The Role of Trehalose for the Stabilization of Proteins - The Journal of

Subscriber access provided by HOWARD UNIV

Article

The Role of Trehalose for the Stabilization of Proteins Christoffer Olsson, Helén Jansson, and Jan Swenson J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b02517 • Publication Date (Web): 02 May 2016 Downloaded from http://pubs.acs.org on May 4, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The Role of Trehalose for the Stabilization of Proteins Christoffer Olssona, Helén Janssonb, Jan Swensona* a b

Department of Physics, Chalmers University of Technology, SE-412 96 Göteborg Sweden

Department of Civil and Environmental Engineering, Chalmers University of Technology, SE-412 96 Göteborg Sweden

Abstract Understanding of how the stabilization mechanism of trehalose operates on biological molecules against different types of environmental stress could prove to gain great advancements in many different types of conservation techniques, such as cryopreservation, or freeze-drying. Many theories exist that aim to explain why trehalose possesses an extraordinary ability to stabilize biomolecules. However, all of them just explaining parts of its mechanism and a comprehensive picture is still lacking. In this study we have used differential scanning calorimetry (DSC) and viscometry measurements to determine how the glass transition temperature Tg, the protein denaturation temperature Tden and the dynamic viscosity depend on both the trehalose and the protein concentration in myoglobin-trehalose-water systems, The aim has been to determine whether these physical properties are related and to gain indirect structural insights from the limits of water crystallization at different concentration ratios. The results show that for systems without partial crystallization of water the addition of protein increases Tg, most likely, due to that the protein adsorb water and thereby reduces the water content in the trehalose-water matrix. Furthermore, these systems are generally decreasing in Tden with an increasing protein concentration,

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

and thereby also an increasing viscosity, showing that the dynamics of the trehalose-water matrix and the stability of the native structure of the protein are not necessarily coupled. We also infer, by analyzing the maximum amount of water for which ice formation is avoided, that the preferential hydration model is consistent with our experimental data.

1.

Introduction

Cryopreservation of biological systems is a common protocol in both biological research and medical practices. Cells and tissues are preserved at cryogenic temperatures, typically at the temperature of liquid nitrogen (77 K), which makes it possible for a prolonged storage time for future use in scientific studies and for clinical purposes. One of the crucial steps in such a protocol is the prevention of formation of ice crystals in the intracellular and extracellular milieus, which, if occurred, would severely damage the cells and tissues by fragmenting cellular components. To avoid ice formation most of the water molecules in the specimens have to be replaced by a cryoprotectant, such as dimethyl sulfoxide (DMSO), sugar alcohols like glycerol, or disaccharides like trehalose, maltose and sucrose1, 2. In fact, the accumulation of a cryoprotectant, such as glycerol, in the body is the reason for why, for instance, various types of tree frogs can survive in climates of longer times of subzero temperatures without coldand freezing-induced damages of proteins and protecting membranes3-5 (and references therein). Another example of an organism that naturally produce a cryoprotecting substance, in this case trehalose, in order to survive a harsh environment is the desert plant Selaginella lepidophylla, which is capable of surviving dry periods in a desiccated state, and resurrect when rehydrated6.

2


Page 3 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


An important property of cryoprotectants is that they are excellent glass-formers, also when they are mixed with water. They can thus turn the cellular environment into a glassy state at cryogenic temperatures, and thereby eliminate detrimental effects of ice crystals. On the other hand, the addition of some cryoprotectants might have toxic effects (by introducing osmotic stresses for example) on the “frozen” cells and tissues7. This is also one of the reasons for why cryopreservation is still an active research field. Another reason is the intense studies of stem cells and of the extensive practices of testtube fertilization and organ transplantations, for which cryopreservation is one of the indispensable techniques. The significant clinical benefits with these efforts essentially render the cryopreservation of biological systems as one of the most important areas in the basic and applied studies of biology and medicine.

Although several substances have been used as cryoprotectants the disaccharide trehalose stands out as particularly excellent for a wide range of biologically important storage properties, such as a high recovery of protein functions after cryostorage and an excellent ability to protect lipid bilayers to low and high temperatures (freeze-thaw) or dehydration8, 9. From the literature it is clear that numerous studies have been focused on the molecular mechanism of the protective properties of trehalose, but despite these efforts the topic is still debated (see e.g. Ref. 8-16). Hence, the reason for why trehalose possesses superior protective properties compared to other similar substances remains to be resolved.

One important aspect of trehalose is its particularly excellent ability to form a glass with a high glass transition temperature (Tg) compared to other disaccharides, such as sucrose, raffinose, lactose or maltose17. For sugar concentrations of about 30 wt% (and more) it has also been shown that trehalose 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 33

has an unusually high destructuring effect on the hydrogen bonded network of water, which, in turn, is related to its great ability to prevent water crystallization18, 19. In fact, the high Tg of trehalose and its superior ability to prevent crystallization of water, compared to other disaccharides, was suggested to be the cause of trehalose stabilizing mechanism20. Trehalose has also an ability to regulate the water content of a low water content amorphous trehalose-water matrix by the formation of crystalline dihydrates (thus maintaining a stable Tg)21.

