Resistance to antimicrobial drugs is a major threat to world health. With the number of new drugs released per year decreasing, this danger continues to increase [1]. Antibiotic resistant pathogens such as Staphylococcus aureus and Pseudomonas aeruginosa are associated with many nosocomial infections [1,2]. Patients with weakened immune systems, arising from underlying diseases such as cancer or cystic fibrosis, are especially at risk [2]. In response to the antibiotic resistance of P. aeruginosa, health providers are using older, less safe, and less effective treatments like colistin to treat infections with S. aureus and P. aeruginosa [1]. New and more successful therapeutic techniques need to be developed to cope with the increase of disease and death caused by these bacteria.

Quorum sensing is a method of chemical signaling by which some microbial species communicate [2-5]. These chemical signals increase in concentration with population density. Responses to quorum sensing can control many functions within a cell including biofilm formation, virulence, growth, antibiotic resistance, motility, and autolysis [2-6]. Important pathogens such as P. aeruginosa and Vibrio cholerae use quorum sensing to regulate these functions [3]. With the concept of quorum sensing in mind, Saeidi et al genetically engineered nonpathogenic Escherichia coli to first detect P. aeruginosa using its quorum sensing molecules and then releasing a protein toxin to kill the pathogen [2].

In vitro, the modified E. coli killed 99% of P. aeruginosa and slowed biofilm formation by approximately 90% [2]. For this approach to be useful, it must work within the human body. This review will cover the potential implementation of this technology into microbes already within the microbiota and into probiotics.

The inspiration for this genetic system was P. aeruginosa itself [2]. For quorum sensing, this bacterium uses an autoinducer type-1 system that is prevalent in other Gram-negative bacteria [2,3]. In this system, AHL molecules are secreted into the environment by the bacterium. When the population density is high enough, AHL concentrations are able to activate a cascade of luxR promoters, which result in microbial invasion [2,3]. When the microbial population density is low, the amount of AHL is dilute and the luxR promoters remain repressed [3].

This quorum sensing system was engineered into a non-pathogenic E. coli chassis [7]. The genetic design, shown in Figure 1, includes a tetR and a luxR promoter [2]. The constitutive tetR promoter continuously releases a LasR protein which binds to any present AHL 3OC₁₂HSL released by P. aeruginosa [2]. The luxR promoter is not activated until a LasR-3OC₁₂HSL complex binds to the promoter itself [2]. The activation of the luxR promoter begins the production of the pyocin and the lysis proteins [2].

Figure 1 This figure represents a schematic of the engineered E. coli within proximity of P. aeruginosa. The constitutive tetR promoter releases a LasR protein which forms an activator complex with a 3OC12HSL molecule. This complex activates the luxR promoter which signals the release of pyocin S5 and the E7 lysis protein. After the lysis protein breaks down the membrane of the E. coli, the soluble pyocin attaches to and kills the P. aeruginosa [2].

When recombining biological sequences the outcome is unpredictable. Thus, parts cannot be used in random combination with each other. This system was therefore built to BioBrick assembly standards [2]. BioBrick standard biological parts are genetic sequences that follow rules for physical and functional conformation, so the behavior of the genetic device is predictable [7]. This saves time and money and increases the probability of success [7].

The detection of P. aeruginosa occurs via the combination of the quorum sensing of P. aeruginosa and the engineered E. coli. P. aeruginosa secretes 1.0x10^-6 M to 1.0x10^-4 M 3OC₁₂HSL which, when proximal to the E. coli, will create the LasR-3OC₁₂HSL complex [2]. The complex then activates the luxR promoter, leading to a release of the pyocin and lysis proteins [2]. Figure 1 illustrates how pyocin will accumulate within the cell membrane until the lysis protein releases the pyocin by degrading the membrane [2].

