#32 from R&D
Innovator Volume 2, Number 4
Catalysis Without Water
Klibanov is professor of chemistry and a member of the
Biotechnology Process Engineering Center at the Massachusetts
Institute of Technology. He
has received many awards, including the 1991 International Enzyme
Engineering Prize. He
is a member of the National Academy of Engineering.
One of the
fastest growing areas of industrial biotechnology is the use of
enzymes in non-aqueous media.
Water has been the conventional reaction medium because it
is an enzyme's natural habitat.
I can still remember people telling me at scientific
meetings, after our first paper on enzymatic catalysis in
anhydrous solvents was published in Science
in1984, "Professor Klibanov, you are doing interesting
work but, frankly, I don't believe enzymes can function in organic
But why bother
using enzymes outside their natural habitat? Because using an enzyme in organic solvents eliminates
several obstacles that limit its usefulness in water.
For example, most compounds that interest organic chemists
and chemical engineers are insoluble in water, and water often promotes unwanted side reactions.
Furthermore, in many aqueous reactions, only a small amount
of product is formed. Also, because of its high boiling point, water is far from
the ideal milieu for product recovery.
once it was established that enzymes can work in organic solvents
with little or no water, R&D in the area surged.
Numerous applications are being commercialized or developed
in the pharmaceutical, food, and specialty chemical industries,
and further development is on the horizon.
Where did our
concept originate? What
is the anatomy of its invention?
Despite a few scattered reports (which were either ignored
or considered unique cases), the idea of using enzymes in pure
organic solvents so flouted the conventional wisdom as to seem
several laboratories, including mine, were doing the next best
thing—using a solvent with water in biphasic mixtures.
In such mixtures, the water and solvent (such as benzene or
ether) remain separate. The
enzymes stay dissolved in the water, since they are insoluble in
the solvent. Substrates
added to the organic solvent diffuse into the aqueous phase,
undergo enzymatic conversion, and the products diffuse back into
the organic phase for recovery.
systems were useful, but they told us little about how effective a
monophasic organic solvent would be by itself.
In 1983, I decided to investigate the relationship of
enzymatic activity to water content in biphasic systems.
I expected a threshold phenomenon, and thought the location
of the threshold might indicate the minimum amount of water needed
by the enzyme.
I suggested that
my graduate student, Alex Zaks, carry out an enzyme reaction in a
biphasic system, starting with a 95:5 mixture of benzene and
water, and then reducing the water content.
At what point would the reaction cease?
A few days later,
Alex came to my office and said that he'd finally removed all the
water, and the enzyme was still catalyzing the reaction! I was surprised and skeptical and asked him about various
controls—he'd done them all and nothing seemed to be skewing the
results. I asked him
to repeat the experiment; he grudgingly agreed—and obtained the
I went to the lab and performed the no-water experiment myself
(without telling Alex). I
found the same result—the
enzyme worked in an anhydrous organic solvent!
We were finally
convinced that the effect was real, and Alex found the phenomenon
working with different solvents, enzymes and substrates.
Not once did he see the threshold of water concentration I
anticipated. I put
several other members of my research group on the project, and
over the years we've learned to activate other enzymes in organic
solvents and have discovered that enzymes exhibit remarkable
properties in these solvents.
For instance, we've seen enhanced thermostability, altered
selectivity, and the ability to catalyze new reactions.
of these properties can be profoundly altered simply by switching
the organic solvent. Hence
the new "enzyme-solvent engineering" represents an
alternative to protein engineering, where enzymes are changed
rather than the reaction medium.
Today, dozens of
laboratories throughout the world are working in this area, and
hundreds of research papers and dozens of patents have appeared.
No longer does anyone doubt that enzymes can work in
Discovery Could Have Been Made Decades Ago
But why did the
discovery wait until the mid-1980s?
After all, the need to carry out enzymatic reactions in
non-aqueous media has long been recognized.
My reading of the literature reveals a problem with general
significance for innovation (or the process of stifling
innovation). Although it is admittedly difficult to reconstruct the
reasoning of scientists from years past, the following thought
process seems to have dominated in the field of enzyme research:
"It would be
nice to use enzymes in pure organic solvents.
But everybody knows enzymes are destroyed in that
environment, so let's not waste time doing something ridiculous.
Instead, we'll start by gradually adding a water-miscible
solvent, say, ethanol or acetone, to an aqueous solution of an
enzyme and see what happens."
As more organic
solvent is added, the enzymatic activity is gradually destroyed;
with 50 to 60 percent solvent, virtually no activity is left.
There is clearly no incentive to proceed to 100 percent
conclusion sounds plausible, but
it's wrong. The
key is this: Although
enzymes work in water (or in water with a very small
percent of solvent), they also work in organic solvents with
little or no water. They
just don't seem to work in intermediate mixtures.
counterintuitive behavior is explained by the fact that enzymes
are in their most stable state in pure water.
In pure organic solvent, they "want" to lose
their 3-dimensional structure, which is essential for activity.
But they can't because they become too rigid (water, which
acts as a molecular lubricant, is absent).
However, in aqueous-organic mixtures, enzymes both want
to and can lose
their structure—and that produces the inactivation.
So much for logical extrapolation!
As a result, I've
started telling students that if it's easy to test a striking but
improbable hypothesis, then test it.
If it works, they will be heroes.
If not, they can keep mum and avoid embarrassment.
How many inventions have been missed simply because
intelligent, knowledgeable researchers decided to forego
To finish the
story, I've noticed there's no stopping innovation once the
critical barriers are down. After
proving that enzymes work in organic solvents, we discovered that
antibodies can do likewise. This
work stimulated us and others to determine other
"strange"—but potentially valuable—milieus for
enzyme activity. These
include supercritical fluids, whose fundamental characteristics
can be markedly altered by pressure, thus affording pressure
control of enzyme performance as well.
Some enzymes need
no liquid phase at all—aqueous or otherwise—and catalyze
gas-phase reactions. For
example, solid alcohol oxidase can react with, and thus detect,
ethanol in the breath or the carcinogen formaldehyde in the air.
Several research groups are now pursuing reactions by whole
microbes in organic solvents.
Thus, complex multi-enzyme pathways (naturally occurring in
the microbial cell) may produce valuable products like
And all of this
stems from an important, albeit humble, experiment that was done for
the wrong reason. We
were merely interested in watching an enzyme lose activity and had
no inkling that the condition we thought would totally kill the
enzyme would actually sustain it. The conventional wisdom is what
we already know.
But our job is to find something new.
Perhaps the only thing we deserve credit for is recognizing
the significance of an unexpected result and exploring, rather
than dismissing, it.