.J., Dou, Y., Saito, M.I. and Olson, J.S. (1994) Stability of myoglobin: a model for the folding of heme proteins. Biochemistry 33, 11767. [6] Jiao, Z.Y., Qian, T., Yu, C.H. and Fang, Z.X. (2013) Acid-induced unfolding process of myoglobin and its mutant under macromolecular crowding conditions. Acta Phys. Chim. Sin. 29, 1785. [7] Palaniappan, V. and Bocian, D.F. (1994) Acid-induced transformations of myoglobin. Characterization of a new equilibrium heme-pocket intermediate. Biochemistry 33, 14264. [8] Hunter, C.L., Mauk, A.G. and Douglas, D.J. (1997) Dissociation of heme from myoglobin and cytochrome b5: comparison of behavior in solution and the gas phase. Biochemistry 1018, 36. [9] Sage, J.T., Morikis, D. and Champion, P.M. (1991) Spectroscopic studies of myoglobin at low pH: heme structure and ligation. Biochemistry 30, 1227. [10] Tang, Q., Kalsbeck, W.A., Olson, J.S. and Bocian, D.F. (1998) Disruption of the heme iron-proximal histidine bond requires unfolding of deoxymyoglobin. Biochemistry 37, 7047. [11] Konermann, L., Rosell, F.I., Mauk, A.G. and Douglas, D.J. (1997) Acid-induced denaturation of myoglobin studied by time-resolved electrospray ionization mass spectrometry. Biochemistry 36, 6448. [12] Fulton, A.B. (1982) How crowded is the cytoplasm? Cell 30, 345. [13] Rivas, G., Ferrone, F. and Herzfeld, J. (2004) Life in a crowded world. EMBO 5, 23. [14] Minton, A.P. (2006) How can biochemical reactions within cells differ from those in test tubes? J. Cell Sci. 119, 2863. [15] Ellis, R.J. (2001) Macromolecular crowding: obvious but under appreciated. Trends Biochem. Sci. 26, 597. [16] Dhara, A., Samiotakis, A., Ebbinghaus, S., Nienhaus, L., Homouz, D., Gruebele, M. and Cheung, M.S. (2010) Structure, function, and folding of phosphoglycerate kinase are strongly perturbed by macromolecular crowding. Proc. Natl. Acad. Sci. 107, 17586. [17] Minton, A.P. (1997) Influence of excluded volume upon macromolecular structure and associations in ‘crowded’ media. Curr. Opin. Biotechnol. 8, 65. [18] Dong, H., Qin, S. and Zhou, H.-X. (2010) Effects of macromolecular crowding on protein conformational changes. PLoS Comput. Biol. 6, e1000833. [19] Zhou, H.-X., Rivas, G. and Minton, A.P. (2008) Macromolecular crowding and confinement: biochemical, biophysical, and potential physiological consequences. Ann. Rev. Biophys. 37, 375. [20] Mukherjee, S., Waegele, M.M., Chowdhury, P., Guo, L. and Gai, F. (2009) Effect of macromolecular crowding on protein folding dynamics at the secondary structure level. J. Mol. Biol. 393, 227. [21] Rajapaksha, A., Stanley, C.B. and Todd, B.A. (2015) Effects of macromolecular crowding on the structure of a protein complex: a small-angle scattering study of superoxide dismutase. Biophys. J . 108, 967. [22] Minh, D.D.L., Chang, C., Trylska, J., Tozzini, V. and McCammon, J.A. (2006) The influence of macromolecular crowding on HIV-1 protease internal dynamics. J. Am. Chem. Soc. 128, 6006. [23] Stagg, L., Zhang, S.-Q., Cheung, M.S. and Stafshede, P.W. (2007) Molecular crowding enhances native structure and stability of alpha/beta protein flavodoxin. Proc. Natl. Acad. Sci. U.S.A. 104, 18976. [24] Samiotakis, A., Stafshede, P.W. and Cheung, M.S. (2009) Folding, stability and shape of proteins in crowded environments: experimental and computational approaches. Int. J. Mol. Sci. 10, 572. [25] Malik, A., Kundu, J., Mukherjee, S.K. and Chowdhury, P.K. (2012) Myoglobin unfolding in crowding and confinement. J. Phys. Chem. B 116, 12895. [26] Sirangelo, I., Piaz, F.D., Malmo, C., Casillo, M., Birolo, L., Pucci, P., Marino, G. and Irace, G. (2003) Hexafluoroisopropanol and acid destabilized forms of apomyoglobin exhibit structural differences. Biochemistry 42, 312. [27] Coleman, C., I-Zanetti, C., Carter, S.A., Vinay Krishna, Grover, R.K., Roy, R., Singh, S. and Khurana, R. (2005) Mechanism of thioflavin T binding to amyloid fibrils. J. Struct. Biol. 151, 229. [28] Xia, Z., Das, P., Shakhnovich, E.I. and Zhou, R. (2012) Collapse of unfolded proteins in a mixture of denaturants. J. Am. Chem. Soc. 134, 18266. [29] Li, Y. and Bolen, D.W. (1995) The peptide backbone plays a dominant role in protein stabilization by naturally occurring osmolytes. Biochemistry 34, 12884.

