Note: Descriptions are shown in the official language in which they were submitted.
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
METHODS FOR IDENTIFYING GENETIC DETERMINANTS ASSOCIATED
WITH MODULATION OF TEST COMPOUND ACTIVITY
Technical Field Of The Invention
This invention relates to methods for identifying particular modes of action
for
test compounds which modulate particular detectable cellular processes.
Background Of The Invention
Methods for identifying compounds which modulate specific cellular pro-
cesses, and characterize the gene products which interact with such compounds,
are
critical for discovering new chemical entities which may be used to develop
future
therapeutic compounds. The development of new therapeutic compounds can
proceed by a variety of methods, but generally falls into one of two
approaches.
In the traditional approach, screening methods are used to identify compounds
2 0 that affect a particular tissue or model, without concern for the specific
target. The
second approach involves the identification of new therapeutic targets, for
example, a
particular cell type or a receptor on a cell surface or present in the
cytoplasm, and
screening compounds to identify those which interact with the identified
targets.
Large collections of compounds, whether synthetically derived or isolated
2 5 from natural sources, have traditionally served as raw material for
screening assays.
With more recent technologies such as combinatorial chemistry and phage
display, it
is relatively straightforward to generate large compound libraries for
testing, typically
containing from about 10,000 to 100,000 or more related or random compounds
for
use in high throughput screening protocols. High throughput screening
techniques
3 0 have been enabled by automation of traditional screening methods, although
screening large numbers of compounds against one or more specific targets can
be a
labor and capital intensive endeavor even when implemented in a high
throughput
fashion. Moreover, these methods rely on specific targets.
-1-
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
Summary of the Invention
One object of the invention is to provide a method for determining the genetic
determinants that can reverse the pharmacological effect of a test compound.
It is an obj ect of this invention to provide methods which enable the testing
of
compounds having activities against a cell type displaying or having a
detectable
process of interest, and identifying a subset of such test compounds which
affect or
modulate the detectable process exhibited by the cell. The activities can be
previously
unidentified. After identification of a subset of test compounds active in
modulating
the detectable process being studied, those compounds, or a subset thereof,
are again
1 o screened against cells exhibiting the desired detectable process, although
in this
second screen the cells have been manipulated to overexpress one or more
heterologous polypeptides. Those cells which overexpress one or more
heterologous
polypeptides which reverse the effect of the test compound on modulating the
detectable process in the cell of interest are then characterized, for example
by
sequencing the nucleic acid that encodes the overexpressed gene product. In
this way,
the mode of action for a test compound can be identified and cellular
processes and
constituents responsible therefor can be correlated with a detectable cellular
process
without requiring a specific target. In addition, the invention dramatically
reduces the
effort involved in screening by providing simultaneous screening for compounds
2 0 active on many different targets in a single assay. Thus, the instant
methods are
particularly suited to efficient screening of multiple cellular targets
against large
numbers of test compounds.
One aspect of the present invention concerns methods of identifying modes of
action for test compounds which modulate a detectable process in a cell. Such
2 5 methods comprise exposing a first pool of cells, e.g., prokaryotic or
eukaryotic cells,
to one or more test compounds under conditions which, in the absence of the
test
compound, allow a detectable process to occur the cells. It is then determined
if
exposure to the test compound modulates the detectable process in the cells or
subset
of the cells, and if so identifying those compounds as active compounds.
Active
3 0 compounds are then exposed to a second pool of cells which overexpress one
or more
heterologous polypeptides under conditions which, in the absence of the active
compound, allow the detectable process to occur. It is then determined if
overexpression of the heterologous polypeptide(s) alters modulation of the
detectable
-2-
CA 02327830 2000-10-06
WO 99/53098 PCTNS99/07802
process mediated by the active compound in the cells. By performing such
methods,
one is able to identify the mode of action for a test compound which modulates
a
detectable eukaryotic cell process.
In certain embodiments, the cells used in the claimed methods are eukaryotic
cells, such as mammalian cells, including canine, feline, ovine, porcine,
equine,
bovine cells, and human cells. One may use human cells that are disease-
specific and
correlate with a particular human disease. In some embodiments of the
invention, the
eukaryotic cells employed to identify which test compounds are active
compounds
(first population) are the same types of cells which overexpress one or more
heterologous polypeptides (second population). In other embodiments,
prokaryotic
cells are employed, such as, for example, bacterial cells, particularly those
which are
pathogenic to humans or livestock.
