Immobilized enzymes have diverse applications in chemical, food, fuel and pharmaceutical processing, as well as emerging applications in therapeutic treatment of disease. Although enzymes are traditionally thought to function only in aqueous solution, many industrially important applications occur in non-aqueous, organic systems. As a specific example, lipases (EC 18.104.22.168) are enzymes that catalyze hydrolytic and transesterification reactions between triglycerides and acyl acceptors. Certain lipases can be used to efficiently catalyze the reaction of triglycerides (e.g. plant oils) with methanol to form biodiesel, a renewable fuel, and glycerol. The glycerol product has applications in pharmaceuticals and cosmetics, and can be used to make bioplastics and other novel products.
The lipase-catalyzed synthesis of biodiesel is particularly suited for an immobilized enzyme strategy. The use of an immobilized lipase eliminates the need for the homogeneous catalyst (e.g. sodium methoxide) that is typically used for biodiesel. The catalyst must be removed from the glycerol stream before it can be further processed. However, traditional immobilization techniques often result in reduced accessibility of the enzyme site to the macromolecular substrate, because the enzyme is immobilized close to the surface.
It has long been known that the addition of a linker or long spacer molecule (a “tether”) between the supporting surface and the enzyme can dramatically improve the activity and stability of enzymes in aqueous solution. However, most commercial linkers are specifically designed to be hydrophilic, improving solubility of the linker-enzyme. This is excellent for such standard enzyme systems, but provides little benefit for enzymes in non-polar environments (e.g. bulk oils or organic solvents). There is clearly a need for improvements in enzyme immobilization techniques for use in non-polar systems. To the best of our knowledge, there have been no studies on the use of long hydrophobic polymers to tether enzymes to surfaces in non-aqueous systems. To address this need, we have developed a method to covalently immobilize lipases from solid surfaces using a hydrophobic polypeptide tether.
Although synthetic polymers such as Pluronic™ surfactants or low molecular weight polyester or polypropylene are obvious choices for a hydrophobic linker, they either lack functional groups at both ends, or are only available as mono- or homo-bifunctional molecules. A heterobifunctional molecule is most desirable for a linker, as it allows a two-step conjugation reaction that prevents polymerization and self-conjugation of the linker and ligand molecules.
Polypeptides are naturally heterobifunctional (possessing both –NH2 and –COOH termini) and can range from extremely hydrophobic and practically insoluble (e.g. poly-L-isoleucine) to very hydrophilic polymers (e.g. poly-L-lysine). Various modifications can be made to the side-chains to further modulate the hydrophobic/hydrophilic properties of a peptide. Further, synthetic homopolymers of amino acids (HPAAs) are available in a variety of molecular weights and composition. Finally, the coupling chemistry for HPAAs is identical to that of the enzyme proteins, thus simplifying the conjugation protocol significantly. Thus, HPAAs can serve as good experimental models to determine whether hydrophobic tethers can improve enzyme activity and stability in non-aqueous systems.
Molecular dynamics simulation and experimental analysis of short HPAAs in aqueous and organic phases indicate that most polypeptides will spontaneously adopt an approximately α-helical conformation in non-aqueous environments. The α-helical rises of HPAA molecules can easily span several nanometers. The helical structure is stabilized by “self-solvation” of the peptide backbone caused by H-bonding between the C=O and N–H groups.
Side-chain charge, terminal-group reactivity, and solubility in organic acids were considered in the choice of amino acids for the HPAA tethers. Amino acids with charged side-chains are intrinsically hydrophilic and have reactive side-chains that would compete with the peptide terminal groups during conjugation. Proline forms very stable PPI and PPII helices, but its N-terminal (2°) pyrrolidine group is only slightly reactive with the coupling reagents used for this study. Polyisoleucine and polyleucine are essentially insoluble in common solvents (except trifluoroacetic acid), and are known to exhibit intrinsic catalytic activity in some cases.
Ultimately, homopolymers of tryptophan, (Trp)n, were chosen for this model experiment because the indole side-chain is relatively hydrophobic. Stacking of the pendant indole rings can also further stabilize the 2° structure of the polytryptophan α-helix in organic solvents. The indole ring also provides strong UV absorbance and dye-binding activity, allowing for easy quantification using standard protein assays (e.g. A280 or BCA). Polyglycine, (Gly)n, was chosen as a control for non-specific effects due to the peptide backbone.
Commercial Candida antartica lipase B (CALB) was dialyzed and covalently immobilized on 1 µm silica microspheres. The silica surface was carboxylated with γ-aminopropyltriethoxysilane (APTES) and succinic anhydride; protein/polypeptide immobilization was performed using standard EDC/NHS coupling chemistry. Treatments included the covalent immobilization of CALB with three different hydrophobic (Trp)n tethers (approximately 4, 10, and 26 nm α-helix length). A polyglycine tether (5 nm non-hydrophobic α-helix) and surface-immobilized enzyme (no tether) served as controls. CALB enzyme was also immobilized by adsorption from aqueous buffer onto C18-modified microspheres (mimicking common commercial preparations). The residual activity following immobilization was analyzed for each treatment in aqueous (hydrolytic activity) and organic phases (transesterification activity). Best-fit parameters for catalytic activity and enzyme deactivation kinetics for each treatment were determined and correlated with tether length.
The hydrolytic activity of immobilized enzyme treatments was determined in aqueous solution by two methods. The first method, commonly used in industry, involves the titration of free butyric acid liberated from tributyrin during hydrolysis. The titration method was compared with the recently described “adrenaline” high-throughput screening (HTS) assay. Free glycerol liberated by the enzymatic hydrolysis of triacetin is reacted with sodium metaperiodate (NaIO4); the residual NaIO4 is back-titrated with adrenaline to form a bright-red adrenochrome.
The organic phase synthetic activity of the immobilized enzyme treatments was determined by two methods. The first method, also an industry standard, uses capillary GC to monitor the enzyme-catalyzed esterification of 1-propanol and lauric acid to form propyl laurate. These results were compared with an HTS assay that utilizes the reaction of 1-propanol with vinyl laurate to form propyl laurate and acetaldehyde. Reaction of the acetaldehyde side-product with NBD-H produces a stable, strongly fluorescent hydrazone compound; the formation of this fluorphore can be followed over time to rapidly determine the transesterification kinetics of a large number of samples.
The use of hydrophobic polypeptide tethers to immobilize enzymes for use in non-aqueous systems has been investigated. Applications for food, pharmaceuticals, and synthesis of fine chemicals are numerous. Efficient production of biodiesel by lipases that have been immobilized by hydrophobic tethers is being studied as a model application. Comparison of the traditional, cumbersome analytical methods with the high-throughput assays will be useful for researchers working with non-aqueous enzyme systems.
The experiments described above provide a simple system to investigate the utility of long hydrophobic, structured tethers for immobilized enzymes in non-aqueous systems. Future research efforts will include the in-situ production of controlled-length hydrophobic tethers for immobilization of lipophilic enzymes in commercial applications by atom-transfer polymerization (ATRP).