Expression without boundaries: Cell-free protein synthesis in pharmaceutical research

International Journal of Pharmaceutics 440 (2013) 39–47

Contents lists available at SciVerse ScienceDirect

International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm

Expression without boundaries: Cell-free protein synthesis in pharmaceutical research Marco G. Casteleijn, Arto Urtti ∗ , Sanjay Sarkhel Centre for Drug Research, University of Helsinki, Finland

a r t i c l e

i n f o

Article history: Received 5 December 2011 Received in revised form 1 April 2012 Accepted 3 April 2012 Available online 11 April 2012 Keywords: Protein engineering Cell free expression Protein expression High throughput Protein development

a b s t r a c t Proteins are an increasingly important class of new drugs. Pharmaceutical proteins are usually expressed in cell based systems in the development phase and in production, and although cell free methods have recently emerged they have not been used widely for therapeutic protein development or production. Cell free expression methodology is well suited for pharmaceutical protein expression and engineering and will probably become more commonly used in the future. Cell free expression allows protein engineering in high throughput format, flexible strategies for glycosylation and chemical conjugation, and allows easy use of unnatural amino acids as building blocks of proteins. Thus, cell free expression can be used to modify protein solubility, stability, and pharmacokinetics of therapeutic proteins. Likewise, it is potentially useful in protein development for biomaterial matrices, nanoparticles, and vaccines. This review illustrates the potential of cell free expression in pharmaceutical protein research and development while highlighting both advantages and limitations of the method. © 2012 Elsevier B.V. All rights reserved.

1. Introduction As a class of drug molecules, pharmaceutical proteins are important and a growing market. In 2007, approximately 1/6 of the total volume of the 600 billion dollar pharmaceutical industry was protein based and the market is expected to grow 7–15% annually (Walsh, 2010). Proteins are endogenous compounds and thus they are in general inherently less toxic than xenobiotics. Based on the data regarding genomics and bioinformatics, it is clear that there are still tens of thousands of human proteins that are still unknown and part of these proteins may turn out to be important biological modifiers with potential therapeutic utility. The fact of the matter is development of protein-based drugs is less costly and faster than launching small molecular weight drugs based on extensive synthesis and screening programs. Despite this, the development of pharmaceutical proteins is not straightforward. Frequently, endogenous proteins as such are not suitable for therapeutics. The reasons being: the protein may have poor watersolubility, physical or chemical stability problems, non-optimal pharmacokinetic properties (e.g. fast proteolytic degradation, poor distribution to the target tissue, short half-life) or increased efficacy is wanted. This situation addresses the need to optimize the protein structure. In fact, 17 out of 23 new biological entities released

∗ Corresponding author at: Centre for Drug Research, University of Helsinki, Viikinkaari 5 E, 00014 University of Helsinki, Finland. Tel.: +358 9 191 59 636. E-mail address: arto.urtti@helsinki.fi (A. Urtti). 0378-5173/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijpharm.2012.04.005

in the EU and USA in the last 4.5 years were engineered (Walsh, 2010). Pharmaceutical protein development requires the ability to express and purify recombinant proteins having desired pharmacokinetics and physicochemical properties. Several iterative rounds of optimization are needed to confer the ‘drug-like’ status to the protein of interest. Alteration of the genetic sequence and chemical modifications of the protein are usually employed to bring about the desired changes. This is a trial and error process and the challenges can stem from lack of expression, low yields to a totally non-functional aggregated protein or with severely compromised physicochemical properties. New expression systems allowing rapid product development (ability to rapidly screen conditions and constructs with optimal protein expression and desired functionality) in a cost effective manner would be of significant benefit. Most importantly, the system should be able to express functional proteins of human origin that are of therapeutic significance. 2. Cell free protein expression Conventional protein expression methodologies rely on bacterial, yeast, insect and mammalian cell systems. These are mature tools that involve genetic manipulation of the specific expression organism where the bacterium Escherichia coli is still the most common expression system (Chou, 2007). Other major systems are the baculovirus mediated insect cell expression (Hunt, 2005) and the mammalian cell line Chinese hamster ovary cells (CHO)

40

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

Fig. 1. Schematic overview of CFPS. The reaction buffer includes amino acids (aa).

(Hacker et al., 2009). Less commonly used systems for pharmaceutical protein production include fungal, ciliate protozoa, algae, plants, and transgenic animals. The prokaryotic expression systems are amenable to high-throughput expression methodologies but lack the versatility of being able to express mammalian proteins (e.g., those requiring post-translational modification). The eukaryotic expression systems on the other hand are capable of expressing post-translationally modified proteins but are rather difficult to integrate into high-throughput methodologies. Thus, a robust expression system that allows expression of eukaryotic proteins and can be readily integrated into a high-throughput platform is highly desirable towards developmental research in therapeutic proteins. Cell free protein synthesis (CFPS), as the name implies, involves recombinant in vitro protein expression without the use of living cells (Fig. 1). Cellular extracts instead, are used for protein translation. The CFPS gained considerable popularity within projects related to structural proteomics and genomics (Sawasaki et al., 2002b). It drew quick attention from the fact that it could synthesize a protein rapidly with an easier downstream purification process. The system was also efficient to synthesize toxic proteins that would inhibit the cellular machinery and limit the protein expression in cell-based systems. It afforded a greater degree of control on the parameters that influenced protein expression (levels, quality) and therefore could be suitably tweaked for the optimization process. The ‘openness’ of the system allows an addition of components during protein synthesis such as liposomes/detergents or

microsomal membranes for membrane protein stabilization. Compared to the cell based methods, in vitro CFPS is considerably faster since it does not require transfection, cell culture and extensive purification. In cell-based methods, proteins expressed in the cytosol have to be recovered by lysing the cells and methods employed for cell lysis may sometimes result in protein denaturation. Further, the CFPS can be performed on a microliter scale making it ideal for high-throughput applications (mutant screening for optimal physicochemical/functional properties). In this review, we explore the possibilities and prospects of the CFPS as a platform for pharmaceutical protein development. CFPS has immense applications in areas not in scope of this review; we refer the readers to excellent texts on ribosomal display (He and Khan, 2005; He and Taussig, 2007) and protein arrays (Chandra and Srivastava, 2010). Cell free extract derived from E. coli was the first known in vitro protein translation system (Nirenberg and Matthaei, 1961). Eukaryotic cell free systems developed later include rabbit reticulocyte lysates (RRL), wheat germ embryos, insect cell lysates and human cell extracts. The cell free crude extract consists of a buffer system with essential ions and the components of translational machinery. This comprises ribosomes for polypeptide translation, tRNAs, aminoacyl-tRNA synthetases, initiation, elongation and termination factors. The components are essential for correct protein folding. Amino acids, co-factors and energy sources are added to the crude extract as supplements and protein translation starts with the addition of the genetic template (DNA, mRNA). Expression using a DNA template can be carried out in a ‘linked’ or ‘coupled’ mode. In the linked synthesis, DNA needs to be transcribed first, followed by an addition of the reaction mixture to the translation mix; in the coupled mode both steps take place in one pot. Purification of the translated soluble protein (if Histagged) can be achieved by centrifuging the reaction mixture (to remove insoluble materials) followed by loading the supernatant to a Ni2+ -resin column (affinity chromatography). The eluted protein sample may further be subjected to ion exchange and/or size exclusion chromatography depending on the purity obtained from affinity chromatography. Since the pioneering work of Nirenberg and Matthaei (1961), the experimental protocols for CFPS have undergone significant modifications. Problems related to limited efficacy of the translational activity due to depleting energy sources have been overcome by the continuous flow cell free translation method (CFCF), wherein, a solution containing amino acids and energy factor is fed to the translation chamber across a bilayer or a semi permeable membrane. A significant development has been the advent of protein synthesis using recombinant elements – the ‘PURE’ system (Shimizu et al., 2001). In the ‘PURE’ system, cell free translation takes place in the presence of purified recombinant translation factors. This method facilitates efficient protein production, incorporation of unnatural amino acids and easy purification. For the purification of the translated protein from the PURE system, the high molecular weight ribosomes are first removed by ultrafiltration (membrane cut-off 100 kDa). The components of the PURE system being His-tagged can be conveniently captured/removed by passing through a Ni2+ column. This allows easy purification of the protein in the native form without any additional tag. Each CFPS system has its advantages and limitations. The choice of the expression system depends on the properties of the protein and the template used for protein expression. To keep within the scope of this review, here we discuss here the two most popular cell free systems based on the E. coli (prokaryotic) and the wheat germ cell extract (eukaryotic). We highlight specific examples from published reports to demonstrate the utility and versatility of the cell free expression system. E. coli S30 fraction is most commonly used for prokaryotic protein expression. S30 extracts are prepared by modified methods originally developed by Zubay (1973). Extracts prepared from E. coli