Other embodiments of the invention concern high throughput screening
methods employing more than one aliquot of the first pool of cells, wherein
each
aliquot is exposed to a different test compound. Other embodiments of high
throughput screening also employ aliquots of the second pool of cells wherein
each
aliquot is exposed to a different compound identified as an active compound in
the
initial pre-screening against the first pool or aliquot of a first pool of
cells. In one
embodiment of the invention, different samples of the second pool of cells
2 0 overexpress one or more different heterologous polypeptides than are
overexpressed
in the other samples. In certain embodiments, high throughput screening
methods
according to the invention are conducted in one or more 96-well microtiter
plates,
although many other formats are also suitable for high throughput screening,
and
largely depend on the automated equipment being employed.
2 5 In certain embodiments, the disclosed methods are used to identify
compounds
which modulate a detectable cellular process in a negative way, for example by
inhibiting (partially or completely) the detectable process in the cells being
screened.
Preferred detectable processes are those which involve measurable
physiological
cellular processes, for example, cell growth, cell growth rate, cell
migration, nucleic
3 0 acid replication, nucleic acid synthesis, protein synthesis, protein
secretion, cell
adhesion, phagocytosis, contact inhibition, and cell death, for example
programmed
cell death or apoptosis. Other measurable physiological processes include
those
involving infra- or extra-cellular localization of a cellular component, or
expression of
-3-
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
a reporter gene. In other embodiments of the invention, modulation involves an
increase in the detectable process exhibited by the cells.
In certain embodiments of the invention, overexpression of one or more
heterologous polypeptides is mediated by a recombinant nucleic acid molecule
introduced into the cells. Preferably, the introduced recombinant nucleic acid
encodes
one or more genes (or functional portions thereof) under the control of a
promoter or
other cis-acting element required for transcription in the cells. However,
promoters or
other transcription activation sequences are not essential, particularly when
the
recombinant nucleic acid being introduced is to be inserted by homologous
recombination into a region of the cell s genome functionally adjacent to a
tran-
scriptional activation sequence sufficient to enable overexpression of one or
more
genes adjacent thereto.
In other embodiments, overexpression may be achieved by upregulating
expression of one or more endogenous genes. Upregulation in such embodiments
may be achieved by addition or removal of one or more chemicals or other
compounds to or from the growth medium. Alternatively, genetic modifications
or
mutations may result in such upregulation, for instance as may occur upon
infection
of the cells by certain viruses.
Recombinant nucleic acids according to the invention can be incorporated in
2 0 gene delivery vehicles, for example, viral vectors, liposomes, or nucleic
acid/condensing agent compositions, although delivery of "naked" nucleic acids
can
also be employed. Exemplary viral vectors include retroviral vectors, adeno-
associated viral vectors, and alphaviral vectors, particularly viral vectors
which are
replication defective. Replication defective recombinant retroviral vectors
which
2 5 comprise one or more heterologous genes whose expression is under the
control of an
efficient promoter or other transcription activation sequence are useful in
the
invention. In certain embodiments, such activation sequences are inducible pro-
moters. Retroviral vectors are particularly useful because they enable stable,
long
term expression. See U.S. patents 5,716,832, 5,591,624, 5,693,522, 5,716,613,
3 0 5,716,826, and 5,662,896, each of which is hereby incorporated by
reference, for a
description of producing replication defective retroviral vectors and
packaging cell
lines therefor. As with retroviruses, the host range of other gene delivery
vehicle
being used can be manipulated to target specific cell types.
-4-
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
Certain embodiments of this aspect of the invention concern overexpression of
a library of polypeptides, such as may be encoded by a library of heterologous
genes,
in the second pool of cells. Such a library can be a cDNA library prepared
from
messenger RNA isolated from cells of the same species as are represented in
the
second pool of cells. In a complete embodiment, the cDNA library is fixlly
representative of all genes expressed in the organism in which the cell was
derived.
