Industrial Biotechnology PDF

Preview

Summary

Includes Pharming...

Description

Industrial Biotechnology This concerns the production of a range of molecules including amino acids, vitamins, antibiotics, commercial, and therapeutic enzymes. A major goal in biotechnology is to reduce the production costs, thereby increasing profit margins. Production costs do reflect the yield of a given molecule. The selling price of products is a strong function of product concentration and consequently, the cost of separation and/or purification. Usually, the reactor cost is typically less than 25% of the total production cost. Water is the cheapest, followed with ethanol and then amino acids and antibiotics (such as penicillin). Bulk enzymes (proteases, enzymes used in detergents and food industry) and diagnostic enzymes (Taq polymerase) are a bit more expensive, as well as monoclonal antibodies (usually recombinant, such as insulin). The most expensive products are therapeutic enzymes, including Factor VIII and Urokinase. Another goal is to improve on what nature can provide and change the properties of a molecule – therefore, redesigning natural products. The improvements can involve empirical changes (random approach, with selection for an improved/new activity). Alternatively, a rational engineered design can be introduced. In general, the way to improve yield is by alleviating or bypassing a rate-limiting step in product synthesis, such as increasing gene dosage and expression (assuming the rate-limiting step concerns transcription and translation). Increasing the concentration of precursors and reducing susceptibility of organism to high (toxic) levels of product can also improve the yield of the product. Additionally, you can inhibit product degradation or facilitate product secretion. A protein is the product of a single gene, and the rate limiting steps for synthesis are transcription and translation. Yield is therefore a function of gene dosage and gene expression. There are factors which influence gene expression: strength of promoter (constitutive or inducible), translation initiation, mRNA secondary structure and stability, codon choice, stability of protein, and the toxicity of protein. To overcome the problems of source availability, some proteins are produced naturally at very low concentrations, so compromising a large scale production; although, sacrifice of producer may be unethical. There is source safety for recombinant proteins, as proteins isolated from human plasma, risk contamination from viruses, prions etc. A cloned gene can also be modified. Different organisms can be used as a cloning hosts, although there are advantages and disadvantages for each one. E. coli can be used due to its ease of manipulation and growth (as well as being cheap), and the promoters and gene regulation are well understood. Although, we don’t usually get export of proteins into the growth medium and there are some overexpressed foreign proteins that form inclusion bodies. These foreign proteins are often degraded and with E. coli, we can’t introduce eukaryotic posttranslational modifications. Besides this, gram-positive bacteria (Bacillus Streptomyces) can be used, due to the proteins exported into growth medium, inexpensive to grow, and there’s a large surface area to volume ration favoring export (Streptomyces). However, manipulation of expression is not easy and there is vector instabilities (Bacillus). Proteases are also secreted, and again, we can’t introduce eukaryotic post-translational modifications. Yeasts can also be used as cloning hosts, as they are also easy and cheap to grow. Additionally, they export proteins into growth medium and they remove N-

