Celiac disease (CD) is an autoimmune disorder that affects approximately 1% of the population [55]. CD is characterized by intestinal villus atrophy after consumption of gluten from wheat, barley, or rye. Patients with CD often experience abdominal pain, diarrhea, malnutrition, fatigue, and a failure to thrive. There is currently no treatment for CD. Patients must live on a strict lifelong exclusion of dietary gluten. Due to the high content of gluten in western diets and poor labeling of gluten content, adherence to a gluten free diet (GFD) is difficult [15].Nearly all the enzymes that can digest the gluten peptide are sensitive to the stomach's low pH . As a result, dietary supplementation with enzymes to digest gluten has yet to produce a viable alternative treatment to a GFD. We propose to use a resident microbe of the human intestinal tract to express a peptidase to digest the immunoreactive gluten fragments. The bacteria, L. reuteri, will colonize the host's intestines and digest the gluten peptides before causing an autoimmune response. To accomplish this task, this thesis describes a food grade, plasmid based system to integrate genes into the genome of L. reuteri. The plasmid system utilizes an origin of replication that requires a protein, RepA, to propagate itself. A helper plasmid provides the RepA protein in trans to an integration plasmid that cannot provide RepA to itself. The integration plasmid carries a homologous region to the genome of L. reuteri allowing for targeted genomic integration. The integration plasmid will not replicate on its own, and will be integrated into the genome if the helper plasmid is absent. To select for these genomic integrants the integration plasmid expresses an erythromycin resistance marker. Using the Cre/Lox system the antibiotic resistance will be removed from the bacterial genome to re-establish the L. reuteri's food grade status. This thesis describes the construction and verification of the above mentioned plasmid tool kit containing the helper, integration, and Cre expression plasmids to integrate genes into the L. reuteri genome.

Probiotics have been defined as “live microorganisms which when administered in adequate amounts confer a health benefit on the host”. The beneficial effects of probiotics in the gut are well described and roles including immunomodulation and colonisation resistance have been documented. Recent reports suggest that topical use of probiotic bacteria may be an effective strategy to promote skin health or inhibit disease. Therefore, in this thesis the potential of probiotics to protect skin from pathogenic bacteria was assessed using primary keratinocytes as a model system, and the skin pathogen, Staphylococcus aureus. The ability of three probiotics, L. reuteri ATCC 55730, L. rhamnosus AC413 and L. salivarius UCC118 to inhibit the growth of S. aureus was tested using well-diffusion assays and spot on the lawn assays. All three probiotics inhibited the growth of S. aureus in well-diffusion assays, though this property was dependent on growth medium. Inhibition of S. aureus growth was principally via the production of organic acids rather than bacteriocin production. Next, to determine whether probiotics could protect keratinocytes, confluent normal human epidermal keratinocytes (NHEK) were infected with S. aureus (106 CFU/ml) in the presence or absence of the probiotic (108 CFU/ml). NHEK viability was measured using trypan blue exclusion assays. L. reuteri had a significant protective effect on NHEK when applied 1h prior to (P=0.0003), or simultaneously with S. aureus (P=0.002). L. reuteri did not however protect NHEK when applied 1h after S. aureus addition. There was no change in the number of viable S. aureus in cell culture assays. To determine whether the protective effect was due to the inhibition of adhesion, NHEK were either pre-exposed to the probiotic for 1h, simultaneously exposed to the probiotic and S. aureus for 1h, or exposed to the probiotic 30 minutes after S. aureus addition for 1h. Pre-exposure of NHEK to L. reuteri (exclusion) and simultaneous exposure to L. reuteri and S. aureus (competition) resulted in significantly less staphylococci adhering to NHEK (P=0.03 and P=0.008 respectively). However when L. reuteri was added after S. aureus (displacement), the number of adherent staphylococci was not reduced. The necessity of S. aureus adherence for the inactivation of NHEK was demonstrated using a α5β1 integrin blocking antibody. Finally, to compare the innate response of NHEK to probiotics with S. capitis and S. aureus, TLR-2, antimicrobial peptide (AMP) expression and IL-8 production were measured. TLR-2 protein (but not mRNA) expression was reduced in the presence of S. aureus (P=0.018). NHEK pre-exposed to S. capitis prior to S. aureus infection however, exhibited elevated TLR-2 protein and mRNA expression (P<0.0001 and P=0.009 respectively). NHEK pre-exposed to L. reuteri prior to S. aureus had no significant change in TLR-2 expression compared to untreated controls. ELISAs demonstrated that IL-8 production was significantly increased in NHEK pre-exposed to L. reuteri prior to S. aureus infection (P=0.0001). In conclusion, L. reuteri protected NHEK from the toxic effects of S. aureus at least partly through competitive exclusion of binding sites on NHEK. Finally, NHEK innate responses to probiotic bacteria were akin to those to the skin commensal, S. capitis. L. reuteri induced expression of a neutrophil chemoattractant, suggesting it could be of importance in priming the innate immune response against S. aureus infections. Taken together, these results suggest that probiotic bacteria could be used prophylactically within skin creams and soaps to prevent S. aureus colonisation and infection in skin.

Non-healing wounds is a growing medical problem, often associated with pathological conditions such as diabetes and peripheral ischemia. A non-healing wound entails a large amount of suffering for the patient, and demands extensive health care resources. In this thesis, a new drug treatment paradigm for wound healing was developed by transforming Limosilactobacillus reuteri R2LC with a plasmid encoding CXCL12 (LB_CXCL12). The drug candidate was tested for safety and biological effects following topical administration to full thickness wounds in both mice and minipigs. In parallel, different techniques, including 2D and 3D measurements, planimetry, and ultrasound, for assessing wound healing were developed and evaluated.   Murine wounds treated with LB_CXCL12 demonstrated increased proliferation of dermal cells, and an increased density of macrophages of which a larger fraction expressed TGF-β. If macrophages were depleted prior to wounding, the accelerated effect on healing was abolished demonstrating a macrophage-dependent mechanism of action. Importantly, the LB_CXCL12 treatment also accelerated wound healing in mice with impaired healing as a result of hyperglycemia or peripheral ischemia, conditions that in humans are associated with development of non-healing wounds. Wounds in minipigs treated with the freeze-dried formulation of LB_CXCL12, upon resuscitation referred to as ILP100, showed accelerated healing both by increased granulation tissue formation and accelerated re-epithelialization. The treatment with ILP100 was well tolerated with no treatment-related deviations in haematology, urinalysis, and histopathology. Further, we found improved detection of thin layers if newly formed epithelial using planimetry and ultrasound compared to 2D photographs, whereas 3D scans accounting for surface curvatures yielded larger wound areas than 2D photographs of the same wounds.  Development of topical treatments for non-healing wounds are limited by the proteolytic environment of the wound that cause degradation of applied molecules. Our developed technology, a new-in-class candidate, overcomes this by continuous on-site delivery and increased bioavailability of CXCL12, resulting in prolonged instruction of local immune cells to stimulate wound healing.