Previous Article | Next Article ![]()
Infection and Immunity, February 2005, p. 872-877, Vol. 73, No. 2
0019-9567/05/$08.00+0 doi:10.1128/IAI.73.2.872-877.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Costi D. Sifri,3,
,
Samuel Goldman,1
Stephen B. Calderwood,3,4 and
Frederick M. Ausubel1,2*
Department of Molecular Biology,1 Division of Infectious Diseases, Massachusetts General Hospital,3 Department of Microbiology and Molecular Genetics,4 Department of Genetics, Harvard Medical School, Boston, Massachusetts2
Received 3 July 2004/ Returned for modification 31 August 2004/ Accepted 5 October 2004
|
|
|---|
|
|
|---|
It has been difficult to identify S. aureus virulence factors due to the unwieldiness of mammalian pathogenesis models. As an alternative to traditional mammalian pathogenesis models, our laboratory and others have utilized a strategy for studying host-pathogen interactions, using the nematode Caenorhabditis elegans as a simple surrogate model host (3). A variety of human bacterial and fungal pathogens kill C. elegans, and there is a remarkably high degree of correlation between pathogen virulence factors required for nematode killing and virulence in vertebrate models (39). This allows genome-wide screening for pathogen virulence factors by a relatively high-throughput screening procedure (15, 16, 23, 26, 29, 42, 43).
Recently, we showed that a variety of S. aureus strains efficiently kill C. elegans and that S. aureus genes known to be important in mammalian pathogenesis are also required for nematode killing (40). As described here, we generated a Tn917 transposon insertion library of 2,950 mutants in S. aureus NCTC 8325 and screened the library for attenuated killing of C. elegans. We identified 10 genes important for virulence in C. elegans, one of which appears to encode a previously undescribed virulence factor important for pathogenicity in a murine renal abscess model.
(This work was presented in part at the 103rd General Meeting of the American Society for Microbiology, Washington, D.C., 18 to 22 May 2003.)
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Strains and plasmid used in this study
|
Nematode killing assay and screening protocol. S. aureus killing plates were prepared as previously described (40). Primary screening was carried out by transferring 20 to 30 L4-stage nematodes to each plate (27), incubating plates at 25°C, and scoring for live and dead worms after 40 h. Subsequent assays were performed with 25 to 35 L4-stage nematodes per plate in triplicate.
Determination of Tn917 insertion site. The DNA segment adjacent to Tn917 insertion sites was determined with genomic DNA (DNeasy tissue kit; QIAGEN) and a nested "arbitrary" PCR technique using the transposon-specific primer Seq1 and the arbitrary primer Arb1a in the first round and the nested primers Arb2 and Int1 in subsequent rounds (primer sequences are shown in Table 2) (34). PCR products were purified (QIAquick PCR purification kit; QIAGEN), sequenced with primer Int2 (Taq DyeDeoxy Terminator Cycle sequencing kit; Applied Biosystems) in concert with an ABI 3700 PRISM automated sequencer, and analyzed with the Vector NTI Suite 7 software package (InforMax, Inc.).
|
View this table: [in a new window] |
TABLE 2. Oligonucleotide primers used in this study
|
(Invitrogen). Plasmid DNA was purified from these strains, and flanking DNA was sequenced with the transposon-specific outward-facing primers Int3 and Seq1.
Transduction of Tn917 insertions.
The general transducing phage
80 was used as previously described (24), and transductants were confirmed by PCR with Seq1 and a mutant-specific primer listed in Table 2.
Murine kidney abscess infections. Bacterial inocula were prepared from overnight cultures grown at 37°C in TS medium with the appropriate antibiotic and diluted to approximately 108 CFU/ml in TS medium. Appropriate dilutions were prepared and plated on TS agar plates to determine the precise inoculum. Female CD1 mice (Charles River Laboratories) were injected intraperitoneally with 0.2 ml of the bacterial suspension, and their health was monitored daily. Five days postchallenge, mice were sacrificed and both kidneys from each mouse were harvested, washed in phosphate-buffered saline, weighed, and homogenized in 2 ml of phosphate-buffered saline and colony counts were determined on TS agar plates with appropriate antibiotics. Animal study protocols were reviewed and approved by the Subcommittee on Animal Studies of the Massachusetts General Hospital.
