Background. Staphylococcus aureus is an important gram positive pathogen that causes a multitude of human diseases. Many of these diseases are associated with the production of staphylococcal exotoxins. Superantigens are a family of staphylococcal exotoxins associated with many of these illnesses. Superantigens are defined by their abilities to induce massive cytokine release from host T cells and antigen presenting cells, leading to the development of toxic shock syndrome (TSS). Superantigens have also been associated with atopic dermatitis, an inflammatory skin disease that often coincides with S. aureus colonization of skin lesions. Superantigen production has been shown to induce corticosteroid resistance in human T cells in vitro (corticosteroids are the most common treatment for atopic dermatitis), thus linking superantigens with more severe cases of atopic dermatitis. Past atopic dermatitis studies have focused on select superantigens. To the best of my knowledge, the work presented in this dissertation represents one of the most comprehensive investigation into superantigens associated with S. aureus strains from patients with atopic dermatitis.
Superantigens may be also be involved in staphylococcal pulmonary infections. Recently, several community-associated strains of S. aureus have been linked with severe and often fatal pulmonary diseases. I have found that these strains produce superantigens such as toxic shock syndrome toxin-1 (TSST-1), and staphylococcal enterotoxins B (SEB) and C. I hypothesize that superantigens are involved with the severity and fatality associated with these infections (likely by causing TSS).
Few treatment options currently exist for preventing staphylococcal toxin production, or neutralizing superantigens in humans. Recent studies have shown exciting evidence that suggests that administration of soluble high affinity T cell receptor subunits (Vβ-TCRs) can neutralize superantigen activity in vitro, and are able to neutralize lethality in a rabbit model of TSS. Further investigations would be useful for determining the full range of toxin-mediated diseases these small peptides could be used to treat. Additional research has demonstrated the potential for using glycerol monolaurate (GML) for inhibiting staphylococcal toxin production in vitro. The ability of GML to inhibit toxin production in vivo has not been studied prior to the work presented in this dissertation.
GML has also been shown to inhibit the growth of several vaginal pathogens such as Gardnerella vaginalis, and Candida albicans, without inhibiting the grown of Lactobacillus crispatus, which is a predominant member of the vaginal normal flora. This finding suggests that GML may be useful as a treatment for vaginal infections such as bacterial vaginosis and vulvovaginal candidiasis.
Methods. PCR was used to determine the superantigen profiles of S. aureus isolates collected from patients with steroid sensitive atopic dermatitis, steroid resistant atopic dermatitis, and vaginal isolates collected from healthy women. In some cases, superantigen production was quantified using Western immunoblots. Antibiotic sensitivity was determined using disk diffusion assays. SCCmec typing for DNA encoding methicillin-resistance was performed using PCR.
To investigate the role of superantigens in severe staphylococcal pulmonary diseases, rabbits were challenged, via the intrabronchial route, with live bacteria (USA200 or USA 400 strains of S. aureus) or purified TSST-1, SEB, or SEC. Some rabbits were hyperimmunized against TSST-1, SEB, or SEC prior to challenge, and some rabbits were administered live bacteria or purified SEB plus soluble high affinity Vβ-TCRs capable of neutralizing SEB superantigenicity.
To investigate the effects of GML on staphylococcal exotoxin production and vaginal inflammation, I initiated a double-blind randomized study in which 225 menstruating women were recruited to donate their own used tampons and then wear study tampons (± GML) on their second day of menstruation for 4-6 hours. Tampon samples were screened for the presence of S. aureus, TSST-1, α-hemolysin, and interleukin-8 (IL-8).
To investigate further the in vitro and in vivo effects of GML on vaginal microbes, I initiated a small pilot study to determine the effects of GML on vaginal Candida sp, Gardnerella vaginalis, and Lactobacillus sp. In this clinical study, women with chronic vulvovaginal candidiasis and/or bacterial vaginosis were recruited to use an intravaginal gel containing 0%, 0.5%, or 5% GML for 2 days. Swabs of vaginal discharge were collected from the women before and after gel use. Swabs were analyzed for the presence of the microbes of interest, as well as for GML content.
Lastly, I investigated severe pulmonary infections caused by USA300 S. aureus strains by challenging rabbits with live bacteria, or purified superantigen. While conducting these studies, I discovered certain USA300 isolates produce a variant form of TSST-1. I used rabbit TSS models, PCR, Southern blots, and superantigenicity assays to begin characterization of the variant form of TSST-1. TSST-1 may cause lethality through mechanisms in addition to superantigenicity. I tested for this lethality by first neutralizing superantigen activity with soluble high affinity Vβ-TCRs, and then administering TSST-1 to rabbits.
