A new type of medicine, custom-made with tiny proteins

Some common life-saving medicines, such as insulin, are made of proteins so large and fragile that they need to be injected instead of ingested as pills. But a new generation of medicine -- made from smaller, more durable proteins known as peptides -- is on its way. In a quick, informative talk, molecular engineer and TED Fellow Christopher Bahl explains how he's using computational design to create powerful peptides that could one day neutralize the flu, protect against botulism poisoning and even stop cancer cells from growing.

I'm a protein designer.

And I'd like to discuss
a new type of medicine.

It's made from a molecule
called a constrained peptide.

There are only a few constrained
peptide drugs available today,

but there are a lot that will hit
the market in the coming decade.

Let's explore what these new
medicines are made of,

how they're different and what's causing
this incoming tidal wave

of new and exciting medicines.

Constrained peptides
are very small proteins.

They've got extra chemical bonds
that constrain the shape of the molecule,

and this makes them incredibly stable

as well as highly potent.

They're naturally occurring,
our bodies actually produce a few of these

that help us to combat
bacterial, fungal and viral infections.

And animals like snakes and scorpions

use constrained peptides in their venom.

Drugs that are made of protein
are called biologic drugs.

So this includes constrained peptides,

as well as medicines like insulin

or antibody drugs like Humira or Enbrel.

And in general, biologics are great,

because they avoid several ways
that drugs can cause side effects.

First, protein.

It's a totally natural,
nontoxic material in our bodies.

Our cells produce tens of thousands
of different proteins,

and basically, all of our food
has protein in it.

And second, sometimes drugs interact
with molecules in your body

that you don't want them to.

Compared to small molecule drugs,

and by this I mean
regular drugs, like aspirin,

biologics are quite large.

Molecules interact when they adopt shapes
that fit together perfectly.

Much like a lock and key.

Well, a larger key has more grooves,

so it's more likely to fit
into a single lock.

But most biologics also have a flaw.

They're fragile.

So they're usually
administered by injection,

because our stomach acid
would destroy the medicine

if we tried to swallow it.

Constrained peptides are the opposite.

They're really durable,
like regular drugs.

So it's possible to administer them
using pills, inhalers, ointments.

This is what makes constrained peptides
so desirable for drug development.

They combine some of the best features

of small-molecule
and biologic drugs into one.

But unfortunately,
it's incredibly difficult

to reengineer the constrained peptides
that we find in nature

to become new drugs.

So this is where I come in.

Creating a new drug
is a lot like crafting a key

to fit a particular lock.

We need to get the shape just right.

But if we change the shape
of a constrained peptide by too much,

those extra chemical bonds
are unable to form

and the whole molecule falls apart.

So we needed to figure out
how to gain control over their shape.

I was part of a collaborative
scientific effort

that spanned a dozen institutions
across three continents

that came together
and solved this problem.

We took a radically different approach
from previous efforts.

Instead of making changes
to the constrained peptides

that we find in nature,

we figured out how to build new ones
totally from scratch.

To help us do this,

we developed freely available
open-source peptide-design software

that anyone can use to do this, too.

To test our method out,

we generated a series
of constrained peptides

that have a wide variety
of different shapes.

Many of these had never been seen
in nature before.

Then we went into the laboratory
and produced these peptides.

Next, we determined
their molecular structures,

using experiments.

When we compared our designed models

with the real molecular structures,

we found that our software
can position individual atoms

with an accuracy that's at the limit
of what's possible to measure.

Three years ago, this couldn't be done.

But today, we have the ability
to create designer peptides

with shapes that are custom-tailored
for drug development.

So where is this technology taking us?

Well, recently,

my colleagues and I
designed constrained peptides

that neutralize influenza virus,

protect against botulism poisoning

and block cancer cells from growing.

Some of these new drugs

have been tested in preclinical trials
with laboratory animals.

And so far, they're all safe
and highly effective.

Constrained peptide design
is a cutting-edge technology,

and the drug development pipeline
is slow and cautious.

So we're still three to five years
out from human trials.

But during that time,

more constrained peptide drugs

are going to be entering
the drug development pipeline.

And ultimately, I believe
that designed peptide drugs

are going to enable us all to break free

from the constraints of our diseases.

Thank you.

(Applause)