Conceived and designed the experiments: AJD RHffC NRW. Performed the experiments: AJD JR. Analyzed the data: AJD PAW. Contributed reagents/materials/analysis tools: AJD PAW MGTH SDB MAQ RWT. Wrote the paper: AJD RHffC.
The authors have declared that no competing interests exist.
The Gram-negative bacterium
To perform the screen we chose the strain
Libraries of recombinant
We screened genomic libraries of
The screen reveals a broadly even distribution of anti-macrophage associated loci across both chromosomes with some notable exceptions (
Outer pair of concentric circles represents both coding strands of the
Several broad functional classes were repeatedly predicted for the anti-macrophage loci identified (
Graph showing the number of CDSs for different functional classes found within anti-macrophage loci identified by the screen on both chromosomes. The majority of CDSs identified within positive loci are associated with exoprotein/ enzymic activity and transport or secretion systems.
Several of the anti-macrophage loci encode putative toxins. For example, BPSS1993 encodes a metalloprotease A, termed MrpA, a 47 kDa protein negatively regulated by QS molecules, a putative hemolysin with homology to
The genome of
There are a total of 105 predicted functional ABC systems encoded within the genome of strain K96243
In order to begin to characterize the range of likely cellular phenotypes caused by this plethora of new candidate anti-macrophage factors we focused on the four positive clusters diagrammed in
Genomic organisation of four anti-macrophage loci isolated by the screen representative of the main functional classes identified. Genes putatively involved in anti-macrophage activity are highlighted in red and roles of other genes in the region are provided in the figure legend based on the original genome annotation and BLASTX analysis. The number of overlapping positive clones isolated by the screen that identify these as regions of interest are shown in blue below the ORF maps. BPSS1263-BPSS1269 encodes NRPS/PKS genes involved secondary metabolism including a putative efflux system and have homology to the SylA producing genes of
Macrophages treated with crude lysate prepared from
Thirty of the anti-macrophage loci contained ‘hypothetical proteins’ whose potential functions cannot be predicted from homology with known virulence factors. BPSL0590 and BPSL0591 are CDSs found in a positive locus on chromosome 1 which encode such hypothetical proteins. However closer examination of protein predictions from these two CDSs does reveal some limited homology to known toxins from other bacterial pathogens suggesting they may encode novel toxins. Position-specific-iterative blast (psi-BLAST) reveals that this putative membrane protein also has a central region similarity to the Rhs associated core sequence (1.98e-10) and to the middle N and C-terminal domains of a
The neighbouring hypothetical protein BPSL0591 also displays predicted homology to a
The third cluster chosen for further phenotypic analysis (
Treatment of macrophages with lysate from clones carrying a putative phospholipase D gene for 24 h results in a loss of cytoskeletal structure and abnormalities within the nucleus (A). In contrast, macrophages treated with lysate containing a hemagglutinin-like protein for 24 h (B) suffer severe alterations in their actin cytoskeleton producing pronounced actin filaments (white arrows). Scale bars = 10 µm.
The fourth cluster chosen for follow up analysis contains two CDSs encoding a putative hemagglutinin and with homology to the large filamentous hemagglutinin precursor, FhaB, (BPSS1727) and hemolysin activator-like protein precursor, FhaC (BPSS1728) of
Finally, to demonstrate how bioactivities from positive gene clusters can be further confirmed and fractionated, we carried out a more detailed analysis of the gene cluster encoding the
Fraction screening of clones containing the SylA-like region reveals that cytotoxicity is not present in the cytosolic fraction. F-actin cytoskeleton staining of the treated cells reveals an intact macrophage monolayer (x10) with cells displaying a normal phenotype. Cytotoxicity is found in the bacterial cell-free supernatant a clear reduction in the number of macrophages in the monolayer can be seen. Higher magnification reveals cytoskeletal collapse and cell shrinkage implying that the anti-macrophage moiety is secreted. Scale bars: x10 = 50 µm, x60 = 10 µm.
