Nonrandom Distribution of Pseudomonas aeruginosa andStaphylococcus aureus in Chronic Wounds�
Mustafa Fazli,1 Thomas Bjarnsholt,1 Klaus Kirketerp-Møller,2 Bo Jørgensen,2
Anders Schou Andersen,2,3 Karen A. Krogfelt,3 Michael Givskov,1and Tim Tolker-Nielsen1*
Department of International Health, Immunology, and Microbiology, Faculty of Health Sciences, University of Copenhagen,Copenhagen, Denmark1; Copenhagen Wound Healing Center, Bispebjerg Hospital, Copenhagen, Denmark2; and
Received 17 July 2009/Returned for modification 24 August 2009/Accepted 30 September 2009
The spatial organization of Pseudomonas aeruginosa and Staphylococcus aureus in chronic wounds wasinvestigated in the present study. Wound biopsy specimens were obtained from patients diagnosed ashaving chronic venous leg ulcers, and bacterial aggregates in these wounds were detected and located bythe use of peptide nucleic acid-based fluorescence in situ hybridization and confocal laser scanningmicroscopy (CLSM). We acquired CLSM images of multiple regions in multiple sections cut from fivewounds containing P. aeruginosa and five wounds containing S. aureus and measured the distance of thebacterial aggregates to the wound surface. The distance of the P. aeruginosa aggregates to the woundsurface was significantly greater than that of the S. aureus aggregates, suggesting that the distribution ofthe bacteria in the chronic wounds was nonrandom. The results are discussed in relation to our recentfinding that swab culturing techniques may underestimate the presence of P. aeruginosa in chronic woundsand in relation to the hypothesis that P. aeruginosa bacteria located in the deeper regions of chronicwounds may play an important role in keeping the wounds arrested in a stage dominated by inflammatoryprocesses.
Chronic wounds, such as diabetic foot ulcers, pressureulcers, and venous leg ulcers, are an increasing problemworldwide. One to 2% of the population in developed coun-tries develops chronic wounds, a condition associated withsevere patient suffering, the loss of employment, a reducedquality of life, and high costs to the health care system (13).Detailed knowledge about chronic wounds is required inorder to develop better wound treatment and managementstrategies.
A normal wound healing process involves four main phases:(i) coagulation, (ii) inflammation, (iii) cell proliferation andrepair of the matrix, and (iv) epithelialization and remodelingof the scar tissue (23). However, chronic wounds are believedto be captured in the inflammatory phase, where persistentinflux and elevated activity of polymorphonuclear neutrophils(PMNs) occur (1). Although PMNs play a critical role in thehost defense and wound healing, they release cytolytic en-zymes, free oxygen radicals, inflammatory mediators, and ma-trix metalloproteases, which cause local tissue damage in thehost (22, 23, 26).
It is known that the microflora of chronic wounds com-prises multiple species. In a bacterial profiling study, Gjøds-bol et al. found that chronic venous leg ulcers harboredStaphylococcus aureus (in 93.5% of the ulcers), Enterococcus
faecalis (71.7%), Pseudomonas aeruginosa (52.2%), coagu-lase-negative staphylococci (45.7%), Proteus species(41.3%), and anaerobic bacteria (39.1%) (12). S. aureus andP. aeruginosa are opportunistic pathogenic bacteria and arewidely known to cause chronic biofilm-based infections intheir hosts. S. aureus is most commonly isolated fromchronic wounds (8, 12, 15, 17) and, in certain situations, canexpress a number of potential virulence factors and surfaceproteins which promote its adherence to the damaged tissueand decrease neutrophil functions and immune responses ofthe host (10, 11). P. aeruginosa often causes biofilm-basedchronic infections and expresses virulence factors, in partic-ular, rhamnolipid, that can eliminate the activity of PMNs(4, 16). A number of studies have demonstrated that P.aeruginosa is frequently present in chronic wounds (12, 17)and have provided evidence that the bacteria are located inaggregates enclosed in extracellular polymeric matrix mate-rial as found in biofilms (17). Furthermore, chronic woundsthat harbored P. aeruginosa were larger than those that didnot, and the healing process also seemed to be more se-verely hindered for those wounds (12, 14, 20).
