Agr function is upregulated by photodynamic therapy for Staphylococcus aureus and is related to resistance to photodynamic therapy


  • Subject section and field; Bacteriology/Antibacterial agents and chemotherapy


Iel Soo Bang, Department of Microbiology and Immunology, Chosun University School of Dentistry, Gwangju 501-759, Korea.

Tel: +82 62 230 6872; fax: +82 62 232 6896; email:


Photodynamic therapy (PDT) has been considered a feasible alternative for antimicrobial therapy of multidrug-resistant pathogens. However, bacterial response mechanisms against PDT-generated photo-oxidative stress remain largely unknown. Herein, it is shown that the accessory gene regulator Agr is involved in Staphylococcus aureus response to photo-oxidative stress generated by laser-induced PDT with the photosensitizer chlorin e6. Transcriptional profiling revealed that sublethal PDT induces a general stress response and also activates Agr-dependent gene regulation. Moreover, mutant S. aureus lacking Agr function showed hypersusceptibility to two independent PDT conditions with higher energy densities, demonstrating Agr-dependent S. aureus resistance against PDT.

List of Abbreviations

chlorin e6


colony forming unit


clusters of orthologous genes


light-emitting diode


optical density


pheophorbide a


photodynamic therapy


reactive oxygen species




standard deviation


tryptic soy broth


wild type

Emergence of multidrug-resistant pathogens has rapidly increased, resulting in concurrent threats to public health. Many studies have shown that PDT can kill a variety of microorganisms, reviving a century-old technology as an alternative antimicrobial approach to controlling infectious diseases caused by multidrug-resistant pathogens [1-4]. In particular, many studies concerning the application of PDT have focused on biofilm-mediated conditions because most such infections are localized and therefore suitable for this form of therapy [1]. In PDT, excitation of nontoxic dyes (photosensitizers) by light sources transfers electrons to molecular oxygen, which generates ROS, including singlet oxygen and oxygen radicals, that can inactivate bacterial macromolecules [5, 6]. Because of its light-dependent production, singlet oxygen-mediated stress has been specifically termed photo-oxidative stress. Studies have focused extensively on bacterial protection from radical-type oxygen species such as superoxide and hydrogen peroxide. However, the mechanism by which bacteria respond to PDT-mediated stresses, including photo-oxidative stress, remains relatively unknown [5].

The purpose of this study was to gain insight into the mechanism of bacterial response to PDT by exploring the genetic response of S. aureus to PDT combined with the photosensitizer Ce6 and laser light, which a previous study demonstrated damages both gram-positive and gram-negative bacteria [7]. Furthermore, based on transcriptome analysis, major regulatory systems for gene regulation of S. aureus during PDT were explored, which led to the deduction of the hypothesis that the function of the accessory gene regulator Agr is necessary for protection of S. aureus from PDT-mediated photo-oxidative stress.


Bacterial strains and culture conditions

A methicillin-sensitive wild type S. aureus (NCTC8325-4) and its isogenic agr mutant strain (PC6911; Δagr::tet) were used in this study [8]. S. aureus cells were grown at 37°C with shaking in TSB (Difco, Franklin Lakes, NJ, USA) supplemented with antibiotic(s) when necessary.

