SEARCH

SEARCH BY CITATION

Keywords:

  • biofilm;
  • type IV pili;
  • extracellular DNA;
  • Pseudomonas aeruginosa;
  • Staphylococcus aureus

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Biofilm infections may not simply be the result of colonization by one bacterium, but rather the consequence of pathogenic contributions from several bacteria. Interspecies interactions of different organisms in mixed-species biofilms remain largely unexplained, but knowledge of these is very important for understanding of biofilm physiology and the treatment of biofilm-related infectious diseases. Here, we have investigated interactions of two of the major bacterial species of cystic fibrosis lung microbial communities –Pseudomonas aeruginosa and Staphylococcus aureus– when grown in co-culture biofilms. By growing co-culture biofilms of S. aureus with P. aeruginosa mutants in a flow-chamber system and observing them using confocal laser scanning microscopy, we show that wild-type P. aeruginosa PAO1 facilitates S. aureus microcolony formation. In contrast, P. aeruginosa mucA and rpoN mutants do not facilitate S. aureus microcolony formation and tend to outcompete S. aureus in co-culture biofilms. Further investigations reveal that extracellular DNA (eDNA) plays an important role in S. aureus microcolony formation and that P. aeruginosa type IV pili are required for this process, probably through their ability to bind to eDNA. Furthermore, P. aeruginosa is able to protect S. aureus against Dictyostelium discoideum phagocytosis in co-culture biofilms.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Cystic fibrosis (CF) is the most common hereditary disease in Caucasian populations (Davis et al., 1996). The defective expression and function of the transmembrane conductance regulator of CF patients alters the viscosity of airway mucus and leads to colonization of the airway by pathogenic microorganisms since infancy. Microbial lung infection is the leading cause of morbidity and mortality in CF patients (Gibson et al., 2003; Harrison, 2007). Coinfections involving different bacteria are common in CF patients and different bacterial species interact both synergistically and antagonistically (Høiby, 1974; Rogers et al., 2004; Wahab et al., 2004; Harrison, 2007). Interactions among different bacterial species might determine CF morbidity and should therefore be investigated (Harrison, 2007).

Pseudomonas aeruginosa and Staphylococcus aureus are two of the major species that colonize CF airways (Harrison, 2007), and they are well known for their tolerance towards antibiotic treatment due to their abilities to form biofilms (Costerton et al., 1995; Stewart & Costerton, 2001; Götz, 2002). The biofilm mode of growth is proposed as the survival strategy of environmental bacteria under antibiotic treatment and immune response in the lungs of the CF patients (Costerton, 2001; Høiby, 2002). Multiple factors such as surface appendages, quorum sensing, motility and extracellular polymer substance (EPS) components [e.g. extracellular DNA (eDNA) and polysaccharides] were reported to be required for biofilm development by different bacteria (Götz, 2002; Rice et al., 2007; Barken et al., 2008). However, it is unclear how these factors contribute to mixed-species biofilm development.

Previous studies provide evidence that genetic adaptation plays an essential role in P. aeruginosa colonization of the airways of CF patients (Smith et al., 2006; Huse et al., 2010; Rau et al., 2010). Mutations in regulator genes such as lasR, mucA and rpoN have huge impacts on P. aeruginosa phenotypes, which include factors involved in biofilm formation (Totten et al., 1990; Davies et al., 1998; Hentzer et al., 2001). Thus, these adaptive mutations might affect the community dynamics and interactions among different bacterial species of the CF respiratory tract.

