Isis 5320: New Combinatorial Approach to Drug Development

Isis 5320 is a new anti-HIV agent developed by Isis
Pharmaceuticals, of Carlsbad, California. It is much too
early to know whether this potential drug will be useful in
AIDS treatment; not even complete animal toxicology data is
yet available. What is most interesting now is the unusual
method of drug development by which this substance was
created.

Isis Pharmaceuticals is best known as a leading developer of
antisense technology. Antisense drugs are designed to target
particular genes and selectively block them, without
affecting other genes. Genetic information is contained in a
long sequence of four different chemical "building blocks,"
called nucleotides, which make up the genetic code. Antisense
drugs provide an exactly complementary sequence to a unique
part of the target gene, so that the drug will bind to and
affect only to that gene. (The nucleotides used by the body
would not work well in pharmaceuticals, however; therefore
chemists have created four artificial building blocks, which
correspond to the four natural ones, to use in assembling
antisense drugs. Different antisense technologies use
different sets of four chemicals for building the drugs.) For
more information on antisense drugs for HIV and for CMV, see
AIDS TREATMENT NEWS #185, October 15, 1993, and #187,
November 19, 1993.

Isis 5320 is not an antisense drug, however; it was developed
by a different approach, called combinatorial drug discovery.
Isis is only one of the companies exploring this kind
approach. It has the advantage of being able to use the four
building blocks it had already created for making antisense
drugs -- and its expertise in the related chemistry-- in this
new way. Its scientists already know that the resulting drugs
(produced as sequences of these four) can work in the human
body.

In combinatorial drug development, "libraries" of thousands
of different potential drugs are created at random, in one
batch. If the library as a whole shows the desired activity
as a drug, various techniques are used to determine which one
or more of the chemicals is responsible. That chemical can
then be synthesized in pure form for further testing.

This approach has at least two advantages. First, thousands
of potential drugs can be screened in one test. And also, the
mechanism of action of the drug does not matter -- which
means that an effective drug can be found even if it works in
a way not expected by the researchers, or in a new way which
is unknown.

Isis 5320 was discovered as follows. First, researchers
decided to look at sequences eight units long; there are
exactly 48 (65,536) such sequences. These sequences were
prepared in sixteen different libraries, each containing
4,096 sequences; these libraries were tested against HIV in
the laboratory. (Each library was created by holding two of
the eight positions fixed, and allowing the other six to vary
at random.) By various chemistry and virology, and some
intelligent guess work, one particular sequence -- Isis 5320
-- was selected for development as a potential HIV treatment.

The next scientific step was to find the mechanism of
antiviral action of Isis 5320; when this substance was first
identified, the researchers did not know how it might work.
It was found to target a part of the virus called the V3
loop, preventing HIV from binding to and entering uninfected
cells. Usually the V3 loop would not be selected as a target
for drug development, because it varies greatly among
different HIV strains. But there always seem to be certain
chemical similarities, which may be essential for it to work
properly, and which are apparently targeted in this case.
Isis 5320 has been tested against a number of clinical and
drug-resistant strains of HIV, and the concentration required
for 50 percent viral inhibition is 0.2 to 2.0 micromolar.

What may be most important here is the early practical
demonstration of combinatorial drug development in AIDS.

References

Wyatt JR, Vickers TA, Roberson JL and others. Combinatorially
selected guanosine-quartet structure is a potent inhibitor of
human immunodeficiency virus envelope-mediated cell fusion.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, USA.
February 1994; volume 90, pages 1356-1360.

Also see Background Information for Isis 5320, prepared by
Isis Pharmaceuticals, for a less technical introduction.