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015

J. Kundu et al. / FEBS Letters xxx (2015) xxx–xxx [30] Miklos, A.C., Li, C., Sharaf, N.G. and Pielak, G.J. (2010) Volume exclusion and soft interaction effects on protein stability under crowded conditions. Biochemistry 49, 6984. [31] Sarkar, M., Li, C. and Pielak, G.J. (2013) Soft interactions and crowding. Biophys. Rev. 5, 187. [32] Politi, R. and Harries, D. (2010) Enthalpically driven peptide stabilization by protective osmolytes. Chem. Commun. 46, 6449. [33] Sapir, L. and Harries, D. (2015) Is the depletion force entropic? Molecular crowding beyond steric interactions. Curr. Opin. Colloid Interface Sci. 20, 3. [34] Terao, T., Yoshika, S. and Jordan, G.M. (1994) The aggregation of bovine serum albumin in solution and in the solid state. J. Pharm. Pharmacol. 46, 182. [35] Arafa, I.M., Al-Khateeb, M. and Atmeh, R.F. (2007) Albumin aggregates: hydrodynamic shape and physico-chemical properties. Jordan J. Chem. 2, 169. [36] Kyte, J. and Doolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105. [37] Zhou, Z., Shen, M., Cao, C., Liu, Q. and Yan, Z. (2012) Opposing influences of ruffling and doming deformation on the 4-N cavity size of porphyrin macrocycles: the role of heme deformations revealed. Chem. Eur. J. 18, 7675. [38] Zhou, Z., Cao, C., Liu, Q. and Jiang, R. (2010) Hybrid orbital deformation (HOD) effect and spectral red-shift property of nonplanar porphyrin. Org. Lett. 12, 1780. [39] Haddad, R.E., Gazeau, S., Pe´caut, J., Marchon, J.-C., Medforth, C.J. and Shelnutt, J. A. (2003) Origin of the red shifts in the optical absorption bands of nonplanar tetraalkylporphyrins. J. Am. Chem. Soc. 125, 1253.

9

[40] Knowles, D.B., LaCroixa, A.S., Deinesa, N.F., Shkela, I. and Record Jr., M.T. (2011) Separation of preferential interaction and excluded volume effects on DNA duplex and hairpin stability. Proc. Natl. Acad. Sci. U.S.A. 108, 12699. [41] Beg, I., Minton, A.P., Hassan, Md.I., Islam, A. and Ahmad, F. (2015) Thermal stabilization of proteins by mono and oligosaccharides: measurement and analysis in the context of an excluded volume model. Biochemistry 54, 3594. [42] Sharp, K.A. (2015) Analysis of the size dependence of macromolecular crowding shows that smaller is better. Proc. Natl. Acad. Sci. U.S.A. 112, 7990. [43] Senske, M., Törk, L., Born, B., Havenith, M., Herrmann, C. and Ebbinghaus, S. (2014) Protein stabilization by macromolecular crowding through enthalpy rather than entropy. J. Am. Chem. Soc. 136, 9036. [44] Benton, L.A., Smith, A.E., Young, G.B. and Pielak, G.J. (2012) Unexpected effects of macromolecular crowding on protein stability. Biochemistry 51, 9773. [45] Sarkar, M., Smith, A.E. and Pielak, G.J. (2013) Impact of reconstituted cytosol on protein stability. Proc. Natl. Acad. Sci. U.S.A. 110, 19342. [46] Miklos, A.C., Sarkar, M., Wang, Y. and Pielak, G.J. (2011) Protein crowding tunes protein stability. J. Am. Chem. Soc. 133, 7116. [47] Politou, A. and Temussi, P.A. (2014) Revisiting a dogma: the effect of volume exclusion in molecular crowding. Curr. Opin. Struct. Biol. 30, 1. [48] Spitzer, J. and Poolman, B. (2013) How crowded is the prokaryotic cytoplasm? FEBS Lett. 587, 2094. [49] Bourgeois, D., Vallone, B., Schotte, F., Arcovito, A., Miele, A.E., Sciara, G., Wulff, M., Anfinrud, P. and Brunori, M. (2003) Complex landscape of protein structural dynamics unveiled by nanosecond Laue crystallography. Proc. Natl. Acad. Sci. U.S.A. 100, 8704.

Please cite this article in press as: Kundu, J., et al. Unusual effects of crowders on heme retention in myoglobin. FEBS Lett. (2015), http://dx.doi.org/ 10.1016/j.febslet.2015.11.015