Alternatively, the library may be less than fully representative of the
organism, as
may be obtained by generating a cDNA library from a specific cell type of an
organism, e.g., a hepatocyte or nerve cell. Moreover, libraries according to
the
invention may be further refined by techniques wherein "housekeeping" genes
common to most or ali cell types of the organism are eliminated by subtractive
cross-
hybridization. In yet another embodiment, the cDNA library is at least
partially
comprised of a custom library containing preselected cDNAs pooled for the
purpose
of conducting the particular assay.
Certain embodiments of the invention concern characterization of a heterolo-
gous polypeptide, or a gene encoding the heterologous polypeptide, which is
found to
modulate the activity of an active compound through its overexpression in the
second
pool of cells. With respect to characterization of polypeptides, various
techniques can
employed, for example antibody- or other high-affinity receptor-based
detection.
2 0 With respect to nucleic acid characterization, hybridization or sequencing
techniques
are typically utilized. Sequencing typically comprises determining at least a
portion
of a nucleotide sequence of the heterologous gene or genes which are
responsible for
overexpression of the heterologous polypeptides which modulate activity of the
active
compound. In some circumstances it is desirable to amplify heterologous genes
prior
2 5 to sequencing. To facilitate amplification, the vector in which the
heterologous genes
are inserted preferably contains a unique amplifiable sequence 5' and 3' to
the region
where the heterologous gene is inserted into the vector. Amplification primers
which
have nucleotide sequences substantially complementary to the unique
amplifiable
sequences can then be used to amplify the heterologous gene insert by an
appropriate
3 o amplification methodology, e.g., PCR (polymerase chain reaction),
transcription-
mediated amplification, ligase chain reaction, or strand displacement
amplification.
In addition, or alternatively, the vector may also contain a "tag" region for
amplification. When tags are used, it is preferred that there is a one-to-one
-5-
CA 02327830 2000-10-06
WO 99/53098 PCTNS99/07802
correspondence between a particular tag and the heterologous gene of unknown
sequence included in the same vector. In this way, a specific tag corresponds
to the
specific heterologous gene carnes in the recombinant nucleic acid.
Another aspect of the invention concerns methods of correlating a gene with a
detectable cellular process, such methods being based on exposing a first pool
of cells
to one or more test compounds under conditions which, in the absence of the
test
compound, allow the detectable process to occur, provided that in the event
more than
one test compound is to be screened, an aliquot of the first pool of cells is
used for
each such test compound. Following exposure of a test compound to the first
pool of
1 o cells or an aliquot thereof, it is determined if one or more of the test
compounds
modulates a detectable process in such cells, and if so identifying those test
compounds) as an active compound. The active compound or compounds are then
exposed to a second pool of cells which overexpress one or more heterologous
polypeptides under conditions which, in the absence of an active compound,
allow the
detectable process to occur. If overexpression of a heterologous polypeptide
in the
cell alters modulation of the detectable process, the heterologous gene or
nucleic acid
molecule encoding the heterologous is then sequenced and correlated with the
detectable process.
2 0 Detailed Description
Definitions
When used in this application, the following terms will have meanings
described below, unless otherwise specifically indicated.
"Mode of action" for a test compound refers to the cellular process or
2 5 processes affected by the test compound when administered to a cell. A
mode of
action is based on the activity of one or more genetic determinants.
"Test compound" refers to any molecule, synthetic or naturally occurnng,
used in the practice of the present invention. Such compounds include, without
limitation, nucleic acids, e.g., oligonucleotides, ribozymes, and antisense
molecules
3 0 (including without limitation RNA, DNA, RNA/DNA hybrids, peptide nucleic
acids,
and polynucleotide analogs having altered backbone and/or bass structures or
other
chemical modifications); proteins, polypeptides, carbohydrates, lipids, and
small
molecule drug candidates. "Small molecules" are, for example, naturally
occurring
-6-
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
compounds (for example, derived from plant extracts, microbial broths, and the
like)
or synthetic organic or organometallic compounds having molecular weights of
less
than about 10,000 daltons, preferably less than about 5,000 daltons, and most
preferably less than about 1,500 daltons.