terminal F-met (which E. coli does not). Like Gram+ bacteria, the manipulation of expression is not easy and the post-translational glycosylation is not identical to mammals. Besides this, filamentous fungi are also used due to its large surface area to volume ratio favoring export into growth medium, and it has multiple gene copies in chromosome, as well as its relatively inexpensive to grow. Although, manipulation is not easy, the post-translational glycosylation is not identical to mammals, and they secrete proteases. Finally, mammalian cells can also be used as they export proteins into growth medium and exhibit authentic post-translational modification. But, manipulation is not easy and the large scale culture is very expensive and is prone to contamination. Case Study: Chymosin Worldwide dairy industry spends over $100 million a year on Chymosin, an acid protease. It’s required for site-specific proteolysis of k-casein, and when k-casein is cleaved by Chymosin (between Phe105 and Met106), it removes glycosylated C-terminus (106-169 peptide). This loss of hydrophilic C-terminal sequence allows aggregation of casein micelles – in other words, ‘milk clotting’. The traditional source of Chymosin is from the 4th stomach (abomasum) of unweaned calf (yielding 10g per calf). Due to high demand, it outstrips this source and so fungal proteases have been used as a substitute. However, it lacks site-specificity and are more thermostable, resulting in ‘bitter’ taste. Hence, recombinant Chymosin is a really good solution. Chymosin is synthesized in calf mucosal cells as preprochymosin. The 16 amino acid signal sequence is removed as prochymosin, which is secreted via the ER-Golgi pathway. In acidic conditions of the abomasum, autocatalytic cleavage removes the 27 amino acid pro-sequence – leaving the active protease. To obtain the gene, cDNA was synthesized from mRNA extracted from abomasum mucosal cells. The cDNA was tailored to remove the sequence, encoding the eukaryotic signal pre-sequence, and engineered into an E. coli multi-copy expression vector. To optimize expression, the distance between the RBS and initiator ATG codons, was manipulated. The formation of inclusion bodies is due to very high local concentrations of recombinant protein, insufficient chaperones/folding enzymes, the formation of disulphide bonds being prevented (due to reducing environment of cytoplasm), there’s no localization to specific organelles (as occurs in eukaryotic cell), and there’s a lack of appropriate post-translational modification resulting in less stable products. [Met]prochymosin accumulates as inclusion bodies in E. coli. These inclusion bodies can be separated by centrifugation. The protein is then solubilized with 9M urea, and the urea is then removed by extensive dialysis. Acidification then promotes autocatalytic removal of the 43 amino acid [Met]pro N terminal peptide. Only 30% of the total synthesized Chymosin can be recovered as active protein. This enzyme was the first recombinant protein approved for human consumption. To avoid formation of inclusion bodies, a secretory host-vector system was required. 1st attempt was using S. cerevisiae with Yep vector, strong promoter, and a prochymosin; fused to invertase signal sequence. The yield is 0.1% and its uneconomic because S. cerevisiae is a poor secretor. Kluveromyces lactis is employed for the commercial production of non-recombinant lactase and it has the GRAS (generally recognized as safe) status. Entire preprochymosin cDNA is cloned between a strong inducible Lac4 promoter and Lac4 terminator. Linearized vector is integrated into K. lactis chromosome (stable in absence of selection). Its growth on lactose-containing whey induces expression and 95% of

2

product can be correctly exported (substantially pure). Its yield is 1g/I, which is 10% total protein, and the product is marketed as Maxiren by Gist Brocades. E. coli for Recombinant Proteins To provide a high gene copy number, plasmid vectors based on ColE1/pBR322 are employed. Consequently, in order to maintain these plasmids and kill plasmid-free segregants (usually faster growing), an antibiotic is added to the fermentation. Vectors include engineered strong inducible promoters (such as tac/trp/T7) on the upstream of multiple cloning sites. With a strong inducible promoter, the next rate-limiting aspect of expression is translation. You need to optimize the distance between the RBS (ribosome binding site) and the start codon. Heterologous genes can contain non0favoured codons for which E. coli produces insufficient tRNA to. The codon bias can be determined by calculating the Relative Synonymous Codon Usage (RSCU) of individual codons. This is done by dividing the observed number of times a particular codon is used, with the expected number if all codons are used with equal frequency. If RSCU is equal to 1, the codon is used without bias, if it’s less than 1, the codon is ‘non-favored’. If it is greater than 1, the codon is favored. Case Study: Interleukin 2 IL-2 plays a pivotal role in inducing an antigen-specific immune response by stimulating growth and differentiation of activated T and B lymphocytes. Therapeutic used is in treatment for renal cell carcinomas and melanomas. In the 399 bp IL-2 gene, only 43% of codons are favored in E. coli. Using gene synthesis, an alternative IL-2 gene with 85% favored codons was synthesized. Further comparisons of the 2 genes showed that the synthesized IL-2 gene supports 8 times higher yield of active IL-2 in E. cli (identical amounts of mRNA). Gene Fusions Heterologous proteins expressed in E. coli retain their N-terminal methionine. If the proteins are overexpressed, they are prone to proteolytic degradation and formation of inclusion bodies. Fusion of heterologous proteins with native proteins can avoid these problems and, in addition, facilitate protein purification. Following purification, the ‘carrier’ protein component is removed by site-specific cleavage. Case Study: The ThioFusion Expression System (Invitrogen) E. coli thioredoxin is a small protein with 2 cysteine residues in close proximity. It normally functions as a carrier of reducing power in biochemical reactions. When it’s overexpressed from plasmid vectors, thioredoxin can accumulate in soluble form to 40% of the total cellular proteins. The accumulation of Thioredoxin at the sites in E. coli, are called ‘adhesion zones’, and it’s selectively released into the medium by rapid osmotic shock treatment of E. coli cells – which simplifies purification. A ThioHis vector is used for gene fusions. Its properties include a multi-copy plasmid and a strong inducible trc promoter (regulated by lac repressor and induced by IPTG). Additionally, it has a His-tag-thioredoxin, which is a polyhistidine sequence engineered into thioredoxin, that permits rapid purification of fusion proteins. The polyhistidine sequence also has a high affinity for divalent metal ions (e.g. nickel) and the resulting