Biofilms are bacterial aggregates enclosed in a self-producedextracellular polymeric matrix (6, 21, 25). In clinical environ-ments biofilms can form on dead or living tissues, mucosalsurfaces, or the surfaces of medical devices in the host. Thebacteria in biofilms often display characteristics different fromthose of their planktonic counterparts, such as increased resis-tance to the activities of the host immune system and toleranceto antimicrobial treatments (7). Such characteristics are im-portant, since biofilms are involved in many chronic bacterialinfections. Recent studies have shown the presence of bacterial
* Corresponding author. Mailing address: Department of Inter-national Health, Immunology and Microbiology, Faculty of HealthSciences, University of Copenhagen, Blegdamsvej 3B, Copenhagen NDK-2200, Denmark. Phone: 45 353 26656. Fax: 45 353 27853. E-mail:[email protected].
� Published ahead of print on 7 October 2009.
4084
biofilms in chronic wounds (9, 15, 17). Although the role ofbiofilms in chronic wounds is not yet fully understood, it isbelieved that their existence may be one of the reasons forimpaired wound healing (4, 16).
We previously demonstrated that there is a lack of correla-tion between the bacteria detected by standard culturing andthose detected directly by peptide nucleic acid (PNA)-basedfluorescence in situ hybridization (FISH) in chronic woundsamples (17). While S. aureus was detected more frequentlyby swab sample cultivation than by PNA-FISH, the oppositewas true for P. aeruginosa. This lack of correlation betweendetection by swab sample cultivation and PNA-FISH may bedue to the ability of the different bacterial species to colo-nize different regions of chronic wounds. Swab sample cul-tivation identifies the microorganisms present in the surfaceregion of the wound but may not detect microorganismslocated inside the wound bed. Accordingly, in the presentreport, we present evidence that S. aureus primarily colo-nizes the region of chronic wounds which is close to thesurface, whereas P. aeruginosa primarily colonizes thedeeper regions of chronic wounds. The ability of P. aerugi-nosa to colonize the deeper regions of chronic wounds maybe due to the ability of this organism to produce virulencefactors which destroy PMNs (4, 16), and it may play animportant role in keeping the wounds arrested in a stagedominated by inflammatory processes.
MATERIALS AND METHODS
Tissue sample collection and preparation. Nine patients diagnosed withchronic venous leg ulcers were included in the study. As described below, fourpatients had S. aureus-containing wounds, four patients had P. aeruginosa-con-taining wounds, and one patient had a wound that contained both S. aureus andP. aeruginosa. Material (4-mm punch biopsy specimens) from chronic venous legulcers (Fig. 1) was obtained with the acceptance of the patients and in accor-dance with biomedical project protocols H-B-2008-023 and KA-20051011, whichwere approved by the Danish Scientific Ethical Board. Wound biopsy materialwas collected by a surgical team before cleansing and surgical preparation of thewound (2). The material was immediately transferred to phosphate-bufferedsaline with 4% paraformaldehyde and stored at 5°C before further preparationfor microscopic investigation. The biopsy material for microscopic investigationwas embedded in paraffin, cut into 4-�m sagittal sections, and mounted on glassslides. Prior to microscopic investigation, the paraffin was removed from thetissue sections by immersing the glass slides twice in xylene (total, 10 min), twice
in 99.9% ethanol (total, 6 min), twice in 96% ethanol (total, 6 min), and threetimes in distilled sterile water (total, 9 min).
PNA-FISH and conventional tissue staining. The deparaffinized tissue sec-tions were analyzed by means of conventional hematoxylin and eosin (H&E)staining and FISH with PNA probes. The PNA probe in hybridization solution(AdvanDx, Inc., Woburn, MA) was added dropwise to each tissue section, whichwas then covered with a coverslip and hybridized in a PNA-FISH workstation(AdvanDx, Inc.), which was covered with a lid, at 55°C for 90 min. Three separatePNA probe solutions were used: (i) a Texas Red (TxR)-labeled P. aeruginosa-specific probe, (ii) a fluorescein isothiocyanate (FITC)-labeled S. aureus-specificprobe, and (iii) a mixture of the TxR-labeled P. aeruginosa-specific probe and theFITC-labeled S. aureus-specific probe. The slides with tissue sections werewashed in a wash solution (AdvanDx, Inc.) at 55°C for 30 min, air dried, mountedwith Vectashield mounting medium with 4�,6�-diamidino-2-phenylindole (DAPI;Vector Laboratories), and covered with a coverslip. The tissue sections wereexamined as described below.