Although the great glass forming properties of trehalose may be a part of the answer to why trehalose is superior in protecting biological molecules against environmental stresses, it appears not to be the complete picture22. This was partly shown by storing a protein in a glassy matrix made of dextran, which gives a higher Tg than trehalose, but nevertheless a higher protein degradation rate than a corresponding sample stored in a trehalose matrix22. Thus, there ought to exist some interaction between a protein and trehalose that allows for the protein to remain in its native state, and thereby avoid denaturing during desiccation, heat shocks, and cold storage11, 22, 23.

From previous studies, see e.g. Refs. 11, 13, 14, 16, 24, 25, it has been suggested that trehalose form structures close to the surfaces of proteins, and that these structures form multiple hydrogen bonds (HB:s) to a water layer on the protein surfaces. In this model11 the water is entrapped in a layer between the protein surfaces and the surrounding trehalose structures. The dynamics of the protein is thus, via the water layer, linked to the more rigid trehalose-water structures, which slow down the protein dynamics 4


Page 5 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


as well as increasing the stability of the protein. However, other studies, e.g. Refs. 12, 22, indicate that trehalose tends to bind directly to the protein surface. This gives a scenario where trehalose molecules replace the surface water (the so-called water replacement model), and through that interaction the protein is stabilized.

In this study these different models of trehalose binding mechanisms, i.e. the water entrapment and water replacement models, are evaluated by use of differential scanning calorimetry (DSC) and viscometry. With DSC the glass transition (Tg) and denaturation (Tden) temperatures of different binary and ternary mixtures of myoglobin, water and trehalose are determined, and viscometry is used to investigate whether the viscosity is related to Tg and Tden. Particularly, we elucidate the role of protein concentration for the stabilizing effect of trehalose in cryopreservation and desiccation. The results indicate that for systems with sufficiently low water concentrations to avoid ice formation, an increasing protein content leads to an increasing stability of the trehalose-water matrix. However, an increase of the protein content in the system also results in a decreasing denaturation temperature, which indicates a destabilization of the protein molecules with increasing protein content. These two effects indicate that the dynamics of the trehalose-water matrix and the stability of the protein are not necessarily correlated for these completely non-crystalline samples. However, for systems that partially crystallize such a correlation was in fact obtained, i.e. the amorphous region of the partially crystalline systems exhibited an increase in Tden as Tg increased. Furthermore, by analyzing the maximum water content for which

5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 33

crystallization occurred, an evidence against water replacement was obtained, and also against a strong water entrapment effect (i.e. water entrapment of multiple water layers).

2.

Materials and Methods

2.1

Sample preparation

For this study lyophilized powdered myoglobin (equine heart) and crystalline trehalose (di-hydrate) was used, both purchased from Sigma Aldrich and used without any further purification. The different systems investigated were based on water:trehalose of molar ratios of 172, 86, 44, 23, 11, and 7. Each of these trehalose-water mixtures also contained between 0 and 70 wt% protein (mProtein/mSample). In table 1 a complete list with details of the sample compositions is shown.

The samples were prepared in a number of parallel and subsequent steps (based on a previously used method10). In the first step, trehalose was dissolved in milli-Q water to a concentration of 20 mM. Due to the relatively low solubility of trehalose at ambient temperatures this step was done under stirring at elevated temperatures. In parallel, lyophilized myoglobin was dissolved, also during stirring, at room temperature to a concentration of 5 mM. In the second step, these both pre-prepared solutions were mixed under stirring over night to obtain a homogenous solution with the desired concentration of trehalose and protein.

6


Page 7 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For the DSC measurements, a certain amount (~20 mg) of each composition was placed in a DSC pan of aluminum and blow-dried at a temperature of around 40 °C until the desired water content was obtained before it was hermetically sealed. The samples used for the viscometry were blow-dried in their respective vials until the desired water content was obtained, and then placed in the viscometer using a syringe.

2.2

Differential scanning calorimetry (DSC) measurements

The DSC measurements were performed using a DSC Q1000 from TA Instruments. Each sample composition was measured multiple times to reduce the sample variability and ensure reproducibility. During a typical DSC scan, the sample was cooled down at a rate of 30 °C/min to -150 °C, at which it was equilibrated for 1-2 minutes. Thereafter the sample was heated up to 100 °C at a rate of 10 °C/min. The glass transition temperature (Tg) was determined on heating and by taking the inflection point of the glass transition step. For some of the samples, which exhibited crystallization during heating, two glass transitions were observed, one at a lower temperature and one that partially overlapped with the melting of the ice. For these samples the determined Tg value refers to the glass transition occurring at the lowest temperature. Denaturation temperatures were determined as the minima of the exothermic dip in the temperature range between 65 °C and 100 °C.