3

protein can be purified by chromatography using a metal chelating resin. Lastly, the His-tag-thioredoxin sequence is immediately followed by an enterokinase cleavage site. An example of recombinant proteins produced in E. coli are Insulin A and B chains, Somatostatin, and Interleukin 2. Random Mutagenesis In protein engineering, there’s 2 approaches. The first approach is a ‘top-down’ approach, which is a sitedirected mutagenesis. This introduces a small number of specific amino acid substitutions, based on the structure and function of protein. Another approach is the ‘bottom-up’ approach, and this is a random mutagenesis. It introduces many random amino acid substitutions, where no structural information of the protein needed. Examples of random mutagenesis include a mutator strain of E. coli, Error-prone PCR, and DNA shuffling (combining error-prone PCR and recombination). The current aim is to emulate the evolutionary algorithm, but collapse the time-scale. There are number of possible protein variants (20N), and there’s an implication that in over 4 billion years, nature has only explored a tiny fraction of these possibilities. Inference that the vast majority of permutations will be inactive. Therefore, choosing an existing enzyme as starting material is important. DNA Shuffling The process of DNA shuffling begins when the double-stranded parental genes are randomly cut into fragments with the enzyme DNase I. These fragments are then reassembled using repeated cycles of the polymerase chain reaction (PCR) without added primers. The PCR reaction has three steps: first, the mixture is heated so that the fragments separate into single strands. Second, the temperature is lowered, allowing the fragment to anneal, or become double-stranded, once again. However, the fragments do not necessarily anneal just as before, and most will anneal with some sort of overhang where the fragments do not overlap (and remain single-stranded). In the third step, the temperature is raised slightly, and a polymerase enzyme fills in overhanging fragment portions with new nucleotides making the fragments longer. These three-step cycles continue until fragments reach the length of the original parental genes. Starting at a baseline where the desired trait is a measurable (albeit low level), in the starting enzyme activity, can improve function in unnatural or extreme environment. This improve activity can be used towards a new substrate. It also allows the tuning of the specificity of the enzyme, and can further increase the functional expression in a heterologous host (codon usage, solubility, and improved stability). An application of DNA Shuffling is the isolation of thermostable enzymes, such as evolved Catechol 2,3dioxygenase (C230). The C230 catalyzes the ring-cleavage of catechol and its derivatives, and it plays an important role in the degradation of pollutant hydrocarbons. There are 2 well studied C230 genes which have been used, such as the nahH and xylE, which are obtained from Pseudomonas putida. The nucleotide sequence identity between nahH and xylE is 80%. When observed in bacterial colonies, the chimeric C230s are expressed with high-temperature stability, and the colonies turn yellow when C230 transformed the catechol into yellow-colored 2-hydroxymuconic semialdehyde.