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

strains deficient in ompT endoproteinase and ion protease activity are used for expressing DNA templates. E. coli S30 extracts produce higher expression levels of protein with circular template of DNA as compared to in vivo where host encoded repressors limit expression levels (Collins, 1979). DNA sequences cloned in plasmids or lambda vectors containing a T7 promoter can be transcribed with an extract containing T7 RNA polymerase. One main drawback of the conventional E. coli cell free system is the short lifetime of the translation machinery resulting in low protein yields. The E. coli cell free system is also limited by its scope of application. Multidomain eukaryotic proteins tend to miss-fold due to the absence of desired translational machinery that facilitates co-translational domain folding. Eukaryotic cell free systems derived from yeast cells, insect cells, tumor cells, rabbit reticulocytes are capable of translating functional proteins but they suffer from low productivity. The wheat germ cell free expression system derived from plant sources offers a convenient alternative for eukaryotic protein synthesis. Developed by Roberts and Paterson (1973), the conventional system had a short translational life span and also required isolating the mRNA before the translational step. Current methods do not require isolation of mRNA and the transcription and translation processes are ‘coupled’ into a one-pot synthesis. Technological advancements led to the development of the ‘split-primer’ method for polymerase chain reaction (PCR) to minimize artifacts within the transcriptional templates (Sawasaki et al., 2002b). This allows direct use of PCR generated linear cDNAs for translation, a desired advantage over the need to clone cDNAs into an expression vector. In the ‘split-primer’ method, four primers are utilized. Primer-3 is designed such that its 3 end has target specific sequence (5 terminal of open reading frame) and the 5 end has a portion of 5 terminal site of the enhancer sequence. Primer-2 comprises the full-length enhancer sequence and a portion of the promoter sequence at its 5 end. Primer-1 contains a portion of the 5 sequence of the promoter with an overlapping sequence at 3 site to form a base pairing at the 5 site of primer-2. Primer-4 is specific for sequence at the 3 UTR of the vector. In the first PCR reaction, the target gene is amplified using primer-3 and primer4. This is followed by a second PCR reaction whereby the amplified target gene is equipped with the enhancer and promoter sequences. Primer-1, primer-2 and primer-4 are utilized in this step. The ‘split primer’ approach mitigates the problem of non-specific amplification and ensures the expression of one complete and active mRNA. The ‘split primer’ method does not require optimization of reaction conditions for individual genes and thus allows high-throughput construction of DNA templates directly from E. coli cells harboring the cDNAs of interest. This is of considerable advantage compared to the time consuming steps in conventional cloning methods. Translation efficiency and stability of mRNA could be enhanced by eliminating the 5 -7 mGpppG (cap) and poly(A)-tail of mRNAs. The ‘bilayer reaction’ was developed to make the system operational in a continuous flow mode. In this method, the substrate mixture was overlaid onto the translation mixture (Sawasaki et al., 2002a). The efficiency of the system could be explained as this arrangement allowed continuous supply of substrates and removal of small byproducts by diffusion through the phase between the translation mixture and substrate mixture. The longevity was ten times longer than the conventional batch mode reaction resulting in a better protein yield. Improvements in the wheat germ cell system have been achieved by blocking the endogenous inhibitory pathways of the translation machinery originating from the endosperm (Madin et al., 2000). Wheat embryos with high stability and activity can be obtained by removing the endosperm contaminants through extensive washing. The cell extract can be stored in a lyophilized form for a long time without compromising its activity. The wheat cell system being of plant origin is free of biohazard and bioethical issues, while also limiting bio-pollution to a minimum.

41

2.1. Incorporation of unnatural amino acids The ability to incorporate unnatural amino acids is an important tool in protein engineering efforts that are directed towards therapeutic applications. In a recent study, Cho et al. (2011) incorporated the non-natural analog p-acetylphenylalanine (pAcF) into distinct locations of the genetic sequence of human growth factor (hGH) to allow for site-specific conjugation with polyethylene glycol (PEG). PEGylated hGH demonstrated improved pharmacodynamics in GH-deficient rats. In a similar approach, replacing methionine with its non-natural analogs allowed site-specific pegylation of interferon ␤ (Allozyne web site) with better dosing convenience and superior tolerability than that of existing agents for the treatment of multiple sclerosis. Although the cited examples relate to therapeutic proteins expressed in cellular systems, they have been highlighted here to draw attention to the fact that such modifications effecting an ‘expansion of the genetic code’ could be easily carried out in the cell free expression system (Hirao et al., 2002). Non-natural analogues bearing reactive side chains have immense potential in structural and functional proteomics. The ability to engineer physicochemical properties of proteins with relative ease would truly expand the utility of the cell free system towards expression and production of proteins of pharmaceutical relevance. 2.2. Disulfide bond formation Due to low cost of production, it is highly desirable to be able to carry out protein expression in E. coli based systems (cell or cell free). One major drawback of E. coli based methods has been the inability to express disulfide bond containing proteins. Since many proteins of pharmaceutical relevance have multiple disulfide bonds that are required for proper folding, this poses a serious limitation on the applicability of the E. coli based system. Kim and Swartz (2004) and Yin and Swartz (2004) successfully demonstrated expression of 40 ␮g/ml urokinase and 60 ␮g/ml plasminogen activator, respectively. They carried out the batch reaction by pretreating E. coli cell extract with iodoacetamide, adding optimized ratios of oxidized and reduced glutathione (control of redox potential) and the disulphide-bond forming enzyme (DsbC). In a recent report, Goerke and Swartz (2008) have highlighted the versatility of the cell free method by expressing a set of nine complex proteins with multiple disulfide bonds. 2.3. Multidomain eukaryotic proteins with high content of A/T and low complexity sequences A major issue of concern in malaria vaccine discovery research is the lack of an ideal recombinant protein synthesis system that can express A/T rich and low complexity sequences of the malaria genome. The faster rate of peptide growth on ribosomes of prokaryotes as compared to eukaryotes hampers the correct folding of the multidomain proteins. Consequently, the expressed proteins have limited solubility and functionality. Multidomain proteins in general when expressed in either E. coli cell based or cell free systems showed tendency to be misfolded and become insoluble (Netzer and Hartl, 1997). Both the multidomain and A/T rich (low complexity) sequence aspects pose a major limitation towards recombinant expression of malarial proteins not only in E. coli based systems (Aguiar et al., 2004; Vedadi et al., 2007) but also in eukaryotic expression systems such as yeast, baculovirus and CHO cells (Tsuboi et al., 2008). The wheat germ cell free expression system has proved extremely useful in successfully expressing 478 out of 567 malarial protein targets in soluble form (Tsuboi et al., 2010). The wheat germ system proved highly suitable for A/T rich malaria genes, not requiring codon optimization. Further exploring its versatility

