NEC is a serious gastrointestinal condition typically seen in premature infants. Symptoms include abdominal distension, bloody stool, and lethargy. A number of studies have demonstrated that probiotic therapy reduced both the incidence and severity of NEC in a study of very low birth weight infants 1415.

IBD encompasses chronic inflammatory conditions of the gastrointestinal tract, characterized by unpredictable and spontaneous periods of remission and relapse. The most common types of IBD are ulcerative colitis (UC) and Crohn's disease (CD), with the prevalence of IBD estimated to be 1.4 million in the United States and 2.2 million in Europe 16. While the precise aetiology of IBD is unknown, genetic susceptibility, imbalances or disruption to the commensal host microflora 17 especially a reduction in the Firmicutes phylum 18 and an abnormal intestinal immune response 1920 are thought to play an important role in disease manifestation. Consequently, probiotics have been utilised in an attempt to re-establish the balance of the host microflora and attenuate an aberrant immune response. In a trial of UC patients investigating the effect of an oral capsule containing Bifidobacteria following treatment with sulfasalazine and glucocorticoids (a standard therapy for UC) it was found that 93.3% of patients in the placebo group suffered a relapse compared to only 20% in the probiotic group. A significant reduction in inflammation was also seen in those administered the probiotic compared to the control group 21. A recent study has shown promising potential for the use of Faecalibacterium prausnitzii as a probiotic with anti-inflammatory properties in the treatment of CD 22. This bacterium was found in lower numbers in patients with recurrent CD. F. prausnitzii and its supernatant was found to have anti-inflammatory effects both in vitro and in vivo, inducing interleukin 10 (IL-10) production in peripheral blood mononuclear cells (PBMC's), reducing IL-8 and NF-κβ (pro-inflammatory compounds) in Caco-2 cell lines and attenuating the severity of induced colitis in mice 22. There is, as yet, little evidence documenting the effectiveness of probiotics in the treatment of CD and as such further research is required. Shanahan 23 suggests that differences in the composition of the host microflora and the locations of CD-associated lesions respectively along the GI tract may indicate one probiotic strain is not sufficient to exert a beneficial effect in different patients. Furthermore, the author poses the question of whether researchers are using the correct probiotic, at a high enough dose and for the correct indication. Overall, only a limited number of studies are available and often results are conflicting, but there is sufficient evidence to warrant further research. Future studies need to be randomized, double-blind placebo-controlled trials encompassing large subject bases and possibly using combination therapies with more than one probiotic strain. Probiotics have also been used in the treatment and prevention of other gastrointestinal disorders including; irritable bowel syndrome (IBS) 24 and Helicobacter pylori associated infection (for a review see; 25), as well as in non-gastrointestinal conditions such as urinary tract infections (UTI's) 26 and atopic diseases 27.

Before bacteria can be considered for use as probiotics, it is recommended that they meet certain selection criteria and possess a number of intrinsic physico-chemical characteristics outlined in a joint report by the FAO and WHO in 2002 28. The report sets out a number of guidelines that should be followed so as to standardise procedures when assessing probiotics for use in foods and validating health claims thereof. The guidelines include: probiotic health effects are usually strain specific 29 so it is important to characterize probiotics to the strain level and with current nomenclature and subsequently deposit them in an international culture collection. It was recommended that in vitro tests be carried out before any subsequent in vivo animal or human trials were initiated. Furthermore, these tests require substantiation with in vivo performance. Common in vitro tests performed on probiotics include resistance to gastric acidity and bile, adherence to mucus and/or human epithelial cells, bile salt hydrolase activity, reduction of pathogen adherence to surfaces and antimicrobial activity against pathogenic bacteria such as Heliocbacter pylori, Salmonella sp., Listeria monocytogenes and Clostridium difficile. The report also outlines some safety considerations when dealing with probiotics. While probiotic bacteria, as a group, are generally regarded as safe (GRAS) organisms, safety tests should include the determination of antibiotic resistance profiles, evaluation of certain metabolic activities such as D-lactate production and bile salt deconjugation (an undesirable action in the small bowel), assessment of side effects in humans trials, post market epidemiological surveillance of adverse effects in consumers, toxin production and haemolytic activity. Furthermore, animal trials should be undertaken, where possible and appropriate, before commencing human studies. There have been very few reports of possible systemic infections attributed to probiotics 303132 and in those rare instances, all occurred in patients with serious underlying medical conditions. However, demonstration of lack of infectivity in immuno-compromised animals is also desirable to reinforce the safety profile of the probiotic strain. Large human clinical trials with probiotics are sorely lacking and the report recommends double-blind, randomized, placebo-controlled trials for humans with a sufficiently large participant base and preferably that the trial be repeated by an independent laboratory to confirm the outcome. The full report can be downloaded at: ftp://ftp.fao.org/es/esn/food/wgreport2.pdf.

