Lisa R. Wilken

Recovery and purification of plant-made recombinant proteins

Plants are becoming commercially acceptable for recombinant protein production for human therapeutics, vaccine antigens, industrial enzymes, and nutraceuticals. Recently, significant advances in expression, protein glycosylation, and gene-to-product development time have been achieved. Safety and regulatory concerns for open-field production systems have also been addressed by using contained systems to grow transgenic plants. However, using contained systems eliminates several advantages of open-field production, such as inexpensive upstream production and scale-up costs. Upstream technological achievements have not been matched by downstream processing advancements. In the past 10 years, the most research progress was achieved in the areas of extraction and pretreatment. Extraction conditions have been optimized for numerous proteins on a case-by-case basis leading to the development of platform-dependent approaches. Pretreatment advances were made after realizing that plant extracts and homogenates have unique compositions that require distinct conditioning prior to purification. However, scientists have relied on purification methods developed for other protein production hosts with modest investments in developing novel plant purification tools. Recently, non-chromatographic purification methods, such as aqueous two-phase partitioning and membrane filtration, have been evaluated as low-cost purification alternatives to packed-bed adsorption. This paper reviews seed, leafy, and bioreactor-based platforms, highlights strategies for the primary recovery and purification of recombinant proteins, and compares process economics between systems. Lastly, the future direction and research needs for developing economically competitive recombinant proteins with commercial potential are discussed.

Evaluation of monoclonal antibody and phenolic extraction from transgenic Lemna for purification process development.

Several pharmaceutical protein products made in transgenic plant hosts are advancing through clinical trials. Plant hosts present a different set of impurities from which the proteins must be purified compared to other expression hosts such as mammalian cells. In this work, phenolic compounds present in extracts of monoclonal antibody (mAb)‐expressing Lemna minor were examined. Two different extraction pHs were evaluated to assess the effect of extraction condition on the concentration of mAb and phenolics in the extracts. The extract prepared at pH 4.5 had an enriched level of mAb relative to native protein when compared to a pH 7.5 extract although similar overall mAb was extracted at either pH. Slightly more mAb was recovered from the pH 3 elution of the pH 4.5 extract run on a MabSelect column than was recovered from the pH 7.5 extract. Phenolic levels in extracts were assessed by spectrophotometry, Folin–Ciocalteu assay and by profiling on RP‐HPLC. The Folin–Ciocalteu assay results did not agree with those obtained by the other two methods. Therefore phenolic levels were quantified by RP‐HPLC comparing the total area of phenolic peaks to those of reference phenolic compounds. The pH 7.5 extract had 22% less phenolics than the pH 4.5 extract. Acidic precipitation of the pH 7.5 extract resulted in further reduction of phenolics originally present in the pH 7.5 extract. The total phenolics present in the extracts were effectively removed by incubation of extracts with a commercially available anion exchange resin, Amberlite IRA‐402. We anticipate that early removal of phenolic compounds will prolong the life of more expensive affinity columns used for the purification of potential pharmaceutical proteins and should therefore be considered in process development involving proteins extracted from transgenic plant hosts. Biotechnol. Bioeng. 2009; 104: 562–571

Factors Influencing Recombinant Human Lysozyme Extraction and Cation Exchange Adsorption

Human lysozyme has numerous potential therapeutic applications to a broad spectrum of human diseases. This glycosidic enzyme is present in tears, saliva, nasal secretions, and milk – sources not amendable for commercial development. Recently, a high expression level of recombinant human lysozyme (0.5% dry weight) was achieved in transgenic rice seed. This paper evaluates the effects of pH and ionic strength on rice protein and lysozyme extractability, as well as their interactions with the strong cation‐exchange resin, SP‐Sepharose FF. The extraction conditions that maximized lysozyme yield and the ratio of extracted human lysozyme to native rice protein were not optimal for lysozyme adsorption. The conditions that gave the highest extracted lysozyme to native protein ratio were pH 4.5 and 100 mM NaCl in 50 mM sodium acetate buffer. At pH 4.5, salt concentrations above 100 mM NaCl reduced the lysozyme‐to‐protein ratio. The best conditions for lysozyme adsorption were pH 4.5 and 50 mM sodium acetate buffer. Lysozyme extraction and subsequent adsorption at pH 4.5 and 50 mM NaCl was an acceptable compromise between lysozyme extractability, adsorption, and purity. The primary recovery of human lysozyme from pH 6 extracts, irrespective of ionic strength, was inferior to that using pH 4.5 with unacceptably low saturation capacities and lysozyme purity. High purity was achieved with a single chromatography step by adjusting the pH 4.5 extract to pH 6 before adsorption. The disadvantage of this approach was the drastically lower saturation capacity compared to adsorption at pH 4.5.