In this study, we used confocal laser scanning microscopy (CLSM) to investigate co-culture biofilms of P. aeruginosa and S. aureus grown in a flow-chamber system. We demonstrated how adaptive mutations in regulator genes of P. aeruginosa affect interactions between P. aeruginosa and S. aureus in co-culture biofilms.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Bacteria and growth conditions

Pseudomonas aeruginosa wild-type PAO1 (Holloway & Morgan, 1986), P. aeruginosa mucA mutant (Hentzer et al., 2001), P. aeruginosa rpoN mutant (Webb et al., 2003), P. aeruginosa pilA mutant (Klausen et al., 2003b), P. aeruginosa pilH mutant (Barken et al., 2008), P. aeruginosa pqsA mutant (D'Argenio et al., 2002), S. aureus MN8 (Yarwood et al., 2004), S. aureus ISP479 (Toledo-Arana et al., 2005) and S. aureus 15981 (Toledo-Arana et al., 2005) were kindly provided by the cited authors and used in the present study. The pDA2 plasmid (An et al., 2006) was used to complement the pilA mutant. Fluorescence-tagged strains were constructed by the insertion of a mini-Tn7-eGFP-Gmr cassette as described (Koch et al., 2001; Klausen et al., 2003b). Escherichia coli strains MT102 and DH5α were used for standard DNA manipulations. Luria–Bertani medium (Bertani, 1951) was used to cultivate E. coli strains. A modified FAB medium (Qin et al., 2007) supplemented with 0.3 mM glucose and 3% of Tryptic Soy Broth (TSB, BD Diagnostics) was used for biofilm cultivation. Selective media were supplemented with ampicillin (100 mg L−1), gentamicin (60 mg L−1) or carbenicillin (200 mg L−1).

Cultivation of biofilms

Biofilms were grown in flow chambers with individual channel dimensions of 1 × 4 × 40 mm at 37 °C. The flow system was assembled and prepared as described previously (Sternberg & Tolker-Nielsen, 2006). Overnight cultures of P. aeruginosa and S. aureus were diluted to an OD600 nm of 0.001. The flow chambers were inoculated by injecting 350 μL of monospecies diluted cultures or P. aeruginosaS. aureus 1 : 1 mixed-species diluted cultures into each flow channel with a small syringe. After inoculation, flow channels were left without flow for 1 h, after which medium flow (0.2 mm s−1) was started using a Watson Marlow 205S peristaltic pump. For DNase I treatment, biofilm medium was supplemented with 20 μg mL−1 bovine DNase I (Sigma) from the beginning of cultivation.

Microscopy and image acquisition

All microscopic observations and image acquisitions were performed using a Zeiss LSM 510 confocal laser scanning microscope (Carl Zeiss, Jena, Germany) equipped with detectors and filter sets for monitoring of green and red fluorescence from general nucleic acid staining SYTO 9 (Invitrogen) and gram-positive specific staining hexidium iodide (Invitrogen) (Mason et al., 1998), respectively. BacLite Live/Dead viability stain (Molecular Probes, Eugene, OR) was used to visualize dead and live cells in co-culture biofilms. Images were obtained using a × 40/1.3 objective. Simulated three-dimensional images and sections were generated using the imaris software package (Bitplane AG, Zürich, Switzerland). For the assessment of statistical significance, five images from different place of biofilms were captured for each case and CLSM images were further analyzed using the computer program comstat (Heydorn et al., 2000). A fixed threshold value and connected volume filtration were used for all image stacks.

Cultivation of Dictyostelium discoideum

Dictyostelium discoideum DH1-10 cells (Cornillon et al., 2000) were grown in sterilized Petri dishes containing 12 mL HL5 medium (Fey et al., 2007) and transferred to a new dish containing 12 mL HL5 medium twice a week. The D. discoideum cell density was determined by counting the cells under microscopy.

Dictyostelium discoideum phagocytosis assay

To assay the protection effects of P. aeruginosa on S. aureus in co-culture biofilms, we challenged the 2-day-old mature monospecies biofilms formed by the P. aeruginosa PAO1 strain, the rpoN mutant, the S. aureus MN8 strain and the co-culture biofilm formed by P. aeruginosa PAO1–S. aureus MN8 with D. discoideum. Briefly, the flows of 2-day-old biofilms were stopped and 200 μL of 2 × 106 cells mL−1D. discoideum were injected into flow chambers. The flow chambers were left without flow for 30 min, after which medium flow was started again. The growth of biofilms and D. discoideum were observed after 2 h of D. discoideum inoculation at room temperature (25 °C).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Biofilm formation by P. aeruginosa and S. aureus in 2-day-old monospecies biofilms

