iGEM Manchester Header

iGEM Manchester - Industry

In order to evaluate the future prospects of our project, it must be analysed from an industrial perspective and compared to the current methods by which the typical L-DOPA supplement treatment is synthesised. Since 1967, L-DOPA supplements have been used for dopamine replacement therapy [1], and they continue to be the standard today. In 2005 the L-DOPA market reached a total sales volume of 250 ton/year, with a market worth of $101 billion [2]. Although our approach produces a fundamentally different product, both treat the same problem and must therefore be considered in comparison for viability as medicines. The key factors for industrial comparison include waste production, efficiency and energy consumption, general safety of reactions, as well as time taken for processes to run to completion. In our analysis we considered all of these aspects.

Current production methods

Figure 1 Summaries of chemical L-DOPA

The industrial processes employed today for the production of L-DOPA are the Hoffmann La-Roche, the Monsanto, the Ajinomoto, and the Sankyo process. The Hoffmann La-Roche, Monsanto and Sankyo processes all are linear, start with vanillin, and through interaction with protected glycine and aldehyde condensation yield a carbon-carbon double bond. Hydrogenation of a double bond in an L-DOPA synthetic precursor produces L-DOPA.

Resolution-based Hoffmann La-Roche synthesis (HRP) was developed in 1960 and is considered a not particularly efficient process, being a linear chain of four reactions (Fig 2). In the first step, Erlenmeyer–Plöchl condensation of vanillin and hippuric acid is followed by phenol acetylation and hydrolysis which gives azlactone (an enamine). In the second step the carbon-carbon double bond is hydrogenated and azlactone is reduced to the corresponding N-benzoyl amino acid. The third step separates the (S)-enantiomer by resolution with half-equivalent (+)-dehydroabietylamine. In the fourth and final step, the (S)-enantiomeric salt is decomposed and deprotected by hydrogen bromide. Even though three steps are highly integrated, the overall yield is low due to the resolution step at 19.2% [1].

Figure 2 The Hoffman La-Roche process

The Monsanto process was developed by a Nobel Prize laureate (2001), Wiliam S. Kowels, with the first pilot synthesis taking place in 1974. The Monsanto process is notable as the first commercial process that uses asymmetric synthesis with transition metal complexes (Fig 3). It totals four steps, the first being the synthesis of a side chain of azlactone by protection of amine and vanillin followed by condensation. Then the resulting enamine is hydrolysed to enamide. This enamide is then asymmetrically hydrogenated to give the desired (S)-enantiomer. Finally, the (S)-enantiomer is deprotected yielding phenols, amides and L-DOPA [1].

Figure 3 The Monsanto process

The Ajinomoto process is a biosynthetic process that was developed by Kumagai’s group at Kyoto University. It is a three-component one-step process that uses a biocatalytic route (Fig 4). Tyrosine Phenol Lyase (TPL) is an enzyme that uses the non-physiological substrate catechol. TPL installs a side chain consisting of an amino group by forming a carbon-carbon bond and carbon-nitrogen bond at the same time from 3 substrates. Since the reaction is reversible, L-DOPA crystallization induced by seeding shifts the equilibrium to the product side [1].

Figure 4 The Ajinomoto process

The Sankyo process was developed in 1990 and introduced commercially in 2008. It is predicted to reach an annual yield of 3000 ton and become the leading process by which L-DOPA is manufactured. It is composed of seven reactions in total, making it the longest process considered here (Fig 5), starting with vanillin being protected by methylation to yield an aldehyde. The newly synthesized aldehyde is condensed into hydantoin, which makes benzylidene hydantoin. Subsequent hydrogenation of benzylidene hydantoin gives a racemic mixture that is hydrated, acetylated, and then the enantiomer of interest is selectively deacetylated via an enzymatic reaction. Hydrolysis and deprotection of two methoxy groups yields L-DOPA [1].

Figure 5 The Sankyo process

General waste

Any industrial process has to be assessed by its strengths and weaknesses, with some criteria being more important than others. Waste and efficiency are key examples of important factors in industrial analysis. In understanding the importance of these factors, and how current industry standards perform in this respect, we can better discern the place of our process in the future of manufacturing.

