BIOFILM PRODUCTION BY CLINICAL STAPHYLOCOCCUS AUREUS AND ITS INHIBITION BY HYPERICUM METABOLITES

Evolution has allowed bacteria to develop sophisticated methods of survival. One of these methods is biofilm production. Biofilms can be best described as complex bacterial communities embedded in a self-producing slime. Once bacteria form these biofilm communities, they become very difficult to treat with antibiotics. Along with biofilm production, another rising concern is antibiotic resistance. Bacterial resistance to many of our current antibiotics is sharply increasing, thereby creating a critical need to develop novel antimicrobial drugs. The most successful strategy for the discovery of new antibacterial agents has been the study of molecules from nature. Most of our current clinical antibiotics derive from metabolites produced by bacteria. Antimicrobial compounds of plant origin also have enormous therapeutic potential. Previous studies have demonstrated that certain plant metabolites have potent inhibitory effects on the growth of pathogenic bacteria. Not only might metabolites from plants help mitigate infectious diseases, but they may also lack adverse side effects often associated with existing antimicrobial agents, including hypersensitivity, allergic reaction, and immunosuppression. Therefore, future efforts to discover new antimicrobial drugs should also include the evaluation of new natural products from both microbes and plants. In this thesis, Chapter 1 describes an investigation to classify and quantify biofilm production in unique clinical strains of methicillin-resistant Staphylococcus aureus (MRSA). Staphylococcus aureus bacteria are responsible for causing a wide range of diseases and many are capable of biofilm production. To date, the largest focus of biofilm research has been on S. epidermidis and Pseudomonas aeruginosa. Few investigators have addressed the basic and central question: “What percentage of our clinical S. aureus bacteria produce biofilms and from what patient source are they most likely identified?” Two hundred and nineteen (n = 219) clinical methicillinresistant S. aureus (MRSA) isolates obtained from patients at the Veterans Affairs Medical Center (VAMC) in Providence, Rhode Island were evaluated. I evaluated biofilm formation using a modified microtiter-plate assay, and used a statistical approach to quantifying biofilm production. The results indicate that biofilm production is most frequently encountered in clinical MRSA from catheter sources. The surface area of these catheters may provide the ideal conditions for biofilm growth, especially in urine. Interestingly, a lesser incidence of biofilm production was observed by MRSA isolates obtained in the nares. Chapter 2 describes a phytochemical investigation of plant metabolites from Hypericum species that inhibit bacterial growth as well as biofilm production in Grampositive bacteria. For this study, seven acylphloroglucinol metabolites were obtained from Dr. Geneive Henry at Susquehanna University. I tested each compound in assays that measure both bacterial growth and biofilm production. The results showed that not only do some of these metabolites inhibit bacterial growth, but they also inhibit biofilm growth at sub-MIC concentrations. Important findings from this investigation included new structure-activity relationships demonstrating the importance of certain functional groups to the antibacterial nature of these metabolites. These results suggest that Hypericum spp. deserve further attention as a source of new antimicrobial agents.

to many of our current antibiotics is sharply increasing, thereby creating a critical need to develop novel antimicrobial drugs.
The most successful strategy for the discovery of new antibacterial agents has been the study of molecules from nature. Most of our current clinical antibiotics derive from metabolites produced by bacteria. Antimicrobial compounds of plant origin also have enormous therapeutic potential. Previous studies have demonstrated that certain plant metabolites have potent inhibitory effects on the growth of pathogenic bacteria. Not only might metabolites from plants help mitigate infectious diseases, but they may also lack adverse side effects often associated with existing antimicrobial agents, including hypersensitivity, allergic reaction, and immunosuppression. Therefore, future efforts to discover new antimicrobial drugs should also include the evaluation of new natural products from both microbes and plants.
In this thesis, Chapter 1 describes an investigation to classify and quantify biofilm production in unique clinical strains of methicillin-resistant Staphylococcus aureus (MRSA). Staphylococcus aureus bacteria are responsible for causing a wide range of diseases and many are capable of biofilm production. To date, the largest focus of biofilm research has been on S. epidermidis and Pseudomonas aeruginosa. Few investigators have addressed the basic and central question: "What percentage of our clinical S. aureus bacteria produce biofilms and from what patient source are they most likely identified?" Two hundred and nineteen (n = 219) clinical methicillinresistant S. aureus (MRSA) isolates obtained from patients at the Veterans Affairs Medical Center (VAMC) in Providence, Rhode Island were evaluated. I evaluated biofilm formation using a modified microtiter-plate assay, and used a statistical approach to quantifying biofilm production. The results indicate that biofilm production is most frequently encountered in clinical MRSA from catheter sources.
The surface area of these catheters may provide the ideal conditions for biofilm growth, especially in urine. Interestingly, a lesser incidence of biofilm production was observed by MRSA isolates obtained in the nares.

