TBP01x Best Practice 2a Semi synthetic artemisinin production
Welcome to this unit on semi synthetic artemisinin production as your real world example
as technology for biobased products. My name is Chris Paddon, I am a scientist at a
company called Amyris in California in the USA. Let me start by giving you the full title of the
talk, it is:  Developing production of the world s most important anti malarial drug using
synthetic biology and fermentation technology . Before getting into the science of the talk
let me tell you a little bit about the company that I work for. The company is Amyris, we
have a proven technology, we have multiple molecules in production, we re operating at
industrial manufacturing scale; the backdrop shows our production plant in Brotas Brazil and
we have a differentiated and proven of business model with multiple investors as listed
across the bottom of the slide here. The company was founded in 2003 by post doctoral
fellows from the university of California, Berkeley and the headquarters are just outside of
San Francisco, the picture is our headquarters there. We ve got almost 400 employees of
whom a good proportion have PhDs and well over 200 issued patents. So that is the
background of the company. Let me talk to you about malaria, the disease that we are
impacting. The map shows the areas in the world were malaria is endemic. As you can see in
the black, all of Africa, South East Asia, South America and a good part of Central America
are impacted by malaria. The statistics are huge, there are over a 100 counties affected
worldwide, over 200 million cases in 2012 and over 600,000 deaths. A sober statistic is that a
child dies of malaria every minute. And the horrible thing is that the vast majority of these
kids, these patients, are children under the age of 5 in sub Saharan Africa. The disease is
transmitted by the female Anopheles mosquito, the parasite is Plasmodium falciparum and
shown in the picture is an infected red blood cell here. There have been drugs for a long
time for malaria, but the problem is that the old drugs are failing. The insert map here,
shows you in the brown the areas were resistance to the old drugs has developed.
Specifically Chloroquines are the most widely used drugs to treat malaria but the Plamodium
parasite is now largely resistant to chloroquine. The successor drugs, which is a combination
of sulfadoxine and purimethamine are also largely ineffective now due to resistance by the
parasite. So what are we to do? Well in 1974 a highly effective drug was identified by
Chinese scientists, it comes from the plant Artemisia Annua, otherwise known as sweet
wormwood. The structure of Artemisinin, the drug, is shown here. And in 2004, the World
Health Organization recommended the use of artemisinin combination therapy for the
treatment of malaria. The word combination therapy means that you always give the drug in
combination with another effective anti malarial drug. The idea is that you have the
artemisinin and that you have another effective anti malarial at the same time. The point of
this is, that you both increase the efficacy of the anti malarial combination but you also
decrease the change of developing resistance to artemisinin by the parasite. There is a
problem though, the price and supply of plant derived Artemisinin has fluctuated greatly
over the years. This shows you the price is US $/kg of artemisinin derived from the plant
over the past decade or so. As you can see, in 2005, after the WHO recommended the
artemisinin use, the price shot up to over a 1000 $/kg. After that it plummeted, and went up
again. In 2009 it was over 600 $/kg and there was a shortage of the drug. Since then it has
fluctuated again and it is not possible to tell what the future will bring. This means that it is
very hard for the ACT manufacturers to plan production. So, what we aim to do is to use
fermentation and chemistry to stabilize the supply. The agricultural supply, as it is currently
exhausted, takes about 10 months from planting the seed to producing the drug. What the
semi synthetic artemisinin approach does is using fermentation and photochemistry to
shorten this time to 3 months. But also it becomes totally predictable, not subject the
vagaries of the agricultural supply. Now, artemisinin is a terpenoid, so what it a terpenoid?
Well you come across terpenoids in your daily live. So examples here: flavors, the hops the
in beer that s a terpenoid. Drugs, the anti cancer drug taxol is a terpenoid, the organge color
in carrot, the pigments these are terpenoids as is rubber, used to make all tiers and many
other products I am sure you re familier with. There are approximalty 50,000 terpenoids, all
made from the five carbon isoprene unit and this shows the structure of the five carbon
isoprene unit. So terpenoids come in units of five. Amyris origins lie in academic research on
engineering microbes to produce terpenes. It all started in 2003 with a paper published by
University of California, Berkeley in the lab of professor Jay Keasling;  Engineering a
mevalonate pathway in Escherichia coli for the production of terpenoids . This lead to the
semi synthetic artemisinin project, which is a Not for profit project funded by the Bill and
Melinda Gates Foundation, it s a collaboration between four partners, the University of
California, Berkeley, my company Amyris, PATH, a Not for profit company based in Seattle
and San Francisco and a French pharmaceutical company named Sanofi. So let s move on to
the science. Let me talk to you first about E.coli, importing the isoprenoid pathway from
yeast and Artemisia annua. I should mention that isoprenoid and terpenoid are
interchangeable terms. So this shows you the basic metabolic pathway of E.coli, from
glucose, down to this product FPP. FPP is the 15 carbon precursor of all the 15 carbon
terpenoids, or isoprenoids. The first step was to express the enzyme from the plant, called
Amorphadiene Synthase which converts FPP to Amorphadiene, which is the hydrocarbon
precursor for artemisinin. This produced some Amorphadiene. The big breakthrough came
with the expression of the entire Mevalonate pathway from yeast in E.coli, producing a
firehose of FPP and much higher production. Using this strain and quite a bit of a
development we at Amyris were able to produce 25 g/L of amorphadiene in E.coli. And this
shows you the result of three fermentation runs producing the 25 g/L target. So let me give
you a prime into some of the important compounds in semi synthetic artemisinin synthesis.
I have shown you amorphadiene, by biological oxidation the plant prod uces artemisinic acid
and that is what we wanted to produce for the production of artemisinin. Artemisinic acid
would be extracted from the fermentation broth and then by synthetic chemistry converted
to artemisinin. So the product that we really wanted to make in the production organisms is
artemisinic acid. So let me introduce you now to an alternative host, yeast. So E.coli vs.
yeast: here are some of the properties of yeast that attracted us; it s a eukaryotic system, it
is simple to engineer, it is industrially scalable, think of beer, wine and alcohol and it can
express the enzyme class, the cytochrome P450s, that converts amorphadiene to artemisinic
acid in the plant. So some work was done on amorphadiene and artemisinic acid production
in yeast and the graph shows you a succession of genetic mutations in yeast that allowed the
production of 150 mg/L of amorphadiene. We were also able to engineer the yeast to
express the plant enzyme that converts amorphadiene to artemisinic acid and to produce a
100 mg/L of artemisinic acid. So let s do a comparison of E.coli and yeast. The aim was to
produce 25 g/L of artemisinic acid, by fermentation. You remember that the E.coli was
producing 25 g/L of amorphadiene, yeast was producing 150 mg/L. So E.coli was producing a
100X more amorphadiene than was yeast. On the other hand artemisinic acid production;
yeast had produced a low concentration but E.coli had produced none and what s more is
that E.coli had a terrible reputation for expressing the cytochrome p450 class that converted
amorphadiene to artemisinic acid. So which way should the project team go? We had a
number of choices, but resources were, as always, limited and we couldn t do everyting. We
could engineer E.coli to make high concentrations of artemisinic acid, but that would require
good expression of the P450. We could engineer yeast to make high concentrations of
artemisinic acid but of course we only made a low concentration of armorphadien but we
had made some artemisinic acid. A third alternative was to use E.coli to make amorphadiene
but to develop the chemistry to convert it to artemisinin, completely novel chemistry. So
what would you do? This is a choice that we re left with. You think what would you do and I
see you in the next unit and we can tell you what we did. Thank you.