CA2328497A1 - Genetics-assisted target evaluation in antibacterial drug discovery - Google Patents

Abstract

The present invention provides a method, which is designated GATE (Genetics- Assisted Target Evaluation), for gene analysis in pathogenic bacteria. GATE functionally classifies genes according to their essential or nonessential character with respect to bacterial growth or pathogenicity. The technology uses a nonessential scorable gene as an internal standard against which the dispensability of a given target gene is assessed by gene disruption by Campbell-type integration. Functional analysis was performed on 26 genes in the Gram-positive pathogen Streptococcus pyogenes. Application of this metho d resulted in a clear identification of essential and nonessential loci: 16 of the 26 genes evaluated in S. pyogenes clearly encoded functions essential fo r bacterial growth, while the remaining 10 encoded functions that were nonessential. The successful application of GATE has also been demonstrated in Staphylococcus aureus.

Description

GENETICS-ASSISTED TARGET EVALUATION
IN ANTIBACTERIAL DRUG DISCOVERY
FSeld of the Invention This applicaxion claims priority under 3 5 U. S. C. ~ 119 from provisional patent application Serial No. 60/085,593, filed May 15, 1998, the entire disclosure of which is incorporated-by reference herein in its entirety.
The present invention is directed to methods for gene analysis in pathogenic bacteria that provide the means to discover and validate new therapeutic targets.
BACKGROUND OF THE INVENTION
The increasing availability of bacterial genome sequence information has created a need for scalable methods by which to assess gene function. Sequence homology should allow assignment of potential function to many gene products. However, functional genetic analysis is necessary to effectively utilize genomic data.
The technology of the present invention allows for the functional classification of genes according to their essential or nonessential character with respect to bacterial growth or pathogenicity.
Gene products identified by this procedure as being indispensable for growth or virulence provide a source of new targets for high-throughput screens aimed at discovery of small molecules which possess bactericidal activity.
Genetics-Assisted Targeted Evaluation (GATE) uses homologous recombination between integrative plasmids bearing target gene fragments and the bacterial chromosome to create gene disruption mutations at defined loci. GATE is a robust procedure for gene knock-out analysis in bacteria that is rapid, reproducible, and scalable to accommodate analysis of large numbers of genes. A further attribute of GATE is its potential for functional genetic analysis in a broad range WO 99160111 PCfIUS99/10919 of pathogenic bacteria, including both Gram-positive and Gram-negative organisms. As a proof of principle, the utility of GATE in two distantly related Gram-positive pathogens of medical importance, Streptococcus pyogenes and Staphylococcus aureus, has been demonstrated.
GATE holds several advantages over prior art methods for determining whether or not a gene is likely to be essential for bacterial growth. Typically, simple failure to obtain transformants with a plasmid construct designed to produce either gene disruption or gene replacement mutations at a locus of interest is interpreted as evidence that the gene is essential .for bacterial growth.
However, failure to obtain transformants is not in itself sufficient for assignment of gene function, due to lack of adequate positive controls for successful transformation and plasmid integration.
Similarly, assessment of gene essentiality based on construction of partial diploids is labor-intensive, requiring not only a full-length clone for a gene of interest, but also development of gene transfer systems for pathogenic bacteria such as, e.g., specialized transducing phage, that can deliver the complementing gene copy to a secondary genomic site.
Thus, there is a need in the art for rapid and efficient methods for establishing the essentiality or nonessentiality of bacterial genes.
Summary of the Invention The present invention provides methods for determining the essentiality or non-essentiality of a bacterial gene of interest, which are carried out by the stpes of:
(a) transforming a bacterial culture in a substantially simultaneous manner with:
(i) a first plasmid comprising a recombination cassette comprising a first selectable marker flanked on its 5' and 3' termini by sequences derived from the gene of interest, WO 99!60111 PCTIUS99/10919 wherein the recombination cassette is capable of being integrated into the genome of said bacteria by homologous recombination; and (ii) a second plasmid comprising a recombination cassette comprising a second selectable marker flanked on its 5' and 3' termini by sequences derived from a gene known to be non-essential for growth of the bacteria, wherein the recombination cassette is capable of being integrated into the genome of the bacteria by homologous recombination;
(b) individually culturing the transformed culture produced in step {a) under selective conditions in which (i) only cells expressing the first selectable marker will survive or (ii) only cells expressing the second selectable marker will survive;
(c) measuring the number of surviving cells in cultures (il and (ii) of step (b); and (d) comparing the measurements made in step (c), wherein the lack of detectable colonies in culture (i) and the appearance of colonies in culture (ii) indicates that the gene of interest is essential for growth of the bacteria.
"Substantially simultaneous" as used herein indicates that the two DNA
preparations used in the transformation are mixed or are otherwise both contacted with the bacterial cells prior to the stimulus (e.g., electric pulse, heat shock, and the like) used to introduce the DNA into the bacterial cells. It will be understood that the 5' and 3' sequences that flank each selectable marker are of sufficient length, i.e., at least about 100 nt each, to allow for homologous recombination into the chromosomal copy of the gene of interest or the non-essential (control gene), respectively.
In one embodiment, the selectable markers are antibiotic resistance genes, and the selective conditions are individual cultures comprising each antibiotic specified by the cognate resistance gene.

BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a schematic diagram showing the use of the GATE assay for target validation in the discovery of new antibacterial drugs.
Figure 2 depicts the advantages of using Streptococcus pyogenes group A
bacteria in GATE assays of the present invention.
Figure 3 is a schematic diagram showing the gene disruption strategy using GATE
according to the present invention.
Figure 4 shows the use of the spell locus encoding the Streptococcus erythrogenus toxin B gene used as an internal standard for classifying test genes in S. pyogenes.
Figure 5 shows the analysis of selected ORFs in S. pyogenes. Colony counts for each cotransformation of the control gene and a test gene are shown.
Figure 6 shows the use of the GATE technology of the present invention in the Gram-positive pathogen Staphylococcus aureus. Successful disruption of the structural gene for a-hemolysin (hla) gene was evidenced by the loss of hemolysis on blood agar and was confirmed by PCR analysis.
DETAILED DESCRIPTION OF THE INVENTION
All patent applications, patents and literature references cited herein are hereby incorporated by reference in their entirety. In the case of inconsistencies, the present disclosure will control.
Gene analysis using GATE is performed by cotransformation of pathogenic bacteria with two independent integrative plasmids which contain a 300-500 by internal fragment of either (i) the target gene, whose mutant phenotype is being investigated, or (ii) a nonessential gene for which there is an easily scored mutant phenotype. Each integrative plasmid is marked with a different antibiotic resistance determinant so that transformants bearing gene disruptions at either the target locus or at the nonessential control locus can be distinguished by selection on appropriate antibiotic media. Gene disruptions are created by a single recombination event between the plasmid-encoded internal gene fragment and the homologous chromosomal sequence, resulting in duplication of truncated copies of the target gene. Determination of gene essentiality/nonessentiality is made by comparing the number of antibiotic-resistant transformants generated by plasmid integration events which inactivate the target with those formed using the control gene.
Use of a nonessential scorable gene as an internal standard for plasmid transformation and homologous recombination is a key feature of this method. This provides a benchmark against which to assess the dispensability of a given target gene for bacterial growth. If it is possible to obtain plasmid integration with a target gene construct at a frequency comparable to that.obtained with the control gene, it is inferred that the target gene is nonessential for growth.
If, on the other hand, it is not possible to obtain plasmid integration at the target locus when in the same transformation mixture gene disruption at the control locus is readily obtained, it is inferred that the target is essential for growth.
Suitable candidates for nonessential scorable genes are easily identifiable in most pathogenic bacteria. GATE was performed in Streptococcus pyogenes using as an internal control a nonessential gene, spell, encoding an extracellular cysteine protease whose activity in bacterial colonies is readily assayed on Petri dishes containing casein as a protease substrate. Similarly, in applying GATE to gene analysis in Staphylococcus aureus, a gene disruption mutation in the structural gene for a secreted lipase (geh) whose expression can be scored conveniently on an indicator medium was used.