7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2.3

Page 8 of 33

Crystallization limits by DSC

In order to determine at which water content a specific sample composition undergoes partial crystallization, five different sets of sample series were prepared, one with water and trehalose, one with water and protein, and three with water and three different ratios between trehalose and protein (nt/np ratios of 17, 51, and 154 trehalose molecules per protein, as also shown in table 2). These samples were cooled in the DSC to -50 °C with a rate of 30 °C/min and then kept isothermal for 2 minutes before heating with a rate of 10 °C/min up to 40 °C. For each sample series the samples were prepared with a successively reduced water content until there was no sign of any crystallization or melting from the DSC scans.

2.4

Viscosity and density measurements

For six of the systems (sample compositions shown in table 3) with relatively high water contents (> 40 wt% water) the dynamical viscosity was determined by using a rolling-ball viscometer (Lovis 2000 ME, from Anton Paar). Densities were determined simultaneously using a U-tube density-measuring module (DMA 4500 ME, from Anton Paar). Both viscosity and density measurements were performed between 10 °C and 40 °C with a temperature step of 5 °C.

8


Page 9 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.


Results

The resulting DSC curves were divided into three different categories; those where water crystallizes during cooling (see Fig. 1(a)), those where water crystallizes during heating (see Fig. 1(b)), and those where no water crystallizes during the entire scan (see Fig. 1(c)). In the cases when part of the water crystallizes during cooling the systems become phase separated into regions of ice and freezeconcentrated solute-water mixtures. Thus, for these water rich freeze-concentrated solutions the amount of non-crystalline water is independent of the water content for a given protein:trehalose ratio. This amount of non-crystalline water is furthermore similar to the amount of water that is perturbed from its bulk-like hydrogen-bonded network structure by the solutes. At intermediate water contents the samples may avoid crystallization during cooling, but still exhibiting a clear melting peak during heating. This implies that such samples must exhibit ice formation (which also yields a micro phase separation) over a broad temperature range when the samples are heated above Tg. Finally, the samples with relatively low water contents do not crystallize at any temperature, since all water molecules in these samples are more or less affected by interactions with the solutes (i.e. trehalose and protein), and therefore not sufficiently “free” to participate in a tetrahedral network structure of ice.

In order to rule out the possibility that crystallization of trehalose occurred during the DSC scans, the melting enthalpy of each sample that exhibited crystallization/melting was measured and compared to the water concentration. It was found that these enthalpy values were directly proportional to the expected amount of ice for freeze-concentrated solutions. If trehalose had contributed to these melting 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

enthalpies the proportionality had been lost and a deviation from the expected melting enthalpies had been observed. Furthermore, the melting always occurred close to 0°C, which further reassured that it was indeed only water which crystallized and melted.

For the sample categories where there is no crystallization during cooling, shown in Fig. 1(b) and (c), there is no ice at Tg (or Tden) and therefore these samples will be denoted as “non-crystalline” and categorized similarly in the analysis of Tg and Tden. Although crystallization occurs for the samples exhibiting the behavior shown in Fig. 1(b), the ice is formed in the temperature range between Tg and the melting temperature. This implies that the water molecules can be assumed to be similarly homogeneously distributed during the glass transition as in samples which completely lack crystallization, i.e. samples following the behavior shown in Fig. 1(c). The remaining samples, exhibiting the behavior shown in Fig. 1(a), will be denoted as “crystalline” and separated from the “noncrystalline” samples in the analysis of Tg due to the fact that these samples are freeze concentrated solutions at Tg.

When analyzing the denaturation temperature the effect of possible ice formation and melting at considerably lower temperatures should have negligible effect since the samples should have enough time to reach equilibrium after the ice has melted. This was also verified by comparing the denaturation process, as seen by DSC, for samples that were only heated to observe the denaturation with samples (of equal compositions) that were first cooled below their crystallization points and then heated above their 10


Page 11 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


denaturation temperatures. No difference in Tden was observed between these two different types of thermal treatments.

The dependence of Tg on the total wt% of protein (mprotein/msample), trehalose (mtrehalose/msample) and water (mwater/msample) are shown in Fig. 2(a), (b) and (c), respectively, for the non-crystalline samples and in Fig. 3(a), (b) and (c), respectively, for the partially crystalline samples. By comparing the two figures it is evident that there is a large difference between the samples depending on whether crystallization during cooling occurs or not. For the non-crystalline samples there is an increase in Tg, whereas for the partially crystalline samples there is a decrease in Tg,, with an increasing concentration of protein.