The term "test conditions" refers to a variation in the environment other than
the presence of a test compound, which perturbs the metabolism or activity of
a test
cell. Examples of test conditions include elevated or depressed temperature,
altered
concentration of nutrients (other than proteins), adhesion or contact
surfaces, gas
concentrations, rate of temperature change, presence or absence of other
cells, and the
like.
"Modulate" means increasing or decreasing a particular activity or detectable
process of a cell. Modulation of an activity ranges from partial to complete.
"Detectable process" refers to a cellular process which is observable or
measurable. Representative examples of such processes include measurable
physiological processes such as cell growth, changes in cell growth rate, cell
migration, nucleic acid synthesis, protein synthesis, protein secretion, cell
adhesion,
phagocytosis, contact inhibition, apoptosis and cell death. The mechanism or
mechanisms used to observe or measure the process depends on the process being
detected. For example, detection may depend on the intracellular or
extracellular
2 0 localization of a specific cellular component, such as a protein, lipid,
carbohydrate, or
nucleic acid.
The term "cell" refers to any cell, and includes both prokaryotic and
eukaryotic cells. "Prokaryotic cell" refers to any cell lacking a membrane-
bound
nucleus, for example, bacterial cells. "Eukaryotic cell" refers to any cell
which has a
2 5 membrane-bound nucleus. Eukaryotic cells used in the practice of this
invention may
be derived from single-celled organisms such as fungi or from mufti-cellular
organisms such as plants and animals, for example, higher animals such as
birds and
mammals, e.g., human, bovine, canine, feline, equine, ovine, and porcine
animals, and
can be normal cells, cell lines, or cells associated with a particular disease
state, for
3 0 example, cancer. The cells used may represent a recognized disease model.
"Pool" of cells refers to a population of cells, and includes cell populations
comprised of different cells types, e.g., prokaryotic and eukaryotic cells, or
different
types of eukaryotic cells, e.g., human hepatocytes and lymphocytes. A pool of
cells
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
can be derived from the same cell source, e.g., cells of a particular cell
line such as
HeLa cells. An "aliquot" of such a pool refers to a subset of the pool, e.g.,
a 100 p.L
sample of a 10 mL culture of cells.
The term "specifically altered" as used herein refers to the alteration of the
activity of a heterologous gene, by changing the level of expression of a gene
(up or
down) found in the first cell type, or the specific activity of its protein
product.
Specific alteration includes overexpression of the gene product, upregulation
and
down-regulation of the gene, inhibition of the gene activity and/or
transcription, and
mutation of the gene that alters the biological activity of its product.
1 o The terms "overexpress" and "overexpression" refers to expression of a
gene
in a cell at a level higher than normally expressed in a cell of that type
under the
particular growth conditions employed. Thus, if a gene is not expressed in the
particular cell type under the growth conditions employed, any expression of
that
gene would constitute overexpression. Overexpression of a gene can be mediated
by
introduction of a heteroiogous gene into the cell, such as by transfection.
Alternatively, overexpression can also be achieved by manipulating the cell to
upregulate expression of the naturally occurring gene in that cell, for
instance by
virus-mediated mutagenesis or exposure to a chemical or other compound which
directly or indirectly leads to upregulation of expression of the gene.
2 0 "Specific inhibition" refers to the inhibition of the activity of a
specific gene.
Instead of overexpressing a gene native to the first cell, one can inhibit the
function of
that gene (where native to the second cell population) specifically, for
example by
using antisense polynucleotides or ribozymes in the second cell population.
This
embodiment is particularly useful for studying the effects of compounds which
act as
2 5 agonists, or by upregulating the activity of a gene.
"Gene" refers to a polynucleotide that encodes a polypeptide, i.e., two or
more
amino acids linked by a peptide bond. Thus, as used herein, "gene" can refer
to the
entire coding region for a protein, in genomic or cDNA form, or an open
reading
frame (ORF), or fragments thereof.
3 0 The term "reporter gene" refers to a polynucleotide that provides a
detectable
signal following transcription. The signal can be direct or indirect, and can
be
transcriptional (for example, by providing a unique or characteristic sequence
detected by hybridization) or translational (for example, by providing a
distinct
_g_
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
surface antigen, a chromophore or fluorescent protein, a chromogenic enzyme,
and
the like).
