The Hangzhou Public Bicycle System has surpassed Vélib as the largest bike sharing program in the world. Not surprisingly, it has sparked great interest in bike sharing in Mainland China. Indeed, Beijing, Tianjin, Hainan, and Suzhou have already launched pilot programs in 2008 and 2009.Nature plays a crucial role in medicine. Natural products, or molecules produced by living organisms, are a rich source of novel molecules that have a myriad of beneficial properties. The idea of using natural products as medicines is not a new concept. In fact, it dates back to ancient civilizations where various plant products were used to manage, treat and prevent disease. Mesopotamia relied on cedar and cypress oils to cure ailments, Egypt had a collection of about 700 plant based medicines detailed in “Ebers Papyrus.” Chinese civilizations also cultivated their own list of natural medicines with the first record dating back to 1100 BC.1 Over time the lists of natural remedies evolved; however, it was not until the 19th century AD when scientists began to identify the actual molecules that had the medicinal effects. Morphine was the first natural product to be isolated from its natural source and sold in 1826.Since, scientists have continued to identify novel natural products with amazing properties, and while the list of natural product medicines is long, some of the most notable are: penicillin , paclitaxel , lovastatin , quinine and cannabidiol .Beyond medicinal natural products, scientists have also identified natural products that can be used as fuels, fragrances and dyes. Nature continues to be a rich source for novel molecules, with some predicting only 10% of the biodiversity that exists on the planet has been discovered. Despite the importance of natural products in everyday life, drying cannabis there has been a paradigm shift from the use of unmodified natural products to semi-synthetic modified natural products, and synthetic compounds. 

This is in part due to the expense associated with identifying and characterizing natural products, and the decision of the Supreme court in Association for Molecular Pathology v. Myriad Genetics, 569 U.S.  which decided that natural products cannot be patented, thereby limiting economic motivation for identifying new natural products. However, it is imperative to realize the myriad of applications for natural products in human health and other industries, and continue to develop methodologies to identify and synthesize this class of molecules. Herein, we focus on methods to produce natural products. Chemical synthesis is the most common method used to produce a molecule, and while this technique can be used to produce almost any molecule, the process is not always economically feasible on an industrial scale. This is often the case for natural products; the chemical synthesis is possible , but the cost of the synthesis process is greater than the cost of extracting the target molecule from a natural source. As a result, the world’s supply of morphine, paclitaxel and cannabinoids as well as an array of other molecules is still dependent on natural sources. However, there are many challenges associated with relying on natural sources to produce these crucial medicines. Plants are slow growing, susceptible to environmental conditions, inconsistent in secondary metabolite production, and secondary metabolites are often present at very low concentrations. A great example is paclitaxel, produced by the pacific yew tree. It requires 3 to 6, 100-year old, pacific yew trees to treat one cancer patient.1 This is simply not sustainable. Additionally, there can also be political implications. The world’s supply of morphine is predominantly produced in Turkey, India and Afghanistan, and shifts in the political climate could lead to shortages or high prices of the essential drug. So while chemical synthesis and natural product extraction are sufficient for now, in order to capitalize on the properties of natural products, we need new cost-effective, consistent and efficient methods to produce natural products.

Several ideas have been proposed to address these challenges outlined in Figure 1-1, such as engineering plants to improve natural abundance, plant cell culture, engineering microbes, cell-free biosynthesis and synthetic biochemistry. Herein, each method listed above is reviewed for the pros and cons. It is important to realize that natural products are unique and different methods will be suitable for different molecules. First, some effort has been dedicated to engineering plants to either improve the abundance of a target molecule, or engineer a plant to make a natural product it would not make normally. Zhang et al were able to engineer tomatoes to improve the production of genestein, a flavonoid with an array of important medicinal properties. Naturally tomatoes produce very low levels of genistein , but Zhang et al improved the amount of genistein to 78 mg/g of dry weight, nearly 0.8% of the tomato DW. Additionally, you could engineer a plant to produce a molecule it would not have produced naturally. In addition to boosting genistein levels, Zhang et al were able to engineer tomatoes to produce resveratrol a compound not naturally produced by tomatoes at levels 100-fold higher than grapes .However genetically engineering plants can be challenging. There are only a few species that have been extensively studied, which limits the number of model systems. Additionally, plant metabolism is complex and it can be difficult to identify molecular components, like transcription factors, that are necessary to boost biosynthesis of the target molecule. This also does not address some of the inherent issues with plant sources, such as cultivation time and susceptibility to mold, pests and environmental conditions. Plant cell culture is an exciting alternative to extracting the compound from the natural source. They can be generated from a variety of plant species by first isolating plant tissue sterilizing the tissue, and plating the sterilized tissue on solid media supplemented with plant hormones and nutrients. The explants proliferate to form a callus . Calluses can then be used to initiate plant cell suspension cultures, which have the ability to produce secondary metabolites. This method has many advantages over extraction from the natural source such as improved sustainability, shorter incubation times, increased production level through metabolic engineering, consistent environmental conditions, and protection from insects and mold. 