Provided that the water content is low enough to avoid any crystallization of water, it is well known that for the binary trehalose-water system an increase in trehalose concentration leads to an increase of Tg, 20, 26

. This is also what is shown in Fig. 2(d). The reason for this is the plasticizing effect of water on

trehalose, which leads to a faster dynamics of trehalose, and subsequently a lower Tg, with increasing water content. In the concentration range in which partial crystallization occurs, the solution becomes freeze concentrated and thereby the concentration of non-crystalline water is independent on the water content, resulting in a constant Tg (at about -40 °C), as shown in Fig. 3(d). Similar findings are observed, for the binary protein-water system (series labeled nw/nt=0 in Fig. 3(a), (b) and (c)), where Tg is independent (about -75 °C) on the protein content in the water concentration range where ice is formed.

11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

For the ternary samples containing ice the decrease in Tg with increasing protein concentration can be explained by the fact that the glass transition of the freeze concentrated protein-trehalose-water solution is a superposition of the glass transitions of freeze concentrated protein solutions (Tg at about -75 °C) and freeze concentrated trehalose solutions (Tg at about -40 °C). This superposition effect is further evident from the width of the glass transition range, which increases continuously with increasing protein concentration from about 10 °C for the binary trehalose-water system to around 70 °C for the binary protein-water system27.

The maximum water contents before water crystallization occurred, either during heating or cooling, were determined for binary samples containing only water and trehalose or only water and protein, respectively, and also for ternary samples containing nt/np ratios of 17, 51, and 154 trehalose molecules per protein. The values are presented in table 2. Hence, the table gives both the water content where the system no longer exhibited crystallization during cooling (i.e. going from Fig. 1(a) to Fig. 1(b)) and no longer exhibited any crystallization (i.e. going from Fig. 1(b) to Fig. 1(c)), respectively. These latter water contents, presented in the left column of table 2, can be considered as a rough estimate of how much water the solutes (i.e. trehalose and myoglobin) are able to perturb from behaving bulk-like. As an example it can be seen that in the binary system containing water and protein, the protein is capable of binding up to 25 wt% water (nw/np ≈ 326) before any water molecules crystallize. By comparing the amount of water the two binary systems can perturb with the ternary mixtures, it can be concluded that

12


Page 13 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


there is no detectable collaborative effect of the three-component protein-trehalose-water system being able to perturb more or less water than expected from the corresponding two-component systems.

The concentration dependences of the denaturation temperature are shown in Fig. 4(a) as a function of wt% protein in the sample and in Fig. 4(b) as a function of wt% trehalose in the sample. From these data it is clearly seen that the denaturation temperature decreases with increasing protein concentration. Although, as previously reported10, the addition of trehalose to water yields a pronounced increase of Tden, an increase of the protein content decreases Tden for a given water:trehalose ratio. In order to determine whether this increase of Tden can be explained by an increasing viscosity, we performed viscosity measurements of samples with different trehalose and protein contents. Dynamic viscosities and densities of six of the samples (at 20°C) are presented in table 3, where it is shown that the viscosity increases for all increments of either trehalose or protein. It is also worth noting that although the viscosity and density of all samples increase with increasing protein content, there is a stronger increase of the viscosity with increasing protein content for samples containing more trehalose.

By combining the analysis of the denaturation temperature with the viscosity study, it is clear that a protein is not necessarily stabilized by an increased viscosity, because even though trehalose increases both Tden and the viscosity, an increasing protein concentration lowers Tden although the viscosity is increased. However, for the highest protein contents investigated in this study it can be seen in Fig. 4 (b)

13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

that this the trend of a decreasing Tden with increasing protein concentration changes for protein contents above 40-50 wt% protein (most clearly seen for the sample series with nw/nt –ratios of 7, 11, and 86).

4.

Discussion

The most valuable structural insights from this study come from the measurements of the maximum water concentration before crystallization occurs, see table 2. For water contents below these critical concentrations it is clear that the solute molecules perturb the structure of water sufficiently to avoid the occurrence of bulk-like water, and consequently minimizes the risk for crystallization at low temperatures. This perturbation of the structure of water is caused both by interactions with the solute molecules as well as by the geometrical confinements effects the solute molecules apply on the water18, 28

. By analyzing how the amount of perturbed water changes with the trehalose:protein ratio it is

possible to gain structural insights about how the water interacts with the protein and trehalose molecules, and maybe even more important, whether water is excluded from the protein surface by a direct protein-trehalose interaction. In our case we found that the binary samples of trehalose in water and protein in water completely avoided crystallization for water contents up to 36 wt% and 25 wt% respectively. If the solute molecules in the protein-trehalose-water system are surrounded by water in a similar way as in the corresponding two-component systems it is expected that the maximum water content should be a weighted average of the values of the binary systems. This is also what is found, as shown in table 2. On the other hand, if, for example, trehalose had displaced water molecules at the protein surface by direct bonding to the protein, then the displaced water would be “free” and bulk-like 14


