July 1, 2013 (Vol. 33, No. 13)
Automation and advances in solid- and liquid-phase approaches are but a few factors driving costs down, according to this article from our July 1 issue.
The continued strong demand for peptides as research tools and lead compounds in drug discovery is fueling technology development in high-throughput synthesis and purification.
Examples of peptide-based cancer therapeutics include luteinizing hormone-releasing hormone (LHRH) agonists to decrease testosterone production in prostate cancer, somatostatin analogues, and novel peptides that can specifically or preferentially bind to or penetrate tumor cells and offer opportunities for targeted therapies and for delivering radioactive or cytotoxic compounds to tumor tissue.
In addition to synthesizing anti-tumor peptides, American Peptide Company is developing cell penetrating peptides (CPPs) designed for cellular import of therapeutic molecules such as plasmids, DNA, siRNA, PNA, proteins, peptides, and nanoparticles. CPPs can form chemical linkages with their drug cargo, or they may form stable, non-covalent complexes with drugs. They are short peptides, composed of fewer than 40 amino acids, and share common features such as positively charged amino acids, hydrophobicity, and amphipathicity.
One approach the company is using to facilitate peptide synthesis is “click chemistry,” a modular strategy designed for rapidly combining small subunits.
“Click chemistry provides a number of avenues for peptide/protein modifications and could be combined with other techniques to make complex structures and multicomponent functionalized systems,” according to John McKinley, senior marketing manager. “Applications of click chemistry in peptide science include chemical ligation, cyclization, and bio-conjugation, imaging, synthesis of peptidomimetics based on traizole backbone, and conformational and backbone modifications.”
Optimizing for Speed
Automated peptide synthesizers can enable high-throughput, scalable production of high purity peptides for research or clinical applications.
“The speed of automated peptide synthesis will depend both on the optimization of reaction parameters and the throughput of the instrument used,” says James Cain, Ph.D., applications manager at Protein Technologies.
Variation of factors such as coupling and deprotection reagents, reagent excesses and concentrations, solvent selection, resin types, and resin loading make it possible to obtain high purity products with relatively short total reaction times, Dr. Cain explains.
Highly active coupling reagents—such as HATU and HCTU, for example—may not be well-suited for use on some robotic multiple peptide synthesizers due to long reagent dispensing times.
Dr. Cain points to human b-amyloid (1-42) as an example of a peptide that is difficult to synthesize and is of research and commercial interest. Its synthesis is difficult due to the high hydrophobicity of the C-terminal segment and tendency for on-resin aggregation.
Traditional approaches for making this peptide might require 50 to 60 hours of synthesis time, according to Dr. Cain. By optimizing a number of variables, it is possible to synthesize 24 human b-amyloid (1-42) peptides on the company’s Symphony X™ automated synthesizer in less than 14 hours using short reaction times and the instrument’s fluid delivery system.
In this example, the variables modified included use of low-loaded resin, a modified deprotection mixture containing 2% DBU added to the standard 20% piperidine in DMF, and HCTU as coupling reagent. Further efficiency is achieved through simultaneous addition of reagents to multiple vessels and washing of valve blocks and other components at the same time.
The Symphony’s IntelliSynth UV monitoring system, which captures UV readings every 10 seconds during the deprotection reaction, is particularly helpful for difficult sequences, such as the poly-alanine (Ala)10K, which commonly exhibits severe aggregation after the addition of the fifth residue, according to Dr. Cain.
Under standard deprotection conditions, longer deprotection times and more repetitions were required for full removal of the Fmoc protecting group after the onset of aggregation. More efficient removal could be achieved by adding 2% DBU to the deprotection mixture.
Infrared heating can also be incorporated into the design of the Symphony X. It has been shown to speed the synthesis of certain difficult sequences, such as an analog of ACP(65-74), in which the alanines have been replaced with sterically hindered aminoisobutyric acid (Aib) residues.
According to Dr. Cain, “While very pure product can be obtained using long coupling times at room temperature, the addition of IR heating allows for a ninefold reduction in the coupling time for the most difficult steps.”
