Finding Medicine Where You Least Expect It | Christina Smolke | TEDxStanford

Translator: Ivana Krivokuća
Reviewer: Mile Živković Medicines are essential
for our quality of life, and yet, we lack
safe and effective medicines for many of our most pressing
diseases and conditions. Many of these limitations
are very closely tied to how we get our medicines. What you’re looking at is digoxin, which is a medicine
that’s used to treat heart failure. This medicine is obtained
from the foxglove, which is an ornamental plant
that is commonly found in gardens. In fact, over half of our medicines
are obtained from nature, and a very large fraction of those
are obtained from plants. You can imagine that relying on nature
to provide our most important medicines comes with many challenges. The first is that nature is imperfect. These plants take
a very long time to grow, oftentimes over the course of a year, and during that time, they’re subjected
to variations in weather, to disease and pests, all of which can affect the amount
and the quality of medicine that’s available in any given year. The second is that most
of our most valuable and needed medicines are not going to be provided by nature. These plants are not making
these compounds as medicines for us; they’re making these compounds
for their own purposes, oftentimes for defense,
and what that means is that there’s just a large set of medicines
that we’re not going to be able to source from nature. The third is that most of the people
on this planet, still to this day, have insufficient access to medicines. With all of these challenges,
you might wonder wouldn’t it be better if we could actually just go in
and make our own medicines. We took that question and we said, “Okay, let’s start with one of the most
complex family of medicines out there”. This is a class of compounds
that has evaded decades of research in the laboratory in terms of trying
to arrive at better solutions. We said, “Could we develop an approach
that’s grounded in biotechnology, that would address
a lot of the challenges that arise when we rely on nature
to source our medicines?” What you’re looking at is the opium poppy. It’s from this plant that we get
many of our most effective pain medicines. We used this as our starting point. I want to start by discussing
the scale of what’s required to source medicines from nature. There’s about 800 tons of opiates
that are sourced annually to make our painkillers. What this equates to
in terms of land usage is about 100,000 hectares of land
that’s farmed for poppies. You can think about this roughly
as the size of Rhode Island. Because poppies require
a very specific climate to grow, and also because of the regulation
associated with growing these crops, what it means is that
there’s a very small number of geographical regions on our planet
where poppies are grown. Over half of these come from Australia. Finally, still to this day,
the large majority of people on our planet have insufficient access to pain medicines
to treat moderate and severe pain. Let’s look in a little bit closer at what this manufacturing platform
actually looks like. There are many steps from the point
at which we get or medicinal opioids to the point of the poppy. Essentially, this plant is producing
one particular compound, morphine, and it’s making this compound
as a very, very small fraction of the overall plant material. So we have to go into the plant and basically remove this compound
so that we can use it as a medicine. That will require us to use
fairly toxic chemicals to do that. Because these plants are being grown
oftentimes, very far distances from where the medicine
is actually being used, we then have to take
very large volumes of narcotic material and transport it very long distances to where the medicines
will ultimately be used. Then, finally, because the poppy is not
making the most valuable pain medicine, we have to take the compound
that we get from poppy – morphine – and then subject it further
to chemical processing steps to allow it to be converted
into higher value, more effective pain medicines. The next challenge that arises is that this medicinal plant
is not just making opiates; it’s making a number of other compounds that we have great interest in
to treat other types of diseases such as cancer and neurodegenerative
disorders and hypertension. So we’d like to be able
to go into this plant and retrieve these molecules as well. But the plant is making these
as a very complex mixture. It’s making all these different
compounds together. That adds a lot of cost
and complexity in terms of, once we retrieve the compounds,
trying to separate them, so that we have purified amounts
of each of the compounds that we can then use as medicines
and ultimately those challenges will limit our access
to those other types of medicines. Then finally, as many of us are aware of, this type of manufacturing
feeds directly into an illegal market, where poppies can be farmed
and fed into the heroin market. It’s also the case that we have abuse
of these medicinal opioids. We started this journey
about ten years ago. We asked a seemingly simple question, which was: could we shift
from a manufacturing platform that was based on growing
vast amounts of a drug crop, and instead ask
a simple single-celled organism, like baker’s yeast, to make compounds
that we could use as medicines directly, that would have improved properties. I want to start with an overview
of what this process actually looks like. Again, what we’re essentially trying to do
is move from the sort of one organism, one compound that that organism
has evolved to make, to one in which we actually go in
to the broader natural world and begin to look
at the different organisms that are present on our planet
and search for activities that we believe would be of interest
in making the perfect medicine. We can begin to look
at different organisms, we can pull out activities
that we believe will be of interest, and then we can take those activities
that are distributed in different organisms throughout nature, and begin to combine them
into a new single organism, in this case, baker’s yeast. At the end of that, what we’ll have
is an organism that now has these activities combined
in ways that you don’t find in nature. We can now grow this organism
very inexpensively, and over a period of several days. This organism will be able
to provide for us new sets of medicines
that we can’t find in nature. Now that we’ve kind of
looked at it at a high level, let’s zoom in to some of the details
of what this actually means. At this level, it’s useful
to think of the process that’s going on inside the yeast as this really complex
molecular assembly line, where this assembly line is essentially
taking the sugar that we feed yeast as a building block,
and then it’s building it up into a complex medicinal compound
that we can then use. The first challenge that happens is that,
in addition to sort of coaxing yeast into making this
really complex assembly line, the next challenge we’re confronted with
is the fact that oftentimes many of the components
that we would like to be able to use to build this assembly line, we simply don’t know what they are;
we haven’t found them. This presents us with our first challenge. How do we build within an environment
that we just don’t know actually, what we’re asking the yeast to make? The way that we do this
is that we go back into nature and we began to go and search
the genomes of those different organisms. Essentially, what we’re doing
is we’re looking at the DNA sequences of many different organisms, and that sequence essentially
holds the directions for how those organisms
make different activities that we believe will be of interest
to make new medicines. We can use computational tools
that will search those DNA sequences and essentially identify its subsets
of the DNA sequence that we believe encodes
interesting activities. Once we identify
those subsets of sequences, we can simply take that DNA sequence,
move it into our yeast host, and begin to test directly
whether the activity is actually what we hoped it would be and whether it will work
to build our medicine. This leads to the second challenge
that then comes up. You can imagine that plants and yeast
are very different organisms. We’re basically taking components that have evolved
over many thousands of years to work within plants,
which is a very particular environment, and asking them to move
directly over into yeast and give us the desired function. The challenge that arises
is that oftentimes that just doesn’t work. The components aren’t made correctly,
they don’t function within the yeast host, and then this presents us
with the next challenge, which is: how do we sort of combat against
these thousands of years of evolution that have worked to very specifically tune
and perfect the activity of these components
within one environment, and be able to rapidly move it over
into our other organism of interest? The way that we solve this
is we essentially develop the tools that allow us to recode
the instructions and directions so that yeast can read them. What this essentially means
is if we return to the bioactivities that I talked about, that we identify
from different organisms in nature, we have to go in and take
that DNA sequence information, and basically completely rewrite it. We’re rewriting it so that yeast now
is able to read it and make the active components
within its own environment. They make them in a way
that they can then assemble into these assembly lines within yeast, to be able to efficiently convert
the building blocks to the medicinal compounds. At the end of this process, what we have
is a very engineered organism. The yeast that are able to convert
sugar to medicinal opioids have been engineered to make
23 synthetic proteins, and these have been taken
from six different organisms. What’s notable about this engineering feed
in and out of itself is that we’re now at the stage
where we can go into nature, pull large numbers of proteins
from different organisms, and move them into
an entirely different organism from any of them that we pulled from,
to create the perfect beast. In doing this, we’re now able
to move beyond the molecules that opium poppy gives us naturally, and directly make improved medicines
for our therapeutic use. The importance of this technology then is not just that we’ve developed
an improved process for making one particular class
of a very important medicine, such as painkillers, but that in tackling this challenge for one of the most complex family
of molecules out there, we’re now able to rapidly
apply this technology to build any molecule of interest, and even more importantly, we can apply the technology
to go beyond what we find in nature to create better medicines
with improved therapeutic activities, and reduce negative side effects. What you essentially
have here is a platform that is changing our relationship
with medicines and with nature. The final question is: is this it, are we able
to completely declare victory? The answer is: not yet. But this is a long journey, and there are
many steps within this journey. The incredible news is that we’re now
at the point with the technology that within the next year we’ll be able
to partner with pharmaceutical companies to bring this technology to market
in a rapid time frame, that will impact people’s lives. But it’s also really important
to understand that there’s a lot of very important work
that’s left for us to do to be able to take this technology
and broadly develop new medicines and make them available
to everyone on this planet. Thank you. (Applause)