Image acquisition and analysis. Microscopic observations of the tissue sec-tions were performed with an epifluorescence microscope (Olympus, Hamburg,Germany) or a TCS-SP5 confocal laser scanning microscope (Leica Microsys-tems, Mannheim, Germany) equipped with an argon laser and a helium-neonlaser for excitation of the fluorophores. Multichannel simulated fluorescenceprojection images were generated by using the IMARIS software package (Bit-plane AG, Zurich, Switzerland) and were further processed for display by usingPhotoShop software (Adobe). Subtraction of the background from the imageswas performed with the IMARIS software package to remove the host tissueautofluorescence. The images were converted to eight-bit gray-scale images byusing ImageJ (version 1.41o) software (http://rsb.info.nih.gov/ij/index.html), andthe moment calculator tool (http://rsb.info.nih.gov/ij/plugins/moments.html) ofthe same software was used to locate the center of mass of the bacterial popu-lation displayed on the images.
Statistical evaluation. To evaluate whether the data obtained from the dis-tance measurement of P. aeruginosa and S. aureus aggregates to the woundsurface were statistically significant, an unpaired t test was performed. P valuesof �0.05 were considered significant. The statistical program Stat-View (SASInstitute Inc., Cary, NC) was used to calculate P values.
RESULTS
Initially, we identified P. aeruginosa and S. aureus in biopsymaterial from chronic wounds by the use of PNA-FISH withspecies-specific probes. On the basis of those identities, weselected five wounds containing P. aeruginosa and five woundscontaining S. aureus (one of the wounds selected containedboth P. aeruginosa and S. aureus). In order to study the spatialdistribution of P. aeruginosa and S. aureus in these wounds, wecut five sections sagittally at 50-�m intervals from each woundbiopsy specimen and performed PNA-FISH of these sectionswith P. aeruginosa- and S. aureus-specific probes. We subse-quently acquired images at three different regions on eachsection by confocal laser scanning microscopy (CLSM). Thebacteria were predominantly present as large aggregates. Toget a measure of the distance of the bacteria to the woundsurface, we located the center of mass of the bacterial aggre-gates identified on each image by using the moment calculatortool of ImageJ software and measured its distance to thewound surface. This analysis showed that the S. aureus aggre-gates were located close to the wound surface, whereas the P.aeruginosa aggregates were located deeper in the wound bed(P � 0.0001) (Table 1). The centers of mass of the S. aureusaggregates were primarily located at a distance of 20 to 30 �mto the wound surface, whereas the centers of mass of the P.aeruginosa aggregates were primarily located at a distance of50 to 60 �m to the wound surface (Fig. 2). Figure 3 showsrepresentative CLSM images of the locations of P. aeruginosaand S. aureus in the chronic wounds. The range of the distri-bution of P. aeruginosa and S. aureus was limited, so colocal-
FIG. 1. Sampling region on a chronic venous leg ulcer. Biopsyspecimens were taken from a central region within the wounds. Thearrows point to a representative sampling region.
VOL. 47, 2009 P. AERUGINOSA AND S. AUREUS IN CHRONIC WOUNDS 4085
ization of the two bacterial species was rare. In order to visu-alize host cells and bacteria in the wound biopsy specimens, weperformed combined PNA-FISH and DAPI staining as well asH&E staining of the biopsy specimens from wounds containingP. aeruginosa or S. aureus. As shown in Fig. 4, the analyzedsections from wounds with P. aeruginosa had a higher numberof PMNs than the sections from wounds with S. aureus, sug-gesting that wounds infected with P. aeruginosa may have ahigher degree of inflammation than wounds infected with S.aureus.
DISCUSSION
Although the microflora of chronic wounds is polymicrobialand heterogeneous, S. aureus and P. aeruginosa are among thebacteria that are most frequently isolated from these wounds(8, 12, 15). In the present study, we characterized the distribu-tion of P. aeruginosa and S. aureus in nine chronic wounds: fourwounds with S. aureus, four wounds with P. aeruginosa, and onewound with both S. aureus and P. aeruginosa. Analysis of theimages obtained by PNA-FISH and CLSM indicated that P.aeruginosa was located significantly deeper in the wound bedthan S. aureus.
