Dr. Barry holding the rare Polynesian
Storm-Petrel afloat on a two week voyage through the
tropical South Pacific in search of rare seabirds. The team
landed on many fantastic atolls and islands throughout
French and British Polynesia.
In our Special Topics analysis of tuberculosis research
over the past decade, the work of Dr. Clifton Barry, III,
ranks at #1 by total cites, based on 53 papers cited a
total of 5,178 times. His work ranks in the top 1% in the
field of Microbiology in
Essential Science IndicatorsSMfrom
Dr. Barry is a Senior Investigator in the Laboratory of
Clinical Infectious Diseases as well as Chief of the
Tuberculosis Research Section at the National Institute of
Allergy and Infectious Diseases in Bethesda, MD.
Below, ScienceWatch.com correspondent Gary
Taubes talks with Dr. Barry about his highly cited work in
What initially prompted your interest in
My first knowledge or interest in TB had absolutely nothing to do with
anything other than if you look at the bacteria with the eyes of a chemist,
which I did, you see some really cool stuff, some really interesting
complex natural products that decorate the surface. For a chemist, this is
a very attractive organism, regardless of the importance of the disease.
How would you describe your approach to TB,
considering your chemistry background?
Well, as always in science, a lot of the things you do first on any problem
are the things you're relatively comfortable doing. My formal training was
in organic chemistry, but I quickly strayed from the straight chemistry
approaches and more into the biology side of things. Nowadays, I'm even
more on the clinical side. I actually have patients in clinical
trials—when I was a graduate student studying the fine details of
some enzyme, I would have thought it inconceivable that this is what I'd
end up doing. It's still inconceivable.
What prompted you do the research that led to your
highly-cited 1998 Science paper, "Inhibition of a
Mycobacterium tuberculosis beta-ketoacyl ACP synthase by
isoniazid" (Mdludi K, et al., 280: 1607-10, 5 June
Isoniazid is one of the most important TB drugs. Of the 7.8 million people
who get diagnosed with TB, virtually all of them will receive isoniazid.
It's a hugely important drug. We had no effort in our lab in that drug. It
was an area I was only peripherally interested in. I had read a lot of
stuff on it, when a paper that predated ours came out in Science
from another lab. It was basically claiming to have figured out the
mechanism for isoniazid resistance.
Dr. Barry at the
opening of an NGO
he helped found in
I heard the PI of that paper give a talk on it at a meeting in Colorado,
and he told a story that made absolutely no sense in the context of
everything else that had happened in the field. I remember getting up to
ask a question, being completely terrified at the time to do so, and asking
that since there were conflicting reports dating back 20 years, how did he
reconcile those reports? His answer, effectively, was that he wasn't even
aware of those reports. That got me thinking about the problem quite a bit
and motivated me to take on the project.
When we did, it turned out to be unbelievably complex, much more so than
anybody had ever thought. In fact, there is still disagreement in the field
about the details and this remains an open question. Here's one of the most
important drugs used globally to control TB, and we still don't really
understand what it is about this drug that makes a tuberculosis bacterium
die. That topic has carried us through the last decade or so with several
of other big lines of investigation, including a Science paper we
just published a couple of months ago, describing another class of drugs
The underlying theme to all of these is that many of these very
simple-appearing compounds are, in fact, a kind of Trojan horse, ways in
which we slip something innocuous in to the bacteria and then it kills
itself. We have different ways of understanding how that happens with
different drugs, but basically it kills itself by acting on something
completely innocuous. That 1998 Science paper was our first brush
with that concept. It isn't about intelligent design of the drug. We don't
pick an enzyme and make a better and better inhibitor that turns into a
drug—what the pharmaceutical industry is really good at. With
bacteria, we really have to get them to accept something and then kill
themselves, and we don't know why they do it.
Why do you think that paper then has been so highly
In part because it was so controversial at the time, parts of it were
certainly over-simplified and still seen through a lens where we wanted
there to be one single target. In part also because the paper described for
the first time the enzymes that biosynthesize one of the most important,
genus-defining molecules in mycobacterial cells—the mycolic acids.