Photodynamic therapy conditions

For analysis of the genetic response of S. aureus to PDT, a sublethal PDT condition comprising Ce6 and a diode laser was chosen; this inhibits growth of S. aureus [7]. Mid log-phase (OD600 nm ∼ 0.8) cells of S. aureus grown at 37°C with shaking in TSB were washed and resuspended with PBS (OD600 nm = 0.01), then incubated with Ce6 (5 μM) in the dark for 15 mins at room temperature. Half of the Ce6-treated cells were immediately mixed with phenol/ethanol solution (5% phenol, 95% ethanol, v/v) to stop bacterial transcription and the other half were subjected to PDT by irradiation with a diode laser (15 J/cm2; Geumgwang, Daejeon, Korea) at a wavelength of 664 nm for 25 s before mixing with the same solution. Preparation of Ce6 was essentially the same as described in a previous report [7]. For lethal PDT conditions, in addition to Ce6 and laser PDT set at higher energy densities, another PDT set comprising the photosensitizer Pa and a red LED (635 nm wavelength; Philips Luxeon Lumileds, San Jose, CA, USA) was also employed. Pa was prepared as described previously [2, 9, 10]. Mid log-phase cultures of S. aureus cells prepared as described for the sublethal PDT condition were treated with Ce6 (10 μM) and Pa (0.5 μM), and incubated for 15 mins and 1 hr, respectively. The cells were then spread onto Luria–Bertani plates in the dark. Diode laser and LED of two different energy densities (10 and 30 J/cm2) were used to irradiate 1 cm diameter circles on the surface of the bacterial lawn on the agar plates for Ce6-treated cells and the entire area of bacterial lawn for Pa-treated cells, respectively. CFUs were enumerated after incubation of the plates overnight.

cDNA microarray analysis

Total RNA was purified from three independent biological samples using an RNA mini kit (Qiagen, Valencia, CA, USA) after cell lysis with lysozyme (50 mg/mL) and glass beads (Lysing Matrix B; MP Biomedicals, Irvine, CA, USA). cDNA microarray analysis, including cDNA synthesis, hybridization, and scanning, was performed using GeneChip S. aureus arrays (Affymetrix, Santa Clara, CA, USA) according to the manufacturer's protocol. Data were analyzed using Affymetrix Command Console v1.1 and normalized using MAS5 software (Affymetrix) that determines the reproducibility of the gene probes detected together with statistical significance.

Hydrogen peroxide susceptibility test

Bacterial susceptibility to hydrogen peroxide was determined by measuring growth kinetics after challenge with hydrogen peroxide. S. aureus cells grown overnight in TSB media were diluted in PBS to A600 nm = 0.01 and then equal amounts of cells were inoculated into wells of microplates containing fresh TSB either supplemented with hydrogen peroxide or not. Growth kinetics was measured by determining the OD (A600 nm) at 37°C during incubation in a Bioscreen C Microbiology Microplate reader (Growth Curves, Helsinki, Finland).


Photodynamic therapy induces Agr-dependent gene regulation

In order to investigate transcriptional changes in S. aureus in response to PDT, cDNA microarray analysis was performed after exposing WT S. aureus to a sublethal condition of PDT comprising Ce6 and a diode laser, as described in Materials and Methods. This PDT condition was selected by determining the maximum concentrations of Ce6 that do not change the CFUs of S. aureus before and after irradiation with a diode laser (data not shown). Statistically significant changes in transcription levels of S. aureus genes by PDT were found (Table S1). There was a greater than twofold difference expression of 312 genes, including 136 and 176 genes upregulated and downregulated, respectively (Table S1). The functional categories of PDT-regulated genes were grouped based on analysis of the COGs database (; Fig. 1). Excluding 73 genes not included in any COGs, 239 genes were sorted according to this COGs classification. Except for genes for nucleotide transport (F) and recombination/repair processes (L), genes involved in cell metabolism (COG groups C, E, G, H, I, P, Q) were up-regulated much more frequently than down-regulated, whereas there were more down-regulated than up-regulated genes for cell division (D), envelope biogenesis (M) and translation (K). These findings suggest that PDT treatment causes a bacterial stress response that alters the priority of bacterial physiology toward maintaining survival rather than growth. In addition, microarray analysis showed that significant numbers of genes differently expressed by PDT belong to the regulon of the accessory gene regulator Agr. The findings presented in Table 1, which shows representative Agr-regulated genes identified in previous studies [11, 12] and also differentially expressed in this microarray analysis, demonstrate that PDT activates Agr-dependent gene regulation. The combination of semi-quantitative RT-PCR data validating PDT induction of Agr-dependent genes (Fig. S1) and the results of the microarray experiment suggest that Agr regulator has a role in staphylococcal response to PDT-mediated photo-oxidative stress.