Following oral administration, probiotics must survive transit through the gastrointestinal tract, facing host-associated stresses such as the low pH environment of the stomach (which can be as low as pH 1.5 when fasting) 33 as well as bile and elevated osmolarity in the intestine. Once in the gut they must be able to colonise and proliferate and exert a beneficial effect on the host. In addition to host-associated stresses, probiotics encounter technological stresses during processing, formulation and packaging and must survive in sufficient numbers for an extended period of time during their shelf-life, often at refrigeration temperatures. They should also possess good organoleptic properties and be phage resistant 34. Potential probiotic strains need to be both physiologically and technologically robust, but even the toughest strains are restricted in the variety of food applications to which they can be applied. In addition the most promising and clinically relevant probiotics are unfortunately often rendered unusable due to their physiological and technological fragility 34. Pre-exposing probiotic strains to stresses such as sodium chloride (NaCl), elevated temperature, bile and low pH can increase their survival and viability 3536. This pre-exposure to sub-lethal stresses can significantly increase their survival following subsequent exposure to lethal stress. It has been demonstrated that exposing Bifidobacterium adolescentis to 47°C for 15 minutes prior to a lethal heat shock increased the strain's heat tolerance 128-fold 37. Furthermore it has been shown that acid-resistant strains can be produced by subjecting acid-sensitive strains to prolonged exposure (16 hours) to pH2.0 38. The acid-resistant derivatives were seen to obtain a cross-protective effect from acid exposure, growing better in the presence of both bile and NaCl, whilst also having an increased fermentative ability and enzymatic activity. Such treatments however can result in a significant decrease in cell yield, as well as cellular activity and process volumetric productivity 39. A recent study compared the stress tolerance profiles of eight probiotic strains following microencapsulation challenged with acid, bile and heat stress 40. Microencapsulated strains survived three logs CFU/ml better than free probiotic control strains. Increased survival was also observed for acid and mild heat treatment in the microencapsulated strains. Other strategies such as immobilized-cell technology 41 have also been shown to enhance the stress tolerance profiles of certain probiotic strains. However further research is required before industry standards are reached 39 and research on technologies such as microencapsulation and its benefits on the stability and release of bioactive compounds in the gastrointestinal tract 42.

Whilst it has been clearly demonstrated that probiotics can be effective therapeutics in certain cases 48152126 results are often conflicting and can vary between and within individuals. This is due in part to different modes of action of probiotics as well as strain specific effects. Coupled with this, the increase in antibiotic resistance due to the indiscriminate use, overuse and misuse of antibiotics and the emergence of the so-called 'superbugs' (multi-antibiotic resistant strains), the need for novel and alternative therapies is paramount. The economic burden to the medical care sector in the United States for the treatment of patients with infections caused by antibiotic resistant organisms is estimated to be four billion $US annually 43. In addition, the fact that no new antibiotic classes have been discovered 44 and that pharmaceutical companies have severely reduced investment and in some cases, completely abandoned antimicrobial research and development 45, reinforces the point that radically new and innovative therapies are urgently needed. The design, creation and genetic modification of probiotic stains exclusively tailored to target a specific pathogen or toxin thereof, or as vaccine and drug delivery vehicles is a promising and rapidly expanding area of research.

