AmericanChemicalSociety.com

Petroporphyrins are found in all fossil fuels, with concentrations from relatively low in light crude oil, to extremely high in Boscan heavy crude (~10³ ppm) and oil shale (~10⁴ ppm). They exist as metal complexes, primarily with nickel and vanadium, although others (iron, gallium, copper and manganese) are known. These metallo-petroporphyrins pose a significant challenge to the petroleum industry, as their degradation results in the fouling of catalyst beds utilized in oil refining, thus increasing the cost of petroleum products and/or making certain sources of oil unusable. While removal of petroporphyrins is a conceptually easy solution to this problem, the challenge is that there are no known petroporphyrin-specific binding materials or degradation catalysts to accomplish this.

Human heme oxygenase (hHO-1) is a microsomal enzyme that catalyzes the regiospecific oxidation of heme to biliverdin, carbon monoxide, and free iron. It is a relatively small (33 kDa), easily expressed and purified protein that has been the focus of numerous studies given its physiological importance in maintaining iron-homeostasis as the initial enzyme involved in the heme degradation pathway. As free heme is detrimental to cells due to its ability to initiate free-radical chemistry, binding of heme to HO is tight, as evidenced by a K_d of ~0.3 – 1.0 mM (depending on the isoform). Additional studies have also shown the ability of HO to bind other metalloporphyrins, including Ni, Mn, Co, and Sn containing protoporphyrin IX, with similar affinities, making it an excellent starting point for developing a petroporphyrin-specific binding material using modern protein engineering methods such as directed evolution (DE). Although petroporphyrins are not specific targets of heme-binding proteins such as heme oxygenase, evolving this family toward these substrates would yield a new protein family: “petroporphyrin-binding proteins (PBPs).” As heme oxygenase has already demonstrated an ability to bind both heme and non-naturally occurring tetrapyrroles with high affinity, altering substrate specificity from heme to petroporphyrin represents a goal well within the capabilities of direction evolution.

The primary focus of the research performed to-date has been to develop a functional Yeast Surface Display (YSD) directed evolution system to evolve heme oxygenase to specifically bind and catalytically degrade petroporphyrins to facilitate their removal from crude oil. The general approach is as follows: hHO-1 was displayed as a fusion to the Aga2p mating agglutinin protein, which is in turn linked by two disulfide bonds to the Aga1p protein that itself is covalently linked to the yeast cell surface. Both protein expressions (Aga2p-hHO-1 fusion and Aga1p) are controlled by a galactose-inducible GAL1 promoter. As constructed, this specific YSD methodology employs several advantages over other evolution strategies, including i) properly maintaining the yeast cellular machinery; ii) quality control provided by the eukaryotic expression as the yeast endoplasmic reticulum ensures that only properly folded proteins reach the cell surface; and iii) automated screening using Fluorescence Activated Cell Sorting (FACS) to confirm hHO-1 expression. The Aga2p-hHO-1 fusion protein was expressed with HA and c-myc epitope tags. Surface localization of hHO-1 was confirmed by labeling the yeast cells with an anti-c-myc monoclonal antibody and a fluorescent reagent that in turns binds to this antibody. The fluorescent signal exhibited for the hHO-1 YSD samples is evidence that the enzyme is displayed on the yeast surface wall since only the fusion protein is expressed with the terminal c-myc tag. Control experiments lacking the YSD fusion protein confirmed the absence of secondary fluorescent probe binding sites. Isolation of the hHO-1 released from the yeast cell wall was accomplished by treating the cells with dithiothreitol (DTT) to disrupt the disulfide linkages.

To construct the YSD system, the gene sequence of hHO-1 was inserted into the pCTCON vector. The primer designs were selected as to provide significant homology at each end between the insert containing the hHO-1 sequence and the yeast surface display plasmid. To yield higher amounts of the insert, PCR was carried out using the Finnzymes PhusionTM High-Fidelity DNA Polymerase with standard conditions. The insert was purified using PelletPaint®, a commercial methodology based on the typical ethanol precipitation for DNA purification. The digested plasmid and the insert containing the sequence of hHO-1 were obtained, mixed, and taken up by the yeast using electroporation. The EBY 100 yeast strain was utilized for hHO-1 expression, and contains the Aga1 gene integrated under the control of a galactose inducible promoter. Electroporation was performed using 25 mF (capacitance) and 0.54 kV (voltage), which yields an electric field strength of 2.7 kV/cm with 0.2 cm cuvettes (time constant ~ 18 ms). As yeast are able to carry out a homologous recombination, the resulting pCTCON plasmid now containing the hHO-1 gene was reassembled in vivo. In order to obtain a successful hHO-1 expression via YSD, a previously reported multistep expression methodology was performed: After induction, the cells were first incubated for 30 minutes with the non-fluorescent labeling reagent chicken anti-c-myc IgY. A secondary fluorescent reagent, anti-chicken IgG Alexa488, was then added. Once the yeast cells were properly labeled, they were analyzed using FACS, a high-throughput library screening tool used for the rapid analysis of thousands of variants by employing a fluorescent probe. Here, FACS confirmed the expression hHO-1 on the surface of the yeast surface wall.

As we have successfully constructed our YSD system, we have now begun to employ error prone PCR in the creation a starter library of hHO-1 variants for automated FACS screening against metallo-petroporphyrins. Taq polymerase has a natural low rate of error insertion at 10^-4 to 10^-5 per base added. However, the addition of a low concentration of Mn²⁺ can raise the AT → TA transversion rate to equal the GC → AT transition rate, while the overall rate of error insertion can also be increased by raising the concentration of MgCl₂ present in the system. These conditions together have increased the error insertion rate to a probability of 0.013 to 0.02 per base added, or minimally 100-fold. This library will be coupled with the FACS methodology described above to provide rapid screening of petroporphyrin-binding to hHO-1 variants using fluorescence quenching.