Cell culture has been particularly fruitful for the production of paclitaxel, a powerful anticancer drug. As previously stated it would take 6, 100 year old trees to produce enough paclitaxel to treat one patient, however the plant cell cultures are able to produce 150 mg/L in 6 weeks, a dramatic improvement. Plant cell culture is well suited for the production of paclitaxel due to the complexity of the bio-synthetic pathway, so generating cell lines from cells that naturally produce paclitaxel is a relatively simple process. Some additional examples natural products produced by plant cell culture include: scopolamine , protoberberines rosmarinic acid , shikonin and geraniol .Despite the benefits of plant cell cultures there are still several challenges. The cultivation time is still 2-3 weeks, which is rather long compared to microbial cultivation times. Further genetic instability and physiological heterogeneity leads to variability and unpredictability in secondary metabolite production. The secondary metabolites also remain intracellular which can make it difficult to reach high titers. As this technology develops further it may be possible to address some of these issues. For example, Wilson et al are working to reduce genetic instability and rescue necrotic calluses in order to improve production of paclitaxel in plant cell cultures. However, some challenges are more difficult to address such as improving the titer of toxic compounds. While plant cell culture is an important method for natural product biosynthesis, specifically paclitaxel biosynthesis, the method is not broadly applicable to other natural products due to the complexity of plant metabolism. In addition to plants, microbes are a rich source of natural products. Both fungi and bacteria naturally produce an array of useful molecules. In fact, several FDA approved antibiotics are produced via microbial fermentation. However, it is also possible to engineer microbes to produce natural products from other organisms such as plants. Microbes are easy to manipulate, they have relatively short cultivation times , curing cannabis and are easier to culture than plant cells. As a result, there is an entire field of research dedicated to engineering microbes to produce nonnative natural products. Due to the vast number of studies conducted in this field, this review will only focus on several examples of microbial production of bio-fuels, terpenes, alkaloids and polyketides. In addition to being a rich source of medicines, nature also produces molecules that play an important role in everyday life. Due to the need for energy security and limits on the world’s petroleum supply there is a need for sustainable liquid fuels for transportation. Microbes naturally produce alcohols, isoprenoids, and fatty acids, which can be used as fuels or easily converted into fuels. 

As a result, a great deal of research has focused on engineering microbes to improve the natural production of bio-fuels. Microbial production of isobutanol and farnesene are reviewed below. Alcohol-derived bio-fuels are the ideal candidate to replace gasoline used in cars. Isobutanol has a high energy density low vapor pressure, and a high octane rating . Therefore, it can either be blended into gasoline or replace petrochemicals altogether. There have been several studies published that discuss the microbial production of isobutanol in various organisms, such as: E. coli, C. glutamicum, S. cerevisiae, C. acetobutylicum, R. eutropha, and S. elongatus. The highest titer achieved was an engineered strain of E. coli. Atsumi et al engineered E. coli to produce 22 g/L of isobutanol. First, they over expressed the enzymes in the isobutanol pathway shown in Figure 1-2. Then, they deleted non-essential genes that would divert precursors and intermediates out of the isobutanol pathway. Finally, they improved the flux of pyruvate into the isobutanol pathway by using a non-native enzyme with better kinetic parameters. A later study by Baez et al using a similar E. coli strain obtained titers of 50 g/L by continuously removing isobutanol from the culture, demonstrating the effects of isobutanol toxicity on E. coli cultures. While alcohol based bio-fuels are a possible alternative, they are too expensive to compete with the low cost of gasoline. Iftiters could be improved further, the cost of the bio-fuels would decrease, however this is unlikely due to the toxicity of alcohols at high concentrations. The highest microbial production of a bio-fuel reported thus far is a sesquiterpene derived bio-fuel, which can be converted into diesel fuel and jet fuel. Amyris was able to engineer yeast to produce the sesquiterpene farnesene at titers that exceed 130 g/L by re-engineering yeast central metabolism to direct sugar into the isoprenoid pathway. However, these production levels are still relatively low, when considering the energy demand of US transportation. At this titer, it would require 23.5 L of yeast culture to produce 1 gallon of farnesene. To replace the amount of jet fuel used in the US airlines in 2018 with farnesene, 1.2 billion liters of yeast would need to be cultivated and processed per day a total of 420 billion liters of yeast per year. In addition, while this titer is significantly higher than most reported, the cost of producing farnesene via microbial production is still more expensive than petrochemicals . The real challenge with bio-fuels is not engineering the microbes, but competing with the very low prices of petrochemical products. While bio-fuel production in microbes is generally too expensive to compete with fossil fuels, this is not the case with all natural products. Due to the challenges associated with chemical synthesis, a low natural abundance and the variability with plant based production of complex natural products, microbial fermentation is a plausible alternative for the production of high value natural products. Microbes have been engineered to produce an array of compounds, spanning three of the major classes of natural products: terpenoids, alkaloids and polyketides. Terpenes and terpenoids are a large diverse class of natural products derived from the mevalonate or methylerythritol phosphate pathway . The products of the MEP and MVA pathways are isopentyl pyrophosphate and dimethyallyl pyrophosphate , which serve as the core building blocks for terpene biosynthesis . For example, the condensation of DMAPP and IPP produces geranyl pyrophosphate , the precursor to monoterpenes. Condensation of GPP with IPP produces farnesyl pyrophosphate , which is the precursor to sesquiterpenes, like farnesene mentioned above. The condensation of two FPP molecules yields squalene , which is the precursor to cholesterol. Terpenes are classified by the number of isoprene units which range from monoterpenes to polyterpenes .