In support of this, lactobacilli possessing desirable properties such as inherent immunogenicity, bile resistance and the ability survive and proliferate in the gastrointestinal tract 46, have been successfully employed as oral, mucosal vaccine delivery vehicles. Such vaccine delivery vehicles offer an advantage over traditional live attenuated pathogenic strains in that there is no possibility of reversion to a virulent phenotype which always remains with the attenuated pathogenic strains 46. The majority of pathogens initiate their initial infection at a mucosal surface and vaccination against such pathogens is best accomplished through mucosal vaccination. Mucosal vaccines offer a number of functional advantages 47 as well as practical benefits; they are non-invasive, easy to administer and do not require the presence of medically trained personnel, a significant advantage in the developing world 48. Lactobacilli expressing Tetanus toxin fragment C (TTFC) have been shown to be able to elicit a positive immune response 49. Such a strategy could be applied to the expression of other toxins from enteric pathogens, thus providing immune protection. In addition, a recombinant Lactococcus lactis strain producing human IL-10 has been investigated in a human clinical trial for treatment of Crohn's disease 50.

As previously mentioned, diarrheal diseases are responsible for significant morbidity and mortality worldwide and treatment is becoming increasingly more difficult due to the rise in antibiotic resistance. Paton and co-workers described the development of designer probiotics for the prevention of gastrointestinal infections using a strategy involving the expression of host cell receptor-mimics on the surface of probiotic strains which can bind to the pathogen itself or neutralize and mop up secreted toxin 5152 (Figure 1). This approach has a number of advantages; (i) the probiotic can be administered orally (ii) numerous human receptors recognized by enteric pathogens or their toxins are well characterized (iii) preventing pathogen adherence would inhibit development of infection (iv) sequestration of a toxin would prevent clinical presentation of symptoms so the pathogen can be removed by the host immune system and possibly most importantly (v) this therapeutic strategy does not apply a selective pressure on the pathogen with the of development of resistance extremely unlikely. Resistance usually results from a reduction in bacterial cells numbers due to the selective pressure and this does not occur with receptor mimic therapy.

Research in the same lab has resulted in the construction of a harmless Escherichia coli strain engineered to express two galactosyl-transferase genes from Neisseria gonorrhoeae. This resulted in a modified lipopolysaccharide (LPS) that mimics the Shiga-toxin (Stx) receptor and was found to effectively bind both Stx1 and Stx2. The engineered bacteria were shown to be 100% effective in treating mice infected with normally fatal shigatoxigenic E. coli (STEC) 53. In another study by the same research group, a probiotic was developed for the treatment of travellers' diarrhea caused by enterotoxigenic E. coli (ETEC). Using a similar strategy to that mentioned above, an E. coli strain was engineered to produce a chimeric LPS receptor mimic capable of binding a heat-labile enterotoxin. In vitro tests showed more than 93% of the toxin could be neutralized by the recombinant strain, while in vivo studies showed the recombinant strain provided protection against fluid loss due to the toxin in rabbit ligated ileal loops 54. A molecular mimicry strategy was also used to create a recombinant probiotic for the treatment and prevention of cholera. Cholera, caused by the bacterium Vibrio cholerae, is a serious intestinal infection characterized by severe, watery diarrhea resulting in rapid fluid loss and dehydration which can lead to death in just a few hours without treatment. Transmission is via the fecal-oral route, usually due to consumption of contaminated drinking water and the disease is epidemic in many developing countries. Current treatment is administration of a standard oral rehydration solution (ORS) to replace lost fluids, salts and electrolytes in combination with antibiotics in some cases. This is usually an effective treatment but must be administered promptly following infection. A similar strategy was used to design a probiotic strain with an altered LPS which terminates in a structure that mimics the G_M1 ganglioside terminus, which is the binding receptor for cholera toxin (CT) 55. The bacteria were found to be able to bind CT, abolish >99% of its cytotoxicity and absorb more than 5% of its own weight of toxin. Murine studies demonstrated that administration of the recombinant probiotic immediately following infection with V. cholerae increased survival rates of the mice. Furthermore, when administration of the probiotic was delayed until after establishment of V. cholerae infection, all mice in the probiotic group survived compared to 1 out of 12 for the control group 55. Such powerful and significant results reinforce the potential of such alternative therapies. Further research to improve and fine-tune this approach may be necessary before such therapies can be applied to humans. For example it has been suggested that the introduction of genes to aid gastric transit or promote improved gut colonization would reduce dose regimes and thus costs, while the use of food grade bacteria such as lactococci and lactobacilli for receptor mimicry may also be possible 51. As mentioned earlier, evidence is lacking for the recommendation of probiotics for use in the treatment of C. difficile. However designer probiotics may offer a viable alternative 56. The bacteriocin lacticin 3147 has been shown to have significant antibacterial activity against C. difficile 57 however it is acutely sensitive to gastric acidity in vivo 58. This sensitivity might be overcome by cloning bacteriocin production (and resistance) genes into a suitable host, such as a lactobacillus which could survive stomach passage and deliver the intact bacteriocin to the point of infection. Furthermore, designer probiotics meet the criteria laid out by McFarland 59 for novel approaches to manage C. difficile 56606162. With regard to inflammatory bowel disease, there have been some studies involving recombinant probiotics with some good preliminary results being seen in the treatment of induced- colitis in animal models 606162. However, more information is needed on the exact causes of such conditions before suitable treatments can be fully developed for use in human trials.