Atom Economy, developed in 1991 by Barry M. Trost with the goal of achieving “synthetic efficiency in transforming readily available starting materials to the final target” [3], is a key metric in determining the efficiency of a synthetic process and evaluating alternative strategies.

Table 1 Atom economy of L-DOPA synthesis processes

An important limitation of Atom economy is that it remains fixed, regardless of the chosen catalyst and other reaction parameters; it is therefore important to recognize that an atom economy analysis will not always determine the greenest approach [3]. The higher AE value, the more economical the process is and therefore the cheaper it is to run. From the table above, the Ajinomoto process is the most economical, compared to Hoffmann La-Roche, Monsanto and the Sankyo processes. The AE of the Hoffmann La-Roche process is lower that that of the Sankyo process because the chemical input in the former is larger: hippuric acid has a larger molar mass than hydantoin (179.17 g/mol vs 100.08 g/mol).

Synthetic ideality is a complex factor that enables the definition of the plan efficiency for a chemical process. Since the Hoffman La-Roche chemical process is closer to practice, its synthetic ideality is higher that that of the Sankyo process. The Sankyo process is the least integrated out of the processes listed above, as well as being the most hazardous with its production of the neurotoxic ozone-layer depleting chemical – methyl bromide. The Ajinomoto process is the most efficient regarding synthetic ideality, in addition to being convenient to operate, since there is no material concern or toxic by-product mitigation [2]. The Ajinomoto process is also a biotechnological process, which allows us to estimate similar parameter values for AE and synthetic ideality for our biotechnological production of L-DOPA using Tyrosine Hydroxylase or Tyrosinase.

Two other metrics that are useful for waste comparison are Process mass intensity and Waste Stream analysis. The former is a factor recommended by the ACS Green Chemistry Institute Pharmaceutical Round Table and measures the mass efficiency from input, encouraging cost reduction at the very start of the process, defined by:


PMI can give a realistic picture of a process, providing a “process mass property of a contributing chemical in a synthesis plan” as long as every chemical is tagged in analysis. It is a useful factor that represents the contribution of different reagents to the final product and identifies productivity issues as well as predicting global impact of a specific improvement to the system. Waste stream (WS) is the flux of waste from the point of generation to the point of its final disposal or treatment, “the characteristics of waste stream inventory will help to identify hazardous streams, evaluate options for solvent and energy recovery, estimate the needs for waste management, and develop integrated treatment based on site capacity” [1].

Figure 6 PMI of the Hoffman La-Roche process

Hoffmann La-Roche has 12 WSs: 10 aqueous WSs (3 solvent-free, 7 organic or solvent-contaminated) and 2 solid WSs with a total waste for all WSs being 48240.33 kg per 100 kg of L-DOPA synthesized; 81.27% of which (39205.02 kg) is contaminated water, primarily being produced in WS 2-2 (Cyclisation, isolation; 19255.60 kg). The Deprotection step (4-1) generates hazardous MeBr shown in WS 4-7 that is needed to be handled separately in addition to the existing WSs arising directly from synthesis. DMF and benzoic acid must also be handled as hazardous waste, decreasing the mobility of the process. Recognized improvements include: recycling of the (+)-dehydrobietylamine separating agent and the biodegradation of ethanolamine [1].

Table 2 Waste Streams for the Hoffman La-Roche process

Figure 7 PMI of the Monsanto process

The Monsanto process has 10 WSs: 2 solvent-free aqueous streams, 9 organic or organic-laden WS and 3 organic-laden aqueous WS, producing a total waste for all WSs of 10373.28 kg per 100 kg of L-DOPA synthesized; with 58.62% of this waste (6081.13 kg) being contaminated water. Organic compounds include acetone from WS 2-1 and isopropanol and water from WS 3-1, all of which are regulated by CWA and can be recycled or used for energy recovery due to simple WS composition. The Scrubbing of MeBr produced in WS 4-1 gives WS 4-5 that is similar to WS 4-7 of the Hoffmann La-Roche process. Since WS 4-5 is the largest and needs pre-treatments, it makes up the majority cost of the overall process [1].

