Challenges in the large scale production of peptides
Peptides, which are small proteins typically containing a chain of up to about 50 amino acids, have long been seen as promising therapeutic agents. Over 60 are currently on the market and are generating sale of more than $13 billion/year, over 1.5% of total drug sales.1 Four - Glatiramer, Leuprolide, Goserelin and Octreotide -have already crossed the $1 billion/year mark. Table 1 shows some of the top selling peptide drugs.
Table 1 - Top selling peptide drugs
Sales revenue ($)
2015 projection ($)
Patent expiry date
3.9 billion (2012)
2.2 billion (2010)
1.14 billion (2010)
1.12 billion (2010)
751 million (2010)
780 million (2010)
1.19 billion (2011)
650 million (2010)
The Medicines Company
132 million (2010)
400 million -peak sale
Direct thrombin inhibitors
Aeterna Zentaris &
~500 million - peak sale
230 million (2010)
Sustained growth expected
The number of peptide drugs entering clinical trials is increasing steadily: it was 1.2/year in the 1970s, 4.6 in the 1980s, 9.7 in the 1990s and 16.8 in 2000s. Currently about 140 peptides are in clinical trials and more than 400 in pre-clinic development (Figure 1). From 2000 onwards, peptides entering clinical study were most frequently for cancer (18%) and metabolic disorders (17%).2 More than 100 pharmaceutical and biotech companies are now pursuing peptide drug discovery programmes.
There are several reasons for the tremendous increase of interest in peptide-based therapeutics. The most important is the major advances in drug delivery systems for peptides in the last ten years. Certain peptides can also pass through cell membranes and therefore can be used as carriers for targeted drug delivery. The use of peptides is also growing in medical diagnostics, nutraceuticals, anti-microbials and cosmetics.
The CMO sector is under tremendous pressure to meet the current need for large quantities - about 1,400 kg/year - of peptide APIs. Of this total, about 40% is produced by CMOs and their share is expected to rise drastically by 2017, when a number of peptide drugs become generic and more peptides are approved. Moreover, pharmaceuticals companies are expected to rely more on CMOs for large-scale manufacture of peptides in an attempt to cut production costs.
Therefore, it is time to focus on improving current manufacturing methods in order to increase productivity without compromising quality. Big peptide CMOs are upgrading their capabilities by acquiring large-scale manufacturing equipment, strategic positioning of their facilities around the world and innovative manufacturing methods.
A number of chemistries and methods have been developed for building peptide chains. The classical method, liquid-phase peptide synthesis (LPPS), involves reactions in solution using protected amino acids and requires the isolation of product after each coupling and de-protection stage from solution.
Solid-phase peptide synthesis (SPPS) relies on assembling peptide chain on an insoluble polymeric support, thereby simplifying the process. Protected amino acids are added one by one to the resin beads after each de-protection step and require no intermediate isolation. When the desired length of sequence is synthesised on the resin, the peptide is cleaved off and subsequently purified using chromatographic techniques. A mixed - hybrid or convergent - strategy involving both LPPS and SPPS also is used in the manufacture of certain peptides.
Obviously, shorter chain (<15 amino acid) peptides are significantly more economical to synthesise in the high purity required for drug applications. Large-scale manufacturing of such peptides is generally performed by LPPS. Although LPPS relies on multiple steps, solution chemistry allows bulk manufacturing of peptides more economically than SPPS technology. However, LPPS is not a viable option for the synthesis of longer peptides, due to the complexity of the process.
In the last ten years, SPPS has emerged as the method of choice for most peptides. It can be performed using two different techniques, a Boc protection strategy and Fmoc chemistry. The latter has become more popular due to the simplicity of the process and because the use of hazardous hydrogen fluoride can be avoided.
The availability of orthogonal protecting groups in peptide synthesis has opened new methods for making complex peptides using SPPS technology. Different types of polystyrene-based resins are available for SPPS, the most commonly used being Wang, 2-chlorotrityl chloride, Rink amide and Merrifield resins. Fmoc-protected amino acids are widely commercially available.
Both SPPS and LPPS can be used in tandem for making long and complex peptides. For example, several short fragments could be synthesised via LPPS methods, followed by assembling them on a solid phase resin to make a longer chain. In some cases, larger peptide fragments can be synthesized on a solid phase resin, followed by cleavage of the protected peptide from the resin and assembling the protected fragments in solution.
Purification & isolation
Once synthesised, a peptide has to be purified, usually to over 95%, to meet the end use specifications. Reverse phase HPLC (RP-HPLC) using hydrophobic interactions is seen as the most powerful method for peptide purification. It is characterised by the use of a stationary phase and an aqueous mobile phase containing an organic solvent, such as acetonitrile or an alcohol.
Various chromatography media have been used for large-scale purifications of peptides on reverse phase resins. Among the most popular are those based on C4, C8 and C18 alkyl chains attached to a silica surface.