Page 15 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and thereby undergo crystallization at low temperatures. As a consequence, this would lead to a lower maximum concentration of non-crystalline water in the three-component system than given by a weighted average of the two binary systems, since this would give rise to fewer sites on both types of solute molecules for the water molecules to bind to. This finding excludes the possibility that the water replacement model12, 22 is a good structural description of the protein-trehalose-water system. Rather, our results suggest that a layer of water molecules (with a similar amount of water as in the binary system lacking trehalose) is trapped at the protein surface, and surrounded by a trehalose-water-matrix similar to what is found in the corresponding two-component system without protein. Hence, the protein molecules are preferentially surrounded by water molecules, in consistency with the preferential hydration model, as first proposed by Belton and Gil11. Their FTIR results showed that water molecules adsorb to the protein surface preferentially over trehalose. Similar findings have been obtained from MD simulations by Cottone et. al.13, 14, Lins et.al.16 and Corradini et. al24. Generally the results show that trehalose form structures around protein molecules and that water is entrapped between the structures and the protein surface. However, even if the present study gives no conclusion regarding the existence of such structures, the results show that if such structures exist there is no indication for an excessive water content at the protein surface. Thus, the water molecules entrapped between the protein and the trehalose-water matrix do not seem to exceed the amount of non-crystalline water surrounding the protein in the binary system, because that had most likely resulted in more non-crystalline water than the weighted average of the two binary systems.

15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

Further support for that the local environment of non-crystalline water around the two types of solute molecules appears to be the same as in the two binary systems is found for the partially crystalline samples shown in Fig. 3. Due to the freeze-concentration of these solutions Tg is independent on the water content for a given trehalose:protein ratio, but the figure shows also that for different trehalose:protein ratios Tg is a weighted average value of the Tg-values of the freeze concentrated binary solutions, which are about -75 °C the protein-water system and about -40 °C for the trehalose-water system. This latter finding is the expected concentration dependence of Tg for the ternary system if the local structure of water around the solutes is the same as in the corresponding binary systems. Hence, the Tg-values shown in Fig. 3 provide further support for the structural scenario proposed above.

At lower water contents, where no ice is formed during cooling, Tg exhibits completely different concentration dependences, as shown in Fig. 2. In this case Tg increases with increasing contents of both protein and trehalose, see Fig. 2(a)-(c) for ternary systems and Fig. 2(d) for the binary trehalose-water system. The increase of Tg for increasing trehalose concentration, at a given water:protein ratio, is a natural consequence of the plasticizing effect water has on trehalose and other solutes28, as discussed above. The increase of Tg with increasing protein concentration is more interesting, particularly since Tg of the protein-water system is lower than for the trehalose-water system at maximum concentrations of non-crystalline water. The most plausible explanation of this increase of Tg with increasing protein content is that some of the water in the trehalose-water matrix moves to the added protein molecules, in

16


Page 17 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


accordance with the preferential hydration model discussed above

11,16

. This leads to a lower water

content in the trehalose-water matrix, which, as mentioned above, increases Tg.

The macroscopic viscosity of a system is directly related to the structural relaxation process on a molecular length-scale, and therefore it is directly related to Tg and also thought to be correlated to the biomolecular stability20,

29

. In the case of varying the trehalose content this study confirms these

expected relations by comparing the values of Tg and Tden (see Figs. 2(d) and 4(a), respectively) with the viscosities presented in table 3. The results show clearly that both Tg and Tden, as well as the viscosity, increase with increasing ratio of trehalose to water, in agreement with several previous studies (see for example Ref. 10). An increase of the protein content at a constant water:trehalose ratio causes also an increase of both Tg and the viscosity, as shown in Fig. 2 (b) and table 3, respectively. However, in the case of Tden the trend is generally the opposite for all of the samples except for the one of the highest protein concentrations for which Tden decreases with increasing protein concentration (see Fig. 4(b)). So, the question is why the protein stability can be reduced despite an increase in viscosity? One likely reason for this unexpected behavior is that the protein stability also depends on the trehalose:protein ratio. When the protein content increases each protein molecule will be surrounded by fewer trehalose molecules, and thereby the stabilizing interaction between trehalose and the protein is reduced. The exact nature of the stabilizing effect of trehalose on protein is still unclear, but most likely the trehalose molecules act as a physical barrier between different protein molecules and preventing protein-protein interactions. Thus, if the number of trehalose molecules around each protein molecule decreases, both unfolding and aggregation of the protein simplifies, leading to a reduced stability of the protein, which has been further discussed in more detail in for example Refs. 30-32. However, at higher protein concentrations (~40-50 wt% protein) there are indications for that Tden follows the expected trend of an increase with increasing protein content (most clearly seen for the sample series nw/nt=7, 11, and 86 in 17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 33