Figure 1.

PDT-regulated gene expression in S. aureus. Genes identified as PDT-regulated by cDNA microarray analysis were grouped by functional category classified according to the COGs database. Genes, the transcription of which changed more than twofold, were analyzed. The proportion of PDT-regulated genes was calculated by dividing the number of genes involved in each category by the entire number of genes regulated by PDT. Genes not involved in any COGs category were excluded. The codes for COGs category are listed at the bottom of the figure.

Table 1. Agr-dependent genes expressed differentially under PDT
Probe Set IDGeneORF No.DescriptionFold changeRole category
sa_c6688s5833_a_atgehSACOL2694Lipase12.1General function prediction only
sa_c5184s4484_a_athutUSACOL2324Urocanate hydratase6.8Amino acid transport and metabolism
sa_c6206s5387_a_at SACOL2571Conserved hypothetical protein6.7Unknown
sa_c5182s4480_a_athutISACOL2323Imidazolonepropionase5.2Secondary metabolites biosynthesis, transport and catabolism
sa_c3448s2976_a_atpckASACOL1838Phosphoenolpyruvate carboxykinase4.8Energy production and conversion
sa_c1147s928_a_atpyrRSACOL1210Pyrimidine operon regulatory protein4.3Nucleotide transport and metabolism
sa_c6521s5689_a_atarcDSACOL2655Arginine/ornithine antiporter2.2Amino acid transport and metabolism
sa_c6513s5681_a_atarcC2SACOL2654Carbamate kinase2.1Amino acid transport and metabolism
sa_c2399s1991_a_atcap5CSACOL0138Capsular polysaccharide synthesis enzyme−2.6Carbohydrate transport and metabolism; cell wall/membrane/envelope biogenesis
sa_c2346s1974_a_atcap5ASACOL0136Capsular polysaccharide synthesis enzyme−2.7Cell wall/membrane/envelope biogenesis
sa_c2385s1987_a_atcap5BSACOL0137Capsular polysaccharide synthesis enzyme−3.0Cell cycle control, cell division, chromosome partitioning

Staphylococcus aureus mutant lacking Agr function is hypersusceptible to photodynamic therapy

To test whether Agr response in PDT is required for staphylococcal resistance to PDT, agr mutant and parental WT S. aureus were subjected to lethal PDT conditions. In this experiment, to investigate the role of Agr under general PDT conditions, two independent sets of PDT, Ce6/laser and Pa/LED were employed as described in the Materials and Methods. As shown in Figure 2a, agr mutant cells in zones exposed directly to laser irradiation and even in regions surrounding laser-irradiated zones were barely detectable under PDT conditions that WT cells survive. Pa/LED PDT also dramatically decreased the numbers of agr mutant when energy densities were increased, whereas most WT cells were not significantly affected by this treatment (Fig. 2b). These results clearly demonstrate that Agr regulator is required for S. aureus survival when exposed to PDT.

Figure 2.

Susceptibilities of S. aureus lacking Agr function to PDT and hydrogen peroxide. WT and its isogenic agr mutant S. aureus were cultured in TSB throughout this study. (a) Mid log-phase cultures washed and diluted in PBS (OD600 nm = 0.02) were incubated with or without Ce6 (10 μM), and irradiated with a diode laser at two different energy densities (outlined by dotted circles) as described in the text. (b) Cell preparation for PDT was the same as described in (a). Pa-treated cells were irradiated with LED as described in the text. In (a) and (b), the upper figures show representative results from three independent experiments and the lower tables present the means +/− SD of number of CFUs in laser- and LED-irradiated regions from three independent experiments. ND, not determined. (c). The same amounts of WT and agr mutant cultured overnight in TSB media (OD600 nm = 0.02) were inoculated onto microplates containing fresh TSB media in the presence or absence of hydrogen peroxide (1 mM). Growth was monitored on a Bioscreen C Microbiology Microplate Reader. Data shown are the means ± SD from three independent experiments. UT, untreated.