The term patho-biotechnology was coined to describe the concept of exploiting pathogenic bacteria or more precisely, exploitation of their stress adaptation, host evasion and virulence or virulence-associated characteristics for beneficial use in the biotechnology and food industries and in medicine 63. The patho-biotechnology concept encompasses a number of different areas. Firstly, it involves the use of pathogens such as Listeria monocytogenes as novel vaccine and drug delivery vehicles 64. This may be approached either by the use of conditional auxotrophic mutants or the selective elimination of key virulence factors. Secondly, such a strategy may involve the isolation of certain immunogenic proteins from specific pathogens thus removing the necessity of using the pathogen itself as the carrier vehicle 65. The final area applicable to the patho-biotechnology approach and the main focus of this chapter is the introduction of stress survival genes from pathogenic bacteria into non-pathogenic probiotic strains 63. Probiotic microorganisms encounter an identical set of stress conditions as pathogens upon encountering a host (Figure 2). Thus from a human point of view, an undesirable element from a pathogen (e.g. genes that aid survival in stressful conditions such as gastric acid, bile, low iron, increased osmolarity) if introduced to a probiotic could prove to be beneficial by increasing its resistance to host-associated stresses as well as its technological robustness and clinical efficacy, a distinct advantage with many potentially promising probiotic strains 34.

L. monocytogenes serves as an ideal candidate for the patho-biotechnology concept 66. Its genome has been fully sequenced 67, L. monocytogenes is amenable to genetic manipulation 68 while physiologically it is a robust pathogen capable of resisting numerous stresses 69 while also eliciting a strong host immune response 70.

The patho-biotechnology approach has been successfully employed to increase the stress tolerance of the probiotic bacterium Lactobacillus salivarius UCC118 71. L. salivarius has been shown to have desirable therapeutic properties 72 and has recently been shown to protect mice from L. monocytogenes infection by the production of the bacteriocin, Abp118 73. However the bacterium is technologically fragile. In an effort to improve the physiological robustness of the strain the betL gene of L. monocytogenes (encoding the compatible solute betaine uptake system, BetL) was cloned into L. salivarius UCC118 71. Bacteria accumulate compatible solutes such as betaine, either from their environment or by de novo synthesis, to counter osmotic stress. BetL has been shown to increase L. monocytogenes tolerance to salt 74, low temperature 75 and pressure stress 76, as well as increasing viability in certain foods 77. Thus, it was postulated that the introduction of betL into L. salivarius UCC118 might improve the strain's tolerance to a number of stresses. Indeed the strain harbouring the betL gene (betL⁺) showed a significantly higher growth rate at 7% NaCl compared to the control strain lacking bet (betL^-) (Figure 3). Also, when exposed to low temperature stress, betL⁺ survival was 2 logs greater at -20°C and 0.5 logs greater at -70°C compared to wild type strain 71.