42

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

towards expression of eukaryotic multidomain low complexity region sequences is certainly a matter of further research. 2.4. Expression of protein complexes, toxic proteins and membrane proteins The cell free expression system is of significant benefit for expressing protein complexes consisting of hetero subunits as it allows co-translation of multiple mRNAs to co-translate and form the active protein complex. Matsumoto et al. (2008) successfully expressed active yeast tRNA methytransferase by co-translating the mRNAs for protein subunits Trm8 and Trm82. Efforts to express the individual protein subunits and then reconstitute the complex, did not yield an active Trm8–Trm82 heterodimer. Restriction endonucleases present significant challenges for cellular recombinant expression. PabI, a restriction enzyme from the hyper-thermophilic archae Pyrococcus abyssi, is cytotoxic to the cells upon recombinant expression. Strategies to overcome the toxicity by using tightly repressible expression systems and expression of the cognate methyltransferase were not successful. PabI was successfully expressed in the wheat germ cell free system in the native and SeMet-labeled forms for structural studies by Xray crystallography (Watanabe et al., 2010). Another toxic protein, the human microtubule binding protein (MID1) causing the Opitz syndrome due to specific mutations, is a difficult target for recombinant expression. Mutants of MID1 are difficult to express because they interfere with the cellular metabolic pathways and inhibit cell division. Expression systems such as E. coli, Pichia pastoris, insect cells and mammalian COS cells failed to produce MID1 protein. The rapid translation system (RTS) cell free synthesis based on E. coli S30 extract (marketed by Roche pharmaceuticals), was successful in producing MID1 (Betton, 2003). Membrane proteins constitute nearly 30% of the mammalian genome and are physiological drug targets for more than half of the available ones. Consequently, membrane proteins are subjects of intense investigation in structural, functional proteomics and pharmaceutical research. However, significant challenges remain associated with regards to their recombinant expression and purification. Suboptimal bacterial lipid composition limits the proper insertion and folding of eukaryotic membrane proteins in the E. coli expression system. Overexpression might lead to issues of toxicity and aggregation. Also, extraction of proteins from membrane and reconstituting into membrane-like-‘mimetics’ (e.g., detergents, lipids) to obtain a functional protein is often cumbersome and difficult. Cell free systems provide an attractive alternative to synthesize membrane proteins either as precipitates or translated directly in the presence of surfactants or preformed liposomes. E. coli multidrug transporters (EmrE, SugE, TehA and YfiK) were expressed using a modified E. coli S30 extract at high levels as precipitates and then solubilized in detergent micelles (Klammt et al., 2004). Cell-free expression of proteins facilitated labeling with 15 N isotope for nuclear magnetic resonance (NMR) studies. EmrE when expressed in the presence of detergent, localized predominantly in the soluble fraction (Elbaz et al., 2004). In a recent report, wheat germ cell free extract was utilized to translate human stearoylCoA desaturase in the presence of unilamellar liposomes (Goren and Fox, 2008). The integral membrane protein was incorporated within the liposomes (>90% of total protein). Co-translation of the desaturase with human cytochrome b5 resulted in both proteins being transferred into the liposomes. An important class of drug targets is comprised of the single largest membrane proteins family: the G-protein coupled receptors (GPCRs). Mechanistic understanding of the signaling pathways mediated by the GPCRs is of immense importance towards drug discovery and development. Overexpression of GPCRs in E. coli generally leads to accumulation of this protein as inclusion bodies, and

the expression level attained of functional GPCRs in mammalian or insect cells is usually quite low. Ishihara et al. (2005) utilized the E. coli S30 extract to express the human ␤2 adrenergic receptor (␤2AR), the human M2 muscarinic acetylcholine receptor (M2) and the rat neurotensin receptor (NTR) as a fusion with thioredoxin at their N-terminus. The proteins were obtained in amounts of 150–200 ␮g/ml in only 1 h of the reaction time with 30 ␮l reaction mixture; in the absence of the fusion partner (thioredoxin) no protein expression was detected. In a similar approach, porcine vasopressin receptor type 2 (V2R) containing a small amino terminal T7 tag could be synthesized in a continuous exchange cell free set up producing 3 mg/ml of reaction mixture (Klammt et al., 2005). We have briefly reviewed the CFPS highlighting its utility with numerous examples. Currently, pharmaceutical protein expression is mostly dependent on cell-based systems. The CFPS despite its advantages (summarized in Table 1) have not been utilized widely for pharmaceutical protein research primarily due to cost related issues. However, with the projected growth of the biopharmaceutical industry, we believe that the CFPS would emerge as a viable alternative expression system amenable to high throughput protein engineering and recombinant synthesis of difficult targets. 3. Protein engineering in high throughput systems Advances in the development of pharmaceutical proteins could greatly benefit by adapting high throughput (HTP) and protein engineering methodologies. Here we will review HTP strategies in relation to protein development. Importantly, the cell-free approach is suitable for protein engineering using HTP expression. 3.1. Principle of high throughput expression Classic protein production has been a result of simple expression strategies, which more often than not, are based on trial and error. For example, in protein production using the bacterium E. coli, these efforts were summarized in a consensus protocol consisting of over 10,000 cases (Graslund et al., 2008). However, this not a tailored solution, while for other expression systems such a protocol is not available. Overall, one has to consider the limitations of recombinant technologies: (i) production of proteins that do not affect the physiology of the cells, (ii) cloning of a gene into a suitable vector/transformation, (iii) growing of cells and induction of gene product, and (iv) the quality control of correct folding and functionality (Endo and Sawasaki, 2003; Georgiou and Segatori, 2005). As highlighted earlier, cell free systems omit the first three problems. In order to move to higher throughput in protein production, three issues need to be addressed: (i) finding the right genetic construct(s) which facilitates the expression of a properly folded (active) protein more swiftly; this includes evaluating mRNA properties, (ii) obtaining enough protein needed for evaluation, and (iii) a purification strategy for HTP purifications and subsequent large scale purification (not always). A parallel strategy based on small volume expressions is often adapted (Fig. 2), mostly in combination with automated liquid handling, sampling, and protein specific inline evaluations. In the following sections these three issues are discussed in more detail. 3.2. Genetic constructs Multi-parallel protein expression is needed to increase the throughput in order to overcome the exponential increase for the production of poorly characterized proteins (Hunt, 2005) prior to scaling up (Peti and Page, 2007). Since HTP is rarely needed, most technologies provide modest, but significant increase in throughput. Traditionally, genetic constructs are expression plasmids that transfect eukaryotic cells or transform bacterial cells, and can

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

43

Fig. 2. Process of high throughput protein expression. In the grey boxes are cellbased expression specific required steps. The dashed arrow indicates pre- or postpurification protein assays that can be performed in cell lysates due to post-expression chemical modification to the protein (most often not possible in cell-based systems). Example of a fusion partner (or tag) for: (1) imaging, (2) enhance protein solubility/expression, and (3) purification.