GATE can be applied to any pathogenic bacterium that undergoes homologous recombination and for which gene delivery procedures (e. g. , electrotransformation, natural competence, or conjugation) can be established. GATE is well-suited for analysis of large numbers of genes. The approach requires only a modest amount of DNA sequence information in order to create a gene disruption, because homologous recombination can proceed efficiently in many bacteria using cloned internal gene fragments as short as 300-500 bp. To increase the through-put with which mutants can be produced, a series of integrative vectors (such as, e.g., derivatives of Stratagene's pCR-ScriptT~"Cam) which (i) permit rapid cloning of blunt-end, PCR-generated gene fragments, and (ii) contain antibiotic-resistance markers selectable in a wide variety of Gram-positive bacteria (such as, e.g., erythromycin, kanamycin, or spectinomycin resistance genes) has been constructed. GATE
also incorporates high through-put methods for verifying the genotypes of mutant strains. PCR
procedures have been developed that utilize crude lysates of individual bacterial colonies and thus permit the determination as to whether or not a gene disruption event occurred at the correct chromosomal location.
GATE has been successfully implemented in a functional genetic analysis performed on 26 predicted genes in the Gram-positive bacterial pathogen S. pyrogenes.
These targets predominantly included S. pyrogenes homologs of genes of known function in other bacteria, e.g., genes involved in essential cellular processes such as transcription, translation, DNA replication, and cell wall biogenesis. Representative genes which are suspected to be nonessential for growth in rich medium, including S. pyrogenes homologs of virulence factors known in other bacteria, have also been examined. In a few cases, functional analysis was performed on genes which were essential in certain bacteria and nonessential in others, such that the outcome of the experiment in S. pyogenes could not be predicted in advance.

In cotransformation experiments, integrative vectors bearing an internal fragment of the target gene and the spell nonessential scorable control gene were introduced into S. pyogenes by electroporation and gene disruption events at target or control loci were selected for on rich antibiotic media. Inferences as to gene function in vitro were made based on colony counts as described above.
It was possible to readily distinguish essential and nonessential loci: 16 of the 26 predicted genes that were evaluated clearly encoded functions essential for bacterial growth, while the remaining 10 encoded functions that were nonessential. An added benefit of GATE as an approach to functional analysis is the fact that, in addition to classifying genes according to their essential or nonessential character for growth, mutant strains can be assayed later for alternative phenotypes in vitro such as, e.g., growth under conditions of oxidative stress, altered osmolarity or pH, or iron deprivation. This procedure can be used to identify functions involved in pathogenicity.
The present invention is described below in working examples which are intended to further describe the invention without limiting its scope.
EXAMPLE I
In order to test whether a particular gene is essential or nonessential in S.
pyogenes, integrative plasmid vectors were used that contain different antibiotic resistance markers (Tao, et al., 1992). These plasmids cannot replicate in S. pyogenes because they do not contain a streptococcal origin of replication (Tao, et al., 1992). Integrative plasmids used in this study consisted of the PCR
Script vector (Stratagene) which had been modified by the addition of antibiotic resistance markers.
The kanamycin and erythromycin resistance genes were generated by PCR
amplification of plasmids pSF151 and pVA891.1, respectively (Tao, et al., 1992) using Pfu polymerase (Stratagene). To construct PCRScript-kan, conferring kanamycin-resistance, the following oligonucieotide primers were used: S'-GATCCCATGGGCGAACCATT-3' and 5'-GATCCCATGGAATTCCTCGT-3' . To construct PCRScript-erm, conferring erythromycin-resistance, the oligonucleotide primers were: 5'-GATCCCATGGCGAAATGATA-3' and 5'-GATCCCATGGGGCGCTAGGG-3' . PCR-generated antibiotic resistance genes were ligated into the Nco I site of the PCRScript vector. A 300-500 by internal fragment corresponding to each gene of interest was generated by PCR amplification with Pfu polymerase (Stratagene) using S.
pyogenes genomic DNA template (S. pyogenes strain 2F-3MB-1512) and gene-specific primers were designed using DNA sequence data in the public domain (Streptococcus pyogenes genome sequence database, University of Oklahoma). The PCR reaction mixtures contained 300 ng of genomic DNA
template, 100 ng of each primer, 0.2mM dNTPs, and 2.5 units of Pfu polymerase.
The reactions were amplified for 30 cycles (94°C, 1'; 50°C, 1'; 72°C, 3'). Each gene fragment was ligated into the Sma I site of PCRScript-erm in the presence of Srfl to prevent vector recircularization. An internal gene fragment of the nonessential control gene (spell) was ligated into the Sma I site of PCRScript-kan in the same manner.
Cotransformation of S. pyogenes was performed to determine whether a particular gene is essential or nonessential. Ten micrograms of each plasmid were combined, ethanol precipitated, and dried under vacuum. Transformation of S. pyogenes was by electroporation (McLaughlin and Ferretti, 1992). When preparing electrocompetent cells, I00 ml of early log phase cultures were harvested by centrifugation at 3000 g for 10 minutes at 4°C. The supernatant was saved and cell pellets were resuspended in 5 ml of the spent media. The cell suspension was subjected to heat shock at 42° C for 9 minutes, after which cells were collected by centrifugation and washed 2X in 10 ml of 15 % sterile glycerol. Final pellets were resuspended in 0.6 ml of 15 %
glycerol. For transformation, 200 ml of electrocornpetent cells were added directly to the tube containing the dried WO 99!60111 PCT/US99/10919 plasmid DNA pellet. The transformation mixture was transferred to a chilled sterile electroporation cuvet and a single pulse was applied (1.75kV, 25~cF capacitance, 400W
resistance). Cell suspensions were placed on ice for 30 minutes, then transferred to 10 ml THY broth and incubated at 37°
C/5 %C02 for 2-3 hours. Cells were harvested by centrifugation and transformants were resuspended in 1 ~1 of THY broth. Half of the cell suspension was plated in 100 ~1 aliquots on THY agar plates containing kanamycin ( 100 ~uglml), the remaining 0.5 ml was plated on THY
agar plates containing erythromycin (1 ~,g/ml). Plates were incubated at 37°C/SICOz for 48 hours.
Colonies were counted to determine whether the test gene is essential or nonessential.
If the test gene is nonessential, the number of colonies on the test plates (erythromycin) should be comparable to the control plates (kanamycin). If the test gene is essential, the erythromycin plates should have no colonies, while the control kanamycin plates should contain significant numbers of colonies (typically 200-300) indicating integration into the nonessential control gene.
Disruption of spell was scored visually on agar containing 1 % casein. Wild-type S.
pyogenes colonies are ordinarily surrounded by a clear halo due to spell-encoded proteolytic activity, while spell mutants defective in protease activity produce a greatly diminished halo. Disruption of test genes was verified molecularly by direct PCR amplification of representative transformant colonies using plasmid and gene-specific primers.
Using the above methods, 26 S. Pyogenes open reading frames (ORFs) were analyzed to determine their essentialitylnon-essentiality. The results are set forth in Table I
below.