We first investigated monospecies biofilms formed by the wild-type P. aeruginosa PAO1 strain, mucA mutant, rpoN mutant and three widely used and well-characterized S. aureus strains MN8, ISP479 and 15981. With the TSB-supplemented medium, the PAO1 strain formed flat, tightly packed biofilms with little heterogeneity (Fig. 1a), while the mucA mutant formed biofilms with mushroom-shaped microcolony structures (Fig. 1b) in accordance with previous studies (Hentzer et al., 2001). The rpoN mutant formed biofilms with loosely packed irregular microcolony structures (Fig. 1c). All the tested S. aureus strains formed biofilms consisting of loosely packed microcolony structures with a relatively smaller surface coverage on the glass substratum than biofilms formed by the P. aeruginosa strains (Fig. 1d–f).

image

Figure 1.  Two-day-old biofilms of Pseudomonas aeruginosa PAO1 (a), mucA mutant (b), rpoN mutant (c), Staphylococcus aureus MN8 (d), ISP479 (e) and 15981 (f). The P. aeruginosa was visualized by staining with SYTO 9 and the S. aureus was visualized by staining with hexidium iodide. The central pictures show horizontal CLSM optical sections and the flanking pictures show vertical CLSM optical sections. Scale bars=20 μm.

Download figure to PowerPoint

Biofilm formation by wild-type P. aeruginosa and S. aureus in co-culture biofilms

We next studied co-culture biofilms formed by P. aeruginosa PAO1 and S. aureus MN8, ISP479 and 15981, respectively. In the co-culture biofilms, PAO1 eventually covered the S. aureus strains, and together, they formed biofilms with firmly packed microcolony structures (Fig. 2, first row). comstat analysis showed that during mixed-species biofilm formation, PAO1 was more abundant than the S. aureus strains. The ratios of total biomass of PAO1 to MN8, ISP479 and 15981 were 2.42 (± 0.45) : 1, 2.65 (± 0.42) : 1 and 2.85 (± 0.35) : 1, respectively. To investigate the composition of the firmly packed microcolonies in co-culture biofilms, we used green fluorescent protein (GFP)-tagged P. aeruginosa PAO1 to grow co-culture biofilms with S. aureus instead of using SYTO 9, which could stain both species. We observed that both P. aeruginosa and S. aureus exist in the firmly packed microcolonies of co-culture biofilms (Supporting Information, Fig. S1).

image

Figure 2.  Two-day-old biofilms of Pseudomonas aeruginosa PAO1–Staphylococcus aureus MN8 co-culture (first row), P. aeruginosa mucAS. aureus MN8 co-culture (second row) and P. aeruginosa rpoNS. aureus MN8 co-culture (third row). The P. aeruginosa was visualized by staining with SYTO 9 (first column, green) and the S. aureus was visualized by staining with hexidium iodide (HI) (second column, red). The third column shows the combined images of the first and second columns. To avoid redundancy, the images of the P. aeruginosaS. aureus ISP479 and P. aeruginosaS. aureus 15981 co-cultures are not shown. The central pictures show horizontal CLSM optical sections and the flanking pictures show vertical CLSM optical sections. Scale bars=20 μm.

Download figure to PowerPoint

Biofilm formation by P. aeruginosa mucA mutant and S. aureus in co-culture biofilms

In co-culture biofilms formed by the mucoid P. aeruginosa mucA mutant with S. aureus MN8, ISP479 and 15981 strains, respectively, P. aeruginosa mucA only weakly associated with S. aureus (Fig. 2, second row). Small S. aureus microcolonies were found on the substratum of the flow chambers. comstat analysis showed that during the mixed-species biofilm formation, the mucA mutant was much more abundant than the S. aureus strains. The ratios of the total biomass of the mucA mutant to MN8, ISP479 and 15981 were 5.58 (± 0.99) : 1, 5.82 (± 2.16) : 1 and 5.72 (± 1.48) : 1, respectively. The mucA biofilms were highly similar with or without co-cultivation with S. aureus.

