Use of Microcalorimetry to Study Protein Stability

Differential scanning calorimetry (DSC) is used in the pharmaceutical, polymer and biotechnology industries to study the effect of temperature on a variety of solids or liquids, including those that are organic, inorganic, synthetic or bio-based. The technique works by measuring the heat flow absorbed by or released from a substance while undergoing heating or cooling at a controlled rate. Thermodynamic data, such as heat capacity, and temperature and enthalpy events, such as phase transitions, conformation changes, chemical reactions, adsorption and desorption, are among the applications for which DSC is used.

This article provides 1) an overview of microcalorimetry system components and 2) results of a study comparing data from various samples to investigate the effect of different processing conditions, compositions and formulations.

Calorimetry sensors

Calorimetry sensors that utilize high-sensitivity thermoelectric materials and state-of-the-art temperature control systems to obtain limits of detection of less than a microwatt are called microcalorimeters, and are able to detect either very small thermal events in pure materials or larger thermal events that have evolved from only one of the elements within a formulation, such as highly diluted protein solutions.

When heated, a protein in solution undergoes unfolding, or denaturation. This has an endothermic effect on a DSC trace because heat is required to break the protein’s intramolecular links. Once unfolded, the protein may aggregate, a reaction that is typically exothermic because it involves the formation of intermolecular links and decreases the protein’s energy.

Microcalorimetry sensors such as those from SETARAM Instrumentation (Hillsborough, N.J.) are based on Peltier elements, the most sensitive thermoelectric elements available. As shown in Figure 1, the sample holding cell is sandwiched between two Peltier elements, permitting continuous heat flow measurement of the sample and its environment. In order to correct the sample signal for the heat capacity of the cell in which it is held, it is coupled to a reference cell, which is also sandwiched between two Peltier elements.

Figure 1 – Schematic of the calorimetric block of a microcalorimetry sensor. The sample holding vessels are fitted into the cavity of a metallic part. This part is sandwiched between two thermoelectric elements that measure the heat flow exchange between the vessels and their surroundings.

The cells shown in Figure 1 are the most commonly used for protein stability studies and have an available sample volume of 850 µL. The Hastelloy C276 cell body contains the sample, and is tightened due to a stopper and elastomer O-ring (up to 20 bar). The cells can be inserted into and removed from the sensor with each new experiment, which facilitates cleaning and helps prevent most problems related to cross-contamination. This is important, particularly with proteins that can aggregate or form gels.

A temperature-control system is a key component of a microcalorimeter. In operation, heat-transfer fluid is pumped into a double-stage thermostat that allows it to be controlled. The double-stage thermostat has two temperature-control loops that provide heating and cooling power and temperature stability sufficient for the instrument to operate at temperatures as low as –40 °C and as high as 200 °C. After being pumped into the thermostat, the heat-transfer fluid is pushed into the calorimetric block containing the sensor, sample and reference cells, which can be brought to the desired temperature. Finally, the heat-transfer fluid is pumped into an expansion tank and can then loop back to the pump.

Impact of formulation on the unfolding of RNase A

The thermally induced structural modifications of a protein depend not only on its origin and extraction and purification conditions, but also on environmental conditions such as pH, salt content and solvent type.

In the present example, ribonuclease (RNase), a type of nuclease that catalyzes the degradation of RNA into smaller components, was tested under varying conditions. Bovine pancreatic RNase A from Worthington Biochemical Corp. (Lakewood, N.J.) was chosen as one of the classic model systems in protein science.

The samples tested were 5 mg/mL aqueous solutions of RNase A, 0.15 M NaCl, pH 7. Urea, in a concentration of 3 M, was added to one of the solutions. About 750 µL of each solution was injected into two distinct µSC Evo cells from SETARAM. For comparison purposes, similar volumes of the corresponding buffer solutions were introduced into the reference cells. After the cells were inserted into the microcalorimeter, they were heated to 10–90 °C at 1 K/min-1.

The results of this test are shown in Figure 2. An endothermic peak associated with the unfolding of the enzyme was recorded at different temperatures. The peak temperature (Tm) corresponding to the point at which half of the proteins in the solution are already unfolded is shown for both peaks. The 10.3 °C shift of Tm to lower values in the presence of urea corresponds to a destabilization of the formulated protein.

Beta-lactoglobulin unfolding, aggregation and decomposition

Globular proteins are widely used in the food and pharmaceutical industries because 1) their unique folded structure allows binding of smaller molecules (such as active principles or flavors) and 2) they are water soluble. Thus they can transport interestingly active but insoluble molecules. Beta-lactoglobulin (β-LG) is the major whey protein found in the milk of cows and sheep. This well-characterized globular whey protein exists mainly in its dimeric form between pH 3 and 7.

Solutions were prepared from commercial β-LG (Sigma-Aldrich, St. Louis, Mo.) in aqueous buffer (pH 4 and 7) with concentrations of 7.5 and 25 g/L and sodium chloride content of 0 or 0.5 M. The samples were heated from 30 °C to 160 °C at 0.5 K/min in standard cells.