Table 1 FUNCTIONAL ANALYSIS
OF 26 S. o eves ORFs Homolo Ex acted Role Essential* Nonessential*

rA DNA re lication X

dnaG DNA re lication X

s B ~ virulence X

sc A virulence X

hasp virulence X

virR virulence X

murC cell wall s nthesisX

murD cell wall s nthesisX

murE cell wall s nthesisX

recA recombination X

ftsZ cell division X

ftsH cell division X

fabD fatty acid X
s nthesis nB enicillin bindin X

b 2x nicillin bindin X

oB transcri tion X

oC transcri Lion X

oD transcri don X

secA rotein secretionX

1 B rotein secretionX

fb c iron traps ort X

ob GTP bindin X

inlA virulence X

ppi peptidylproiyl X
isomerase tufA translation sod oxidation stress *growth on rich medium As can be seen from the data in Table I, 16 genes were identified as being essential and non-essential.
References:
McLaughlin, R. E. and Ferretti, J. (1992) In: Methods in Molecular Biology, Vol.
47: Electroporation Protocols for Microorganisms 10 Tao, L., LeBlanc, D.J. and Ferretti, J. (1992) Gene 120, 105-110.

Claims

Claims:

1. A method for determining the essentiality or non-essentiality of a bacterial gene of interest, said method comprising:
(a) transforming a culture of said bacteria in a substantially simultaneous manner with:
(i) a first plasmid comprising a recombination cassette comprising a first selectable marker flanked on its 5' and 3' termini by sequences derived from said gene, wherein said recombination cassette is capable of being integrated into the genome of said bacteria by homologous recombination; and (ii) a second plasmid comprising recombination cassette comprising a second selectable marker flanked on its 5' and 3' termini by sequences derived from a gene known to be non-essential for growth of said bacteria, wherein said recombination cassette is capable of being integrated into the genome of said bacteria by homologous recombination;
(b) individually culturing the transformed culture produced in step (a) under conditions in which (i) only cells expressing said first selectable marker will survive or (ii) only cells expressing said second selectable marker will survive;
(c) measuring the number of surviving cells in cultures (i) and (ii) of step (b); and (d) comparing the measurements made in step (c), wherein the lack of detectable colonies in culture (i) and the appearance of colonies in culture (ii) indicates that the gene of interest is essential for growth of the bacteria.