Biofilm formation by the P. aeruginosa rpoN mutant and S. aureus in co-culture biofilms

We further studied co-culture biofilms formed by the P. aeruginosa rpoN mutant with S. aureus MN8, ISP479 and 15981, respectively. In co-culture biofilms, the P. aeruginosa rpoN mutant weakly associated with S. aureus and formed biofilms with loosely packed microcolony structures (Fig. 2, third row). There was very little S. aureus biomass embedded inside the microcolonies of rpoN mutant, and it seemed that S. aureus could not even colonize the substratum where no P. aeruginosa biofilm was located (Fig. 2, third row). These results indicate that the P. aeruginosa rpoN mutant lacks components mediating S. aureus microcolony formation. comstat analysis showed that during the mixed-species biofilm formation, the rpoN mutant was much more abundant than the S. aureus strains. The ratios of the total biomass of the rpoN mutant to MN8, ISP479 and 15981 were 100.29 (± 17.07) : 1, 95.86 (± 8.57) : 1 and 98.1 (± 14.1) : 1, respectively.

Pseudomonas aeruginosa type IV pili are involved in microcolony formation of co-culture biofilms

The P. aeruginosa rpoN mutant is defective in the formation of flagellin and pilin (Ishimoto & Lory, 1989; Totten et al., 1990), which are the essential components for the synthesis of flagellum and type IV pilus, respectively. The P. aeruginosa cell surface appendages flagella and pili and their mediated motilities were shown to be important factors for biofilm structure development (Klausen et al., 2003a, b; Barken et al., 2008). Moreover, the rpoN monospecies biofilm structures are similar to biofilm structures formed by the pilA mutant from our previous studies (Klausen et al., 2003b). We therefore examined the effects of P. aeruginosa type IV pili on microcolony formation in P. aeruginosaS. aureus co-culture biofilms. Because we observed that there was no significant difference among the three tested S. aureus strains in both monospecies and mixed-species biofilms, we chose the MN8 strain for the subsequent biofilm studies. The P. aeruginosa pilA mutant, which is unable to produce type IV pili, was found to be unable to associate with S. aureus MN8 to form microcolonies in co-culture biofilms and tended to outcompete S. aureus MN8 (Fig. 3a). The ability of the P. aeruginosa pilA mutant to associate with S. aureus MN8 and form mixed-species microcolonies in co-culture biofilms could be restored by complementation in trans with the pilA gene on the pDA2 plasmid (Fig. 3b).

image

Figure 3.  Two-day-old biofilms of Pseudomonas aeruginosa pilA–Staphylococcus aureus MN8 co-culture (a), P. aeruginosa pilA(pDA2)–S. aureus MN8 co-culture (b), and P. aeruginosa pilH–S. aureus MN8 co-culture (c). The P. aeruginosa was visualized by staining with SYTO 9 and the S. aureus was visualized by staining with hexidium iodide. The central pictures show horizontal CLSM optical sections and the flanking pictures show vertical CLSM optical sections. Scale bars=20μm.

Download figure to PowerPoint

To further examine the role of P. aeruginosa type IV pili in the interactions of P. aeruginosaS. aureus co-culture biofilms, we used the P. aeruginosa pilH mutant in our study. The P. aeruginosa pilH in-frame deletion mutant showed an increased level of surface piliation and slightly reduced twitching zones in an agar stab plate assay (Barken et al., 2008). In co-culture biofilms, the size of the P. aeruginosa pilHS. aureus MN8 mixed-species microcolonies was increased compared with the size of the P. aeruginosa PAO1–S. aureus MN8 mixed-species microcolonies (Fig. 3c). These results suggest that the level of P. aeruginosa surface piliation has an important impact on microcolony formation in the P. aeruginosaS. aureus co-culture biofilms.

