<p>Enzymes are proteins with the capacity of catalysing various reactions. Nowadays two types of enzymes, proteases and lipases, are available for use in detergent formulations for household and industrial laundry washing. Proteases are capable of catalysing the hydrolysis of proteins while lipases enable the hydrolysis of glycerol esters, the main component in fats and non-mineral oils. In this study, two enzymes each representing one of these categories were examined: Savinase <sup>TM</SUP>, an extracellular serine protease of the subtilisin family from <em>Bacillus lentus</em> and Lipolase <sup>TM</SUP>, an extracellular lipase from the thermophilic fungus <em>Humicola Lanuginosa</em> S-38. To <em></em> improve the soil- and stain removal in laundry washing the enzyme has to adsorb onto the soiled fabric and hydrolyse one or more of the stain components. This thesis describes a study on the mechanisms by which these enzymes adsorb at various solid-liquid interfaces from buffered solutions in the absence of other components that are found in common detergents.<p>Chapter 1 gives a general introduction. It contains an overview of the physico-chemical characteristics of the two enzymes and a summary of the present insight into the protein adsorption. In Chapter <em>2</em> the investigation of the adsorption measurement techniques for Savinase is reported together with methods used to inhibit the enzyme with the small organic inhibitor PMSF (phenylmethanesulfonyl fluoride) to suppress autolysis.<p>Several techniques are known to measure the protein concentration in solution and, hence, to determine the adsorption. However, the measurement of a <em>proteolytic</em> enzyme, such as Savinase, at such low concentrations as used in laundry washing systems proved to be a serious complicating factor. Proteolytic enzymes hydrolyses proteins and as the enzyme itself is a protein it is capable of self-destruction or cannibalism, so-called autolysis. Of the four well known methods for measurement of the protein concentration only two proved to give reliable results. The radio-active tracer technique could not be used for the measurement of Savinase adsorption because the unlabelled enzyme adsorbed preferentially over the labelled one. The latter was shown to be degraded by autolysis during the labelling reaction and subsequent storage. The competition between the numerous protein fragments resulted in a surface area- and volume- dependent adsorption comparable to the adsorption from polydisperse polymer solutions. The detection limit of the UV 280 <em></em> nm extinction method was too low to determine the rising part of the adsorption isotherm. The Lowry colouring method gave reliable results but it can not be used to discriminate Savinase in a protein mixture. For the measurement of Savinase adsorption onto protein-soiled surfaces the so-called "AAPF" method proved to be applicable. This method is based on the detection of the proteolytic activity of Savinase on AAPF (N-succinyl-L-Alanyl-L-Alanyl-L-Prolyl-L-Phenylalwiine-p-Nitroanilide). As autolysis of active protease becomes more rapid at higher enzyme concentrations this method is restricted to the concentration regime of 0.1 μg ml <sup>-1</SUP>to 30/μg ml <sup>-1</SUP>.<p>In laundry washing detergent systems the enzymes have to adsorb onto soiled cloth in order to remove stains. These practical "solid" surfaces proved to be very complex as they are mostly porous, irreproducible and difficult to characterise. Therefore, the examination of the adsorption mechanism of both enzymes, the main part of this study, is performed using well characterised, non-porous and non-hydrolysable solid surfaces. The enzyme adsorption at these solid-liquid interfaces has been approached in a similar way as protein adsorption is generally studied. In Chapter 3 the adsorption of Savinase on glass and on several polystyrene latices is reported at various pH-values, ionic strengths and temperatures. The adsorption of Savinase at solid-water interfaces was found to be driven by electrostatic interactions between the surface and the enzyme, dehydration of hydrophobic interfaces and lateral interactions between the adsorbed Savinase molecules. It was also concluded that the enzyme adsorbed as a sphere and did not unfold upon adsorption.<p>As the actual adsorption depends strongly on the electrostatic interactions between the surface and the enzyme the contributions of electrostatic lateral repulsion and dehydration of hydrophobic parts of the surface to the free energy must be of the same order of magnitude. The range of the lateral repulsion extended far beyond the Debye length. The relative importance of the electrostatic interaction resulted in a decrease of adsorption on a negatively charged surface with an increase in pH. Under attractive electrostatic conditions the adsorbed amount decreased with increasing ionic strength. The uptake of an extra Ca2+ ion in the weak calcium binding site of Savinase increased the adsorption at a negatively charged interface, probably because of the increased positive charge on the enzyme. The weak binding site proved to be selective for calcium ions as magnesium ions were not sequestered.<p>The adsorption of Savinase is dynamic, i.e. protein molecules replaced already adsorbed ones, although it was not reversible against dilution. On a hydrophobic surface was it proved to be reversible towards changes in pH and ionic strength. The adsorption of inhibited Savinase was temperature, independent in contrast to that of active Savinase at 30°C which was strongly determined by enzymatic autolysis.<p>In Chapter 4 the influence of the properties of Savinase on its adsorption on PS-latices and glass and on the interaction with chromatographic column materials is examined. Therefore, a set of six closely related protein-engineered variants of Savinase were studied. The variants differed from the naturally occurring Savinase in their electrostatic properties such as the isoelectric point and the number of charged amino-acids. On negatively charged PS-latex differences in electrostatic interaction dominated the differences between the adsorption of the various Savinase variants. The electrostatic properties of the variants could not be described completely on the basis of either the net charge of the protein or the mean surface potential. Altering the primary structure of the protein brought about changes that could not be described properly without specifically considering the distribution of the charged residues over the protein surface. Hydration of the enzyme was a less important factor for the adsorption on hydrophobic PS-latex especially in the case of electrostatic attraction.