Statistical analysis. Nematode survival was calculated by the Kaplan-Meier method, and survival differences were tested for significance by use of the log-rank test, comparing mutant and wild-type survival curves (GraphPad Prism, version 3.0). Differences in colonization of murine kidneys by S. aureus mutant strains and wild-type S. aureus were tested for significance with an unpaired two-tailed t test (GraphPad Prism, version 3.0). P values < 0.05 were considered significant.
|
|
|---|
Determination of insertion site and phenotypic confirmation by transduction. The insertion sites of the Tn917 transposons in the 21 selected mutants were determined by an arbitrary PCR or a plasmid marker rescue method (7, 16, 34) as described in Materials and Methods. BLAST analysis (4) against the S. aureus 8325 genome and the annotated S. aureus N315 genome identified the insertion sites and associated disrupted open reading frames (ORFs) in all 21 mutants (22, 25). As shown in Table 3, eight distinct genes were directly disrupted, three mutants contained insertions within 200 bp of the 5' end of three different genes, and one mutant contained a Tn917 insertion site in intergenic sequence that was downstream of two convergently transcribed genes. Twelve mutants (indicated in Table 3) with mutations corresponding to insertions in 12 different genes or chromosomal locations were chosen for further analysis. These clones resulted from Tn917 transposition and not whole plasmid integration, since all mutants were tetracycline sensitive. DNA blot analysis showed that each of these 12 mutants contained a single Tn917 insertion (data not shown).
|
View this table: [in a new window] |
TABLE 3. Attenuated S. aureus mutants identified by screening in C. elegans
|
80 (24). As shown in Table 3, among the 12 transduced insertions, 10 maintained their attenuation after transduction. Figure 1 shows that these 10 mutants have significantly decreased virulence toward C. elegans. All 10 mutants grew similarly to wild-type S. aureus in TS medium, except for mutant 30A5, which contains Tn917 inserted into the pyrAA gene (data not shown).
![]() View larger version (18K): [in a new window] |
FIG. 1. C. elegans killing by transduced S. aureus 8325 mutants. Shown are the fractions of nematodes killed after 39 and 43 h of feeding on S. aureus 8325(pLTV1) and transduced isogenic Tn917 insertion mutants. Following the growth of bacteria on TS agar plates, 25 to 35 L4-stage hermaphroditic N2 worms were transferred to each plate. Nematode death was determined after 39 and 43 h of feeding. The data represent the mean + standard deviation (error bar) of three sets of plates and are representative of three independent experiments.
|
![]() View larger version (17K): [in a new window] |
FIG. 2. Survival of C. elegans on S. aureus Newman mutants. Shown is the survival of nematodes fed S. aureus Newman (wild type, closed squares; n = 77) and transduced isogenic mutants 6A5 (nagD, open diamonds; n = 79), 7G12 (citG, open squares; n = 80), 15G12 (SA1241, x; n = 79), 28G12 (5' yacA, open circles; n = 76), 29C3 (5' fbp, open triangles; n = 71), and 30A5 (pyrAA, open inverted triangles; n = 76). Survival was plotted by the Kaplan-Meier method. Pairwise comparisons (log-rank test) by strain were as follows: Newman versus 6A5, P < 0.0001; Newman versus 7G12, P < 0.0001; Newman versus 15G12, P not significant; Newman versus 28G12, P < 0.0001; Newman versus 29C3, P < 0.0001; and Newman versus 30A5, P < 0.0001.