Pyocin is a bacteriocin, or antimicrobial peptide, specifically made by and active against P. aeruginosa [2]. Although there are three types of pyocins, the S type pyocin was chosen because it is highly soluble and virulent against P. aeruginosa but not E. coli [2]. S type pyocin is a two protein complex [2]. The actual killing of the target is completed by the larger protein, while the smaller one protects the host against self-killing [2,8]. Pyocin S5 kills by first attaching to the lipopolysaccharide on the outer membrane and then piercing through the bacterial outer membrane, cell wall, and cytoplasmic membrane [8,9]. This depolarizes the membrane potential sufficiently to cause cell death [8,9].

It has not been proven whether pyocin is toxic to humans, or even mammals. In an experiment from 1976, pyocin was used to treat mice with burns, resulting in no evident toxic or allergenic effect from the pyocin [10]. However, bacteriocins are generally low in toxicity; some are even used to extend food preservation [11,12].

As shown in Figure 1, the pyocin reaches the P. aeruginosa by lysing the E. coli via the E7 lysis protein [2]. This lysis protein is found in E. coli [2,13]. It damages both the inner and outer membrane of E. coli to release the pyocin [2,13]. It also has been shown to activate outer membrane phospholipase A which helps to make the membrane more permeable and thus releasing more pyocins [2,13].

P. aeruginosa secretes a concentration of 1.0x10^-6 M to 1.0x10^-4 M 3OC₁₂HSL in proximity to infection [2]. In order for the E. coli to be activated, it has to be able to sense this range. In the first experiment, susceptibility of the E. coli was tested using a range of synthetic 3OC₁₂HSL levels. The supernatants from this test were filtered and added to plates of P. aeruginosa. Growth of P. aeruginosa was inhibited by extracts from E. coli exposed to 3OC₁₂HSL levels between 1.0x10^-6 M and 1.0x10^-4 M. Supernatants from P. aeruginosa cultures were then mixed with E. coli cultures instead of synthetic 3OC12HSL. Extracts from these cultures inhibited growth of P. aeruginosa, proving that the engineered E. coli can respond to the 3OC₁₂HSL at the concentrations produced naturally [2].

To demonstrate killing by E. coli, experiments were conducted using both motile and biofilm stages of P. aeruginosa [2]. The results of these can be seen in Figure 2. For both tests, the final and incomplete systems of the E. coli were co-cultured with chloramphenicol resistant P. aeruginosa. The “incomplete” systems of E. coli were genetically incomplete, missing the genes for either the release of pyocin or the lysis protein. These systems were used as controls for the experiment. Another control was the use of a pyocin S5 resistant strain of P. aeruginosa. For the tests with motile stage of P. aeruginosa, E. coli inhibited growth of the P. aeruginosa up to 99% while the negative controls did not reach above 20% inhibition [2].

Figure 2 Inhibition with engineered E. coli.Figure 2A shows the percentage of motile P. aeruginosa survival. The engineered E. coli was mixed with chloramphenicol-resistant P. aeruginosa and PAO1 (from which pyocin S5 was derived). Then incomplete systems (missing S5 or E7) of the E. coli were mixed with the chloramphenicol-resistant P. aeruginosa. The results show that the engineered E. coli inhibited P. aeruginosa growth by 99%. Figure 2B shows the percentage of P. aeruginosa biofilm survival after 18 hours of exposure to the E. coli (with the same mixture constraints as in Figure 2A). The results show the E. coli inhibited biofilm growth by almost 90% [2].

In a pathogenic state, P. aeruginosa will form a complex biofilm [2,4]. To test whether the E. coli might inhibit an infection, it was co-cultured with a P. aeruginosa biofilm, revealing that further growth of the biofilm was inhibited by almost 90%. The negative controls did not even reach 30% inhibition [2].

