discover new classes of small molecules which inhibit growth of
M. tuberculosis.
To this end, we developed PROSPECT (primary screening of
strains to prioritize expanded chemistry and targets), in which
compounds are tested in a phenotypic assay for growth inhibition
of a pool of genetically barcoded M. tuberculosis hypomorphs
(mutants depleted in an essential gene product) [5]. Hypomorphs’
chemical hypersensitivity enables sensitive detection of inhibitors
which elude conventional phenotypic screening, while the vector of
sensitivities of each hypomorph to a particular compound, known
as the Chemical Genetic Interaction Profile (CGIP) can be used to
assign mechanism of action to bioactive small molecules [5, 6].
The engineering of conditional M. tuberculosis hypomorphs has
been previously described in detail [7–10], therefore we focus on
their application to PROSPECT. In this chapter, we describe the
principles behind PROSPECT and a protocol for collecting CGIPs
in high-throughput and interpreting them for mechanism of
action.
1.1 Overview
of PROSPECT
The conditional hypomorphs used in PROSPECT are proteolytically regulated with a tetracycline-repressible promoter [9] (see
Fig. 1). Each hypomorph is marked by a 20-nucleotide barcode
sequence (which identifies the depleted gene product) integrated
into the chromosome. This genetic barcode is flanked by common
primer binding regions, so that the hypomorphs can be combined
into a single pool; the final readout of the assay is polymerase chain
reaction (PCR) amplification of chromosomal barcodes followed
by next generation sequencing (NGS) and counting of barcodes
(see Fig. 2). We confirmed that barcode counts are a reliable proxy
for relative hypomorph abundance [5].
The hypomorphs are pooled with a similarly barcoded wildtype M. tuberculosis H37Rv strain at equal proportions before being
exposed to a compound library arrayed in 384-well plates (see Note
1). After incubation for 14 days, DMSO (from an aqueous 10%
(w/v) DMSO stock) is added to each well for to a final concentration of 5% (v/v) and the infectious agent is inactivated by heating at
80
C for 2 h. Bacterial cell lysis is achieved by heating at 98
C for
10 min. The chromosomal barcodes are PCR amplified using primers containing 5
0 -overhangs incorporating well- and plate-specific
barcode sequences which are later used to map sequencing reads
back to screening position.
The amplified material is cleaned up using Solid Phase Reversible Immobilization (SPRI) beads before submitting to NGS on the
Illumina platform. The end result is a FASTQ file which can be
parsed for strain barcode counts which are mapped back to screening plate positions and compound identities. Finally, strain counts
are analyzed to interpret chemical-genetic interactions and mechanism of action.
324
Eachan O. Johnson and Deborah T. Hung
M. tuberculosis.
To this end, we developed PROSPECT (primary screening of
strains to prioritize expanded chemistry and targets), in which
compounds are tested in a phenotypic assay for growth inhibition
of a pool of genetically barcoded M. tuberculosis hypomorphs
(mutants depleted in an essential gene product) [5]. Hypomorphs’
chemical hypersensitivity enables sensitive detection of inhibitors
which elude conventional phenotypic screening, while the vector of
sensitivities of each hypomorph to a particular compound, known
as the Chemical Genetic Interaction Profile (CGIP) can be used to
assign mechanism of action to bioactive small molecules [5, 6].
The engineering of conditional M. tuberculosis hypomorphs has
been previously described in detail [7–10], therefore we focus on
their application to PROSPECT. In this chapter, we describe the
principles behind PROSPECT and a protocol for collecting CGIPs
in high-throughput and interpreting them for mechanism of
action.
1.1 Overview
of PROSPECT
The conditional hypomorphs used in PROSPECT are proteolytically regulated with a tetracycline-repressible promoter [9] (see
Fig. 1). Each hypomorph is marked by a 20-nucleotide barcode
sequence (which identifies the depleted gene product) integrated
into the chromosome. This genetic barcode is flanked by common
primer binding regions, so that the hypomorphs can be combined
into a single pool; the final readout of the assay is polymerase chain
reaction (PCR) amplification of chromosomal barcodes followed
by next generation sequencing (NGS) and counting of barcodes
(see Fig. 2). We confirmed that barcode counts are a reliable proxy
for relative hypomorph abundance [5].
The hypomorphs are pooled with a similarly barcoded wildtype M. tuberculosis H37Rv strain at equal proportions before being
exposed to a compound library arrayed in 384-well plates (see Note
1). After incubation for 14 days, DMSO (from an aqueous 10%
(w/v) DMSO stock) is added to each well for to a final concentration of 5% (v/v) and the infectious agent is inactivated by heating at
80
C for 2 h. Bacterial cell lysis is achieved by heating at 98
C for
10 min. The chromosomal barcodes are PCR amplified using primers containing 5
0 -overhangs incorporating well- and plate-specific
barcode sequences which are later used to map sequencing reads
back to screening position.
The amplified material is cleaned up using Solid Phase Reversible Immobilization (SPRI) beads before submitting to NGS on the
Illumina platform. The end result is a FASTQ file which can be
parsed for strain barcode counts which are mapped back to screening plate positions and compound identities. Finally, strain counts
are analyzed to interpret chemical-genetic interactions and mechanism of action.
324
Eachan O. Johnson and Deborah T. Hung