A lot of people still shake their heads and say we still don't understand
how this drug kills bacteria. It poisons many, many processes in a cell
simultaneously from what appears to be a very simple compound. As the
beautiful studies from James J. Collins's group at Boston University have
elegantly shown recently in other bacteria ,even for drugs with a single
target, we don’t understand the molecular mechanisms involved in
death in detail for simple agents, let alone complex ones like isoniazid.
What did the paper actually conclude?
That first paper, the one the PI talked about in Colorado, pointed to one
particular enzyme and said that was the target. Our paper said it's not
that target; instead there was an alternative target in the same pathway
that was affected by the action of that drug. Subsequent papers, some as
recent as last year, have gone on to identify still other targets of the
drug. In a sense, we were as wrong as everyone else; they're all clearly
important and contribute to cell death. But the bug isn't a simple little
machine. There are subtleties and complexities in the way its metabolism
rearranges itself when something happens.
The analogy I like to use is that it's like a modern aircraft with all its
redundancies and failsafes. The bacterium has encountered all these things
before, so sorting out simple cause and effect is extremely difficult. It's
even tricky to identify what event leads to death when they do die. It's
not as if we can identify this enzyme and say if we target this, the
bacteria dies. That just never appears to work.
Another of your highly cited papers is a 2000 paper
in Nature, "A small-molecule nitroimidazopyran drug candidate
for the treatment of tuberculosis" (Stover CK, et al.,
405: 962-6, 22 June 2000). What was that claiming and why was
that so influential?
That paper was really the first description of a set of compounds, or a
series of molecules, that have really important, dramatic effects against
the bacteria. Again, I think that one is most relevant and very highly
cited primarily because it's translated into drugs that are actually being
tested in human beings right now. That molecule described in 2000 is
currently in phase II clinical studies. Another molecule in that series is
also in phase II and probably five or six behind it that are rapidly
approaching phase I trials. At this point, I'm actually optimistic that we
may see a new drug come out of that series. That's big news. There hasn't
really been a new drug like this for TB since rifampicin was introduced 40
Like a lot of good antibacterials, that drug came originally from a natural
product isolated in the 1950s and then it went through a lot of chemical
optimization, the end result of which was that 2000 Nature paper.
A company named PathoGenesis in Seattle elaborated on that original
compound, decorated it with different chemical groups, and came up with
that molecule pa824. That paper has a lot of brand recognition, which is
why it's highly cited. That paper was important because it really broke
this class of drugs open. That particular molecule may or may not be a drug
but something in that class will be. I'm as sure about that as that as you
can be in this business.
One reason we published that paper in Nature, as opposed to some
chemistry journal, is that this was the first time anybody articulated this
idea that TB lesions might be hypoxic—that they may be low in oxygen,
like cancer lesions. And the activity of drugs against hypoxic TB might
actually be important in determining patients' outcomes. That concept
represented a major breakthrough although great microbiologists, starting
with Larry Wayne at the VA Hospital in Long Beach, California, had been
saying this for quite some time.
Did you get lucky with that compound or was this a
well-planned assault on the bacteria?
No, that wasn't at all well planned—it was dumb luck. There are
parallels to cancer here that are amazing. One of the things people do with
cancer when they have a significant tumor is try to measure the oxygen
tension. If that's low, the prognosis is usually very bad for the patient.
You have what's called an anoxic tumor, or hypoxic tumor. So for a long
time companies have tried to make things that would work under low-oxygen
conditions in tumors.
Just by accident a group working at Ciba-Geigy in Hindustan was making this
class of compounds, looking for better tumor radiation hyper-sensitizers
and it just got entered into the corporate archive. They were screening
against everything, running a TB screen, when they discovered this compound
had anti-TB activity. PathoGenesis was led by a great combination of a
microbiologist with sharp insight (C. Ken Stover) and chemist that
wasn’t locked in to the notion that every discovery project had to
have an enzyme target (Bill Baker) and they just picked up on that and took
things quite a bit further chemically. That was complete dumb luck on our
part and it's the story of the whole modern pharmaceutical
industry—compounds made for one reason turned out to have some other
Did you expect these two papers to be so
influential, or did it come as a surprise?