be generated by several cloning strategies. An overview by Hunt (2005) of widely used systems for the rapid cloning for the creation of hundreds of genes nicely summarizes the pros and cons. In short: while efficient, reversible and directional cloning is straight-forward in systems such as the Gateway system (Invitrogen, Carlsbad, USA) or the Ligase Independent System (Merck Biosciences, Nottingham, UK), one major drawback is the introduction of additional amino acids in the genetic construct due to the recombination event needed to create the genetic construct. This is not an issue while using the TOPO cloning system (Invitrogen) or the In-fusion system (BD Biosciences, Palo Alto, USA), however both systems are expensive, and even more so in a HTP setup. Additionally, the backbone of the genetic construct used for protein production is of importance. Obviously the plasmids have an origin, the promoter region and the ribosomal binding site (RBS) which are of importance for protein production (Kraft et al., 2007). Furthermore, the choice of the selection marker could influence glycosylation (Greve et al., 1982; Popolo et al., 1986) or hinder the cell growth, thus reducing protein yields (Neubauer et al., 2007). Another element of the expression plasmid is the choice to include a fusion partner, or tag, to the protein, i.e. the attachment of a polypeptide or protein as an extension of the backbone of

the protein (Lesley, 2001; Peti and Page, 2007). These fusion partners provide functionality either for imaging (detection), increasing expression or to enhance the solubility of the expressed protein (Fig. 2). For example, imaging tags could be applied to visualize whether soluble protein has been expressed in the small cultivation volume, while a solubility/purification tag such as MBP can overcome the formation of insoluble aggregates during protein folding. MBP, and the more often used polyhistidine tag of 2–16 residues (His-tag), can be used as a purification tool, in combination with HTP technologies, such as 96 well under-pressure chamber/96 well column format or with use of magnetic beads and a 96 well magnetic plate holder. If needed these fusion partners can be cleaved of by site-specific proteases for example: TEV protease, thrombin or factor Xa (Jenny et al., 2003). Other examples are the Green Fluorescent Protein (GFP) for detection and the Strep-Tag for purification. The uses of tags and removal of them is nicely described in reviews by Hunt (2005) and Arnau et al. (2006). Cell free expression systems can make use of the above mentioned genetic systems, for example the Gateway system has been applied to produce over 33,000 proteins in CFPS (Goshima et al., 2008). In general these strategies and genetic constructs are used to generate initial plasmid libraries, including synthetic genes.

44

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

Additionally, genes of interest stored in expression plasmids can be copied by means of PCR, including a promoter of choice, and added to a CFPS system to initiate protein synthesis. PCR can also be used to produce fusion proteins for CFPS (Ahn et al., 2011). PCR methods allow expression of the same gene in different cell free systems without the need for re-cloning or recreating (randomized) engineered protein libraries. By diluting a library to a single DNA molecule using the SIMPLEX method one molecule of DNA can be brought to expression by use of CFPS (Koga et al., 2003). Another advantage of the open nature of CFPS and protein secretion systems is the ease of chemical conjugations or tag cleavage in the reaction mixture, media or purification columns, while the most commonly used cell based systems, mammalian and E. coli, need cell lysis. 3.3. Protein yields High protein yields are important in HTP approaches since parallel expression screening often requires small volumes (Fig. 2), and even though some performance and functionality assays need only nanogram amounts of protein, in general more than one assay is needed to evaluate the functionality of the protein in terms of pharmaceutical use and safety. Thus performance and functionality screening often needs micrograms of functional protein. Overall, the choice of the expression system is based on the available means of the laboratory or production facility, modification needs of the protein produced, and on the “best-in-class bioprocess-compatible product gene control system” – wish list as proposed by Weber and Fussenegger (2007). Most items on this list address cell-based specific problems, cell survival, and optimal genetic constructs, and are avoided by the use of CFPS. However, CFPS uses DNA or mRNA as the inducer molecule to trigger protein production, and thus template preparation is important. Due to advances in lowering the energy substrate costs, large quantities of high quality purified DNA is now the limiting factor in terms of cost and production (Carlson et al., 2011). One solution may be DNA rolling circle amplification (Kumar and Chernaya, 2009) or the use of HTP vector preparation techniques (e.g. the Gateways system). In addition, the use of unstructured translation-initiation sequences or “universal mRNA sequences” enables protein synthesis in cell extracts from multiple organisms (Swartz, 2009), which addresses earlier problems for eukaryotic CFPS. In terms of protein yield, either as yield per cell or total yield, the product formation must be determined experimentally, and may either be growth associated or non-growth associated. The specific product rate formation, qp (kg(product) kg−1 (cells) h−1 ) is given by the formula: qp = qs Yp/s , where qs (kg(substrate) kg−1 (cells) h−1 ) is the specific substrate rate and Yp/s (kg kg−1 ) is the yield coefficient describing how much substrate is converted to product. The specific product rate formation can also be expressed in terms of growth associated or non-growth association by means of the Luedeking–Piret model: qp = ˛ + ˇ. Here  is the specific growth rate and ˛ and ˇ are constants that determine the extent of growth association; ˇ = 0 gives a complete growth association, and ˛ = 0 gives a complete non-growth association (Enfors and Häggström, 2000). Protein product formation is mainly growth associated, at least in wild type cells. A low growth rate, for example in E. coli, may hinder product formation due to the maintenance cost of the cell. On the other hand, fast product formation may hinder the correct folding of proteins. These problems are related to protein production in living cells, but not to CFPS. We have compared the protein expression yield of cell based system versus CFPS (Table 2) in regards to the expression of biologically active granulocyte-macrophage colony-stimulating factor (GM-CSF). Not only are the yields much higher in CFPS, but the process is also current good manufacturing practice compliant,

applicable for antibody production (yield ranging from 300 to 1000 mg/L), and scalable from 0.3 up to 100 L (Zawada et al., 2011). The scalability of E. coli cultivations in the range of microliters to one hundred liter has been only addressed recently for E. coli expressions (Siurkus et al., 2010), but here the overall yields were lower compared to CFPS. In regard to production cost, CFPS has become competitive when compared to the cheapest cell based system E. coli. In a recent review by Carlson et al. (2011) an overview of yield expressed as mg/$ substrate has been included. Here the protein yield per production cost is presented based on the prize of nucleotides, secondary energy source, and required enzymes for energy regeneration. In recent studies (Calhoun and Swartz, 2005; Zawada et al., 2011) production cost in E. coli based CFPS were reduced approximately 4000 fold compared to a decade ago (Kim and Swartz, 2000; Spirin et al., 1988). In comparison production cost in terms of media components for E. coli (3.3 mg/$ substrate), P. pastoris (0.62 mg/$ substrate), and Drosophila S2 (0.008 mg/$ substrate), as presented by Vermasvuori et al. (2009), are respectively: ∼4-, 225-, and 1750-fold higher than recent CFPS examples. Based on these few examples CFPS has become competitive with E. coli based protein production. This major advancement has resulted in the first cell-free GMP facility at Sutro Biopharma (South San Francisco, USA) for the production of pharmaceutical proteins (Swartz, 2012). In conclusion, the concerns about cell free expression systems stated in the review by Hunt (2005), in regards of scalability, flexibility, and efficiency, have been recently addressed. The cell free systems circumvent several cell growth related issues related to protein yields and fulfill all requirements of the best-in-class bioprocess-compatible product gene control system” – wish list (Weber and Fussenegger, 2007). Therefore, in respect to final protein yield CFPS is the best in class, and now that costs of reagents to perform CFPS are competitive with cell based systems, the CFPS may slowly replace cell based systems in pharmaceutical protein research and production.

3.4. Purification strategy The recovery of protein from the expression system of choice starts at a genetic construction level, as pointed out earlier. In classic biochemistry, protein purification is protein dependent, based on its size, hydrophobicity, isoelectric point, specific binding to an immobilized ligand, and its relative stability in relation to pH, temperature and salt concentration. Therefore, protein purification using classic techniques is a trial-and-error process and protein specific. By use of fusion partners to the protein, specific purification tags can be utilized to harmonize the purification of proteins with the same tag. However, though this process is more straightforward, a protein specific protocol still needs to be optimized in most cases. Overall protein purifications can be done on several targets at the same time in these ways, but true HTP purifications on a massive scale are still tedious and costly. Therefore, screening of several lead candidates via a protein specific assay, before optimizing the purification protocol, is still needed, and is often the bottleneck in acquiring proteins with wanted properties.

4. Pharmaceutical protein engineering HTP approaches, combined with cell-free expression systems have the capacity to test a large number of protein variants against a given end-point, thereby enabling more efficient protein optimization. In the following paragraphs the importance of pharmaceutical protein engineering is demonstrated with examples.

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

45

Table 1 Comparison of the different cell free protein expression systems.