Nowadays, with the popularisation of dynamic axial compression (DAC) columns for chromatography and the availability of excellent, relatively affordable reversed phases, the purification of peptides has become easier and more accessible than ever before. A 20 cm column can yield several kgs/month of purified
peptide.After purification, peptides are isolated and dried using lyophilisers. As well as bottle lyophilisers, peptide manufacturers use large commercial-scale tray lyophilisers to increase productivity and improve control of drying parameters.
SPPS reactor, 20 cm RP-HPLC column and 200 litre tray lyophiliser at CPC’s GMP peptide manufacturing unit
Challenges in manufacturing
Peptide manufacturing technology has matured and major manufacturers can now produce peptides APIs in hundreds of kilograms scale. With more peptide drugs poised to enter the market and many more in trials, pressure is mounting on CMOs to offer peptides more economically while still adhering to GMP and without compromising quality and integrity. Moreover, the complexity of peptides, in terms of size, modifications, conjugation methods, stability and purity requirements, is expected to increase. This scenario is a challenging prospect for manufacturers. From a synthetic point of view, the resins currently being used for SPPS are not re-used and are expensive compared to other raw materials.
Although the prices of common amino acids protected with Fmoc group has come down in recent years, many of the D-amino acid and unnatural amino acid derivatives used as building blocks will increase product cost. Large quantities of resins (only 0.5-1.0 mmol of reaction sites are available per gram of resin) and excesses of amino acids (2 to 3 equivalents) are used in peptide synthesis.
Solvents such as dichloromethane and dimethylformamide are used in large excess for resin swelling and washing steps, thereby producing a considerable amount of waste. Typically, a minimum of 5 ml solvent/gram of resin is needed for each washing step, which means that when the synthesis of a 20 amino acid is performed on a 1 kg of resin about 1,000 litres of solvent waste are generated.
Waste disposal and solvents amount to approximately 10% of the total production cost. At the end of the synthesis, the peptide is cleaved from the resin using trifluoroacetic acid followed by precipitation with large excess of ether, further increasing the amount of waste. The purity of the synthesised peptide largely depends on the efficiency of synthetic process, its own intrinsic nature and the quality of raw materials used. Incomplete coupling and de-protection during synthesis will lead to deleted sequences and impurities.
Purification is the most critical step in the manufacturing process. Poor solubility and the aggregation of certain peptides could create problems during purification, so careful consideration of peptide properties before purification is crucial. Peptide modifications like cyclisation and disulfide bond formation are accomplished before purification.
Diluted (1 mg peptide/ml) solutions of peptide after such processing make the purification step more difficult, not only because of the huge volume of solution to be pumped to the HPLC column but also due to the presence of polymeric materials formed via oxidation. Removing closerunning impurities from the peptide of interest often requires multiple purification steps using different buffer system and processing. Purification accounts for more than 20% of the production cost of peptides. The price of solvents, such as acetonitrile for RP-HPLC purification, often fluctuates, further increasing purification costs. These solvents are generally not recovered or reused. The freezing and drying of large quantities of peptide solution after purification adds considerable further cost. The manufacture of peptide APIs requires full compliance with GMP regulations and FDA guidelines. Raw materials have to be qualified and all manufacturing takes place in controlled areas. Purification and lyophilisation are performed in Class 100,000 clean room areas followed by packaging in Class 10,000 clean rooms.
Figure 2 shows a process diagram for the synthesis of peptides by SPPS. No major technology has overtaken SPPS since Merrifield first introduced it, mainly because of its ability to synthesise long chain peptides conveniently in high yield. SPPS is quicker than conventional step-by-step synthesis and can be fully or semi-automated. In large-scale peptide manufacturing, it allows semiautomation for solvent delivery, coupling reaction and waste management, whereas LPPS is limited to the synthesis of shorter peptides and takes a longer time for production. Other technologies, such as fermentation and recombinant methods, have been introduced recently in the manufacture of peptides such as cubicin and liraglutide. These methods can easily be scaled up to produce large quantities, but their wider use is limited by such issues as the use of specific enzymes in fermentation and the difficulty in introducing unnatural amino acids via recombinant expression. Improving SPPS and LPPS technology is necessary to achieve low-cost manufacturing of peptides. New methods for reducing waste generation and improving coupling efficiency and overall yield will drive the process costs down. The recycling of solvents is a viable option during large-scale synthetic processes. The initial investment in a distillation plant may be high, but once it is set up the system can be used to regenerate solvents, which will reduce waste disposal costs as well. Efficient supply chain management and consistent quality of raw materials from trusted vendors is necessary in large-scale manufacturing.