The term "heterologous polynucleotide" refers to a polynucleotide that is
foreign to the host cell, or that is native to the host cell but operatively
associated with
a promoter other than its native promoter. "Heterologous polynucleotides"
further
include complementary and antisense sequences capable of inhibiting
expression, and
mutated sequences affecting the biological activity of the product. For the
purposes
of this patent, "native" polynucleotides include polynucleotide sequences
capable of
specifically inhibiting transcription andlor expression of a gene or cDNA
found in the
first cell type. A "heterologous polypeptide" is a protein or polypeptide
product of a
heterologous gene.
An "inducible expression system" refers to a nucleic acid the expression of
which is regulated. Under certain conditions, the genes) encoded by such a
system
may be induced, for instance, by the addition of a chemical or some other
change in
environmental conditions.
A "library" of heterologous genes refers to a collection of two or more such
genes, preferably at least about 10 genes, more preferably at least about 20,
most
preferably about 100 or more. As applied to cDNA, such a library can be
prepared
from total RNA, or preferably mRNA, from one or more cell types. A "fully
2 o representative library" is one containing all genes expressed in a
particular cell type,
developmental stage, or organism, for example. Preferred cDNA libraries
include
those which are "subtractive," in which certain mRNAs are removed by a cross
hybridization reaction designed to remove genes that do not contribute to the
attributes of a specific cell type which distinguish it from other cell types.
As a result,
2 5 such a library will generally lack members common to all or other cell
types, for
example "housekeeping" enzyme genes.
As used herein, "high throughput" refers to screening techniques wherein
more than about 10, preferably more than about 100, and more preferably more
than
about 1,000, compounds are screened in an automated fashion in a single
experiment
3 0 according to the methods described herein.
General Methods
-9-
CA 02327830 2000-10-06
WO 99/53098 PCTNS99/07802
In the practice of the claimed method, the order of genomics driven drug
discovery is reversed. Genomics approaches usually provide large sets of genes
(sometimes all genes , when the organism is entirely sequenced) suspected to
be
involved in a particular cellular process. It also provides tools to
investigate these
genes and prioritize them with respect to the process of interest. This allows
one to
focus on a few novel and useful genes around which drug screens can be
designed.
In the practice of the invention, one first identifies a set of compounds that
affect a process of interest. Genomic techniques are then used to identify
sets of genes
that modulate the effect of these compounds. This concept can be applied to
many
different situations. The requirements are (1) a relevant cell type, (2) a
scorable
phenotype relevant to the ultimate therapeutic goal, preferably amenable to
high-
throughput screening, and (3) a method to introduce gene-specific alterations
in either
the level of expression of a gene or in the specific activity of the encoded
gene
product. Thus the method comprises two parts: first, one identifies a set of
compounds
affecting the process of interest, preferably in the cell of interest,
preferably using the
most relevant readout for the desired outcome, and then one identifies a set
of genes
(for example all genes) that modulate the effect of these compounds. One can
use the
same cell type, and the same phenotype as a readout.
Relevant cell types can be prokaryotic or eukaryotic, such as a mammalian
2 o cell line, and can be primary cultures, permanent cell cultures, and the
like. In cases
in which a particular cell type is known to be involved in a disease process
under
investigation, one can use primary cells of that type, or a model cell can be
substituted. The second pool of cells employed in screening can be any type of
cell,
but is preferably eukaryotic.
2 5 The scorable phenotype is preferably relevant to the cellular process
under
investigation, and preferably simple to observe. For example, if the therapy
under
investigation is cancer metastasis, the phenotype can be anchorage-independent
growth. Alternatively, the phenotype can be read by employing a hybridization
array,
and directly determining the concentration of mRNAs produced in response to
the test
3 0 compounds. Alternatively, the scorable phenotype can be expression of a
reporter
gene or cell surface antigen, which permits one to label cells with antibodies
or other
binding ligands, and sort the cells using FACS, panning and the like.
-10-
CA 02327830 2000-10-06
WO 99/53098 PCT/I1S99/07802
The second pool of cells comprises at least one heterologous polynucleotide
found in the first cell (and thus a possible target of the test compound),
where
expression of the heterologous polynucleotide is specifically altered in
comparison to
the first cell. The heterologous polynucleotide can be provided on a plasmid
or other
non-genomic vector, or can be integrated into the second cell population's
genome
(for example, by using a retroviral vector, homologous recombination, and the
like).