Table 3 Waste Streams for the Monsanto process

Figure 8 PMI of the Monsanto process

The Ajinomoto process generates 5 WSs: 1 solid WS and 4 aqueous WSs, with the total waste for all WSs being 3879.37 kg per 100 kg of L-DOPA synthesized; with 59.34% of this waste (2301.85 kg) being contaminated water. The Ajinomoto process is characterised by the absence of any hazardous chemicals. The solid WS is recyclable and four aqueous WSs are treatable by biological methods, with the main challenge being compliance with the standards of discharge set by Chemical Oxygen Demand (COD) and Biochemical Oxygen Demand (BOD) [1].

Table 4 Waste Streams for the Ajinomoto process

Unfortunately, information for Sankyo's WSs were unavailable, and although we did learn from one of our industrial interviews that accurate extrapolation does require a full view of the industrial process, changes to the synthesis may be made in scaling up for manufacturing. For example, side reactions may increase, as physical parameters change with increasing volume of reagents, causing diminished purity that requires further treatment before shipping. Consequently, we could not get an accurate estimate for the waste caused by the Sankyo process.

c.E factor:
Another metric that ties together numerous waste components in the broader context of production is the Environmental factor (E factor), defined as:


It characterizes the synthetic process on a practical level, proving useful for assessing the contribution of separate elements as it uses the mass of the products as a common denominator, with the number of Waste streams directly affecting the E factor [1]:


Figure 9 E factor of Hoffman La-Roche WS

Each waste stream can be characterized by its E factor; the higher the E factor, the higher the waste produced to synthesise a particular chemical, and thus the greater the negative environmental impact. The ideal E factor is zero. Unlike atom economy, which evaluates the theoretical waste of raw materials expected to be generated in a chemical process, the E factor shows the amount of waste produced in practice during the process, considering every piece matter involved in a chemical reaction, as well as yield, reagents, solvent losses, all process aids and energy. Water is not typically included in E factor calculations in order to prevent skewing, and give results that act as a better metric for comparison [4].

Figure 10 E factor of Monsanto WS
Figure 11 E factor of Ajinomoto WS

Based on the information in table 5, and knowledge of fermentation, we can conclude that biotechnologically based processes are the more environmentally friendly when compared to chemical synthesis. A shift in production of L-DOPA toward biotechnological solutions would help achieve the target set at the World Summit on Sustainable Development (WSSD) for chemical synthesis to stop methods leading to adverse effects on human health and the environment by 2020 [5]. E factor is a relatively new metric that considers the economic value of eliminating waste and avoiding the use of toxic/hazardous chemicals as opposed to a purely yield based analysis, moving in the right direction for creating a more sustainable chemical industry for the future.

Table 5 Summarising the E factors of L-DOPA synthesis methods[1]

Methyl Bromide and safety/environment
Another important factor in deciding on an industrial process are health and safety, both of employees, the general public and the subsequent cost to treat harmful by-products. As mentioned above in table 5, all the chemical syntheses processes produce methyl bromide (MeBr): a volatile organic neurotoxin regulated by the US Clean Water Act, Clean Air Act, Resource Conservation and Recovery Act with permissible exposure of only 20 ppm. Common warnings concerning exposure include: Acute toxicity (oral, dermal, inhalation), respiratory sensitization, germ cell mutagenicity, carcinogenicity, reproductive and multiple organ toxicity, toxic to aquatic life, lethal.

Table 6 Amount of MeBr produced in L-DOPA synthesis

MeBr gas (boiling point 4C˚) can be treated by either pyrolysis or scrubber decomposition. Thermal decomposition (pyrolysis) is the primary method for dealing with volatile organic compounds, in the case of MeBr, however, hydrobromic acid and other harmful decomposition byproducts are formed depending on temperature, time of exposure and gas mixture components. The other treatment option is scrubber decomposition, in which MeBr is reacted with ethanolamine to produce safe products [1]. Unfortunately, despite the health and environmental advantages, this method is not employed in all the manufacturing processes yet, due to economic and infrastructure problems.

Figure 12 Scrubber decomposition

In 2006 the World Meteorological Organization and the National Oceanic and Atmospheric Administration and the National Aeronautics and Space Administration published research that showed that methyl bromide significantly depletes the Earth's stratospheric ozone layer [6]. Depletion of the stratospheric ozone layer allows UV-C and UV-B light to reach the planet’s surface, increasing risk of melanoma and non-melanoma skin cancers, eye cataracts and weakened immune systems as well as damaging plastics, ocean ecosystems as well as the health of plants and animals [7]. The half-life of the photochemical lysis of MeBr is between 4 and 20 months, making it a molecule that poses significant threat to the recovery of ozone layer [8].