Yield Glycosylation Large protein Membrane protein Endogenous proteins Reaction condition Disulphide bonds Cost Biohazard

E. coli

Rabbit reticulocytes

Wheat germ

Insect cells

Human cells

High No No Yes No 37 ◦ C Yes Low Yes

Low Yes Yes Yes Yes 30 ◦ C No High Yes

High No Yes Yes No 4–30 ◦ C Yes Low No

Low Yes Yes Yes Yes 27–30 ◦ C Yes High Yes

Low Yes Yes Yes Yes 30 ◦ C Yes High Yes

Protein engineering is often used as a tool when investigating protein stability (Eijsink et al., 2004), altering or improving the catalytic properties of enzymes (Jestin and Kaminski, 2004), and understanding its functionality (Nardella et al., 2004; Russell, 2000; Schliebs et al., 1997). One concern is the fundamental difference between “laboratory stability” of small pure proteins that unfold reversibly and completely at high temperatures, and “industrial stability”, which is usually governed by partial unfolding followed by irreversible inactivation (e.g. aggregation). Moreover, directed evolution (applying iterative rounds of randomizing parts or the whole gene to alter functional elements of a protein) is a trade-off between the desired effect and stability of the enzymes (Hamamatsu et al., 2006) although later reports by Reetz and coworkers seem to circumvent this effect by iterative, saturation mutagenesis (Reetz et al., 2006, 2009). These reviews clearly show the difficulty of de novo design of proteins. An excellent overview of protocols on protein engineering can be found in the ‘Protein Engineering Handbook’, by Lutz and Bornscheuer (2008). Several examples of protein engineering can be found in relation to pharmaceutical proteins: Proleukin, recombinant IL-2 mutant Cys125Ser (Kim et al., 2007; Walsh, 2010) and Pr Kepivanc, a delta KGF mutant, Wild Type (WT) minus 23 N-terminal amino acids (Hsu et al., 2006), are substantially more stable than the WT. A classic example of rational protein design is the increase in activity by monomerisation of insulin, hereby avoiding dissociation by dilution. In the WT the insulin-hexamers and insulin-dimers dissociate by dilution from the injection site (Humalog® Eli Lilly, Noble et al., 1998). Other examples include: (i) a DNAseI with 35 fold higher activity involved in the treatment of cystic fibrosis by increasing the local electrostatic attraction towards DNA by adding six positively charged residues (Pan and Lazarus, 1998); (ii) IFN-␣ con-1 is an in silico designed 88% homolog of human IFN-␣-2 where 20 amino acids were altered with a higher activity than WT (Ozes et al., 1992), which was the basis for even more active variants (Brideau-Andersen et al., 2007); and (iii) increasing the plasma half life has been engineered by fusing therapeutic proteins to albumin (e.g. IFN-␣-2b, single chain antibodies (MM-111) (Elsadek and Kratz, 2012), insulin, hirudin, growth hormone (Chuang et al.,

2002)). Clean protein constructs are also very important in order to avoid immunogenicity (Morihara et al., 1979; Zuppinger et al., 1987). In therapeutic antibody research over the last 25 years several humanized (clean constructs), chimeric, and IgG related products have seen the light (Beck et al., 2010). Currently bispecific mAb, recombinant polyclonal antibodies (e.g. Sym001, Symphogen A/S), and newly engineered protein scaffolds (e.g. Ecallantide, Kalbitor/DX-88, Dyax) are among the new strategies to expand this specialized field of protein engineering (Walsh, 2010). Future challenges include increasing productivity of current cell lines (i.e. alternative systems), more stable IgGs, and proving the clinical efficacy of new protein scaffolds. CFPS has the potential to fulfill these needs (Zawada et al., 2011). Glycoengineering of proteins is another increasingly important field in pharmaceutical protein development (Walsh, 2010). Many advances are currently being made to the expression systems (Durocher and Butler, 2009) to enable site-specific modifications of glycoproteins (Henderson et al., 2011), hyper-glycosylation (Pisal et al., 2010), and subsequent chemical conjugations (Walsh, 2010). Future regulation of glycosylated pharmaceutical proteins may change towards requirements of a more uniform protein production and controlled glycoengineering process. Here we envision the use of cell lysates of various cell based systems already applied for glycoengineering, which could be either native- or metabolically engineered-cell lines. In a recent study a prokaryote-based translation/glycosylation CFPS was developed, which helps to avoid glycoprotein heterogeneity (Guarino and Delisa, 2011). Proteins produced in CFPS can also be modified posttranslationally by chemical methods. This field of research is relevant for optimizing pharmaceutical proteins, but outside the scope of this review. For further reading on this subject several insightful reviews have been published (for example: Canalle et al., 2010). Concluding, there is an apparent need for many pharmaceutical proteins to be modified. Here the open nature and applicability for HTP platforms of CFPS seems ready to be applied in pharmaceutical protein development and production.

Table 2 Comparison of example protein yields in different expression systems. Protein b

mGM-CSF rhGM-CSFc rhGM-CSF rhGM-CSF rhGM-CSF rhGM-CSF Anti-IL13R␣1 Fab antibody Anti-IL-23 scFv a b c

Cultivation mode

Cultivation mediaa

Final yield [mg protein/L media]

References

P. pastoris; fed-batch S. cerevisiae; batch Insect cell (Sf9); batch E. coli; batch E. coli; fed-batch CFPS; batch CFPS; batch CFPS; batch

Basal salt–glycerol media YNB− trp/YPD TNM-FH M9 Complex (E. coli cell lysate) (E. coli cell lysate) (E. coli cell lysate)

200–250 50–60 11–45 20 400 700 300 1000

Sainathan et al. (2005) Price et al. (1987) Chiou and Wu (1990) Libby et al. (1987) Das et al. (2011) Zawada et al. (2011) Zawada et al. (2011) Zawada et al. (2011)

Compositions of the media are described in the corresponding references. Mouse-granulocyte-macrophage colony-stimulating factor. Recombinant human-granulocyte-macrophage colony-stimulating factor.