Fig. 4(b)) and increasing viscosity. A possible reason for why the trend changes at high protein contents is that both effects, i.e. the stabilizing effect of trehalose and the stabilizing effect of an increase in viscosity, are present and competing with each other. Thus, at low protein concentrations the stabilizing effect of trehalose dominates, leading to a decrease of Tden with increasing protein content, while at high protein concentrations the stabilizing effect of an increase in viscosity dominates (in line with the ideas of Green and Angell20), resulting in an increase of Tden with increasing protein content. Since Tg was found to increase with increasing protein concentration over the whole protein concentration range (in the case of the non-crystalline samples) an anomalous relation between Tg and Tden is observed, as shown in Fig. 5. In this figure we used the minimum values (0) = -76.8°C and (0) = 70.5°C of the binary protein-water system as reference points. The figure shows clearly that there is no simple relation between Tg and Tden, which most likely is due to that Tg for most of the samples (i.e the samples with a substantial concentration of trehalose) is mainly related to the dynamics of the trehalose-water matrix, whereas Tden is determined by the stability of the protein, and these two properties do not necessarily need to be related.

From the present study it is evident that indirect structural information of value for understanding the stabilizing role of trehalose can be gained from a combination of DSC and viscosity measurements. The experimental data exclude the model that the protein is stabilized by direct interaction with the trehalose molecules. Rather, our results suggest that the protein surface is mainly covered by 1-2 molecular layers of water and that the stabilizing trehalose molecules are generally located outside this hydration layer, as illustrated in Fig. 6. Of course, this is only a schematic picture of a plausible structure, and for a more detailed structural model other more direct structural probes are required, such as neutron diffraction experiments in combination with structural modeling. 18


Page 19 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5.


Conclusions

In this study we have shown that substantial insights about a typical model system for cryopreservation can be gained by systemic determinations of how the glass transition temperature Tg, the protein denaturation temperature Tden and the dynamic viscosity depend on the concentrations of the three components myoglobin, trehalose and water. By determining the highest water concentrations before crystallization occur for different protein:trehalose ratios it was possible to exclude the possibility of a water replacement model, where a substantial number of trehalose molecules bind directly to the protein surface. Instead, the protein molecules seem to be hydrated by 1-2 molecular layers of water in a similar way as for proteins in pure water at low temperatures where the excessive water crystallizes. However, despite that the trehalose molecules do not generally interact directly with the protein molecules, by physical bonding, it is evident from studies of how Tden depends on the protein concentration that the trehalose-water matrix has a pronounced stabilizing effect on the protein. This stabilizing effect increases substantially with both the amount of trehalose-water around each protein (for a given water:trehalose ratio) as well as the concentration of trehalose in this surrounding solution. It is clear that a lack of trehalose molecules between neighboring protein molecules, and consequently a high probability of close protein-protein interactions, is detrimental for the stability of the protein. This study has also shown that there is no simple relation between Tg and Tden, because Tg is mainly related to the dynamics of the trehalose-water matrix, whereas the stability of the protein, as determined by Tden, is a much more complex property, depending on several parameters as discussed above. 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Acknowledgement This work was financially supported by the Swedish Research Council.

Corresponding author Jan Swenson, email: [email protected] , phone number: +46 31 772 56 80

20


Page 20 of 33

Page 21 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Captions

Figure 1. Three characteristic DSC curves. The insets show a zoom in of the glass transitions. (a) Sample crystallized during cooling. (b) Sample does not crystallize during cooling but crystallize partially during heating followed by immediate melting. (c) No crystallization during the entire DSC cycle.

Figure 2. Glass transition temperature dependencies for the systems that do not crystallize during cooling (i.e. similar to those in Fig. 1(b) and 1(c)). Tg plotted against the total weight percentage (wt%) of: (a) protein, (b) trehalose, (c) water and (d) trehalose (for samples without protein). Dashed lines are shown as a guide for the eye to illustrate the concentration dependences.

Figure 3. Glass transition temperature dependencies for partially crystalline systems that crystallize during cooling (DSC-curves similar to the example in Fig. 1(a)). Tg plotted against the total weight percentage (wt%) of: (a) protein, (b) trehalose, (c) water and, (d) trehalose (for samples without protein). Dashed lines are shown as a guide for the eye to illustrate the concentration dependences.

Figure 4. Denaturation temperature as a function of (a) protein wt%, and (b) trehalose wt%, for different water:trehalose ratios.

Figure 5. Δ plotted against Δ . Indicating a non-linear correlation between increasing glass transition temperature and protein stability. Δ

=

− (0),

21


− (0) and Δ =


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 33

Figure 6. Schematic structural illustration of the protein-trehalose-water system. The preferential hydration of the protein molecules is illustrated by typically 1-2 molecular layers of water at the protein surface. The trehalose molecules and the remaining water molecules are almost randomly distributed between the hydrated protein molecules.

22


Page 23 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1.