Chapter 2 describes a phytochemical investigation of plant metabolites from
Hypericum species that inhibit bacterial growth as well as biofilm production in Grampositive bacteria. For this study, seven acylphloroglucinol metabolites were obtained from Dr. Geneive Henry at Susquehanna University. I tested each compound in assays that measure both bacterial growth and biofilm production. The results showed that not only do some of these metabolites inhibit bacterial growth, but they also inhibit biofilm growth at sub-MIC concentrations. Important findings from this investigation included new structure-activity relationships demonstrating the importance of certain functional groups to the antibacterial nature of these metabolites.   allowing them to better persist in the normally hostile environment of tissue and blood. Additionally, the physical barrier imposed by biofilms impedes antibiotic effectiveness and host defenses (Ceri, et al., 1999;Donlan, 2000;Oie, Huang, Kamiya, Konishi, & Nakazawa, 1996).

Inhibition of Bacterial Growth and Biofilm Production by Metabolites from
It is well known that Staphylococcus epidermidis produce biofilm, and it is estimated that greater than 60% of these isolates are capable of this production (Arciola, et al., 2001;Arciola, et al., 2002;Christensen, Simpson, Bisno, & Beachey, 1982;Cramton, Gerke, & Gotz, 2001;Cramton, Ulrich, Gotz, & Doring, 2001;de Araujo, et al., 2006;Donlan, et al., 2001). However, much less is known about biofilm producing abilities in MRSA, particularly with clinical isolates. Therefore, it was the intent of this project to 1) measure biofilm formation by clinical strains of MRSA obtained from patients at the Veteran Affairs Medical Center (VAMC) in Providence, Rhode Island, 2) assess the frequency of biofilm production in clinical MRSA isolates, and 3) classify the relationship of biofilm producing MRSA to the source of patient obtainment (blood, nares, tissue, urine, and catheters). epidermidis RP62A served as the non-biofilm control.

Biofilm production in methicillin-resistant
Biofilm formation was evaluated and quantified using a modified microtiterplate assay (Christensen, et al., 1985;Stepanovic, Vukovic, Dakic, Savic, & Svabic-Vlahovic, 2000) and growth conditions were optimized for biofilm production in staphylococci (Cramton, Gerke, et al., 2001;Cramton, Ulrich, et al., 2001;LaPlante & Woodmansee, 2009). Based on optical density measurements (OD 610 ), bacterial films were classified into the following categories: biofilm production or no biofilm production using the method established by Stepanovic et al. (Stepanovic, et al., 2000). Statistical Analysis Software (SAS) (Version 9.1) was used for all analyses, and a P-value of ≤ 0.05 was considered statistically significant. Overall, 30.6% (67/219) of the MRSA isolates were determined to be biofilm producers. Biofilm production was most common among isolates obtained from catheter sources (52.3%, P=0.0005). Of interest, isolates obtained from the nares were significantly less likely to produce biofilm (18.6% formed biofilm; P=0.0198). Interestingly, MRSA isolates obtained from patients' nares were significantly less likely to be biofilm formers compared to isolates obtained from the other patient sources. We are not aware of this finding previously discussed in the literature.

RESULTS
Another study found that 60% of methicillin-resistant Staphylococcus epidermidis (MRSE) isolates obtained from the nares were able to produce biofilm (de Araujo, et al., 2006).
In our facility, we found that cultures taken from catheter samples produced significantly more biofilm than other sources. These results are similar to those found by a Japanese study, which concluded that the biofilm-forming capacities of MRSA isolates from catheter-related cases were significantly greater than those from catheterunrelated cases (Ando, Monden, Mitsuhata, Kariyama, & Kumon, 2004). Their studies suggest that MRSA colonization and infection of the urinary tract may be promoted by hla, hlb, and fnbA gene products (Ando, et al., 2004). However, many genes can be responsible for biofilm growth in S. aureus (Houston, Rowe, Pozzi, Waters, & O'Gara, 2011;Kaito, et al., 2011). Our clinical MRSA isolates were not tested for biofilmproducing genes because of this lack of standardization.
Urinary surfaces provide attractive sites for bacterial colonization as well as antibacterial resistance due to the gentle flow of warm nutritious urine (Stickler, 2008). The capacity of a microorganism to establish and form a biofilm on a given surface depends of the nature of the surface in question (Donlan & Costerton, 2002).
When a device such as a urinary catheter is exposed to bodily fluids such as urine, various components adsorb onto the surface and form a conditioning film. This film essentially covers the surface and becomes the bona fide interface where microbial interaction takes place. In the case of biofilm formation the steps are bacterial attachment, microcolony formation and build up of biofilm. It follows that the nature of the material constituting the catheter determines the composition of the conditioning film, which in turn influences which microorganisms can attach (Ferrieres, Hancock, & Klemm, 2007).
In conclusion, based on the data from our study, a correlation may exist between biofilm production and catheters as the source. This may be explained by the