The specific alteration selected will depend on the cellular process under
study, and
the desired or suspected mode of action of the test compounds. Overexpression
of the
heterologous polynucleotide is conveniently provided by inserting the coding
sequence operatively associated with a strong promoter, which can be either
regulated
or constitutive, and can be native or heterologous to the host cell. Suitable
promoters
include those for, without limitation, Tet, GAL, ecdysone, baculovirus, ADH,
GAP,
CMV, SV40, metallothionein, hybrid promoters, and the like. Alternatively, the
heterologous polynucleotide can be integrated into the host cell genome in
sufficient
proximity to a native promoter.
Alternatively, a native gene can be specifically inhibited, for example, by
inserting a heterologous polynucleotide that provides an antisense
polynucleotide or a
ribozyme specific for the gene, using methods known in the art. Again, the
coding
sequence can be provided either on a plasmid or other extra-genomic element,
or can
2 0 be integrated into the host cell genorne.
Finally, the heterologous polynucleotide can be mutated, for example by
random mutagenesis, point mutation at an active site, truncation, and the
like, in such
a way that a biological activity of the protein product is altered. For
example, one can
generate a dominant mutation resulting in a protein that retains the original
activity of
2 5 the native protein but no longer binds the test compound. Expression of
the mutant
reverses the compound-induced phenotype (See, e.g., G. Barnes et al., Mol Cell
Biol
(1984) 4(11):2381-88). If desired, a plurality of different mutated sequences
can be
used, and in fact can be used to simultaneously determine the active portions
of the
sequence along with identifying the corresponding gene as a target of the test
3 0 compound.
The second pool of cells is screened for response to the test compounds
identified as active with the first cell type. Cells in the second pool that
are capable of
counteracting or reversing the activity of the test compound to a detectable
degree)
-11-
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
are identified, and the heterologous polynucleotide responsible for the
activity is
identified by any suitable means. The heterologous polynucleotide can be
identified
by sequencing.
The information obtained is useful for a number of purposes. For example,
the genes identified as capable of reversing the activity of a compound are
candidate
targets of the compound. If the compound is a microbicide, these genes may
form
part of an organism's resistance mechanism. Further, such information could be
used
to design combination therapies that affect different genes (and different
cellular
processes), making it more difficult for resistance to arise.
Example 1
(Antifungal drug development)
The yeast Saccharomyces cerevisiae was used as a target for antifungal
research. The scorable phenotype employed was growth inhibition in the
presence of
test compounds. The method of introducing gene-specific alteration is to
transform
high copy libraries selecting for compound resistance.
(A) Identification of antifungal compounds.
A set of compounds exhibiting antifungal activity was identified by screening
a diverse collection of compounds against four fungal species: Saccharomyces
cerevisiae, Candida albicans, Aspergillus fumigatus, and Cryptococccus
neoformans.
Approximately 100,000 compounds in DMSO were screened in a 96-well microtiter
format. The screening conditions are summarized in Table 1:
Table l: Screenine Conditions
S. cerevisiaeC. albicansC. neo ormansA. ~ umi
atus
Inoculum OD 0.08 0.002 0.2 1:600
[compound] 20 30 20 30
ml
DMSO 2% 3% 2% 3%
Incubation 30C 30C 30C 37C
tem erature
C
Incubation 24 20 46 46
time
hrs
The result of the screen in summarized in Table 2:
Table 2: Screening results
-12-
SI~BST1TUTE SHEET (RULE 2~
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
Species Compounds Hits (#) Compounds active
screened _ a ainst >1 s ecies
# #
A. umi atus 93660 1919 1045
G albicans 94156 1920 1351
C. neo ormans92990 3158 1451
S. cerevisiae95263 1685 1205
Of the compounds screened, 359 were active against all four species.
S. cerevisiae was the only cell type used in the subsequent step of the method
because it is most amenable to genetic manipulations. The other species were
included
at this stage because they are important pathogens and maximize the relevance
of the
set of compounds identified by this screen.