CS Bio provides peptide manufacturing services, including custom and GMP capabilities, and offers medium- and large-scale automated synthesizers. For many years, the largest automated peptide synthesizer CS Bio offered was 50 L. Now demand for its large-scale 300 L synthesizer—which can produce kilograms of peptides and is validated for cGMP production—continues to grow, not only in the biotech industry but also more recently among pharma companies, which are developing internal peptide therapeutics groups, notes CS Bio’s director of operations, Jason Chang.
“It’s easy now to get good quality raw materials. People are trusting the chemistry, they are doing much larger syntheses, and they are getting much larger peptide batches in one shot as opposed to doing more small batches and combining them,” as they had in the past, Chang says.
Faster, High Purity Synthesis
A challenge for the typical peptide supplier that is set up to produce 1 mg to 5 mg quantities on automated synthesizers is the growing demand among researchers developing and running proteomic screens for very small quantities (as little as 20 nanomoles) of hundreds or thousands of peptides at low cost and fast turnaround, according to Mike Pennington, Ph.D., at Peptides International.
“We are seeing requests for increasingly complex peptides—multidisulfides, protein-peptide conjugates, and RNA-peptide conjugates, for example—longer peptides, especially for large-scale drug development; and small peptides with multiple disulfide bonds.”
In addition, he reports growth in the development of more complex combinatorial products such as vaccines having six, 12, even 18 peptide subcomponents and conjugates.
To improve product purity, Dr. Pennington describes greater reliance on specialized chemistries such as the use of pseudoproline derivatives and dipeptide building blocks that facilitate synthesis “through difficult stretches that tend to aggregate” and make it possible to produce peptides with 60 to 70 residues.
From a quality-control perspective, the addition of UPLC technology has “pushed the envelope of peptide technology to solve problems during synthesis,” he says. Errors in synthesis that would result in a deletion peptide that would have been difficult to resolve on conventional reverse-phase HPLC can be detected with UPLC, contributing to optimization of the synthetic chemistry and the use of specialized building blocks.
Peptides used for clinical applications and made under cGMP guidelines require low levels of endotoxin in the final product. Endotoxins are removed during HPLC purification, explains John McKinley of American Peptide, and subsequent downstream processing must be performed under controlled, clean room conditions.
“The use of tray lyophilizers will improve product handling since both the freezing and drying of purified peptide solution takes place inside the lyophilizer, compared to bottle lyophilizers where pre-freezing of peptide solution in several bottles is required,” he says.
The company installed a 294L-tray lyophilizer equipped with a clean-in-place system in a clean room at its Vista, CA-based GMP facility.
“The temperature of each step is controlled, and the conductivity of the water can be monitored,” McKinley explains. A control system will include full PLC automation with 21 CFR Part 11 compliant operation.
Solid vs. Liquid-Phase Synthesis
Conventional wisdom holds that for API process development and GMP manufacturing, solid-phase peptide synthesis (SPPS) is more cost effective for longer peptide sequences (>10 amino acids) and smaller volumes, and liquid-phase peptide synthesis (LPPS) is more suitable for producing short sequences and large volumes. Hybrid SPPS-LPPS strategies find their niche for long sequences at large volumes.
A cost analysis should take into account raw materials, synthesis, purification, and lyophilization costs. Volume and economies of scale, complexity of the peptide and chemistries, solvent use, and waste produced can all affect manufacturing costs, overall efficiency and productivity, as well as sustainability of the synthetic process and environmental factors.
For SPPS, at any given scale, purification costs drive manufacturing cost, according to Mimoun Ayoub, Ph.D., vp, global business, sales and strategic developments at Peptisyntha, a Solvay company. Critical factors that can be modified to reduce purification costs include time, optimization of the stationary and mobile phases, recycling of solvents, control of process parameters such as load and flow rate, and product yield.
A cost analysis performed by Peptisyntha demonstrates the importance of crude purity on cost. A 10% increase in crude purity will lead to higher purification yields and cost savings of >50% of the API.