CHAPTER 2
The following manuscript has been formatted for submission to the journal Phytomedicine.

Inhibition of Bacterial Growth and Biofilm Production by Metabolites from
Hypericum spp. Bacterial resistance to many of our current antibiotics is sharply increasing, thereby creating a critical need to develop novel antimicrobial drugs (Spellberg et al., 2004). Antimicrobials of plant origin have enormous therapeutic potential. Not only could they help mitigate infectious diseases, but they may also lack adverse side effects often associated with existing antimicrobial agents, including hypersensitivity, allergic reaction, and immunosuppression (Iwu et al., 1999;Mukherjee et al., 2002).   Extraction and isolation procedures. Compounds 3, 4, and 5 were isolated from hexanes and acetone extracts of H. densiflorum as previously described (Henry et al., 2009). Compound 7 was isolated from the hexanes extract of H. prolificum as previously reported (Henry et al., 2006). Aerial parts of H. ellipticum were oven dried at 38 C and ground to a fine powder using a coffee grinder. The plant material (459 g) was extracted sequentially at room temperature using hexanes ( methanol-water (4:1, v/v), isocratic elution, flow rate 10 mL/min) to afford compound 6 (28.0 mg). The structure of compound 6 was established based on NMR spectroscopic and mass spectrometric data (Manning et al., 2011).
The results of these studies add to growing knowledge of the antimicrobial properties of acylphloroglucinol metabolites. Other studies have reported acylphloroglucinol derivatives possessing antimicrobial activity against staphylococci (Gibbons et al., 2005;Henry et al., 2009;Ishiguro et al., 1994;Pecchio et al., 2006;Schempp et al., 1999;Winkelmann et al., 2001;Winkelmann et al., 2003). A study by Socolsky et al. reported that acylphloroglucinols containing an additional chromene substituent possess antimicrobial activity against P. aeruginosa, indicating that certain acylphloroglucinols may possess useful antimicrobial effects against Gram-negative pathogens (Socolsky et al., 2010). While a detailed mechanism of action for the antimicrobial activities of acylphloroglucinols remains to be determined, Hubner and colleagues showed that reduced sensitivity of S. aureus to hyperforin did not lead to a cross resistance against clinically used antibiotics (Hubner, 2003), therefore suggesting a possible unique mechanism of action. Shiu et al. found that certain acylphloroglucinols retained activities against tetracyline, fluoroquinolone, and macrolide drug-resistant strains, thereby also suggesting alternate mechanisms of action (Shiu and Gibbons, 2006). These studies further highlight the promise of investigating acylphloroglucinol derivatives as novel agents to treat bacterial disease.
Several of the metabolites in this study have previously demonstrated biological activities in other medically relevant assays. Compounds 3, 4, 5 and 7 are reported to have antitumor activity with IC 50 values ranging from 4.1 to 36 µM (Henry et al., 2009;Henry et al., 2006). Additionally, compounds 3, 4, and 5 display antioxidant activity and cyclooxygenase (COX) inhibition (Henry et al., 2006).
The potent anti-staphylococcal activity of the Hypericum metabolites suggests that these deserve further consideration as novel antibacterial agents. Staphylococcus aureus is recognized as an important human pathogen able to adapt and evolve in terms of its resistance traits and virulence factors; it is among the most important causes of human infections in both the hospital and community settings (Tang and Stratton, 2010). Clinical antibiotics currently used to treat MRSA infections include daptomycin, linezolid, and vancomycin. It is interesting to note that these clinically used antibiotics have shown MIC values that are similar to our active Hypericum spp.
metabolites (Brunton et al., 2006). These bioactive metabolites further highlight the potential for finding structurally new antibacterial agents from plants, and add to our understanding of structure-activity relationships for phloroglucinols that limit growth and biofilm production by pathogenic bacteria.