|
![]() View larger version (16K): [in a new window] |
FIG. 3. Distribution of Tn917 insertion sites within the S. aureus genome. Shown is a schematic representation of the Tn917 insertion sites of those mutants with reduced virulence in the C. elegans infection model along a linearized S. aureus chromosome. Lines indicate nucleotide insertion sites. An enlarged depiction of the terminus region, containing a majority of the Tn917 insertion sites, is shown above the chromosome.
|
![]() View larger version (16K): [in a new window] |
FIG. 4. Select S. aureus Newman mutants tested in a murine renal abscess model. Groups of five CD1 mice were inoculated intraperitoneally with the following strains: Newman (wild type; 9.8 x 107 CFU), ALC638 (agr sarA; 1.4 x 108 CFU), 6A5 (nagD; 1.4 x 108 CFU), 7G12 (citG; 6.6 x 107 CFU), 28G12 (5' yacA; 5.0 x 107 CFU), 29C3 (5' fbp; 6.0 x 107 CFU), 30A5 (pyrAA; 1.4 x 108 CFU), and 15G12 (ORF SA1241, negative control; 1.6 x 108 CFU). Five days after inoculation, the mice were sacrificed and the kidneys were excised, homogenized, diluted, and plated. CFU of Newman, ALC638, and Tn917 mutants in renal tissue were calculated and plotted after log transformation. Statistical significance was determined using a two-tailed t test; compared to wild-type strain Newman, ALC638 and 6A5 were attenuated in renal tissue bacterial load (P < 0.05). Data are representative of two independent experiments.
|
|
|
|---|
Two STM experiments have been carried out with S. aureus. Mei et al. screened a tagged library in a murine bacteremia model and identified 50 mutants harboring insertions in mainly biosynthetic loci (31). Coulter et al. used three different murine infection models (cutaneous abscess, bacteremia, and wound infection) to identify 237 genes important for in vivo survival, the majority of which also code for biosynthesis-related enzymes (11). Genome-wide transcriptional profiling has also been used to identify virulence-related genes. Dunman et al. identified the set of genes regulated by the important regulatory systems agr and sar, which included many previously identified putative virulence factors, a range of metabolic genes, and many hypothetical genes (13). Similarly, Saïd-Salim et al. defined the regulon of the transcriptional regulator rot, which was found to negatively regulate many secreted proteins and to positively regulate many cell wall-associated proteins (37).
Screening a Tn917 insertion library in a pathogenicity model. In this study, we constructed a Tn917 transposon insertion mutation library in S. aureus NCTC 8325. Screening the library resulted in the identification of 10 unique attenuated mutants (Table 3). Five of these genes (odhA, odhB, braB, opp-2C, and dinG) had previously been identified in the STM screens of Coulter et al. and Mei et al. (11, 31). Additionally, five of these genes (odhA, odhB, opp-2C, pyrAA, and dinG) were found to be upregulated by the agr system based on transcriptional profiling (13). While the genes identified in our screen are associated with metabolic functions in S. aureus, all mutants were able to grow well in TS medium, with the exception of the pyrAA mutant. The disrupted genes can be placed into four functional categories based on the proteins they encode: tricarboxylic acid (TCA) cycle components, nucleic acid metabolism/DNA replication, transporters, and miscellaneous proteins. The same classes of genes were identified previously in the STM screens (11, 31).
The TCA cycle appears to play several critical roles in S. aureus pathogenesis. The TCA cycle is critical for the switch from exponential growth to postexponential growth and the subsequent production of capsule (12) and secretion of extracellular toxins such as
-hemolysin and staphylococcal enterotoxin C (10, 41). In addition, many TCA cycle elements are upregulated by the agr global regulatory system (13). As in other bacterial pathogens, nucleic acid biosynthesis is important for survival within hosts, likely due to the poor availability of purines and pyrimidines. The two transporters identified in this study have been previously identified as being important for in vivo survival in the two STM screens performed in S. aureus (11, 31) and likely reflect the importance of amino acid and peptide uptake in the setting of infection.
