From 2012.igem.org

Preface

Our team first began working on the drug delivery and biofilm projects in parallel. When some of our initial experiments showed promising experiments for the drug delivery idea, we started focusing more of our attention on that project due in part to our smaller team size.

Overview

Bacteria are capable of surviving on many surfaces for extended periods of time, until conditions favor their growth. Recent research has discovered that many pathogenic bacteria are capable of forming resilient “biofilm” colonies that are difficult to eradicate and even more difficult to treat, especially when established within a patient. Many chemical approaches have been applied to preventing biofilm formation and surface fouling, especially in invasive devices such as catheters. However, these devices carry the drawback of loss of function over time (as a result of the depletion of antimicrobial elements from the surface). Furthermore, because the compounds that leach from the surface of chemical-based antimicrobial surfaces may be toxic not only to microbes but also human tissue, their safety in vivo is unknown.

We seek to investigate an alternative approach by seeding surfaces with a non-pathogenic, biofilm forming bacterium that would secrete antimicrobial peptides (AMP) that would inhibit subsequent colonization of the surface by pathogenic microbes. The advantages of this approach over traditional chemical treatments is that the surface would be capable of replenishing AMP levels, preventing loss of function over time. For our project we chose to utilize lysostaphin (lss), an enzyme that is capable of destroying the cell wall of Stapylococcus, a genus of bacteria that are responsible for a large proportion of hospital acquired infections, as well as capable of forming biofilms.

System Design

Our original design incorporated a bacterial “leash” which would prevent stray engineered E. Coli. from surviving should they become detached from our engineered biofilm. The signal for this process is Autoinducer-2 (AI-2), a small quorum signaling molecule that is utilized in many species of bacteria. The presence of AI-2 would drive the expression of lss, as well as the protein mazE, which acts as an antitoxin against mazF, which is constitutively expressed. While E. Coli. is typically not able to produce AI-2, we have also added a synthetase known as luxS, which enables AI-2 production. In the case of luxS, we have also found reports that luxS enhances biofilm formation through the AI-2 quorum signaling molecule.

The secretion of lss would enable the E. Coli. biofilm to kill Stapylococcus bacteria, and inhibit their ability to form biofilms on the surface.

Results

This experiment is not without precedent. This concept was utilized in 1999 to produce a protective biofilm on steel to inhibit the growth of sulfate-reducing bacteria that would have otherwise corroded the steel. The results indicated that the protective biofilm that produced AMPs (Gramicidin S) delayed the onset of corrosion and inhibited the activity of sulfate reducing bacteria for up to 28 days when immersed in growth media, and up to 120 hours in continuously replaced media (i.e. a surface placed in a continuous flow of media).

The assembly of the bacterial leashing system was problematic. The promoter region of the lsr operon (plsr) was obtained from biobrick BBa_K117002 (Ashwin, make the part number a clickable link to: http://partsregistry.org/Part:BBa_K117002:Design), however, we are not able to conclude if this is a problem with the biobrick or with the rest of the triple ligation that we attempted.

However, we were able to evaluate the effectiveness of lss expressed by BL21 in killing Staphylococcus epidermidis, a non-pathogenic strain of Staphylococcus that forms prolific biofilms. This experiment confirms that engineered bacteria are capable of expressing lss without killing themselves. This is a crucial step in assembling our system.

Towards the end of this project, we attempted to evaluate the effect that luxS had on biofilm formation in BL21 E. coli. unfortunately, our results were inconclusive. We believe that the expression vector utilized for luxS, pET-26b(+), which contains a pelB secretion signal, may be the culprit. We are continuing our experiments, and plan to remove the pelB secretion signal in the near future.

Lastly, we have also obtained a strain of E. Coli Nissle 1917, a non-pathogenic strain of E. Coli that has attracted attention for its low immunogenicity and lack of endotoxin production, an important trait for any engineered bacteria that may come in contact with the human body to have. We have found that chemical competency of Nissle 1917 can be achieved, and plasmid DNA can be transformed and expressed with acceptable efficiency, and hope to implement our system in the Nissle 1917 chassis.