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2.

24


Page 24 of 33

Page 25 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3.

25



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4.

26


Page 26 of 33

Page 27 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5.

27



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6.

28


Page 28 of 33

Page 29 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Tables Table 1. Compositions of samples investigated by DSC. It should here be noted that the water contents include the two water molecules from each di-hydrated trehalose molecule. 1 The samples in this series with the highest amounts of water crystallized during heating. Water molecules per trehalose (nw/nt)

Trehalose molecules per protein

Crystallized during cooling?

172 86 44 23 11

Protein concentration ranges (wt% of total mass): 9 to 50 10 to 70 17 to 50 9 to 41 9 to 44

5 to 51 13 to 93 15 to 78 33 to 233 41 to 326

Yes Yes Yes Yes No1

7 N/A 4 to 121

23 to 60 20 to 67 N/A

41 to 124 N/A N/A

No Yes Partially

Table 2. Maximum water content before different samples begin to crystallize during cooling and heating, respectively. Sample composition

Maximum water content (wt%) before crystallization occurs during heating

Maximum water content (wt%) before crystallization occurs during cooling

Protein and water

25

30

Trehalose and water

36

40

Trehalose:Protein. nt/np=154

33

38


30

35


28

33

29



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 33

Table 3. Viscosity and density at various water:trehalose:protein ratios at 20°C. nw/nt

Water wt%

Trehalose wt%

Protein wt%

Density g/cm3

Dynamic Viscosity mPa s

86 86 86 23 23 23

82 75 61 55 50 45

18 16 14 45 41 38

0 9 25 0 9 17

1.034 1.046 1.086 1.205 1.212 1.220

1.304 1.626 4.971 11.83 19.52 40.44

30


Page 31 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1) Sousa, R. Use of Glycerol, Polyols and Other Protein Structure Stabilizing Agents in Protein Crystallization. Acta Crystallographica Section D 1995, 51, 271-277. (2) Storey, B. T.; Noiles, E. E.; Thompson, K. A. Comparison of Glycerol, Other Polyols, Trehalose, and Raffinose to Provide a Defined Cryoprotectant Medium for Mouse Sperm Cryopreservation. Cryobiology 1998, 37, 46-58. (3) Storey, K.; Storey, J. Biochemical Adaption for Freezing Tolerance in the Wood Frog,Rana Sylvatica. J Comp Physiol B 1984, 155, 29-36. (4) Goldstein, D. L.; Frisbie, J.; Diller, A.; Pandey, R. N.; Krane, C. M. Glycerol Uptake by Erythrocytes from Warm-and Cold-Acclimated Cope’s Gray Treefrogs. J Comp Physiol B 2010, 180, 1257-1265. (5) Rexer-Huber, K. M.; Bishop, P. J.; Wharton, D. A. Skin Ice Nucleators and Glycerol in the Freezing-Tolerant Frog Litoria Ewingii. J Comp Physiol B 2011, 181, 781-792. (6) Adams, R. P.; Kendall, E.; Kartha, K. Comparison of Free Sugars in Growing and Desiccated Plants of Selaginella Lepidophylla. Biochemical systematics and ecology 1990, 18, 107-110. (7) Karlsson, J. O. M.; Szurek, E. A.; Higgins, A. Z.; Lee, S. R.; Eroglu, A. Optimization of Cryoprotectant Loading into Murine and Human Oocytes. Cryobiology 2014, 68, 18-28. (8) Crowe, L. M. Lessons from Nature: The Role of Sugars in Anhydrobiosis. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 2002, 131, 505-513. (9) Chiantia, S.; Giannola, L. I.; Cordone, L. Lipid Phase Transition in Saccharide-Coated CholateContaining Liposomes: Coupling to the Surrounding Matrix. Langmuir 2005, 21, 4108-4116. (10) Bellavia, G.; Giuffrida, S.; Cottone, G.; Cupane, A.; Cordone, L. Protein Thermal Denaturation and Matrix Glass Transition in Different Protein-Trehalose-Water Systems. J. Phys. Chem. B 2011, 115, 6340-6346. (11) Belton, P. S.; Gil, A. M. Ir and Raman Spectroscopic Studies of the Interaction of Trehalose with Hen Egg White Lysozyme. Biopolymers 1994, 34, 957-961. (12) Carpenter, J. F.; Crowe, J. H. An Infrared Spectroscopic Study of the Interactions of Carbohydrates with Dried Proteins. Biochemistry 1989, 28, 3916-3922. (13) Cottone, G.; Ciccotti, G.; Cordone, L. Protein–Trehalose–Water Structures in Trehalose Coated Carboxy-Myoglobin. The Journal of chemical physics 2002, 117, 9862-9866. (14) Cottone, G.; Giuffrida, S.; Ciccotti, G.; Cordone, L. Molecular Dynamics Simulation of Sucrose‐and Trehalose‐Coated Carboxy‐Myoglobin. Proteins: Structure, Function, and Bioinformatics 2005, 59, 291-302. (15) Jain, N. K.; Roy, I. Effect of Trehalose on Protein Structure. Protein Science 2009, 18, 24-36. (16) Lins, R. D.; Pereira, C. S.; Hünenberger, P. H. Trehalose–Protein Interaction in Aqueous Solution. Proteins: Structure, Function, and Bioinformatics 2004, 55, 177-186. (17) Taylor, L. S.; Zografi, G. Sugar–Polymer Hydrogen Bond Interactions in Lyophilized Amorphous Mixtures. Journal of Pharmaceutical Sciences 1998, 87, 1615-1621. (18) Lerbret, A.; Bordat, P.; Affouard, F.; Guinet, Y.; Hédoux, A.; Paccou, L.; Prévost, D.; Descamps, M. Influence of Homologous Disaccharides on the Hydrogen-Bond Network of Water: Complementary Raman Scattering Experiments and Molecular Dynamics Simulations. Carbohydrate Research 2005, 340, 881-887. (19) Magazù, S.; Migliardo, P.; Musolino, A. M.; Sciortino, M. T. Α,Α-Trehalose−Water Solutions. 1. Hydration Phenomena and Anomalies in the Acoustic Properties. The Journal of Physical Chemistry B 1997, 101, 2348-2351. (20) Green, J. L.; Angell, C. A. Phase Relations and Vitrification in Saccharide-Water Solutions and the Trehalose Anomaly. The Journal of Physical Chemistry 1989, 93, 2880-2882. 31