We previously investigated samples from 22 chronic venousleg ulcers for the presence of bacteria by standard culturingand PNA-FISH (17). By swab sample cultivation, we foundthat 12 of the wounds were colonized with S. aureus, whereas5 of them were colonized with P. aeruginosa. Conversely, byusing PNA-FISH, P. aeruginosa was detected in nine of thewounds, while S. aureus was detected in only two of them. Ourpresent finding that P. aeruginosa is located in the deeperregions of the wound bed offers an explanation for the differentresults obtained by swab sample cultivation and PNA-FISH.Because the swab sample culture technique detects bacteria inthe upper region of the wounds, bacteria that primarily colo-nize the deeper regions may not be detected.
For a good healing response, the bacterial load of chronicwounds needs to be optimally managed. Topical antimicrobials
can in some cases effectively control superficial bacterial bur-dens if the infection is localized but may not be appropriate forhighly infected wounds. Systemic antibiotics may be effective insome cases of severe infection with tissue invasion (23). Theuse of a nanocrystalline silver dressing was shown to decreasethe superficial bacterial burden, as assessed by surface swabinvestigation, but had no effect on the bacterial burden of thedeep wound compartment, as measured by tissue biopsy (24).Thus, it is of great importance to define the spatial organiza-tion of the bacterial species within a chronic wound for themost effective management of the infection. A relevant pictureof the spatial organization of the bacteria in a chronic woundmight be obtained by using molecular methods, such as dena-turing gradient gel electrophoresis (2, 8) and FISH (16), incombination with traditional culturing of swab as well as biopsysamples.
The biofilm mode of growth provides bacteria characteristicsdifferent from those of their planktonic counterparts, such asprotection against the activities of the host immune system andincreased tolerance to antimicrobial treatments (7). P. aerugi-nosa bacteria in biofilms express quorum-sensing-controlledvirulence factors that can kill or eliminate the activity of hostimmune cells. It has been shown that rhamnolipid, a leukocidaltoxin produced by P. aeruginosa, causes rapid necrosis ofPMNs in vitro (16). Bjarnsholt and colleagues proposed thatrhamnolipid offers a protective shield against the activities ofhost immune cells and demonstrated that aggregates of P.aeruginosa in chronic wounds were surrounded by host cells,possibly PMNs, but were not penetrated (5, 17), similar to whatwas observed in in vitro biofilms of P. aeruginosa overlaid withfreshly isolated PMNs (4). The bacteria in chronic wounds areexpected to compete with each other for the available nutri-ents. The ability of P. aeruginosa to migrate via type IV pili andflagellum-mediated motility in biofilms (3, 18, 19) and to pro-duce virulence factors that can eliminate the activity of hostdefense systems (4, 16) may explain the presence of thesebacteria in the deeper regions of chronic wounds. The destruc-tion of PMNs by virulence factors produced by P. aeruginosa
FIG. 2. Distribution of the distances from the wound surface to thecenter of mass of S. aureus aggregates (light gray shading) or P. aerugi-nosa aggregates (dark gray shading). The distances are average valuesobtained from the analysis of 15 images for each wound sample.
TABLE 1. Average distance of bacterial aggregates to the surfaceof wound samples
Wound biopsyspecimen
Bacterial species detectedby PNA-FISH
Avg distance to woundsurface (�m)b
LGA02 S. aureus 28.3 (6.6)BIJ04 S. aureus 8.8 (1.7)HAH08 S. aureus 28.1 (5.0)M2a S. aureus 26.1 (5.1)Pt17 S. aureus 23.7 (3.7)M3a P. aeruginosa 57.5 (9.4)Pt11 P. aeruginosa 50.0 (13.4)Pt20 P. aeruginosa 53.5 (9.9)Pt23B P. aeruginosa 68.7 (11.2)Pt31 P. aeruginosa 46.1 (6.0)
a Specimens M2 and M3 are biopsy specimens obtained from the same wound.Both S. aureus and P. aeruginosa were detected in these biopsy specimens. In thecase of specimen M2, the distance of S. aureus to the wound surface was ana-lyzed, whereas in the case of specimen M3, the distance of P. aeruginosa to thewound surface was analyzed.
b The center of mass of the bacterial aggregates on each image was located,and its distance to the wound surface was measured. The average distances of thecenter of mass to the wound surface were obtained from 15 images acquired foreach wound sample. The values in parentheses are standard deviations.