  1. Globally speaking, I enjoyed the talk. Thank you for delivering it.

    Nevertheless, I cannot avoid thinking that the arguments were not balanced:

    1) Nature could not defend herself from being considered "imperfect" and being blamed for her slow and dynamic pace;

    2) the cons of the Dr. Smolke approach were not exposed, including the risks of making easier and faster to obtain the molecules.
    _ _ _

    Comment on 1): My opinion is that Nature is not imperfect. It would be true to say that humans are rather imperfect because they do not accommodate Nature as it is. Humans don't own Nature, humans deal with Nature as their only medium for being Humans. And just because Nature cannot be genetically manipulated to answer the expectations of humans, this does not mean Nature is imperfect.

    Comment on 2): I think that one of the drawbacks of having a more productive way of producing morphine is furnishing an easier method to expand the illicit drugs market. Since the plant approach will always be possible somewhere in the world, I don't see how having two ways to reach morphine (via plants or yeasts) will combat/ihnibit the illicit drugs market.
    _ _ _

    My final word is this: I think that there is an ethical issue that was not mentioned, concerning whether it is right or not to undergo such a profound interference in nature state as the one proposed through a free market for DNA manipulation. Based on what is known so far, are the consequences of that Pandora box sustainable?

    Science should be aware that for solving the current problems of the world bigger ones should not be generated, and I become concerned everytime I see scientists marketing new approaches as if they had no drawbacks/weaknesses, i.e. as if they did not bring noxious impacts and problems to the world as well.

    Marcelo Melo
    (chemical engineer)

  2. Yes, I agree that many medicines can come or derive from the plants or nature. Some are ready for curing our sickness and some required some modification process. My so-called modification is to add Plant A to Plant B or more to give the best and right results to our disease. Our Chinese medicine composes of many different kind of plants. It is a good proof. My daughter, Leslie is at Stanford this week for meeting. Hope that she can learn things from you. Yes, you can learn things from her too.

Leave a Reply

(*) Required, Your email will not be published