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 33

(21) Kilburn, D.; Townrow, S.; Meunier, V.; Richardson, R.; Alam, A.; Ubbink, J. Organization and Mobility of Water in Amorphous and Crystalline Trehalose. Nat Mater 2006, 5, 632-635. (22) Crowe, J. H.; Carpenter, J. F.; Crowe, L. M. The Role of Vitrification in Anhydrobiosis. Annual Review of Physiology 1998, 60, 73-103. (23) Allison, S. D.; Manning, M. C.; Randolph, T. W.; Middleton, K.; Davis, A.; Carpenter, J. F. Optimization of Storage Stability of Lyophilized Actin Using Combinations of Disaccharides and Dextran. Journal of Pharmaceutical Sciences 2000, 89, 199-214. (24) Corradini, D.; Strekalova, E. G.; Stanley, H. E.; Gallo, P. Microscopic Mechanism of Protein Cryopreservation in an Aqueous Solution with Trehalose. Scientific Reports 2013, 3, 1218. (25) Cottone, G. A Comparative Study of Carboxy Myoglobin in Saccharide−Water Systems by Molecular Dynamics Simulation. The Journal of Physical Chemistry B 2007, 111, 3563-3569. (26) Bellavia, G.; Cordone, L.; Cupane, A. Calorimetric Study of Myoglobin Embedded in Trehalose-Water Matrixes. J. Therm. Anal. Calorim. 2009, 95, 699-702. (27) Jansson, H.; Bergman, R.; Swenson, J. Role of Solvent for the Dynamics and the Glass Transition of Proteins. The Journal of Physical Chemistry B 2011, 115, 4099-4109. (28) Magazù, S.; Migliardo, F.; Telling, M. T. F. Study of the Dynamical Properties of Water in Disaccharide Solutions. Eur Biophys J 2006, 36, 163-171. (29) Köper, I.; Bellissent-Funel, M.-C.; Petry, W. Dynamics from Picoseconds to Nanoseconds of Trehalose in Aqueous Solutions as Seen by Quasielastic Neutron Scattering. The Journal of Chemical Physics 2005, 122, 014514. (30) Lee, S. L.; Hafeman, A. E.; Debenedetti, P. G.; Pethica, B. A.; Moore, D. J. Solid-State Stabilization of Α-Chymotrypsin and Catalase with Carbohydrates. Industrial & Engineering Chemistry Research 2006, 45, 5134-5147. (31) Cao, X.; Wang, Z.; Yang, X.; Liu, Y.; Wang, C. Effect of Sucrose on Bsa Denatured Aggregation at High Concentration Studied by the Iso-Conversional Method and the Master Plots Method. J. Therm. Anal. Calorim. 2009, 95, 969-976. (32) Panzica, M.; Emanuele, A.; Cordone, L. Thermal Aggregation of Bovine Serum Albumin in Trehalose and Sucrose Aqueous Solutions. The Journal of Physical Chemistry B 2012, 116, 1182911836.

32


Page 33 of 33

Maximumwater content before crystallization

= Protein = Trehalose = Water

Wt% water

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Protein:Trehalose wt:wt ratio

TOC Graphics

33


The Role of Trehalose for the Stabilization of Proteins - The Journal of

Recommend Documents