46

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47

5. Conclusions and outlook Pharmaceutical proteins are an important class of drug molecules, often in need of engineering during development for optimal functionality and physiochemical properties. Due to the potential for HTP, cost-effectiveness and high-level protein production, CFPS offers major advantages in protein drug development. Particularly, the high yield and wide tailoring possibilities of cell free expression are advantages compared to the cell based systems. Some challenges remain, such as: synthesizing any biologically active protein in a universal system, the lack of low-cost scalable eukaryotic CFPS systems and creating proteins containing glycosylation patterns resembling human profiles. However, since major challenges such as yield, cost effectiveness, scalability, production of membrane proteins and proteins with disulphide bridges have been addressed in recent years, we expect the remaining challenges to be tackled in the years to come. By applying a combination of CFPS, HTP and protein engineering, the chances of obtaining a therapeutic protein with optimal properties such as efficacy, stability and solubility, can only be increased. The superb ability to control and modify CFPS systems has led to the establishment of the first pharmaceutical CFPS production facility. We expect that this trend will continue and that CFPS will be a true alternative to cell based protein expression systems. References Ahn, J.H., Keum, J.W., Kim, D.M., 2011. Expression screening of fusion partners from an E. coli genome for soluble expression of recombinant proteins in a cell-free protein synthesis system. PLoS ONE 6, e26875. Aguiar, J.C., LaBaer, J., Blair, P.L., Shamailova, V.Y., Koundinya, M., Russell, J.A., Huang, F., Mar, W., Anthony, R.M., Witney, A., Caruana, S.R., Brizuela, L., Sacci, J.B., Hoffman, S.L., Carucci, D.J., 2004. High-throughput generation of P. falciparum functional molecules by recombinational cloning. Genome Res. 14, 2076–2082. Arnau, J., Lauritzen, C., Petersen, G.E., Pedersen, J., 2006. Current strategies for the use of affinity tags and tag removal for the purification of recombinant proteins. Protein Expr. Purif. 48, 1–13. Beck, A., Wurch, T., Bailly, C., Corvaia, N., 2010. Strategies and challenges for the next generation of therapeutic antibodies. Nat. Rev. Immunol. 10, 345–352. Betton, J.-M., 2003. Rapid translation system (RTS): a promising alternative for recombinant protein production. Curr. Protein Pept. Sci. 4, 73–80. Brideau-Andersen, A.D., Huang, X., Sun, S.C., Chen, T.T., Stark, D., Sas, I.J., Zadik, L., Dawes, G.N., Guptill, D.R., McCord, R., Govindarajan, S., Roy, A., Yang, S., Gao, J., Chen, Y.H., Skartved, N.J., Pedersen, A.K., Lin, D., Locher, C.P., Rebbapragada, I., Jensen, A.D., Bass, S.H., Nissen, T.L., Viswanathan, S., Foster, G.R., Symons, J.A., Patten, P.A., 2007. Directed evolution of gene-shuffled IFN-alpha molecules with activity profiles tailored for treatment of chronic viral diseases. Proc. Natl. Acad. Sci. U.S.A. 104, 8269–8274. Calhoun, K.A., Swartz, J.R., 2005. An economical method for cell-free protein synthesis using glucose and nucleoside monophosphates. Biotechnol. Prog. 21, 1146–1153. Canalle, L.A., Lowik, D.W., van Hest, J.C., 2010. Polypeptide–polymer bioconjugates. Chem. Soc. Rev. 39, 329–353. Carlson, E.D., Gan, R., Hodgman, C.E., Jewett, M.C., 2011. Cell-free protein synthesis: applications come of age. Biotechnol. Adv., http://dx.doi.org/10.1016/ j.biotechadv.2011.09.016. Chandra, H., Srivastava, S., 2010. Cell-free synthesis-based protein microarrays and their applications. Proteomics 10, 717–730. Chiou, C.J., Wu, M.C., 1990. Expression of human granulocyte-macrophage colonystimulating factor gene in insect cells by a baculovirus vector. FEBS Lett. 259, 249–253. Cho, H., Daniel, T., Buechler, Y.J., Litzinger, D.C., Maio, Z., Putnam, A.M., Kraynov, V.S., Sim, B.C., Bussell, S., Javahishvili, T., Kaphle, S., Viramontes, G., Ong, M., Chu, S., Becky, G.C., Lieu, R., Knudsen, N., Castiglioni, P., Norman, T.C., Axelrod, D.W., Hoffman, A.R., Schultz, P.G., DiMarchi, R.D., Kimmel, B.E., 2011. Optimized clinical performance of growth hormone with an expanded genetic code. Proc. Natl. Acad. Sci. U.S.A. 108, 9060–9065. Chou, C.P., 2007. Engineering cell physiology to enhance recombinant protein production in Escherichia coli. Appl. Microbiol. Biotechnol. 76, 521–532. Chuang, V.T., Kragh-Hansen, U., Otagiri, M., 2002. Pharmaceutical strategies utilizing recombinant human serum albumin. Pharm. Res. 19, 569–577. Collins, J., 1979. Cell-free synthesis of proteins coding for mobilisation functions of ColE1 and transposition functions of Tn3. Gene 6, 29–42. Das, K.M., Banerjee, S., Shekhar, N., Damodaran, K., Nair, R., Somani, S., Raiker, V.P., Jain, S., Padmanabhan, S., 2011. Cloning, soluble expression and purification of high yield recombinant hGMCSF in Escherichia coli. Int. J. Mol. Sci. 12, 2064–2076. Durocher, Y., Butler, M., 2009. Expression systems for therapeutic glycoprotein production. Curr. Opin. Biotechnol. 20, 700–707.

Eijsink, V.G., Bjork, A., Gaseidnes, S., Sirevag, R., Synstad, B., van den Burg, B., Vriend, G., 2004. Rational engineering of enzyme stability. J. Biotechnol. 113, 105–120. Elbaz, Y., Steiner-Mordoch, S., Danieli, T., Schuldiner, S., 2004. In vitro synthesis of fully functional EmrE, a multidrug transporter, and study of its oligomeric state. Proc. Natl. Acad. Sci. U.S.A. 101, 1519–1524. Elsadek, B., Kratz, F., 2012. Impact of albumin on drug delivery – new applications on the horizon. J. Control. Release 157, 4–28. Endo, Y., Sawasaki, T., 2003. High-throughput, genome-scale protein production method based on the wheat germ cell-free expression system. Biotechnol. Adv. 21, 695–713. Enfors, S.O., Häggström, L., 2000. Bioprocess Technology Fundamentals and Applications, 1st ed. Högskoletryckeriet, Royal Insitute of Technology (KTH), Stockholm. Georgiou, G., Segatori, L., 2005. Preparative expression of secreted proteins in bacteria: status report and future prospects. Curr. Opin. Biotechnol. 16, 538–545. Goerke, A.R., Swartz, J.R., 2008. Development of cell-free protein synthesis platforms for disulfide bonded proteins. Biotechnol. Bioeng. 99, 351–367. Goren, M.A., Fox, B.G., 2008. Wheat germ cell-free translation, purification, and assembly of a functional human stearoyl-CoA desaturase complex. Protein Expr. Purif. 62, 171–178. Goshima, N., Kawamura, Y., Fukumoto, A., Miura, A., Honma, R., Satoh, R., Wakamatsu, A., Yamamoto, J., Kimura, K., Nishikawa, T., Andoh, T., Iida, Y., Ishikawa, K., Ito, E., Kagawa, N., Kaminaga, C., Kanehori, K., Kawakami, B., Kenmochi, K., Kimura, R., Kobayashi, M., Kuroita, T., Kuwayama, H., Maruyama, Y., Matsuo, K., Minami, K., Mitsubori, M., Mori, M., Morishita, R., Murase, A., Nishikawa, A., Nishikawa, S., Okamoto, T., Sakagami, N., Sakamoto, Y., Sasaki, Y., Seki, T., Sono, S., Sugiyama, A., Sumiya, T., Takayama, T., Takayama, Y., Takeda, H., Togashi, T., Yahata, K., Yamada, H., Yanagisawa, Y., Endo, Y., Imamoto, F., Kisu, Y., Tanaka, S., Isogai, T., Imai, J., Watanabe, S., Nomura, N., 2008. Human protein factory for converting the transcriptome into an in vitro-expressed proteome. Nat. Methods 5, 1011–1017. Graslund, S., Nordlund, P., Weigelt, J., Hallberg, B.M., Bray, J., Gileadi, O., Knapp, S., Oppermann, U., Arrowsmith, C., Hui, R., Ming, J., dhe-Paganon, S., Park, H.W., Savchenko, A., Yee, A., Edwards, A., Vincentelli, R., Cambillau, C., Kim, R., Kim, S.H., Rao, Z., Shi, Y., Terwilliger, T.C., Kim, C.Y., Hung, L.W., Waldo, G.S., Peleg, Y., Albeck, S., Unger, T., Dym, O., Prilusky, J., Sussman, J.L., Stevens, R.C., Lesley, S.A., Wilson, I.A., Joachimiak, A., Collart, F., Dementieva, I., Donnelly, M.I., Eschenfeldt, W.H., Kim, Y., Stols, L., Wu, R., Zhou, M., Burley, S.K., Emtage, J.S., Sauder, J.M., Thompson, D., Bain, K., Luz, J., Gheyi, T., Zhang, F., Atwell, S., Almo, S.C., Bonanno, J.B., Fiser, A., Swaminathan, S., Studier, F.W., Chance, M.R., Sali, A., Acton, T.B., Xiao, R., Zhao, L., Ma, L.C., Hunt, J.F., Tong, L., Cunningham, K., Inouye, M., Anderson, S., Janjua, H., Shastry, R., Ho, C.K., Wang, D., Wang, H., Jiang, M., Montelione, G.T., Stuart, D.I., Owens, R.J., Daenke, S., Schutz, A., Heinemann, U., Yokoyama, S., Bussow, K., Gunsalus, K.C., 2008. Protein production and purification. Nat. Methods 5, 135–146. Greve, J.M., Salzmann, G.S., Roller, R.J., Wassarman, P.M., 1982. Biosynthesis of the major zona pellucida glycoprotein secreted by oocytes during mammalian oogenesis. Cell 31, 749–759. Guarino, C., Delisa, M.P., 2011. A prokaryote-based cell-free translation system that efficiently synthesizes glycoproteins. Glycobiology, http://dx.doi.org/10.1093/glycob/cwr151. Hacker, D.L., De Jesus, M., Wurm, F.M., 2009. 25 years of recombinant proteins from reactor-grown cells – where do we go from here? Biotechnol. Adv. 27, 1023–1027. Hamamatsu, N., Nomiya, Y., Aita, T., Nakajima, M., Husimi, Y., Shibanaka, Y., 2006. Directed evolution by accumulating tailored mutations: thermostabilization of lactate oxidase with less trade-off with catalytic activity. Protein Eng. Des. Sel. 19, 483–489. He, M., Khan, F., 2005. Ribosome display: next-generation display technologies for production of antibodies in vitro. Expert Rev. Proteomics 2, 421–430. He, M., Taussig, M.J., 2007. Rapid discovery of protein interactions by cell-free protein technologies. Biochem. Soc. Trans. 35, 962–965. Henderson, G.E., Isett, K.D., Gerngross, T.U., 2011. Site-specific modification of recombinant proteins: a novel platform for modifying glycoproteins expressed in E. coli. Bioconjug. Chem. 22, 903–912. Hirao, I., Ohtsuki, T., Fujiwara, T., Mitsui, T., Yokogawa, T., Okuni, T., Nakayama, H., Takio, K., Yabuki, T., Kigawa, T., Kodama, K., Yokogawa, T., Nishikawa, K., Yokoyama, S., 2002. An unnatural base pair for incorporating amino acid analogs into proteins. Nat. Biotechnol. 20, 177–182. Hsu, E., Osslund, T., Nybo, R., Chen, B.L., Kenney, W.C., Morris, C.F., Arakawa, T., Narhi, L.O., 2006. Enhanced stability of recombinant keratinocyte growth factor by mutagenesis. Protein Eng. Des. Sel. 19, 147–153. Hunt, I., 2005. From gene to protein: a review of new and enabling technologies for multi-parallel protein expression. Protein Expr. Purif. 40, 1–22. Ishihara, G., Goto, M., Saeki, M., Ito, K., Hori, T., Kigawa, T., Shirouzu, M., Yokoyama, S., 2005. Expression of G protein coupled receptors in a cell-free translational system using detergents and thioredoxin-fusion vectors. Protein Expr. Purif. 41, 27–37. Jenny, R.J., Mann, K.G., Lundblad, R.L., 2003. A critical review of the methods for cleavage of fusion proteins with thrombin and factor Xa. Protein Expr. Purif. 31, 1–11. Jestin, J.L., Kaminski, P.A., 2004. Directed enzyme evolution and selections for catalysis based on product formation. J. Biotechnol. 113, 85–103. Kim, D.M., Swartz, J.R., 2000. Prolonging cell-free protein synthesis by selective reagent additions. Biotechnol. Prog. 16, 385–390.

