Increased Infectivity of Staphylococcus aureus in an Experimental Model of Snake Venom—Induced Tissue Damage

Patricia Saravia-Otten 0 1 3 Jose Mara Gutie rrez 1 4 Staffan Arvidson 1 3 Monica Thelestam 1 3 Jan-Ingmar Flock 1 2 3 0 Departamento de Bioqu mica, Facultad de CCQQ y Farmacia, Universidad de San Carlos de Guatemala , Guatemala City , Guatemala 1 Received 28 November 2006; accepted 28 March 2007; electronically published 18 July 2007. Potential conflicts of interest: none reported. Financial support: Swedish International Development Agency (as part of the NeTropica program); Swedish Research Council (grants 05969 to M.T. , 12218 to J.-I.F. , and 4513 to S.A.). and Cell Biology, Karolinska Institutet , S-171 77 Stockholm, Sweden (jan-ingmar 2 Laboratory Medicine, Karolinska Institutet , Stockholm , Sweden 3 Microbiology, Tumor, and Cell Biology 4 Instituto Clodomiro Picado, Facultad de Microbiolog a, Universidad de Costa Rica , San Jose , Costa Rica Soft-tissue infection is commonly found in patients bitten by Latin American Bothrops snakes. Staphylococcus aureus, which is not present in the mouth of the snake, is frequently isolated from these infections. The effects of B. asper venom on infection with S. aureus were analyzed in a model of infection in envenomated mouse gastrocnemius muscle. Inoculation of 50 colony-forming units (cfu) of S. aureus was enough to cause infection in envenomated muscle, compared with 15 104 cfu without venom. This effect was also achieved by injection of venom myotoxin III (an A2 phospholipase). A sarA mutant strain in which production of extracellular toxins and enzymes is up-regulated and binding of fibronectin, fibrinogen, and other host proteins is downregulated was much less virulent than the corresponding parental strain, indicating that the ability of S. aureus to mask itself with host molecules might be more important than the effects of secreted toxins and enzymes in this kind of infection. - Abscess formation is a common complication found in patients bitten by Viperidae snakes in Latin America, where snakebite envenomation represents a serious public health problem [1, 2]. Abscess formation is a risk factor for amputation in these patients [3, 4] and may be associated with sepsis. Large numbers (2 106 organisms/mL) of bacteria, including anaerobic species (Clostridium species), aerobic gram-negative rods (Enterobacteriaceae), and a small proportion of gram-positive cocci (group D Streptococcus and Micrococcus species), are inoculated with snakebites [5, 6] and have been isolated from the abscesses of bitten patients [79]. Interestingly, Staphylococcus aureus, which is not present in the mouth of the snake but is present on the skin of the human victim, is frequently isolated from these abscesses [810]. Obviously, the relatively few S. aureus on the skin can successfully compete with the bacteria from the snake venom to cause infection at the bite site. Soft-tissue infection after bites inflicted by snakes of the genus Bothrops has been attributed to several components of the venom, whose action might create a favorable environment for the multiplication of bacteria [7, 8, 11]. Local tissue damage caused by B. asper venom includes degradation of the extracellular matrix (ECM), necrosis, hemorrhage, edema, blistering, and a prominent inflammatory response [2]. These effects are caused mainly by locally acting A2 phospholipases and by zinc-dependent metalloproteinases [2, 12, 13] but are also caused by host matrix metalloproteinases (MMPs) that are activated through the action of the venom [14, 15]. The virulence of S. aureus depends on secreted toxins and enzymes and on cell wallassociated proteins that bind to the host ECM and plasma proteins [16, 17]. These virulence determinants are synthesized in a growth phasedependent manner in vitro, with most exoproteins being expressed during postexponential or stationary phases; this is in contrast to the cell wallassociated proteins, which are mainly produced during the logarithmic phase of growth [18, 19]. Two major players in the growth phasedependent regulation of virulence determinants and possibly in the shift from adhesive to invasive phenotype are the global regulators agr and sarA [1921]. Inactivation of the agr locus results in increased production of surface proteins, decreased production of secreted proteins, and reduced virulence [22, 23]. Mutation of the sarA locus generally results in an increase in production of proteases, collagen binding, and protein A and a decreased capacity to bind to fibronectin and attenuation [20, 24, 25]. Little is known about the relationship between snake venom induced tissue damage and infectionthat is, the effect that locally acting toxins from the venom have on the colonization and spread of bacteria. Because toxins present in B. asper venom drastically affect ECM proteins, blood vessels, and muscle fibers, we hypothesized that these components contribute to the ability of S. aureus to cause infection and, consequently, to this common complication of snakebite. We have studied the effect of B. asper venom on infection caused by S. aureus strain DB in mouse gastrocnemius muscle. We have also determined the role of 2 locally acting prototype toxins from the venom and their contribution to infection. Isogenic agr or sarA DB mutants were used to further dissect the factors involved in bacterial colonization of venom-affected tissue in our model. MATERIALS AND METHODS Bacterial strains and cultivation conditions. S. aureus strain DB is a clinical isolate [20]. The mutant strains UAMS-931 (a minocycline-resitant agr mutant) and UAMS-932 (a neomycinand kanamycin-resistant sarA mutant) derived from strain DB were provided by M. Smeltzer (University of Arkansas) [24]. Bacterial strains were cultured in Luria-Bertani (LB) medium (GIBCO) with or without appropriate antibiotics. For the animal infection experiments, S. aureus strains were grown to stationary phase in LB medium at 37 C. The bacteria were harvested by centrifugation, suspended in PBS, and adjusted to 1 108 cfu/mL. Aliquots were frozen at 120 C. B. asper venom and toxins. The venom used was a pool obtained from 140 adult specimens of B. asper collected in the Pacific region of Costa Rica and kept at the serpentarium of Instituto Clodomiro Picado (Universidad de Costa Rica). The venom was immediately lyophilized and stored at 20 C. Myotoxin III (MTIII), an Asp-49 myotoxic A2 phospholipase, and the metalloproteinase BaP1 were isolated from this venom by ion-exchange chromatography on a CM-Sephadex C-25 column (Pharmacia), as described elsewhere [26, 27]. BaP1 was further purified by gel filtration on Sephacryl S-200 and affinity chromatography on Affi-Gel Blue (Bio-Rad), as described elsewhere [13]. Venom and toxin solutions were prepared in sterile apyrogenic PBS and filtered through 0.22-mm-pore membranes (DynaGard; Microgon). Animal model of infection in envenomated mouse gastrocnemius muscle. Injection of snake venom into mouse gastrocnemius muscle has been used as a mod (...truncated)