eDNA is involved in mixed-species microcolony formation of co-culture biofilms

Previous reports have shown that P. aeruginosa type IV pili are able to bind DNA, which is a key component of the biofilm EPS (Whitchurch et al., 2002; van Schaik et al., 2005). We stained the P. aeruginosaS. aureus co-culture biofilms with Live/Dead viability stain and observed populations of dead cells accumulated inside the mixed-species microcolonies in the P. aeruginosa PAO1–S. aureus MN8 biofilm (Fig. 4a and b). We observed the same pattern of localization of dead cells in the P. aeruginosa pqsAS. aureus MN8 co-culture biofilms (Fig. 4c and d). These results indicate that S. aureus dead cells might be a major source of eDNA of co-culture biofilms, because the pqs gene operon was shown to be required for eDNA release of P. aeruginosa biofilms (Allesen-Holm et al., 2006; Yang et al., 2007). We then grew co-culture biofilms of P. aeruginosa PAO1 and an S. aureus atl mutant (Toledo-Arana et al., 2005) defective in producing a major autolysin of S. aureus. We observed the same pattern of mixed-species microcolony formation in P. aeruginosa PAO1–S. aureus atl co-culture biofilms as in the other P. aeruginosa PAO1–S. aureus co-culture biofilms (Fig. S2). This indicated that the dead cells we observed from the mixed-species microcolony structures of co-culture biofilms were not due to the activity of atl autolysin of S. aureus.

image

Figure 4.  Two-day-old biofilms of Pseudomonas aeruginosa PAO1–Staphylococcus aureus MN8 co-culture (first row) and P. aeruginosa pqsA–S. aureus MN8 co-culture (second row). The mixed-species biofilms were stained with Live/Dead viability stain. (a, c) Live channel and (b, d) dead channel. The central pictures show horizontal CLSM optical sections and the flanking pictures show vertical CLSM optical sections. Scale bars=20μm.

Download figure to PowerPoint

To test the hypothesis that eDNA is involved in the type IV pili-mediated interactions in P. aeruginosaS. aureus co-culture biofilms, we challenged the P. aeruginosaS. aureus co-culture biofilms with low concentrations of bovine DNase I. When DNase was added to the medium, the P. aeruginosa PAO1–S. aureus MN8 co-culture biofilms showed a significant reduction in the biomass and sizes of mixed-species microcolonies (Fig. 5). Only very small and thin microcolonies were formed in P. aeruginosa PAO1–S. aureus MN8 co-culture biofilms in the presence of DNase in the biofilm medium (Fig. 5). These results suggest that type IV pili–eDNA interactions might be involved in mixed-species microcolony formation of P. aeruginosaS. aureus co-culture biofilms.

image

Figure 5.  Two-day-old biofilms of Pseudomonas aeruginosa PAO1–Staphylococcus aureus MN8 co-culture grew in bovine DNase I-containing medium. The P. aeruginosa was visualized by staining with SYTO 9 (green) and the S. aureus MN8 was visualized by staining with hexidium iodide (red). (a) Pseudomonas aeruginosa image; (b) S. aureus image; (c) combined image. The central pictures show horizontal CLSM optical sections and the flanking pictures show vertical CLSM optical sections. Scale bars=20 μm.

Download figure to PowerPoint

Pseudomonas aeruginosa protects S. aureus against D. discoideum phagocytosis in co-culture biofilms

We used a D. discoideum phagocytosis model to investigate phagocytosis resistance of the monospecies biofilm and co-culture biofilms. Monospecies biofilms formed by P. aeruginosa PAO1, rpoN, S. aureus MN8 and P. aeruginosaS. aureus co-culture biofilm were inoculated with D. discoideum. We found that monospecies biofilm formed by the P. aeruginosa PAO1 strain was more resistant to D. discoideum phagocytosis than monospecies biofilms formed by P. aeruginosa rpoN and S. aureus MN8 (Fig. 6). In the P. aeruginosa PAO1–S. aureus MN8 co-culture biofilm, S. aureus was protected by P. aeruginosa from D. discoideum phagocytosis due to the formation of mixed-species microcolonies (Fig. 6).