<p>The adsorption on PS-latex was compared with the retention on chromatographic columns and this proved to be useful in the investigation of the relative importance of electrostatic and hydrophobic interactions for protein adsorption. The retention of the Savinase variants on the hydrophobic interaction column (Alkyl-Superose <sup>r</SUP>) showed that the retention time was inversely related to the polar - non-polar area ratio of the protein. The retention of the variants on a cation exchange column could not be described satisfactorily with the mean surface potential of the protein. A first examination of the relation between charge distribution over the protein and retention on a negatively charged column showed that strongly localised alterations of the potential could not be expected to lead to differences in retention.<p>In Chapter 5 the adsorption of Lipolase on glass and various polystyrene latices was examined. The approach was similar to that for Savinase, reported in Chapter 3. The adsorption mechanism of Lipolase also proved to be comparable to that of Savinase. Electrostatic interaction and dehydration of hydrophobic parts of the surfaces are the main driving forces. Under attractive electrostatic interaction between the surface and the enzyme the plateau value of the isotherm corresponds to a monolayer coverage (2.3 mg m <sup>-2</SUP>). Under experimental conditions in which dehydration of the hydrophobic surface is almost compensated by electrostatic repulsion the lateral repulsion between the adsorbed enzymes becomes important and determines the surface coverage of Lipolase. Just as in the case of Savinase it was concluded that Lipolase did not unfold significantly upon adsorption.<p>In the final chapter the examined experimental techniques and the insight into the driving forces for Savinase adsorption obtained, are applied to the adsorption of Savinase at practical solid-liquid interfaces such as polyester, cotton and cotton artificially soiled with the protein BSA (Bovine Serum Albumin). These practical solid surfaces proved to be much more complex than the non-porous, non-hydrolysable, surfaces that were used in the previous chapters. A preliminary characterisation of these interfaces, based upon the adsorption of well known proteins, a determination of the time dependence of Savinase adsorption and the measurement of the isotherms under various conditions have been carried out. The specific surface area of polyester was determined to be 0.27 m <sup>2</SUP>g <sup>-1</SUP>by adsorption of the protein Lysozyme. Lysozyme adsorption on polyester reached equilibrium within 1 hour. Under electrostatically attractive conditions the maximum adsorption of Savinase on polyester corresponds to monolayer coverage. The affinity of Savinase for the polyester surface was determined by electrostatic interaction together with dehydration of the hydrophobic interface.<p>By Lysozyme adsorption measurement, the surface areas of clean and BSA-soiled cotton were determined to be 15 m <sup>2</SUP>g <sup>-1</SUP>and 9 m <sup>2</SUP>g <sup>-1</SUP>, respectively. However, the actual values may differ by as much as 50% from the ones determined in this way as Lysozyme adsorption proved to be surface-specific. Adsorption equilibrium was reached after 8 hours contact time. During this period protein diffuses into the hollow cotton fibres. The adsorption of Savinase on clean cotton resembled that on glass and is mainly determined by electrostatic interaction between the enzyme and the surface. The uncertainty in the surface areas of clean and BSA-soiled cotton prevented the determination of possible preferential adsorption of Savinase onto protein soiled cotton.<p>The adsorption of active and PMSF inhibited Savinase on cotton artificially soiled with BSA differed very strongly even after short contact times. Although further experimental evidence is necessary we conclude that the lower adsorption on BSA-soiled cotton of active Savinase compared to that of PMSF inhibited Savinase was (partially) caused by the decreased affinity of the enzyme for the altered interface. Alteration of the solution by an increased concentration of hydrolysed BSA fragments did not significantly reduce the adsorption.<p>The adsorption mechanism of Savinase and Lipolase indicated above can be compared with that reported for other proteins. In the literature protein adsorption in general is described as governed by a delicate balance between four categories of interactions namely protein-surface interactions (electrostatic and Van der Waals interactions), the dehydration of the protein and solid surfaces, the structural alteration of the protein and the lateral interaction between the adsorbed proteins. As a first step, proteins can be divided into two categories, "soft" and "hard", according to the resilience against structural alterations (as discussed in Chapter 1). The results reported here show that Savinase and Lipolase are even "harder" (more rigid) than the so-called "hard" proteins examined in the literature. The adsorption of "hard" proteins, such as RNase and Myoglobin, onto like charged hydrophobic surfaces is caused by their flattening upon adsorption. This flattening is the cause of an increase in the dehydrated surface area on the interface. This interaction is strong enough to compensate for the electrostatic and lateral repulsion. The rigidity of both enzymes examined here is thought to obstruct this flattening therewith limiting the dehydrated surface area. The absence of structural alterations and the relatively small contribution of dehydration make Savinase and Lipolase adsorption very sensitive to the other two important interactions: the electrostatic protein-surface interactions and the lateral interaction between the adsorbed proteins.
|Qualification||Doctor of Philosophy|
|Award date||28 Oct 1992|
|Place of Publication||S.l.|
|Publication status||Published - 1992|
- two-phase systems