M.G. Casteleijn et al. / International Journal of Pharmaceutics 440 (2013) 39–47 Kim, D.M., Swartz, J.R., 2004. Efficient production of a bioactive, multiple disulfidebonded protein using modified extracts of Escherichia coli. Biotechnol. Bioeng. 85, 122–129. Kim, S.W., Kim, J.B., Lee, W.S., Jung, W.H., Ryu, J.M., Jang, H.W., Jo, Y.B., Jung, J.K., Kim, J.H., 2007. Enhanced protease cleavage efficiency on the glucagon-fused interleukin-2 by the addition of synthetic oligopeptides. Protein Expr. Purif. 55, 159–165. Klammt, C., Löhr, F., Schäfer, B., Haase, W., Dötsch, V., Rüterjans, H., Glaubitz, C., Bernhard, F., 2004. High level cell-free expression and specific labeling of integral membrane proteins. Eur. J. Biochem. 271, 568–580. Klammt, C., Schwarz, D., Fendler, K., Haase, W., Dotsch, V., Bernhard, F., 2005. Evaluation of detergents for the soluble expression of alpha-helical and betabarrel-type integral membrane proteins by a preparative scale individual cell-free expression system. FEBS J. 272, 6024–6038. Koga, Y., Kato, K., Nakano, H., Yamane, T., 2003. Inverting enantioselectivity of Burkholderia cepacia KWI-56 lipase by combinatorial mutation and highthroughput screening using single-molecule PCR and in vitro expression. J. Mol. Biol. 331, 585–592. Kraft, M., Knupfer, U., Wenderoth, R., Kacholdt, A., Pietschmann, P., Hock, B., Horn, U., 2007. A dual expression platform to optimize the soluble production of heterologous proteins in the periplasm of Escherichia coli. Appl. Microbiol. Biotechnol. 76, 1413–1422. Kumar, G., Chernaya, G., 2009. Cell-free protein synthesis using multiply-primed rolling circle amplification products. Biotechniques 47, 637–639. Lesley, S.A., 2001. High-throughput proteomics: protein expression and purification in the postgenomic world. Protein Expr. Purif. 22, 159–164. Libby, R.T., Braedt, G., Kronheim, S.R., March, C.J., Urdal, D.L., Chiaverotti, T.A., Tushinski, R.J., Mochizuki, D.Y., Hopp, T.P., Cosman, D., 1987. Expression and purification of native human granulocyte-macrophage colony-stimulating factor from an Escherichia coli secretion vector. DNA 6, 221–229. Lutz, S., Bornscheuer, U.T., 2008. Protein Engineering Handbook, 1st ed. Wiley-VCH, Weinheim. Madin, K., Sawasaki, T., Ogasawara, T., Endo, Y., 2000. A highly efficient and robust cell-free protein synthesis system prepared from wheat embryos: plants apparently contain a suicide system directed at ribosomes. Proc. Natl. Acad. Sci. U.S.A. 97, 559–564. Matsumoto, K., Tomikawa, C., Toyooka, T., Ochi, A., Takano, Y., Takayanagi, N., Abe, M., Endo, Y., Hori, H., 2008. Production of yeast tRNA (m(7)G46) methyltransferase (Trm8–Trm82 complex) in a wheat germ cell-free translation system. J. Biotechnol. 133, 453–460. Morihara, K., Oka, T., Tsuzuki, H., 1979. Semi-synthesis of human insulin by trypsincatalysed replacement of Ala-B30 by Thr in porcine insulin. Nature 280, 412–413. Nardella, C., Lahm, A., Pallaoro, M., Brunetti, M., Vannini, A., Steinkuhler, C., 2004. Mechanism of activation of human heparanase investigated by protein engineering. Biochemistry 43, 1862–1873. Netzer, W.J., Hartl, F.U., 1997. Recombination of protein domains facilitated by cotranslational folding in eukaryotes. Nature 388, 343–349. Neubauer, A., Soini, J., Bollok, M., Zenker, M., Sandqvist, J., Myllyharju, J., Neubauer, P., 2007. Fermentation process for tetrameric human collagen prolyl 4-hydroxylase in Escherichia coli: improvement by gene optimisation of the PDI/beta subunit and repeated addition of the inducer anhydrotetracycline. J. Biotechnol. 128, 308–321. Nirenberg, M.W., Matthaei, J.H., 1961. The dependence of cell-free protein synthesis in E. coli upon naturally occurring or synthetic polyribonucleotides. Proc. Natl. Acad. Sci. U.S.A. 47, 1588–1602. Noble, S.L., Johnston, E., Walton, B., 1998. Insulin lispro: a fast-acting insulin analog. Am. Fam. Physician 57, 279–286, 289–292. Ozes, O.N., Reiter, Z., Klein, S., Blatt, L.M., Taylor, M.W., 1992. A comparison of interferon-Con1 with natural recombinant interferons-alpha: antiviral, antiproliferative, and natural killer-inducing activities. J. Interferon Res. 12, 55–59. Pan, C.Q., Lazarus, R.A., 1998. Hyperactivity of human DNase I variants. Dependence on the number of positively charged residues and concentration, length, and environment of DNA. J. Biol. Chem. 273, 11701–11708. Peti, W., Page, R., 2007. Strategies to maximize heterologous protein expression in Escherichia coli with minimal cost. Protein Expr. Purif. 51, 1–10. Pisal, D.S., Kosloski, M.P., Balu-Iyer, S.V., 2010. Delivery of therapeutic proteins. J. Pharm. Sci. 99, 2557–2575. Popolo, L., Vai, M., Alberghina, L., 1986. Identification of a glycoprotein involved in cell cycle progression in yeast. J. Biol. Chem. 261, 3479–3482. Price, V., Mochizuki, D., March, C.J., Cosman, D., Deeley, M.C., Klinke, R., Clevenger, W., Gillis, S., Baker, P., Urdal, D., 1987. Expression, purification and characterization of recombinant murine granulocyte-macrophage colony-stimulating factor and bovine interleukin-2 from yeast. Gene 55, 287–293.