As illustrated in Figure 3, a main endothermic peak of unfolding is observed at pH 4.01 and is not significantly affected by the β-LG concentration. In the presence of NaCl, the peak is shifted more than 5 °C (see Table 1), meaning that the protein is stabilized.

At pH 7, the mechanism is more complex, and endothermic thermal effects are detected at high temperatures, in agreement with previous observations.1 This corresponds to the depolymerization of the unfolded protein, leading to the formation of smaller peptides. The exotherms have been attributed to the Maillard reaction between the peptides and polysaccharide impurities in the material. In the presence of NaCl, a thin exotherm occurs during the endotherm of unfolding. This peak is associated with protein aggregation, in particular, to a strong increase in aggregate size.2

Lower protein concentrations

The amount of protein available after synthesis and purification processes is generally small because the tested solutions were highly diluted. Therefore, dividing samples for testing means that even smaller amounts of protein are available. In the present experiment, protein amounts were as low as 160 µg in the cell.

The samples analyzed were 0.2 and 0.5 mg/mL aqueous solutions of superoxide dismutase (SOD) mutant G41D in 20 mM HEPES, with buffer at pH 7.8. Approximately 800 µL of each solution was introduced into two distinct standard cells. Corresponding buffer solutions were introduced into the reference cells. The cells were inserted into the microcalorimeter and heated to 15–100 °C at 1 K/min–1.

Figure 2 – Superimposition of the endothermic peaks corresponding to the unfolding process of the two tested RNase solutions. Significant differences can be observed in terms of peak maximum, i.e., Tm, showing that the enzyme is destabilized in the presence of urea.

The DSC traces in Figure 4 demonstrate an endothermic effect in the 80–90 °C range, corresponding to the unfolding of the enzyme. Both result in significant endothermic peaks with Tm at approximately 84 °C. A peak height as low as –15.9 µW can be detected for the 0.2 mg/mL solution.

Figure 3 – Heat flow versus temperature data of thermal unfolding and aggregation of β-LG under different concentration, pH and salt conditions. On the left, peaks at pH 4.01 exhibit only endotherms linked with the unfolding of the protein and stabilization by the presence of NaCl. On the right, peaks at pH 7 exhibit more complex behavior, with hightemperature degradation and an exotherm of aggregation during the unfolding process in the presence of NaCl.

Thermal stability of antibodies

Antibodies, gammaglobulin proteins found in the bodily fluid of vertebrates, are utilized by the immune system to identify and neutralize foreign objects such as bacteria and viruses. Microcalorimetry can be used to examine the thermally induced unfolding effects of immunoglobulin. The samples analyzed were 10 mg/mL aqueous solutions of various immunoglobulins. Sample (800 µL) was introduced into a µDSC3 Evo batch cell (SETARAM), while a buffer solution was introduced into the reference cell. Both were heated to 30–100 °C at 1 K/min–1.

As reported in the literature,3,4 the CH2 domain of an antibody unfolds at a lower temperature, and the CH3 domain at a higher temperature. Tm for the antigen-binding (Fab) fragment is difficult to predict. In other words, the unfolding process may result in one, two or three peaks.

As shown in Figure 5, the I1G7 unfolding process was more cooperative, the Tm of the first peak being higher. This first peak can be attributed to either the CH2 or the F(ab) domain, while the second peak is associated with the CH3 domain.

Figure 4 – Superimposition of the endothermic peaks corresponding to the unfolding process of the two tested SOD solutions.
Figure 5 – Superimposed DSC data corresponding to the unfolding processes of the tested antibodies. The complex process involves successive unfolding of several domains and is depicted on the DSC signals as shoulders on the main endothermic peak.

Conclusion

Microcalorimetry is a useful technique for studying the unfolding and aggregation effects that occur in aqueous solutions of proteins. As demonstrated in this article, simple data observation and comparison allow determination of the most stabilizing formulation conditions, even in cases involving complex blends.

References

  1. Photchanachai, S. and Kitabatake, N. Heating of β-lactoglobulin A solution in a closed system at high temperatures. J. Food Sci.  2001, 66(5), 647–52.
  2. Unterhaslberger, G.; Schmitt, C. et al. Heat denaturation and aggregation of beta-lactoglobulin enriched WPI in the presence of arginine HCl, NaCl and guanidinium HCl at pH 4.0 and 7.0. Food Hydrocolloids  2006, 20, 1006–19.
  3. Tischenko, V.M.; Abramov, V.M. et al. Investigation of the cooperative structure of Fc fragments from myeloma immunoglobulin G. Biochemistry 1998, 37, 5576–81.
  4. Vermeer, A.W.; Norde, W. et al. The unfolding/denaturation of immunogammaglobulin of isotype 2b and its F(ab) and F(c) fragments. Biophys J.  2000, 79, 2150–4.

Rémi André is director of technology, SETARAM Instrumentation, 7 rue de l’Oratoire, 69300 Caluire, France; tel.: +33 472 102 525; e-mail: [email protected]; www.setaram.com. Link Brown is sales director, SETARAM Instrumentation, Hillsborough, N.J., U.S.A.

Comments