4086 FAZLI ET AL. J. CLIN. MICROBIOL.
FIG. 3. Representative CLSM images of S. aureus (A and B), P. aeruginosa (C and D), and both organisms (E) in chronic wounds. The bacteriawere detected by PNA-FISH with an FITC-labeled S. aureus-specific probe (green) or a TxR-labeled P. aeruginosa-specific probe (red), or amixture of the two probes. Arrows point to the wound surfaces. Bars, 30 �m.
4087
bacteria located in the deeper regions of chronic wounds maybe one of the factors that causes a persistent influx of PMNsand that keeps the wound in an inflammatory stage. However,more research is required before specific bacterial species inspecific modes of growth can be identified as the causativeagents in chronic wounds.
ACKNOWLEDGMENTS
We thank Jette Pedersen from the Bartholin Institute, University ofCopenhagen, and Anne Jørgensen from the Department of Pathology,University of Copenhagen, for help with preparing and microscopy ofthe tissue sections. AdvanDx, Inc., is gratefully acknowledged for pro-viding the PNA probes.
M.G. received financial support for the present study from theDanish Strategic Research Council. T.B. received financial supportfrom the Carlsberg Foundation and the Lundbeck Foundation).
REFERENCES
1. Agren, M. S., W. H. Eaglstein, M. W. J. Ferguson, K. G. Harding, K. Moore,U. K. Saarialo-Kere, and G. S. Schultz. 2000. Causes and effects of thechronic inflammation in venous leg ulcers. Acta Dermatol. Venerol. Suppl.210:3–17.
2. Andersen, A., K. E. Hill, P. Stephens, D. W. Thomas, B. Jorgensen, and K. A.Krogfelt. 2007. Bacterial profiling study using skin grafting, standard cultureand molecular bacteriological methods. J. Wound Care 16:171–175.
3. Barken, K. B., S. J. Pamp, L. Yang, M. Gjermansen, J. J. Bertrand, M.Klausen, M. Givskov, C. B. Whitchurch, J. N. Engel, and T. Tolker-Nielsen.2008. Roles of type IV pili, flagellum-mediated motility and extracellularDNA in the formation of mature structures in Pseudomonas aeruginosabiofilms. Environ. Microbiol. 10:2331–2343.
4. Bjarnsholt, T., P. Ø. Jensen, M. Burmølle, M. Hentzer, J. A. Haagensen,H. P. Hougen, H. Calum, K. G. Madsen, C. Moser, S. Molin, N. Høiby, andM. Givskov. 2005. Pseudomonas aeruginosa tolerance to tobramycin, hydro-gen peroxide and polymorphonuclear leukocytes is quorum-sensing depen-dent. Microbiology 151:373–383.
5. Bjarnsholt, T., K. Kirketerp-Møller, P. O. Jensen, K. G. Madsen, R. Phipps,K. Krogfelt, N. Hoiby, and M. Givskov. 2008. Why chronic wounds will notheal: a novel hypothesis. Wound Repair Regen. 16:2–10.
FIG. 4. Epifluorescence micrographs (A and B) and bright-field micrographs (C and D) showing red PNA-FISH-stained P. aeruginosa cells (A),green PNA-FISH-stained S. aureus cells (B), blue DAPI-stained host cells (A and B), H&E-stained host cells and P. aeruginosa cells (C), andH&E-stained host cells and S. aureus cells (D). Some of the bacteria and host cells are encircled and labeled b and h, respectively. Arrows pointto the wound surfaces. Bars, 35 �m.
4088 FAZLI ET AL. J. CLIN. MICROBIOL.
6. Costerton, J. W., Z. Lewandowski, D. E. Caldwell, D. R. Korber, and H. M.Lappin-Scott. 1995. Microbial biofilms. Annu. Rev. Microbiol. 49:711–745.
7. Costerton, J. W., P. S. Stewart, and E. P. Greenberg. 1999. Bacterial biofilms:a common cause of persistent infections. Science 284:1318–1322.
8. Davies, C. E., K. E. Hill, M. J. Wilson, P. Stephens, C. M. Hill, K. G.Harding, and D. W. Thomas. 2004. Use of 16S ribosomal DNA PCR anddenaturing gradient gel electrophoresis for analysis of the microfloras ofhealing and nonhealing chronic venous leg ulcers. J. Clin. Microbiol. 42:3549–3557.
9. Davis, S. C., C. Ricotti, A. Cazzaniga, E. Welsh, W. H. Eaglstein, and P. M.Mertz. 2008. Microscopic and physiologic evidence for biofilm-associatedwound colonization in vivo. Wound Repair Regen. 16:23–29.