Death from dangerous chemicals was among the global top ten leading causes of death in 2004, and continue to remain a significant problem. The World Health Organisation reported that 4.9 million deaths and 86 million Disability-Adjusted Life Years (DALYs) (8.3% of the global total of deaths and 5.7% of the global total of DALYs in 2004, respectively) were attributable to environmental exposure and management of selected dangerous chemicals in 2004. The annual deaths due to occupational particulates were 375,000, acute poisonings by chemicals were 240,000 and self-poisonings by pesticides were 186,000 - totalling 964,000 deaths and 20,986,153 DALYs, corresponding to 1.6% of the total deaths and 1.4% of the total burden of disease globally. The Strategic Approach to International Chemicals Management (SAICM) on the World Summit on Sustainable Development (WSSD) has set an aim to use and produce chemicals by 2020 in “ways that do not lead to significant adverse effects on human health and the environment” [9]. The deaths caused by dangerous chemicals and processes can be curbed by eliminating processes in the chemical industry that contribute to these statistics, exchanging total syntheses for biotechnological approaches.

Economics and time

Ultimately, however, when businesses consider new approaches for production, the primary focus is on the bottom line, of which the predominant factor is time. All the reactions depicted in the tables below are linearly sequential, meaning reactions cannot run in parallel, and as a result the shortest reaction sequence yields the lowest overhead. Production time is always potentially affected by machine maintenance, warming up, waiting in the machine cycle, breakdown repairs, lack of materials and hygiene [10].

Table 7 Timescale of the Hoffmann La-Roche process (100 kg of L-DOPA)[1]
Table 8 Timescale of the Monsanto process (100 kg of L-DOPA)[1]
Table 9 Timescale of the Ajinomoto process (100 kg of L-DOPA)[1]
Table 10 Timescale of the Sankyo process (7.3 g of L-DOPA)[1]

Despite the Sankyo process producing three times the amount of MeBr, it is by far the fastest way to produce L-DOPA, making it a favorite for companies and producing better margins.

Law, regulation and comparison

Another factor that cannot be overlooked when considering future prospects of our method as a business opportunity is legislation concerning GMO’s and intellectual property rights. Despite the fact that our project is now in the open domain and involved in an open source library of parts (i.e., the iGEM registry), our pathway and genetic engineering of bacteria to produce regulated levels of L-DOPA in the gut had a hypothetical possibility of patenting in the UK and USA [11,12]. Under UK patenting law, the project fits under the class of patentable inventions, with the method of administering L-DOPA being novel and having potential medical applications [13]. In the US, the technology is also eligible for a utility patent according to the classifications described under US patenting law, having to be applied for within 12 months of the UK request as a foreign application [14]. As of the time of writing, no patent currently exists for the use of probiotics for the dispensing of L-DOPA in the gut.

The governments of both the US and the UK currently consider the use and development of synthetic biology in a positive light. In the UK there exist a number of well-established academic and industrial institutions for research on synthetic biology, with many sources of funding available through agencies such as BBSRC, EPSRC, MRC and ERA-NET, as well as many industry-based initiatives being set up to build synthetic biology infrastructure [15]. Although certain bans on GMO crops have been imposed in the UK, no such bans are currently being enforced or considered for probiotics [16]. Similarly, the US have various initiatives to encourage biotechnology growth including organizations and research centers such as DARPA’s Living Foundries ATCG program, NSF ERC and Emerging Frontiers in Research and Innovation (EFRI) program. Another program, the Small Business Innovative Research and the Small Business Technology Transfer (SBIR/STTR) programs, looked for proposals with realistic synthetic biology applications [17]. In 2014, a report “Industrialization of Biology: A Roadmap to Accelerate Advanced Manufacturing of Chemicals” identified key points to further advance synthetic biology research in the US [15]. Similarly, the US has some bans on GM crops, but nothing for probiotics (although the case of genetically modified probiotics has not yet been explored in practice).