Evaluation of alternatives for human lysozyme purification from transgenic rice: impact of phytic acid and buffer.

Producing economically competitive recombinant human lysozyme from transgenic rice demands an inexpensive purification process for nonpharmaceutical applications. Human lysozyme is a basic protein, and thus, cation exchange chromatography was the selected method for lysozyme purification. Similar to other protein production systems, the identification of critical impurities in the rice extract was important for the development of an efficient purification process. Previous adsorption data indicated that phytic acid was probably responsible for an unacceptably low cation exchange adsorption capacity. In this study, we confirm that reducing phytic acid concentration improves lysozyme binding capacity and investigate alternative process conditions that reduce phytic acid interference. Compared with the previous best process, the adsorption capacity of human lysozyme was increased from 8.6 to 19.7 mg/mL when rice extract was treated with phytase to degrade phytic acid. Using tris buffer to adjust pH 4.5 extract to pH 6 before adsorption reduced phytic acid interference by minimizing phytic acid‐lysozyme interactions, eliminated the need for phytase treatment, and increased the binding capacity to 25 mg/mL. Another method of reducing phytic acid concentration was to extract human lysozyme from rice flour at pH 10 with 50 mM NaCl in 50 mM sodium carbonate buffer. A similar binding capacity (25.5 mg/mL) was achieved from pH 10 extract that was clarified by acidic precipitation and adjusted to pH 6 for adsorption. Lysozyme purities ranged from 95 to 98% for all three processing methods. The tris‐mediated purification was the most efficient of the alternatives considered.

Downstream Processing of Transgenic Plant Systems: Protein Recovery and Purification Strategies

Plant-derived recombinant proteins provide alternatives to proteins produced by mammalian, microbial, and insect cell cultures due to significant upstream technological achievements over the past 5–10 years. Plants offer flexibility in both growth methods (open-field, greenhouse, and bioreactor) and host expression systems (seeds, leaves, and cell culture). The diversity of plant production systems provides for numerous commercial applications of plant-derived recombinant proteins but economic viability must be ensured through high expression levels and scalable manufacturing processes. Initial research efforts in plant biotechnology were focused on expression strategies and, thus, upstream production achievements have not be matched by downstream processing advancements. However, case-by-case extraction studies for numerous recombinant proteins led to the development of plant system-based approaches. Other progress includes the development of pretreatment strategies to improve purification efficiency and to reduce downstream processing costs for purification from green tissue homogenates that contain chlorophyll, phenolics, and active enzymes. In spite of all the progress and positive developments made in the last 10 years, continual research and technological breakthroughs in downstream processing are needed to capitalize on the lower production cost of transgenic biomass. This chapter describes general advantages and disadvantages of seed-, leaf-, and bioreactor-based plant systems and strategies used for primary recovery and purification of recombinant proteins.

Process evaluations and economic analyses of recombinant human lysozyme and hen egg-white lysozyme purifications.

Human lysozyme and hen egg‐white lysozyme have antibacterial, antiviral, and antifungal properties with numerous potential commercial applications. Currently, hen egg‐white lysozyme dominates low cost applications but the recent high‐level expression of human lysozyme in rice could provide an economical source of lysozyme. This work compares human lysozyme and hen egg‐white lysozyme adsorption to the cation exchange resin, SP‐Sepharose™ FF, and the effect of rice extract components on lysozyme purification. With one exception, the dynamic binding capacities of human lysozyme were lower than those of hen egg‐white at pH 4.5, 6, and 7.5 with ionic strengths ranging from 0 to 100 mM (5–20 mS). Ionic strength and pH had a similar effect on the adsorption capacities, but human lysozyme was more sensitive to these two factors than hen egg‐white lysozyme. In the presence of rice extract, the dynamic binding capacities of human and hen egg‐white lysozymes were reduced by 20–30% and by 32–39% at pH 6. Hen egg‐white lysozyme was used as a benchmark to compare the effectiveness of human lysozyme purification from transgenic rice extract. Process simulation and cost analyses for human lysozyme purification from rice and hen egg‐white lysozyme purification from egg‐white resulted in similar unit production costs at 1 ton per year scale.