Results and conclusions. S. aureus isolates collected from steroid-resistant atopic dermatitis patients had significantly different superantigen profiles than other isolates. Strains from patients with steroid-resistant atopic dermatitis carried the genes for a greater number of superantigens, were more likely to produce the three major TSS-associated superantigens (TSST-1, SEB, and SEC), and were also more likely to have uncommon combinations of superantigens, compared to isolates from steroid-sensitive patients and vaginal isolates from healthy women. These findings suggest that S. aureus isolates from patients with steroid-resistant atopic dermatitis are selected for on the basis of greater superantigen production. Although this study focused on isolates collected from steroid-resistant, steroid-sensitive, and from the vaginal mucosa of healthy women, it may also be beneficial to include skin isolates collected from healthy individuals as a control. Both methicillin-sensitive and methicillin-resistant S. aureus (MSSA and MRSA, respectively) colonized patients with steroid-resistant atopic dermatitis. While MSSA and MRSA isolates both had unusual superantigen gene combinations, MSSA isolates had a greater number of superantigen genes.
From my studies that focused on the role of superantigens in severe pulmonary diseases, I found that rabbits challenged with viable bacteria developed fatal pulmonary infections. Rabbits previously immunized against purified SEC and then challenged with viable bacteria did not develop fatal disease. Rabbits treated with soluble high affinity Vβ-TCRs for SEB, and then challenged with purified SEB, also did not develop fatal pulmonary disease. In these studies, immunity to superantigens prevents lethality associated with the pulmonary bacterial exposure. Administration of a soluble high-affinity Vβ-TCR to non-immune animals also protected rabbits.
Results from in vivo investigations of the effects of tampons containing GML, on both staphylococcal exotoxin production and vaginal inflammation, revealed that lower amounts of S. aureus and exotoxins were detected in study tampons ± GML than the women's own tampons. I also found lower amounts of exotoxins were present in GML+ than GML- study tampons. The chemokine IL-8 was lower in S. aureus- than S. aureus+ tampons and lower in GML+ than GML- study tampons. Collectively, my studies found that tampons containing GML reduce S. aureus exotoxin production. S. aureus increases production of IL-8 in the human vagina, and GML reduces production of this pro-inflammatory chemokine. My findings suggest adding GML to tampons may provide additional safety over other tampons relative to menstrual TSS.
My investigations into the effects of GML on additional vaginal pathogens demonstrated that in vitro GML concentrations of ≥100 µg/ml were bactericidal for C. albicans, while concentrations ≥5 µg/ml were bactericidal for G. vaginalis. In human studies, use of either control (0% GML) or GML gels (0.5 and 5% GML) did not significantly alter vaginal pH or levels of Lactobacillus. Use of gels containing GML significantly reduced vaginal levels of Candida. The control gel did not significantly alter Candida levels. Use of both the control and the GML-containing gels significantly lowered the levels of G. vaginalis. These findings suggest that an intravaginal gel containing GML may be useful for treating both vulvovaginal candidiasis and bacterial vaginosis, meanwhile not inhibiting the growth of Lactobacillus.
During our studies to investigate the role of superantigens in pulmonary infections caused by USA300 S. aureus isolates, I observed that immunity to staphylococcal enterotoxin-like Q (SEl-Q) offers partial (but not statistically significant) protection from lethal pulmonary infection. Further investigation revealed the presence of a variant form of TSST-1 that contains a large deletion in the tstH gene. This deletion was present in several USA300 clinical isolates. This deletion was predicted to disrupt the ability of variant TSST-1 to stimulate T cells via the Vβ-TCR because variant TSST-1 was predicted to be unable to crosslink the MHC II molecule to the Vβ-TCR. When tested for superantigen activity, variant TSST-1 was found to retain some superantigen activity, and is still highly lethal in rabbits. Additional experiments using soluble high affinity Vβ-TCRs to neutralize full length TSST-1 superantigen activity revealed that full length TSST-1 exposure is still lethal; suggesting that full length TSST-1 can cause lethality through another mechanism in addition to superantigenicity. My findings suggest that lethality due to full length TSST-1 exposure may occur through multiple mechanisms, including superantigenicity, and potentially through direct interaction with the central nervous system.