The bacterium
The genomic locations of the gene clusters encoding anti-macrophage effectors and compounds are worthy of brief discussion before a detailed discussion of the clusters themselves and their potential role in infection biology. Thus although positive loci are distributed relatively uniformly across the two chromosomes of
Given the relatively crude and simple nature of the screen and given that the recombinant
In reference to the genetic composition of the clusters themselves, phenotypic analysis of four regions of interest detected by the screen begin to link activity of the gene products to some of the important phenotypes associated with
Following attachment,
Anti-macrophage activity is also seen in response to positive clones containing a phospholipase D domain protein. A PLD gene encoding a protein with phospholipase D activity is associated with phagosomal escape in
Finally, many of the positive gene clusters are associated with the production of NRPS/PKS systems which are predicted to make small molecules or peptides. Whilst it is often possible to predict the likely structure of the small molecules made via the unique combinations of PKS modules present, the role of these gene products in bacterial virulence is often less clear. We focused on one such positive region in
Current models for studying the factors involved
A combination genomic BAC and fosmid libraries, were used in these experiments. The BAC library was constructed in
Library plates were replicated into 96 well microplates containing 100 µl Luria Bertani medium plus 12.5 µg ml−1 chloramphenicol as the selective antibiotic for both the BAC and Fosmid library clones. Replicate library plates were grown for 24 h at 350 rpm, 37°C and cultures were subsequently harvested by centrifugation at 4,000 rpm for 10 minutes. 80 µl of supernatant aspirated the remaining bacterial pellet and supernatant mixed thoroughly with 80 µl 1 mg ml−1 lysozyme in Phosphate Buffered Saline solution. Plates were then incubated at room temperature for a minimum of 1 h, followed by three freeze-thaw cycles before centrifugation at 3,000 rpm for 10 minutes. 80 µl of the crude lysates were removed and applied to 96 well plates containing confluent monolayers of the BALB/c monocyte macrophage cell line J774-2 (from The European Collection of Cell Cultures, ECACC) in Dulbecco's Modified Eagles Medium (DMEM) supplemented with 10% foetal bovine serum, 5% non-essential amino-acids and 5 µg ml−1 chloramphenicol. Crude lysates and macrophages were co-incubated for 24 h (37°C, 5% CO2). Media on the macrophages was then replaced with phenol red-free DMEM containing an antibiotic cocktail: ampicllin 100 µg ml−1, gentamicin 50 µg ml−1, penicillin 100 U ml−1, streptomycin 100 µg ml−1, kanamycin 100 µg ml−1 and tetracycline 5 µg ml−1 and incubated with the macrophages for 2 h (37°C, 5% CO2) to destroy live bacteria which would otherwise affect the readout of the cell viability assay. Macrophage cell viability was assessed using the XTT assay
BAC and fosmid end sequences were aligned to the
Confluent monolayers of J774-2 macrophages on glass coverslips were treated with equivalent volumes of crude lysates (prepared as described above) from clones identified as containing regions of interest and co-incubated for 24 h. Coverslips were then washed in sterile 1X PBS before fixing with 4% paraformaldehyde (w/v) in PBS for 15 min. Coverslips were then washed in 1 X PBS and immersed in a fresh solution of ammonium chloride in 1X PBS (13.3 mg/ml) for 15 minutes, at room temperature followed by washing in 1X PBS. Macrophages were permeabilised by covering with 0.2% Triton X-100 in 1X PBS for 15 minutes. Staining of the filamentous actin cytoskeleton was carried out with TRITC conjugated phalloidin (Sigma), at a 1/500 dilution in 1X PBS by inverting the coverslip onto a 60 µl drop of the staining solution and incubating at room temperature in the dark for 1 h. Following incubation the coverslips were washed 3×5 minutes in 1X PBS with the first wash containing 0.12 µg/µl−1 Hoechst 33258 (Sigma) to stain the nuclei. A final wash by brief immersion 2X in distilled water is then made and coverslips mounted onto slides using ProLong Gold Antifade (Molecular Probes, Invitrogen) before visualization using fluorescence microscopy. For fraction screening of the SylA homolog a single colony of a BAC library clone containing the SylA region of homology was picked and grown for 24 h in LB plus 12.5 µg ml−1 chloramphenicol. Bacteria were harvested by centrifugation at 7,000 rpm for 5 minutes and the culture supernatant removed. Cell-free supernatant was prepared by filter-sterilising with a 0.22 µm syringe-driven filter unit (Millipore). The cytosolic fraction was prepared by re-suspending the bacterial pellet in 1X PBS and sonicating the mixture. The resultant sonicated product was centrifuged at 10,000 rpm for 15 min. to remove cell debris from the cytosol preparation. Supernatant and cytosol fractions were applied to J774 macrophage monolayers at 1∶5 (v/v) supernatant or cytosol: culture media and co-incubated for 24 h before fixing and staining as described.
Complete inventory of anti-macrophage associated loci identified on
(DOC)
Complete inventory of anti-macrophage associated loci identified on
(DOC)
We thank Dr M. Sarkar-Tyson for supplying copies of the