image

Figure 6.  Phase contrast images of 2-day-old biofilms of Pseudomonas aeruginosa PAO1 (a), Staphylococcus aureus MN8 (b), P. aeruginosa PAO1–S. aureus MN8 co-culture (c) and P. aeruginosa rpoN mutant (d) after a 2-h Dictyostelium discoideum treatment. Scale bars=20 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Interspecies interactions of different organisms in mixed-species biofilms remain largely unexplained, but knowledge of these is very important for the understanding of biofilm physiology and treatments of biofilm-related infectious diseases. In this study, we have examined the interactions between two of the major CF pathogens, P. aeruginosa and S. aureus, in co-culture biofilms.

We first examined the interactions between P. aeruginosa wild-type PAO1, a mucA mutant and an rpoN mutant and different S. aureus strains in co-culture biofilms. Different patterns were observed in co-culture biofilms: P. aeruginosa wild-type PAO1 facilitated S. aureus microcolony formation (Fig. 2, first row); the P. aeruginosa mucA mutant formed mushroom-like microcolonies without affecting the S. aureus biofilm formation (Fig. 2, second row); and the P. aeruginosa rpoN mutant formed loosely packed microcolony structures and did not facilitate S. aureus microcolony formation (Fig. 2, third row). Further studies of P. aeruginosa genes that are regulated by RpoN led to the identification of the roles of P. aeruginosa type IV pili and eDNA in co-culture biofilms.

Our study has shown that P. aeruginosa type IV pili are required for microcolony formation in P. aeruginosaS. aureus co-culture biofilms (Fig. 3). Our P. aeruginosaS. aureus mixed-species biofilm results showed some common features with a previous study about the interspecies biofilms formed by P. aeruginosa and Agrobacterium tumefaciens reported by An et al. (2006). In the P. aeruginosaA. tumefaciens co-culture biofilms, the P. aeruginosa type IV pili also mediated interactions between P. aeruginosa and A. tumefaciens that lead to the formation of large microcolonies (An et al., 2006). We also tested co-culture biofilms of P. aeruginosaStaphylococcus epidermidis and observed similar mixed-species microcolony formation in co-culture biofilms as in the P. aeruginosaS. aureus co-culture biofilms (data not shown). The formation of the firmly packed eDNA-containing microcolonies in the co-culture biofilms may impact on the antibiotic tolerance of the bacterial cells embedded inside the microcolonies (Stewart et al., 2000, 2001; Walters et al., 2003).

In many bacteria, eDNA was shown to contribute to the establishment of in vitro biofilms (Whitchurch et al., 2002; Steinberger & Holden, 2005; Allesen-Holm et al., 2006; Qin et al., 2007; Rice et al., 2007). Most of these studies focused on localizing the eDNA in monospecies biofilms with the help of CLSM or scanning electronic microscopy and identifying the genes involved in DNA release. Here, we have studied the role of eDNA in mixed-species microcolony formation in co-culture biofilms. Our study emphasizes the importance of eDNA as a common biofilm EPS component.

In summary, we have shown that eDNA behaves as an essential EPS material shared by different species in co-culture biofilms, which facilitates interspecies interactions through the formation of mixed-species compact microcolony structures during biofilm development. Further understanding of mixed-species biofilm formation may provide valuable information for the diagnostics and therapeutics of biofilm-related problems in medical and industrial environments.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This work was supported by a grant from the Danish Research Council for Independent Research to L.Y. We would like to thank Dr Matthew Parsek (University of Washington at Seattle) for kindly providing us with the pDA2 plasmid.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Two-day-old biofilms of P. aeruginosa PAO1–Staphylococcus aureus MN8 co-culture.

Fig. S2. Two-day-old biofilms of P. aeruginosa PAO1–Staphylococcus aureus atl co-culture.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
FIM_820_sm_fig-s1.tif1875KSupporting info item
FIM_820_sm_fig-s2.tif1875KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.