Stability and Formulation of Enzymes in Household Liquid Detergents
Henrik Lund
During the last three decades the addition of enzymes to laundry detergents has become a universal tool to lower the energy consumption and to generate a broad consumer relevant cleaning effect. It is, however, a major obstacle to ensure stability of these enzymes, especially in the liquid detergents, where presence of water results in increased interaction between the enzymes and the other components of the detergent. Particularly anionic surfactants are challenging to formulate together with enzymes, since they promote protein unfolding. Another important component in most laundry detergents are the various types of chelating agents. These have a number of different functions, such as improving the action of the anionic surfactants by binding calcium and magnesium ions (i.e. reducing water hardness). However, some of the most commonly used enzymes within the detergent industry (i.e. amylases and proteases) are dependent on bound calcium ions to maintain conformational stability and function; thus, the presence of both chelating agents and enzymes in a liquid detergent presents a stability challenge. Additionally, some enzymes themselves can affect the stability of other enzymes negatively. In particular proteases require special attention due to their (auto)-proteolytic properties.
Although a considerable amount of research has been put into understanding protein-surfactant interactions and protease (auto)-proteolysis, these studies are usually performed in dilute conditions with little resemblance to the physical properties of a fully formulated liquid detergent. As a result, very few studies exist where enzymes are studied under conditions reflecting an actual laundry detergent. One reason is that mechanistic studies are hampered by the difficulties faced with examining enzyme inactivation in a fully formulated detergent, where one or more of the components in the detergent give too much interference in many commonly used analytical techniques to study protein structure and/or stability. The process of formulation therefore involves extensive shelf-life stability studies, in which residual enzyme activities are correlated with formulation variations.
The overall purpose of this thesis is to address some of the issues mentioned above, concerning protein stability in liquid detergents, and how to study this – using three common detergent enzymes: a protease, an alpha-amylase and a lipase. The thesis begins with a general introduction to protein stability in liquid detergents, highlighting some of the challenges one may encounter in this particular field of enzyme application. After a brief overview of the enzymes used and their function, a more in-depth description of liquid laundry detergents follows. Next, a chapter concerning general principles of protein stability is presented. This is followed by a chapter where highlights from current literature on protein –surfactant interactions, with emphasis on liquid laundry detergents, are reviewed. Finally, preceding the Results section, a description of some of the commonly used analytical techniques considered in the present work is presented.
Of these techniques, differential scanning calorimetry (DSC) was found to be particularly useful and was subsequently used in a number of applications by predicting the impact of individual detergent components on enzyme functionality. Thus, we found a strong linear correlation between DSC data, in particular Tmax (temperature at peak maximum of the transition () from the folded to unfolded protein state) and enzyme activity studies with correlation values: 0.98 (protease), 0.99 (amylase) and 0.98 (lipase), respectively. Thus, a higher Tmax for the same enzyme in a particular formulation is directly proportional to longer storage stability. These results suggest a new way of greatly accelerating this type of formulation studies. It can allow estimation of enzyme compatibility with a specific formulation on a daily, rather than the present weekly or monthly, basis.
Additionally, the three commonly used Ca2+ chelating compounds citrate, DTPA (diethylene triamine pentaacetic acid) and HEDP (1-hydroxyethylidene-1, 1-diphosphonic acid) were studied with regard to their impact on protease and amylase stability in buffer and in a model liquid detergent. Enzyme stability was characterized by differential scanning calorimetry (DSC) and activity studies and correlated to the chelator Ca2+ interaction properties. The results show that a chelating agent’s ability to reduce water hardness and its Ca2+ affinity are in reality two separate aspects in the context of their use in liquid detergents. In the presence of DTPA, which binds Ca2+ strongly, stoichiometric surplus of free Ca2+ is required to re-establish amylase and protease stability to that of a formulation without DTPA. For the weaker chelators HEDP and citrate, enzyme stability depends on the total concentration of Ca2+ and not stoichiometric balancing. The results underline the importance of Ca2+ in liquid detergent formulation and suggest how proper balancing of small amounts of chelating agents and Ca2+ can be used to improve overall enzyme stability.
A common denominator for the three enzymes studied in this thesis is that they have been found to irreversible unfold in liquid detergents. Based on this finding, and by the use of DSC, the thermal unfolding transitions for the amylase and protease were shown to be fully kinetically controlled, following a well described two-state kinetic model. Using the model the kinetic parameters describing the transitions (rate constant (k), activation energy (EA), and Tmax) were determined for each of the enzymes in buffer and detergent-containing solutions. A modelling study was then carried out to determine the possibility of predicting enzyme half-life by extrapolating the kinetic parameters obtained at the transition unfolding temperature to more realistic storage temperatures. Using this approach the stability of the protease was found to be surprisingly predictable when correlated to actual incubation studies, whereas the prediction for the amylase surpassed the actual storage stability data by several orders of magnitude. Possible explanations for the apparent discrepancies are discussed.
Finally, one important, and yet unsolved aspect of protein-surfactant interactions is concerned with the nature of the surfactant species leading to protein unfolding: is it the surfactant monomer or the surfactant micelles that are causing unfolding of proteins? Results obtained with a protease in dilute surfactant solutions below, at, and above the critical micelle concentration (CMC) suggest that the surfactant monomer is the primary unfolding agent. In relation to these results it is shown how co-addition of non-ionic surfactants to concentrated anionic surfactant solutions significantly enhances enzyme thermal stability, presumably by decreasing the concentration of surfactant monomers that can interact with the proteins.
In conclusion, the work presented reviews the complexity involved when dealing with stability of proteins in complex matrices such as liquid laundry detergents. DSC, contrary to many other commonly used protein characterization techniques, can be successfully used to examine the impact of the most important detergent components (i.e. surfactants, proteases, calcium and chelators).