Aqueous Fractionation of Dry-Milled Corn Germ for Food Protein Production

To improve economics, dry-grind corn ethanol plants are transitioning to fractionation processes to produce higher value coproducts from non-fermentable fractions. Dry-milled corn germ, which contains a significant amount of non-fermentable components (protein, oil, and fiber), is a potential source of higher value food protein. However, the relatively low protein concentration in dry-milled germ currently hinders its use as a starting material for the production of germ protein extracts and concentrates. Germ wet milling is a recently proposed method designed to increase germ protein and oil concentration by water soaking and wet fractionation of dry-milled germ. In this study, the effect of soaking conditions (pH, temperature, time) on germ composition was determined to identify the best conditions for producing germ with high protein content and high protein dispersibility index. The soaking and subsequent processing of dry-milled germ increased protein content from 15 to 21 %, reduced starch from 33 to 9 %, and increased oil content from 16 to 39 %. The effect of soaking and subsequent processing conditions on germ protein quality was also evaluated to determine the suitability of using the defatted germ as the starting material for protein concentrate. Soaking dry-milled germ at 25 °C at either pH 6.0 or pH 9.0 maintained the protein dispersibility index but soaking at pH 3.0 and/or at a higher temperature were detrimental. Results indicated that the yield of corn protein concentrate prepared by isoelectric precipitation from defatted germ flour was significantly affected by the protein dispersibility index.

Analysis of Recombinant Human Serum Albumin Extraction and Degradation in Transgenic Rice Extracts

Transgenic plant systems have successfully been used to express recombinant proteins, including rice seed‐expressed recombinant human serum albumin (rHSA), without the risk of contamination of human pathogens. Developing an efficient extraction process is critical as the step determines recombinant protein concentration and purity, quantity of impurities, and process volume. This article evaluates the effect of pH and time on the extraction and stability of rHSA. The amount of rHSA in clarified extract after 60 min of solubilization increased with pH from 0.9 mg/g (pH 3.5) to 9.6 mg/g (pH 6.0), but not over time as 10 min was sufficient for solubilization. Total soluble protein in extracts also increased with pH from 3.9 mg/g (pH 3.5) to 19.7 mg/g (pH 6.0) in clarified extract. Extraction conditions that maximized rHSA purity were not optimal for rHSA stability and yield. Extraction at pH 3.5 resulted in high purity (78%), however, rHSA degraded over time. Similar purities (78%) were observed in pH 4.0 extracts yet rHSA remained stable. rHSA degradation was not observed in pH 4.5 and 6.0 extracts but higher native protein concentrations decreased purity. Strategies such as pH and temperature adjustment were effective for reducing rHSA degradation in pH 3.5 rice extracts. Low temperature pH 3.5 extraction retained high purity (97%) and rHSA stability. While seed‐expressed recombinant proteins are known to be stable for up to 3 years, the degradation of rHSA was notably extensive (56% within 60 min) when extracted at low pH.

Green microalgae biomolecule separations and recovery

Microalgae biomass has garnered significant attention as a renewable energy feedstock and alternative to petroleum-based fuels. The diverse metabolism of green microalgae species additionally provides opportunities for recovery of products for feed, food, nutraceutical, cosmetic, and biopharmaceutical industries. Recently, the concept of using microalgae as part of a biorefinery model has been adopted in place of refinery methods focused on recovering one target product. This has led to producers exploring co-production of high value and high volume products in an effort to improve process economics. With numerous potential products and applications, the biomass source or specific strain must be carefully selected to accumulate extractable levels of the target molecule(s). It is additionally imperative to understand the morphology and metabolism of the selected strain to cost-effectively manage all stages of commercial production. This review will focus specifically on microalgae in the division of Chlorophyta, or green algae and their extracellular matrices (ECM), potential for commercial products, and finally describe a holistic approach for biomolecule extraction and recovery. Additionally, cell disruption and fractionation methods for recovery of biomolecules for commercial products are highlighted along with an alternative method, aqueous enzymatic processing for multiple biomolecule extraction and recovery from green microalgae. An emphasis is placed on connecting the morphological characteristics of microalgae ECM or organelle membranes to implications on separation and purification technologies.

Commercial Opportunities and Challenges for Protein Products from Corn

To improve economics, dry-grind corn ethanol plants are transitioning to fractionation processes to produce higher value co-products from non-fermentable fractions. Dry-milled corn germ is one non-fermentable fraction that is a potential source of higher value food protein. However, low oil and protein purity currently restrict the use and value of dry-milled germ. Germ wet milling is a novel method developed to improve germ purity (protein, oil) by water soaking followed by wet processing of dry-milled germ. In this study, we determined the effect of soaking conditions (pH, temperature, time) on germ composition to identify the best conditions for producing germ (HPHPG) with high protein and high protein dispersibility index (PDI). The soaking and subsequent processing of dry-milled germ increased protein content from 15 to 21%, reduced starch from 33 to 9%, and increased oil content from 16 to 39%.