Importantly, 5 out of the 10 genes (odhA, odhB, braB, opp-2C, and dinG) identified in our in vivo screen in C. elegans were shown previously to have reduced virulence in murine infection models (11, 31). We employed a murine kidney abscess model to investigate the role in mammalian virulence of the remaining five mutants that had not previously been characterized in a murine infection model. Of these, the nagD mutant was attenuated at the P = 0.03 level. The role of nagD (SA0790) in S. aureus virulence is not clear. It shares a high degree of homology with B. subtilis yutF (50% identity, 68% similarity) and E. coli nagD (31% identity, 50% similarity). In E. coli, Peri et al. and Plumbridge independently cloned a locus involved in N-acetylglucosamine catabolism, termed nag (35, 36). However, nagD, the 3'-most member of the nagBACD operon, is not required for N-acetylglucosamine catabolism nor does it appear to contribute to the regulation of the operon (35, 36). In S. aureus and B. subtilis, the N-acetylglucosamine catabolism operon consists of nagAB only and the B. subtilis nagD homologue yutF is in a distinct three-gene operon of unknown function (yutDEF).
Although four of the five mutants that we tested were not attenuated in the kidney abscess model, this does not exclude the possibility that they may be attenuated in other animal models. For example, of the 237 mutants Coulter et al. found that were attenuated in at least one infection model, only 10% of the recovered mutants were attenuated in all three models (11). Similarly, inactivation of the aconitase gene does not affect mouse mortality when S. aureus is inoculated intraperitoneally but does cause attenuation in a subcutaneous infection model (41).
Recently it was shown that S. aureus strain 8325, in which we constructed the Tn917 library, harbors an 11-bp deletion of the rsbU gene that decreases expression of the alternate sigma factor
B, which affects the expression of a variety of virulence-related genes (18). Indeed reconstruction of rsbU+ results in diminished agr expression and diminished expression of the secreted proteins SspA and Hla but increased nematocidal activity (20, 40). Therefore, to ensure that any transposon insertions showing attenuated C. elegans killing were not specific to the 8325 background, we confirmed the phenotype of selected mutants in the S. aureus Newman strain.
If the Tn917 insertion sites in our library were distributed randomly, then our library of 2,950 mutants would represent
1.5 times the coverage of the nonessential regions of the genome. However, as shown in Fig. 3, Tn917 appears to have a strong insertion bias near the terminus of the S. aureus chromosome. Assuming that the distribution of Tn917 insertions as a whole is similar to that found among attenuated mutants, we estimate that we have screened less than one-third of the nonessential genes in S. aureus. Future screens may benefit from the development of an alternative gram-positive transposon system, such as a mariner-based system, which has been shown to insert at random in a variety of organisms (1, 9).
In summary, we have shown that C. elegans can be used relatively efficiently to identify S. aureus virulence factors. Six out of 10 genes identified in C. elegans are relevant for mammalian pathogenesis, including the nagD gene, which was not previously known to encode a virulence factor. Future screens in C. elegans using a fully saturated mutant library would likely identify additional genes involved in mammalian pathogenesis.
|
|
|---|
This work was supported by a postdoctoral fellowship from the Howard Hughes Medical Institute and NIAID grant K08 AI053677 to C.D.S. and by a research grant from Aventis Pharmaceuticals to F.M.A. and S.B.C.
J.B. and C.D.S. contributed equally to this work. ![]()
Present address: Division of Infectious Diseases and International Health, University of Virginia Health System, Charlottesville, VA 22908. ![]()
|
|
|---|
B activity depends on RsbU in Staphylococcus aureus. J. Bacteriol. 183:1843-1852.
B modulates virulence determinant expression and stress resistance: characterization of a functional rsbU strain derived from Staphylococcus aureus 8325-4. J. Bacteriol. 184:5457-5467.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»