(B) Identification of sets of genes modulating the effect of antifungal
compounds.
In this experiment, a set of genes capable of reversing the effect of some of
1 S the previously identified antifungal compounds (i.e., provide resistance
to the drug)
was identified. It is known that overexpression-based resistance to an
antimicrobial
compound can occur through a variety of mechanisms.
MIC determination in solid medium: For each compound to be tested, the
minimum inhibitory concentration in solid medium was determined. The test was
performed in 24 well plates containing 1 ml of YPD agar. The compound was
serially
diluted from 128 ~tg/ml to 0.5 p,g/ml. Each well was inoculated with 50,000
cfu. The
plates were incubated for 24 hours at 26°C. The lowest concentration of
compound
capable of inhibiting the growth of the inoculum is the MIC.
Generation of pools of overexpressor cells: Here, two yeast genomic DNA 2-
micron plasmid libraries capable of overexpressing random genes were employed.
One of these uses the yEP24vector (Carlson and Botstein, Cell (1982) 28:145),
the
other library is in the pRS203 vector (gift of Philip Hieter). Transformants
(200,000
and 275,000 for the two libraries respectively) were generated, which
correspond to
more than a 50-fold coverage of the genome. The transformants were generated
using
the Lithium Acetate transformation procedure (Rose et al., Methods in Yeast
Genetics,
(Cold Spring Harbor Laboratory Press, 1990)). After 3 days of growth at
26°C the
transformant colonies were recovered from the plate and frozen in aliquots at -
80°C.
-13-
SUBSTITUTE SHEET (RULE 26)
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
Selection for resistance: Approximately 1 x 10' colony forming units from
one of the transformant pools were plated on 60 mm Petri dishes containing 8
ml of
YPD agar, including 2x the previously determined MIC concentration of the
compound to be tested. Resistant colonies that emerged after 3 days gowth on
solid
YPD agar containing the compound were harvested and pooled. Plasmid DNA was
recovered from the yeast cells as described by Rose et al., supra.
Amplification of
this material was achieved by transformation and subsequent isolation of
plasmid
DNA in E. coli.
Identification of genes responsible for the resistance using DNA microarrays:
Approximately 200 ng of plasmid DNA was labeled with the fluorescent
nucleotide
analog Cy3-dUTP (Amersham Pharmacia PA53022) using a commercially-available
nick translation kit (Amersham Pharmacia N5500). The resulting labeled DNA was
purified and used to probe a DNA microarray as described by Eisen and Brown
(Meth
Enzymol, in press). Arrays were prepared as described by Shalon et al., Genome
Res.(1996) 6:639, incorporated herein by reference in full, using PCR products
generated using gene PAIRS primers (Research Genetics, Huntsville AL). This
micro-
array contains over 930 yeast genes including all genes known to be essential
for
vegetative growth, all ABC transporters, as well as other genes known to be
involved
in chemical resistance or sensitivity. A positive hybridization signal in this
2 0 experiment is indicated that the corresponding gene (or a gene linked to
it) is
responsible for the compound resistance. When a positive signal was
identified, the
corresponding ORF was cloned, as well as its close genetic neighbors on an
expression construct under the control of the yeast ADH1 promoter. Each of
these
constructs was transformed back into yeast and the resulting transformants
were
2 5 tested for increased resistance to the compound.
Results: Using this procedure, one or more resistance genes were identified
for
at least half the compounds tested (see Table 3). The most commonly identified
genes
are likely to be involved in the transport of the compound in and out of the
cell. This
effect can be direct (for example, SNQ2, a multidrug transporter of the ABC
family of
3 0 transporters) or indirect (for example, YAP1 and YAP2, two transcription
factors
known to activate multidrug resistance genes). This is very useful information
in the
context of antifungal development, because it indicates potential resistance
mechanisms against the drug. In some cases, the isolated gene can be the
molecular
-14-
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/07802
target for the drug. For example, in Table 3 below, the ERG24 gene may be the
target
of compound CmpdG. In yet other cases, some of the genes isolated by this
method
can be indicative of the mode of action of the compound, without necessarily
being
the molecular target of the compound. This is the case with the DDI1 and the
MAG1
gene providing resistance for CmpdC. These genes are induced by DNA damage and
the fact that they provide resistance for the compound may indicate that this
compound is a DNA damaging agent.