When it comes to implementing fermentation to produce our probiotics, the feedback we received from industry was generally positive, although it foresaw problems with its implementation in an industrial process. If we used our process as a standard way of fermenting L-DOPA, it could not compete with the chemical synthesis methods already in place; however, as a probiotic the outlook is clearly more positive. As fermentation is efficient, generally needs no purification and does not deal with high temperatures or pressures, the energy and waste treatment costs are attractive. The product is also better, with a better target for delivery and therefore better overall pharmacological performance and a strong advantage in the potential to patent the technology.

However, the overheads are comparatively expensive for fermentation; this is aggravated by a drop off in production as a result of spontaneous mutants, for which some sort of selective pressure solution would have to be developed. Despite this, the raw material is much cheaper, as well as being sustainable, giving an economic advantage via future proofing as well as protecting against the potential government legislature restricting waste output and energy consumption found in competing processes. Industrialists interviewed were also worried about the loss of profit in producing a probiotic solution that was self-sustainable, suggesting a less stable strain which would require repeat dosing.

With 50% of children born today expected to live to 100 [18], the late onset disease of Parkinson’s becomes an ever more prevalent problem, and despite economic and commercial factors working against us, we hope that our greener, more sustainable, patient-friendly alternative would find a new niche in the Parkinson’s treatment market.

[1] Li, T,. and Li, X,. (2014) Comprehensive Mass Analysis of Commercial L-Dopa Manufacture. Green Chemistry. 16 (9), pp. 4241-4256.
[2] Inamdar, S,. Surwase, S,. Jadhav, S,. Bapat, V,. and Jadhav, J,. (2013) Statistically optimized biotransformation protocol for continuous production of L-DOPA using Mucuna monosperma callus culture. Springer Plus. 2 (1), pp 1-9.
[3] Dicks, P., and Hent, A,. (2015) Chapter 2 Atom Economy and Reaction Mass Efficiency. In: Sanjay K. Sharma Green Chemistry Metrics A Guide to Determining and Evaluating Process Greenness. India: Springer International Publishing. pp 17-44.
[4] Roger Sheldon (2015) The E factor. [online] Available at: [Accessed 25 Aug. 2015].
[5] United Nations Environment Programme (2015). Harmful substances and hazardous waste. [online] Available at: [Accessed 25 Aug. 2015].
[6] US Environmental Protection Agency (2011). Methyl Bromide Questions and Answers. [online] Available at: <> [Accessed 25 Aug. 2015].
[7] United Nations Environmental Programme (2003). Montreal Protocol talks to consider exemptions from methyl bromide ban. [online] Available at: <> [Accessed 25 Aug. 2015].
[8] Organisation for Economic Co-operation and Development Screening Information Data Set (2002). Screening Information Data Set Initial Assessment Report for 13th SIDS initial assessment meeting. [online] Available at: [Accessed 25 Aug. 2015].
[9] Kaj, M,. Pierre, Q,. and Cyrille-Lazare S,. (2013). Costs of Inaction on the Sound Management of Chemicals. Kenya: United Nations Environmental Programme. United Nations Environmental Programme.
[10] Dudbridge, M. (2011). Handbook of Lean Manufacturing in the Food Industry. Oxford: Blackwell Publishing. N/A.
[11] GOV.UK (2015). Before you apply. [online] Available from: [Accessed: 6th Sept. 2015]
[12] GOV.UK (2015). Patents step by step. [online] Available from: [Accessed: 6th Sept. 2015]
[13] GOV. UK (2015). The Patents act 1977. [online] Available at: [Accessed 6 Sept. 2015]
[14] USPTO (2014). General information concerning patents. [online] Available at: [Accessed 6 Sept. 2015]
[15] Research Councils UK (2014). A synthetic biology roadmap for the UK. [online] Available at: [Accessed 6 Sept. 2015]
[16] Organic consumers association (2015). Countries with GE Food/Crop bans. [online] Available at: [Accessed 1 Sept. 2015]
[17] ERA-SynBio (2015). Synthetic Biology in the USA. [online] Available at: [Accessed 3 Sept. 2015]
[18] Mallet, N,. and Dupret, D,. (2011). Parkinson's Disease. Biological Sciences Review. 24 (1), pp 9-13.