Two possible application methods are the use of engineered E. coli as a probiotic or to genetically engineer a microorganism from the niches within the human body known to contain P. aeruginosa during disease. The number of microbes inhabiting the human body is large: for every human cell, there are ten microbial cells [14]. This community of microbes, known as the human microbiome, provides improved nutrition and resistance to infection. Common sites for these communities include the human oral cavity, vagina, and gut [14]. The microbiome is related to probiotics, or the addition of non-pathogenic microbes to the human body for beneficial reasons [14]. These proposed applications are long term because the microbiome is complicated and little is known about ecological competition or balance. Concerns with the addition of “new” bacteria to the human body include but are not limited to identification, resistance, genetic stability, viability, pathogenicity, and allergic reactions [15]. The more information regarding toxicity and interactions between microorganisms within the body will lead to fewer complications when the flora is modified [15,16].

The engineered E. coli has potential application as a probiotic. If ingested, the E. coli would need to survive, colonize the human gut, and reach the infected area [15,17,18]. Many industries use microorganisms as probiotics. Baur et al used lactic acid producing bacteria as a probiotic for the skin [17]. Lactic acid producing bacteria are members of the gut microbiome and increase immune responses during suppressed conditions and repress the immune system during an inflammatory response [17]. Skin can be very sensitive to many irritants such as chemicals, UV radiation, and pathogens [17]. Baur et al used Lactobacillus johnsonii and Lactobacillus paracasei for use as probiotics in food, and pharmaceutical, cosmetic, or topical products [17]. The mouse model revealed that when the skin is stressed, the immune system can be either over or under active, and the addition of the Lactobacilli can regulate the immune response to within normal expectations [17]. L’Oreal has applied for a patent using this technology in their products [19]. Cosmetics can contain a number of skin irritants like preservatives, dyes, and perfumes [19]. L’Oreal’s goal is to use at least one microorganism to reduce the occurrence of irritant contact dermatitis [19].

The Baur et al study shows that bacteria with a sequenced genome, like L. johnsonii or E. coli, are easier to manipulate to achieve the function desired since a lot of information can be revealed within its genetic coding [15] More genetic information can lead to higher genetic stability, or a decrease in the potential to be altered by evolutionary processes, such as horizontal gene transfer [15]. This could prevent the possibility of contracting antibiotic resistance, making it more difficult to remove the probiotic if a problem would occur [15,16]. Problems with this approach involve viability, colonization, pathogenicity, and allergic reactions [15]. These bacteria would never have lived in the human body before, and therefore, knowledge about reactions with the human body is extremely limited. For treatment, it is required that the bacteria colonize the needed niche of the body [15]. However, bacteria will not merely colonize a human niche because they are “supposed” to [15]. With foreign bacteria, how a new organism will react with the body is less predictable; the potential for damage by toxins or allergic reaction is much higher [15]. Whether the bacteria would live, colonize, be harmful, or trigger immune responses can be tested most safely through an animal model [15].

The other option for implementing the technology is to use a microorganism that is commonly found in the human body. The following two studies are examples using this approach for medical applications.

Infectious diseases are becoming more widespread due to multi-resistant strains of bacteria. P. aeruginosa, one of the most lethal nosocomial pathogens, causes chronic illness and death in many cystic fibrosis patients. Saeidi et al genetically engineered E. coli to sense and kill P. aeruginosa. E. coli sense the chemicals released by P. aeruginosa; then signal in response to the accumulation and release of pyocin via lysis proteins. The preferred approach to apply the technology is modifying microorganisms that are already prevalent within the human body. This method has a greater possibility of viability and colonization, and less risk of reactions from pathogens or allergens.

Because of the complex organization of the human microbiome, the development of this approach will take some time, as there is still much to learn about human body as a bacterial ecosystem. Further understanding of the microbiome will perhaps lead to superior functionality or faster implementation. When complete, this technology holds great commercial potential since P. aeruginosa plagues immune-compromised patients with acute and chronic infections [1,2]. As many as half of P. aeruginosa infections result in death, and 61% of cystic fibrosis patients carry these bacteria [9]. This may be the answer to a new way to fight infections.