10. Fedtke, I., F. Gotz, and A. Peschel. 2004. Bacterial evasion of innate hostdefences: the Staphylococcus aureus lesson. Int. J. Med. Microbiol. 294:189–194.
11. Foster, T. J. 2005. Immune evasion by staphylococci. Nat. Rev. Microbiol.3:948–958.
12. Gjødsbol, K., J. J. Christiensen, T. Karlsmark, B. Jørgensen, B. M. Klein,and K. A. Krogfelt. 2006. Multiple bacterial species reside in chronic wounds:a longitudinal study. Int. Wound J. 3:225–231.
13. Gottrup, F. 2004. A specialized wound healing center concept: importance ofa multidisciplinary department structure and surgical treatment facilities inthe treatment of chronic wounds. Am. J. Surg. 187:38S–43S.
14. Halbert, A. R., M. C. Stacey, J. B. Rohr, and A. Jopp-McKay. 1992. Theeffect of bacterial colonization on venous leg ulcer healing. Australas. J.Dermatol. 33:75–80.
15. James, G. A., E. Swogger, R. Wolcott, E. Pulcini, P. Secor, J. Sestrich, J. W.Costerton, and P. S. Stewart. 2008. Biofilms in chronic wounds. WoundRepair Regen. 16:37–44.
16. Jensen, P. Ø., T. Bjarnsholt, R. Phipps, T. B. Rasmussen, H. Calum, L.Christoffersen, C. Moser, P. Williams, T. Pressler, M. Givskov, and N.Høiby. 2007. Rapid necrotic killing of polymorphonuclear leukocytes is
caused by quorum sensing controlled production of rhamnolipid by Pseudo-monas aeruginosa. Microbiology 153:1329–1338.
17. Kirketerp-Møller, K., P. O. Jensen, M. Fazli, K. G. Madsen, J. Pedersen, C.Moser, T. Tolker-Nielsen, N. Hoiby, M. Givskov, and T. Bjarnsholt. 2008.Distribution, organization, and ecology of bacteria in chronic wounds.J. Clin. Microbiol. 46:2717–2722.
18. Klausen, M., A. Heydorn, P. Ragas, L. Lambertsen, A. Aaes-Jørgensen, S.Molin, and T. Tolker-Nielsen. 2003. Biofilm formation by Pseudomonasaeruginosa wild type, flagella and type IV pili mutants. Mol. Microbiol.48:1511–1524.
19. Klausen, M., A. Aaes-Jørgensen, S. Molin, and T. Tolker-Nielsen. 2003.Involvement of microbial migration in the development of complex multi-cellular structures in Pseudomonas aeruginosa biofilms. Mol. Microbiol. 50:61–68.
20. Madsen, S. M., H. Westh, L. Danielsen, and V. T. Rosdahl. 1996. Bacterialcolonization and healing of venous leg ulcers. APMIS 104:895–899.
21. Matsukawa, M., and E. P. Greenberg. 2004. Putative exopolysaccharidesynthesis genes influence Pseudomonas aeruginosa biofilm development. J.Bacteriol. 186:4449–4456.
22. Nagase, H., and J. F. Woessner, Jr. 1999. Matrix metalloproteinases. J. Biol.Chem. 274:21491–21494.
23. Schultz, G. S., R. G. Sibbald, V. Falanga, E. A. Ayello, C. Dowsett, K.Harding, M. Romanelli, M. C. Stacey, L. Teot, and W. Vanscheidt. 2003.Wound bed preparation: a systematic approach to wound management.Wound Repair Regen. 11:S1–28.
24. Sibbad, R. G., A. C. Browne, P. Coutts, and D. Queen. 2001. Screeningevaluation of an ionized nanocrystalline silver dressing in chronic woundcare. Ostomy Wound Manage. 47:38–43.
25. Whitchurch, C. B., T. Tolker-Nielsen, P. C. Ragas, and J. S. Mattick. 2002.Extracellular DNA required for bacterial biofilm formation. Science 295:1487.
26. Yager, D. R., and B. C. Nwomeh. 1999. The proteolytic environment ofchronic wounds. Wound Repair Regen. 7:433–441.
VOL. 47, 2009 P. AERUGINOSA AND S. AUREUS IN CHRONIC WOUNDS 4089