“Having said that, it is important for a manufacturer to select the right synthesis approach (SPPS, LPPS, or hybrid) to achieve a robust and cost effective manufacturing process taking into account the volumes of the API, the crude delivered by the synthesis approach, and timelines,” says Dr. Ayoub.
He points to several strategies for improving the robustness of synthetic processes, including persilylation technology, phenyl-oxy-carbonyl (Phoc) chemistry, or tetraphenyl borate (TPB) technology used for synthesis of arginine-containing peptides, and urethane-protected N-carboxyanhydride building blocks.
Peptisyntha utilizes all of these technologies to enable cost-effective production by increasing yield and purity, reducing the number of chemical steps, avoiding racemization, and allowing for synthesis of difficult sequences.
Jon Holbech Rasmussen, director of global development, Polypeptide Group, considers the choice between classical solution phase vs. solid-phase synthesis for producing peptides 10 amino acids or shorter “a very interesting exercise.” Faced with the efficiencies of the solid phase, if we can understand and master solid-phase synthesis, and take some of the advantages intrinsic to solution-phase synthesis and transfer them to the solid phase, then we can push the limits of when the liquid and solid-phase synthesis methods reach the break-even point.
“You have to go into rather large volumes before solution phase becomes economically attractive,” notes Rasmussen, who dismisses the idea that solid-phase synthesis is only good for small scale and long peptides.
“It can be highly efficient for synthesizing shorter peptides at larger volumes. This change in thinking is related in part to better manufactured, more consistent solid phase resins and to a better understanding of what is actually going on in the resin during the synthesis reactions. By adopting this change in thinking we will be able to exploit the boundaries of this technology.”
The goal now is to determine which factors should guide technology selection for synthesizing a particular peptide. It is no longer a clear-cut, volume-dependent selection. One way of looking at it is to say that solid phase is merely one kind of protecting group.” From that perspective, it is only a matter of applying the available chemistries.
Rasmussen cites an example of a recent synthesis project in which “we were working with a peptide with a high number of unusual amino acids, and we realized that the classical solution phase was probably less efficient because it could only run dilute.
“Forcing the synthesis onto a diphasic solid phase system, we could increase the effective concentration of the synthesis, and that is really what drives cost. You can have a 5 cubic meter reactor, but if you can only fill it up with grams of material because the material is insoluble then it doesn’t really matter,” he says.
Furthermore, Rasmussen challenges the long-held belief that SPPS necessarily has heavy solvent consumption. Solvent consumption is linked to concentration, he explains, and optimizing wash cycles and exposure of amino acids to the growing peptide chain can reduce solvent use.
Polypeptide Group is increasingly using modeling to predict, design, and assess synthetic processes. The company has developed an in silico method that allows them to input a peptide sequence and generate information such as the potential for aggregation, a pH curve, or a molecular weight fingerprint, and to predict potential problems that might arise during synthesis.
Similarly, an impurity predictive program can predict 90% to 95% of impurities.
The biggest change in the peptide synthesis market in the past three to five years, according to Jason Chang of CS Bio, has been the increase in peptide modifications.
“It is rare that we get a project for a cGMP large peptide that is just a ‘straight’ peptide,” he says.
They more typically are pegylated or radiolabeled, have lipid additions or attached sugars, or are DNA/RNA-peptide combinations. Stapled peptides are becoming increasingly popular.
“Peptide science has come a long way,” he continues, for the purpose of making peptides more stable in the circulation and more viable as a drug candidates. Companies are even taking another look at “old” peptide drug candidates that may have been put on the shelf due to poor pharmacokinetics or bioavailability.
Manufacturing these modified peptides can introduce some challenges, particularly related to solubility when performing HPLC-based purification in aqueous solutions. But, “it is all within the realm of organic chemistry,” as is peptide synthesis is as well, “so it’s very compatible,” says Chang. “The chemistry is feasible, and because of that is has been a real boon for the industry.”