Biomedical Engineering Reference
In-Depth Information
growth of selected strains of bacteria on seeded agar plates. A positive response in such an assay is
observed as a zone of inhibition of growth around a test sample deposited on the Petri plate. In most
cases the diameter of the zone of inhibition is directly related to the concentration of the antibacte-
rial agent in the test sample, so that some degree of quantii cation of the response is also possible in
this simple test. Such assays provide no indication of the mode of action of the active principle, but
are generally quite effective tools for following fractionation processes that lead to pure compounds.
A crucial extension of this in vitro whole organism assay is another phenotypic test in which the
isolated material is administered to rodents that have been challenged with a lethal infection of the
target bacterium. The positive end point of the in vivo rodent test is survival beyond that of the
control animals. Positive results in the in vivo model of infection provides the critical information
that the compound has sufi cient drug-like properties to penetrate the normal xenobiotic defenses of
the host animal and reach the target population of infecting bacteria.
There are many such whole organism models that have been used for drug screening. Among
the simplest of these are those related to infectious diseases, including the aforementioned antibac-
terial system with various classes of pathogenic agents, as well as those designed for antifungal,
antiviral, and antiparasitic agents. In the quest to i nd new effective agents against cancer, animal
models of disease remain a mainstay of the process. Similar models are the norm for advancing
the development of drugs in terms of understanding the efi cacy, tolerability, metabolism, and
long-term effects. These systems are rarely used for high-throughput screening of crude natural
products because they require substantial resources for its maintenance. Therefore, most live ani-
mal models are used to verify the efi cacy of compounds that have been isolated with the aid of a
simpler in vitro assay.
As will be discussed in the following section, screening against isolated target biomolecules,
such as enzymes or receptors is now favored for high-throughput screening operations. In the case
of screening mixtures of natural products, however, the whole organism approach offers tremendous
advantages. The discovery of a novel secondary metabolite that confers a positive response in a whole
organism screen provides the opportunity to discover a new target, and potentially a new mechanism
of action. In current parlance these studies are often referred to as “Chemical Biology” or “Chemical
Genetics” (see also Chapter 4). Specii cally, in forward chemical genetics a small molecule, in our
case a natural product, is employed to probe for their cellular targets. Typical experiments include the
creation of afi nity binding reagents or afi nity matrices that include the small molecule of interest,
and these systems are used to i sh out target macromolecules from cellular components. Molecular
targets for rapamycin and geldanamycin were found by such methods. Once the target macromol-
ecules are verii ed, additional mechanistic studies are developed to understand the relationship
between the binding partners and the disease process.
6.4.3 T
ARGET
-B
ASED
S
CREENING
Molecular biology has provided the tools to engineer and produce macromolecular targets of drug
action. If it is believed that the inhibition of a particular cell-signaling process will mediate the
development of disease, then the isolated enzyme, or receptor that is responsible for the signaling
can be used as a target for screening. Alternatively, a selective whole cell screen can be employed
that is designed to respond by providing some measurable signal as a result of the interaction with
a particular target. Owing to developments in automation for such assay systems, hundreds of thou-
sands of compounds can be conveniently tested for activity in a short period of time. Natural prod-
uct mixtures may also be tested in these highly automated systems. With mixtures, particularly
crude extracts, there is the potential for signii cant interference with the assay. Such interferences
include nonspecii c inhibition of targets by ubiquitous classes of natural products like fatty acids, or
the presence of a highly potent cytotoxic agent that kills the host cell designed for the specii c assay.
These issues are not insurmountable, but require diligence in evaluating screening results. Some
solutions for these issues are presented in Section 6.4.5.