4

Synthetic Biology This occurs in three steps: the parts, to devices, and to systems. The ‘part’ is a DNA sequence encoding some part of the genetic machinery, including (but not limited to) promoters, RBS, protein coding regions, and terminators. A ‘device’ is a group of parts that work together for specific functions, such as protein production, sensing/reporting, measurement, signal inversion, cell signaling, and cell motility. Finally, the synthetic ‘system’ can combine parts derived by directed evolution and/or DNA shuffling. Diversity generation and screening/selection are continuously in process, which then leads to improving enzyme performance (stability, pH activity profile, substrate specificity, stereoselectivity and activity on novel substrates), which could lead to engineering abiological reactions – such as artificial enzymes, or through expansion of reaction space for natural enzymes. Traditional genetic engineering, based on parts nature can provide, is enhanced by using parts obtained by directed evolution. The potential outcomes are far-reaching with microbial system for applications and products including biofuels, vaccines, pharmaceuticals, clean water and food products. The global synthetic biology market reached nearly $2.1 billion in 2012 and $2.7 billion in 2013. This market is expected to grow to $11.8 billion in 2018 with a compound annual growth rate (CAGR) of 34.4% over the five-year period from 2013 to 2018. Glycoproteins Whereas protein synthesis is template dependent (from DNA/mRNA), glycosylation is not. Authentic glycosylation of mammalian proteins (needed for activity/stability/localization) occurs in mammalian cells. Oligosaccharides can be attached to a nitrogen atom (N-linked) of asparagine in the sequence AspX-Thr/Ser. Alternatively, they are attached to an oxygen atom (O-linked) of either Thr or Ser. A core unit is added, onto which a variety of additional sugars can be added. A protein is synthesized on the rough endoplasmic reticulum. In the endoplasmic reticulum, the Nglycosylation precursor Glc3Man9GlcNAc2, is transferred to the protein. After trimming off the precursor, the protein is transferred to the Cis-Golgi by vesicular transport. The glycostructure is trimmed and extended as the protein traverses the Golgi complex from cis- to trans-Golgi. Transportation takes place by the vesicles, and when the glycostructure is complete, the glycoprotein may be excreted by way of a secretion vesicle that fuses with the cell membrane. Specific glycosylation patterns are different between cell lines. CHO cells do not synthesize bisecting Nacetylgoucosamine structures. Mouse cell lines can add terminal galactose units that are immunogenic in humans. Tissue Plasminogen Activator (tPA) Although mammalian cells are expensive and difficult to grow, they can be used to provide authentic, correctly modified, and fully active proteins. The first product was Tissue Plasminogen Activator (tPA), which was administered to heart attack patients. These tPA are proteases that cleaves plasminogen to release active plasmin – a protease that degrades fibrin (which forms blood clots). Although, tPA has a complicated pattern of folding and is glycosylated. The tPA cDNA was cloned into an SV40-based expression vector. After transfection of mammalian cells, gene dosage was increased by ‘dhfr amplification’ using methotrexate (a competitive inhibitor of dihydrofolate reductase). Recombinant tPA is purified from the growth medium.

5

The cloned cDNA for human tPA has a tPA signal sequence a coding sequence for mature tPA. This is cleaved with HindIII and EcoRI, and ligated to the expression vector, with dhfr. Once the expression vector is complete, its introduced into mammalian cells. The cells then express low levels of tPA, so stable transformants are selected by growth in HAT medium. Methotrexate is used to increase the expression of tPA, and afterwards, the cultured cells can be transferred to the fermentator, and the tPA is secreted into culture medium. Some recombinant pharmaceutical proteins approved for general medical use are produced commercially via animal cell culture. This includes FSH for infertility and Erythropoietin for Anaemia. Interferon-B are also used for multiple sclerosis. Pharming This is the production of pharmaceutical proteins in the milk of transgenic mammals. It has high production capacities, as during a typical 5-month lactation period, 1 sheep can produce 2-3 L milk per day. If the recombinant protein is expressed at a level of 1g/L, a single sheep can produce excess of 20 g product per week. Through traditional milking method, it’s easy to collect the protein. There are also low production costs as it avoids costly fermentation of mammalian cells. Lastly, it’s easy to scale up by increasing the size of flock through breeding programmes or through cloning. The gene of interest (GOI) is in a vector with B-Lactoglobulin promoter. This is microinjected into the pronucleus of a sheep ovum. It is then implanted into the foster mother, where transgenic progeny can be identified by PCR. The GOI is expressed only in mammary tissue and the GOI protein is secreted into the milk. By traditionally milking the sheep, you can collect milk containing the GOI protein. Further, allowing the fractionation of milk proteins, then to GOI protein. Case Study: Human α1-antitrypsin The Human α1-antitrypsin is a 394 amino acid glycoprotein normally present at 2 g/l in plasma, and it’s synthesized primarily in liver. There’s deficient of this in patients with emphysema and the replacement therapy for this requires 200g/patient/year. A traditional source is human plasma, although it is an insufficient supply and risk of prions/virus contamination. Glycosylation is required for plasma stability. The yield depends on gene copy number integration site. Milk-derived protein from this is fully Nglycosylated. Glycoengineering Glycoproteins produced by mammalian cells/pharming are a mix of glycoforms, but only a subset of which are active. To produce recombinant proteins with very specific glycosylations, engineered strains of the yeast Pichia pastoris have been developed. Therefore, humanizing glycosylation in yeast can occur. Firstly, the process of knocking out native α-1,6-mannosyltransferase (Och1P) occur. Then it can be fuse and express human...