Table 3: Active compounds and affected genes.
Com ound Gene Descri tion
CmpdA YAP l Transcription factor involved in drug
resistance and
oxidative stress res onse
CAD 1 Transcri tional activator involved in
multidru resistance
SN 2 ABC traps ort rotein
CmpdB YAP1 Transcription factor involved in drug
resistance and
oxidative stress res onse
CmpdC PRP39 U1 snRNA-associated protein required for
commitment of
re-mRNA to s licin athwa
MAG1 DNA-3-methyladenine glycosidase; excises
alkylation-
dama ed DNA base
DDI1 DNA dama a inducible ene
Cm dD SN 2 ABC trap ort rotein
CLB 1 G 1 c clip
SUI3 Translation initiation factor eIF2 subunit
CmpdE YAP 1 Transcription factor involved in drug
resistance and
oxidative stress res onse
SN 2 ABC traps ort rotein
CmpdF YAP1 Transcription factor involved in drug
resistance and
oxidative stress res onse
SN 2 ABC traps ort rotein
Cm dG ERG24 C14 sterol reductase er osterol bios thesis
POP3 Involved in rocessin of rRNA and tRNA
recursors
RAP 1 DNA-binding protein with repressor and
activator
activities
Cm dH PDRI Transcri tion factor involved in dru resistance
Cm dI RRP43 R uired for 3' rocessin of ribosomal 5.8S
rRNA
RNR4 Ribonucleotide reductase
-15-
SUBSTI1UTE SWEET (RULE 26)
CA 02327830 2000-10-06
WO 99/53098 PCT/US99/0780Z
Exam 2
(Adhesion)
A library of 100,000 test compounds is individually screened against an
aliquot of the fibroblasts to identify compounds which alter the cells'
ability to adhere
to a substrate, e.g., a plastic culture dish. Le., the detectable process
being studied in
this instance is adhesion to a substrate. The test compounds that modulate the
cells
adhesion ability are identified and termed "active compounds." Each of the
active
compounds is then individually screened against each of about 15,000
fibroblast
clones (representing the fibroblast-specific clones currently available from
the
LM.A.G.E. Consortium), each of which overexpresses a different human gene
(delivered via a recombinant retroviral vector). Those clones or pool samples
which
alter the ability of the corresponding active compound to modulate, e.g.,
inhibit, cell
adhesion are then isolated and the heterologous genes) therein is
characterized. In
this case, fibroblasts with inhibited adhesion are eluted from the substrate,
and those
remaining (having adhesion restored by the overexpressed gene) are examined.
If the
cells have been cultured and tested individually, the overexpressed gene will
already
be known.
Alternatively, the overexpressing cells can be labeled , for example with a
2 o polynucleotide marker unique to each overexpressed gene, and the genes in
the
adherent cells identified on the basis of their markers. Alternatively, the
overexpressed gene can be sequenced directly. In these cases, the fibroblasts
can be
pooled, and assayed simultaneously.
2 5 Example 3
(CMV Assay)
Cytomegalovirus (CMV) contains a large number of genes, many of which are
still uncharacterized. There are a number of CMV-susceptible cell lines, and
laboratory strains of CMV available for study.
3 0 HFF cells are infected with the RC256 strain of CMV (Hippenmeyer and
Dilworth, Antiviral Res (1996) x:35-42), and are then contacted with a large
variety
of test compounds. Compounds that demonstrate activity against CMV (i.e.,
which
prevent cell death due to CMV infection) are identified and used in the second
phase.
-16-
CA 02327830 2000-10-06
WO 99/53098 PCTNS99/07802
HFF cells are transfected with each of the 175 CMV genes individually, using
retroviral vectors that provide for overexpression of a CMV gene, and are then
infected with CMV RC256 in the presence of the selected compounds identified
in
phase 1. In most cases, the compound will still protect the cell line from the
effects of
CMV infection. However, the CMV gene overexpressed in some cell lines will be
able to overcome the effect of the test compound, resulting in death of that
cell line
from CMV infection. The effective genes are then identified, and are possible
targets
of the test compound.
-17-