Photodynamic therapy produces two types of reactive oxygen species: singlet oxygen and radical-type oxygen species. To examine indirectly which type of oxidative stress corresponds with Agr-mediated responses to PDT, susceptibility of agr mutant to hydrogen peroxide, which promotes chemical reactions that generate typical radical-type oxygen species inside bacteria, was tested [6]. Figure 2c shows that the agr mutant grows better than do WT cells in hydrogen peroxide-containing TSB cultures, implying that, during PDT, Agr is responsible for resistance to photo-oxidative stress mediated by singlet oxygen rather than by radical-type oxygen species. Although Agr is reportedly involved in regulation of staphylococcal stress responses [13], whether the Agr system directly responds to and detoxifies singlet oxygen warrants further studies.


Gram-positive S. aureus has a variety of virulence factors for inactivating the host defense system, causing a diverse range of community-acquired and nosocomial infections from minor skin and soft tissue infections to life-threatening diseases such as toxic shock syndrome, osteomyelitis and bacterial endocarditis. Regulatory networks for gene transcription of virulence determinants in S. aureus are complicated and quite divergent among various strains, including clinical isolates [14]. As a global transcription regulator, Agr plays central roles in gene regulation of S. aureus. The Agr system consists of about a 3 kb locus containing a two component response regulator system and autoinducing ligand for quorum sensing [15]. Both in vitro and in vivo, hierarchical induction of the Agr system controls expression of numerous genes, including genes for virulence, in a cell density-dependent manner which promotes the virulence of S. aureus in animal models of infectious disease [12, 16-19]. Generally, virulence factors of S. aureus include bacterial surface proteins that confer adhesion and evasion of host responses and secreted exoproteins that degrade tissues or defensive antibacterial factors of the host. Activation of the Agr system down-regulates genes associated with bacterial adhesion and up-regulates genes for exoprotein production [12]. This reciprocal feature of Agr-directed gene regulation is related to biofilm formation, a major factor in persistent S. aureus infections. Induction of the Agr system detaches cells in established S. aureus biofilms by increasing extracellular protease activity, whereas pharmacological blocking of Agr or introduction of an agr mutation induces recovery of biofilm activity [20, 21]. Therefore, it has been proposed that the Agr system may regulate the acute versus persistent virulence function of S. aureus [22]. In fact, production of exoproteins and subsequent formation of abscess requires activation of Agr system only in the initial stage of S. aureus infection [16]. In contrast, clinical isolates from patients with persistent S. aureus bacteremia express Agr at much lower levels and these strains form biofilm better than do laboratory strains [23, 24]. Furthermore, S. aureus lacking agr or having weaker Agr function becomes a dominant subpopulation of biofilm during mixed infections in murine abscesses and wounds, prevails as a strong former of biofilm in medical device-related infections and even confers survival advantage during vancomycin treatment [25-29]. These reports concerning Agr-linked biofilm formation in persistent S. aureus infections warrant practical application of the finding of this study that PDT is effective against agr mutant S. aureus.

In conclusion, this study provides information about the genetic response of S. aureus to PDT as well as evidence for a novel function of the Agr system in resistance to photo-oxidative stress. The results presented herein may contribute to further understanding of bacterial responses to PDT-mediated photo-oxidative stress and facilitate the clinical application of PDT as a potent antimicrobial therapy for controlling multidrug-resistant pathogens, especially biofilm-mediated persistent infections that greatly limit the activity of host defenses or antimicrobial drugs.


The authors would like to thank Professor Simon J. Foster, University of Sheffield, Sheffield, UK, for providing the S. aureus strains used in this study. This study was supported by a grant from the Korea Healthcare Technology Research and Development Project, Ministry for Health, Welfare and Family Affairs, Korea (A080160) and by the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (R13-2008-010-00000-0).


All authors declare no conflict of interest associated with this study.