Overall, the synthetic process for peptides can be improved by using high quality raw materials, careful management and recycling of solvents, and using efficient coupling agents and convergent strategies. For example, a difficult peptide with more than 50 amino acids can be synthesised by fragment condensation in the solid phase, thereby achieving better yields and purity compared to step-wise SPPS method. For industrial-scale purifications on reverse phase resin columns, the shape and size of the particles in the stationary phase are important considerations. Spherical particles are preferred to irregularly shaped ones. Extensive use of columns under DAC sometime results in the formation of irregularly shaped particles. Particle size is another important characteristic of bonded silica phases that strongly influences column efficiency. It is important to check the columns periodically for efficiency when they are in continuous use for a specific project. RP-HPLC is an efficient but expensive procedure for the purification of peptides. It is good idea to subject crude synthetic peptide to a ‘capturing’ step in which most of the impurities are removed. This can be achieved, for example, by applying the crude peptide solution to an ion exchange column. A fraction highly enriched with the desired peptide is obtained after this initial purification removing, for instance, those byproducts generated in the final de-protection of the peptide.
This strategy is very economical for the prepurification of a dilute solution (~1 mg/ml) containing disulfide-bonded peptide that has been subjected to oxidative folding reactions. The initial pre-purification will remove polymeric materials and many other impurities, enabling a concentrated solution of peptide to be loaded onto a RP-HPLC column for a second purification. For example, using this method, a 200 litre solution obtained after oxidation with a purity of 30% can be loaded onto an ion exchange (IEX) column to get a 20 peptide solution with over 75% purity. No organic solvents are used in IEX processes and the overall purification cost for this technology is lower than using two RP-HPLC purification steps. Polymeric materials and unwanted impurities can also ruin RP-HPLC columns if loaded directly without prepurification. It is also easier to clean and regenerate IEX columns than RP-HPLC columns. Another way to cut production costs is the recycling of acetonitrile. The suitability of other cheaper alternatives, such as methanol or isopropanol, could also be evaluated in certain cases. Lyophilisation costs could be reduced by concentrating the peptide solution collected from purification columns by evaporation, the elution at higher concentration of organic eluent or membrane filtration techniques.
In recent years researchers have been exploring various methods to improve peptide synthesis technology. In a modified LPPS strategy, a procedure involving the coupling of amino acids and peptide acids, instead of the usual amino esters and peptide esters, to slight excesses of pentafluorophenyl active esters in a THF/water solvent mixture has been examined. Due to their poor solubility, peptide acid intermediates are easily isolated in high purity by acidification under controlled conditions and the removal of excess active esters by selective extraction. Unlike classical repetitive solution phase procedures, this approach does not require time-consuming neutralisation reactions and reduces significantly the number of operation units needed to obtain peptide intermediates.3
Another study uses a new technology platform called membrane-enhanced peptide synthesis (MEPS), which advantageously combines organic solvent nanofiltration with LPPS. A first amino acid is linked to a soluble polyethylene glycol anchor. Through subsequent repeated coupling and deprotection steps, the peptide is extended to the desired length. The residual by-products and excess reagents after each reaction are removed by diafiltration through a solvent-stable membrane, which retains the peptide.4
The purity of the peptides produced by MEPS is reportedly higher than that of those produced by SPPS, under the same conditions. MEPS benefits from the advantages of LPPS, while avoiding the purification steps that have until now made it practically difficult.
A group of Japanese researchers has designed an efficient method for the synthesis of peptides bearing an amide at the C-terminal. This method involves the attachment of a C-terminal protecting group bearing long aliphatic chains, followed by the repetition of simple reaction and precipitation steps with the combined advantages of LPPS and SPPS. Using this method, a hydrophobic peptide was successfully synthesized in good yield and high purity, which cannot be obtained satisfactorily by SPPS.5
Very recently a soluble tag-assisted LPPS was successfully established, based on simple hydrophobic benzyl alcohols, which can easily be prepared from naturally abundant materials. Excellent precipitation yields were obtained at each step, combining the best properties of LPPS and SPPS. This approach can also be applied efficiently to fragment coupling, allowing the chemical synthesis of several bioactive peptides.6
Conclusions & outlook
With increasing demand for large quantities of peptides and their greater complexity in terms of size, modifications and purity requirements, it is necessary to look for better ways of manufacturing peptides. Further advances and improvement in synthetic peptide chemistry, solid support materials, purification technology and isolation methods are needed to tackle current challenges in the large-scale manufacturing of peptides, especially high production costs and complex modifications. Major CMOs have already started focusing on these areas in an effort to supply highly active peptide APIs in larger quantities needed for future needs at affordable prices. Proper infrastructure development is also needed to meet future demands for several thousand kilos of peptides.
Strategically placing manufacturing units around the world adhering to FDA and GMP guidelines a viable way of cutting down manufacturing costs without compromising quality and regulatory guidelines. The availability of high quality raw materials in bulk at reduced prices in these regions as well as highly qualified workforce is crucial for successful operation of peptide manufacturing units in these regions.
1. A.M. Thayer, Chem. Eng. News 2011, 89, 13-20
2. Peptide Therapeutics Foundation, Development Trends for Peptide Therapeutics Report, San Diego, 2010
3. C. Meneses et al., J. Org. Chem. 2010, 75 (3), 564-569
4. S. So et al., Org. Process Res. Dev. 2010, 14 (6),