"...people are more and more
terrified, as they should be, of
drug-resistant TB, completely blowing out all
of our decades and decades of work in TB
I'm always shocked when people actually read our stuff. In both of these
cases, I remember thinking, "Well, this is almost a complete waste of time
to submit it to x, to Science or Nature or something like
that. It has no chance of getting in." I still feel that way when we send a
paper off to Science or Nature. I still do it, though. I
have this fantasy that we'll get something in, and every once in a while it
happens. So, no, I'm definitely not that far-sighted to be able to tell if
we've done something really that important. A lot of our real influence on
the field has been because we’re a highly collaborative group and
we’ve been lucky enough to team up with some of the best scientists
that work in TB and drug discovery. Most of my highly cited papers
represent collaborations—I’ve always been a good team player.
It's been almost a decade since then; how has the
field of TB research evolved in that time?
Well, there's been an exponential increase since then in people interested
in the area. To some extent it's driven by the fact that funding has also
gone exponentially higher; it's also driven by the fact that people are
more and more terrified, as they should be, of drug-resistant TB,
completely blowing out all of our decades and decades of work in TB control
programs. Back when that paper came out, there was probably one group in
the world seriously working on TB drugs. Now you couldn't count the number
of groups that have some serious efforts in this.
I remember that back in 1997, I was asked by Gordon Research Conference
people to put together a new Gordon conference on TB. I said I wanted to do
it on TB drug development. I got a very skeptical letter back saying, "Are
you sure? You have to get 90 people to come to the meeting or we won't hold
it again." So I piled my entire laboratory into a van, took everybody, got
all my friends to come, and barely squeaked by with 90 people. This year,
the meeting is already sold out with 160 people attending and nearly 100
more on the waiting list to get in—and the meeting is still a month
away. It gives you some sense how the field has grown and how thinking has
How has your thinking about TB drug development
evolved over the years?
Mine has changed dramatically. I used to think this would be pretty simple
stuff, all about chemistry, and how you can sort of understand what the
essential enzyme was in the bacteria and move from that to a drug. Now I've
devoted much more effort to understanding how these bacteria are complex
systems, and really the only way to intelligently attack this is to either
give up and say it's all luck, just keep playing the lottery until you win,
or try to delude yourself into thinking you actually understand what's
going on and continue to apply the scientific method to it. I choose the
latter; otherwise I'd just go off and watch birds for the rest of my life.
Could you afford the bird-watching option?
No, I can't, but I'm hoping to retire early enough to have a second career
as an ornithologist.
What would you like to convey to the general public
about your work?
That's easy. The biggest message is that TB is not a fixed problem. The
general perception still, in the United States and Western Europe, is that
TB was my grandfather's disease or their generation's and that it's fixed.
Three million people a year die from TB and 10 million get infected. The
way drug resistance comes about in this organism is really scary. I have
seen enough people dying from this disease that I can tell you it is not
your grandfather's disease.
A lot of great people work in this field now, but it still has a tiny
fraction of the funding of HIV/AIDS, and it kills more people than
HIV/AIDS. It's just that they're the wrong people. They're people with
brown skin and black skin and they don't live here. But that doesn't mean
the disease won't be here in our lifetime.
There's definitely a perception in many countries that TB is a disease of
poverty, something to be ashamed of, that only beggars get it in places
like India. That's absolutely not true, this disease strikes down college
students and middle-class workers as efficiently as the impoverished. Way
back when, it used to be a disease of aristocrats and poets and famous
people—people suffering for their art were the ones who got it.
Thoreau died from it, Orwell died of it, Chopin and Schrödinger died
of it, Chekhov, Keats...the list is unbelievable. It was this romantic
thing in the 19th century—a disease associated with
aristocracy and the arts. In our century, it has this reputation as
something dirty, a little secret that poor people die from. But all that
can change very quickly if we don't do something about it.
Clifton E. Barry, III, Ph.D.
Tuberculosis Research Section
Laboratory of Clinical Infectious Diseases
National Institute of Allergy and Infectious Diseases
National Institutes of Health
Bethesda, MD, USA