47

Reetz, M.T., Kahakeaw, D., Sanchis, J., 2009. Shedding light on the efficacy of laboratory evolution based on iterative saturation mutagenesis. Mol. Biosyst. 5, 115–122. Reetz, M.T., Wang, L.W., Bocola, M., 2006. Directed evolution of enantioselective enzymes: iterative cycles of CASTing for probing protein-sequence space. Angew. Chem. Int. Ed. Engl. 45, 1236–1241. Roberts, B.E., Paterson, B.M., 1973. Efficient translation of tobacco mosaic virus RNA and rabbit globin 9S RNA in a cell-free system from commercial wheat germ. Proc. Natl. Acad. Sci. U.S.A. 70, 2330–2334. Russell, N.J., 2000. Toward a molecular understanding of cold activity of enzymes from psychrophiles. Extremophiles 4, 83–90. Sainathan, S.K., Tu, L., Bishnupuri, K.S., Han, M., Li, A., Newberry, R.D., McDonald, K.G., Crimmins, D.L., Houchen, C., Anant, S., Dieckgraefe, B.K., 2005. PEGylated murine Granulocyte-macrophage colony-stimulating factor: production, purification, and characterization. Protein Expr. Purif. 44, 94–103. Sawasaki, T., Hasegawa, Y., Tsuchimochi, M., Kamura, N., Ogasawara, T., Kuroita, T., Endo, Y., 2002a. A bilayer cell-free protein synthesis system for high-throughput screening of gene products. FEBS Lett. 514, 102–105. Sawasaki, T., Ogasawara, T., Morishita, R., Endo, Y., 2002b. A cell-free protein synthesis system for high-throughput proteomics. Proc. Natl. Acad. Sci. U.S.A. 99, 14652–14657. Schliebs, W., Thanki, N., Jaenicke, R., Wierenga, R.K., 1997. A double mutation at the tip of the dimer interface loop of triosephosphate isomerase generates active monomers with reduced stability. Biochemistry 36, 9655–9662. Shimizu, Y., Inoue, A., Tomari, Y., Suzuki, T., Yokogawa, T., Nishikawa, K., Ueda, T., 2001. Cell-free translation reconstituted with purified components. Nat. Biotechnol. 19, 751–755. Siurkus, J., Panula-Perala, J., Horn, U., Kraft, M., Rimseliene, R., Neubauer, P., 2010. Novel approach of high cell density recombinant bioprocess development: optimisation and scale-up from microliter to pilot scales while maintaining the fed-batch cultivation mode of E. coli cultures. Microb. Cell. Fact. 9, 35. Spirin, A.S., Baranov, V.I., Ryabova, L.A., Ovodov, S.Y., Alakhov, Y.B., 1988. A continuous cell-free translation system capable of producing polypeptides in high yield. Science 242, 1162–1164. Swartz, J.R., 2009. Universal cell-free protein synthesis. Nat. Biotechnol. 27, 731–732. Swartz, J.R., 2012. Transforming biochemical engineering with cell-free biology. AIChE J. 58, 5–13. Tsuboi, T., Takeo, S., Iriko, H., Jin, L., Tsuchimochi, M., Matsuda, S., Han, E.-T., Otsuki, H., Kaneko, O., Sattabongkot, J., Udomsangpetch, R., Sawasaki, T., Torii, M., Endo, Y., 2008. Wheat germ cell-free system-based production of malaria proteins for discovery of novel vaccine candidates. Infect. Immun. 76, 1702–1708. Tsuboi, T., Takeo, S., Arumugam, T.U., Otsuki, H., Torii, M., 2010. The wheat germ cellfree protein synthesis system: a key tool for novel malaria vaccine candidate discovery. Acta Trop. 114, 171–176. Vedadi, M., Lew, J., Artz, J., Amani, M., Zhao, Y., Dong, A., Wasney, G.A., Gao, M., Hills, T., Brokx, S., Qiu, W., Sharma, S., Diassiti, A., Alam, Z., Melone, M., Mulichak, A., Wernimont, A., Bray, J., Loppnau, P., Plotnikova, O., Newberry, K., Sundararajan, E., Houston, S., Walker, J., Tempel, W., Bochkarev, A., Kozieradzki, I., Edwards, A., Arrowsmith, C., Roos, D., Kain, K., Hui, R., 2007. Genome-scale protein expression and structural biology of Plasmodium falciparum and related apicomplexan organisms. Mol. Biochem. Parasitol. 151, 100–110. Vermasvuori, R., Koskinen, J., Salonen, K., Siren, N., Weegar, J., Dahlbacka, J., Kalkkinen, N., von Weymarn, N., 2009. Production of recombinant HIV-1 nef protein using different expression host systems: a techno-economical comparison. Biotechnol. Prog. 25, 95–102. Walsh, G., 2010. Biopharmaceutical benchmarks 2010. Nat. Biotechnol. 28, 917–924. Watanabe, M., Miyazono, K., Tanokura, M., Sawasaki, T., Endo, Y., Kobayashi, I., 2010. Cell-free protein synthesis for structure determination by X-ray crystallography. Methods Mol. Biol. 607, 149–160. Weber, W., Fussenegger, M., 2007. Inducible product gene expression technology tailored to bioprocess engineering. Curr. Opin. Biotechnol. 18, 399–410. Yin, G., Swartz, J.R., 2004. Enhancing multiple disulfide bonded protein folding in a cell-free system. Biotechnol. Bioeng. 86, 188–195. Zawada, J.F., Yin, G., Steiner, A.R., Yang, J., Naresh, A., Roy, S.M., Gold, D.S., Heinsohn, H.G., Murray, C.J., 2011. Microscale to manufacturing scale-up of cell-free cytokine production – a new approach for shortening protein production development timelines. Biotechnol. Bioeng. 108, 1570–1578. Zubay, G., 1973. In vitro synthesis of protein in microbial systems. Annu. Rev. Genet. 7, 267–287. Zuppinger, K., Aebi, C., Fankhauser, S., Herz, G., Zurbrugg, R.P., Schopfer, K., Neri, T.M., 1987. Comparison of human and porcine insulin therapies in children with newly diagnosed diabetes mellitus. Diabetologia 30, 912–915.