Furthermore, significantly higher survival rates for betL⁺ were observed following both freeze and spray drying treatment compared to the control. Higher survival rates were also observed following high pressure processing (300 and 350 MPa). These results clearly demonstrate the potential of patho-biotechnology for improving the technological robustness of probiotic microorganisms. In addition to improving a strain's resistance to stresses encountered during food manufacture and storage, the patho-biotechnology concept has also been applied to tailor improved probiotic resistance to host specific stresses, significantly improving colonization, persistence and clinical efficacy. In support of this, the betL gene from L. monocytogenes was cloned into the probiotic bacterium Bifidobacterium breve UCC2003 78. B. breve UCC2003 strains expressing betL were shown to exhibit significantly increased tolerance to simulated gastric juice (pH 2.5) as well as osmotic stress. In addition, following successful colonization of the murine intestine, B. breve UCC2003 betL⁺ strains were recovered at significantly higher levels in the faeces, large intestine and caecum of inoculated mice. Finally, the B. breve UCC2003 betL⁺ strain was shown to protect against L. monocytogenes infection in a murine model. L. monocytogenes was recovered in significantly lower numbers from the spleens of mice fed with B. breve UCC2003 betL⁺ compared the control strain (which lacked betL) 78. To the best of our knowledge this study provides the first direct evidence for improved therapeutic efficacy using a patho-biotechnology based approach.

Following these initial experiments, subsequent research was carried out in relation to improving the bile tolerance of two common probiotic strains. The authors cloned the bile exclusion system, BilE (genes bilEA and bilEB) from L. monocytogenes into both L. lactis NZ9000 and B. breve UCC2003 79. The BilE system functions as a cellular bile exclusion system and aids gastrointestinal transit and persistence in L. monocytogenes 80. It was therefore postulated that heterologous expression of the BilE system would increase bile tolerance and subsequent gastrointestinal persistence of the probiotic strains. Indeed, the authors found that both L. lactis and B. breve expressing bilE exhibited a 2.5 logs greater survival rate compared to the wild type when grown in porcine bile at concentrations similar to that found in the intestine. Also, both bilE⁺ strains persisted for longer in the gastrointestinal tract and murine faeces (Figure 4A), while B. breve bilE⁺ was recovered at significantly higher numbers (2 logs greater) directly from the murine intestine compared to the wild type (Figure 4B). Furthermore B. breve bilE⁺ significantly reduced L. monocytogenes numbers in the liver following oral infection with the pathogen. These results are significant in that increased bile tolerance also conferred increased gastrointestinal persistence which may enhance probiotic efficacy in therapeutic models 79.

An alternative approach to using pathogen-derived genetic elements to improve the physiological robustness of potential probiotics involves a relatively new and rapidly expanding area of scientific research; metagenomics.

Metagenomics is the culture-independent analysis of the collective genomes of a population of microorganisms. The field of metagenomics is relatively new, with the first international conference on the topic being held in 2003. A metagenomic analysis involves the functional and sequence-based analysis of collective microbial genomes in an environmental sample 81. Metagenomics can be used to study as yet uncultured microbes, which represent more than 99% of the population in some environments. The total number of prokaryotic cells on earth has been estimated at 4-6 × 10³⁰ 82, with the majority of these remaining unknown to science. This diversity represents a vast genetic pool that may be utilized for the discovery of novel genes, entire metabolic pathways and their products 83. Functional metagenomic analysis is based on the construction of clones containing metagenomic DNA in a surrogate host and subsequent screening of the clones for the expression of a desired trait or phenotype. Functional metagenomics has identified novel antibiotics 84, Na⁺/H⁺ anitporter membrane proteins 85, esterases 86, proteases 87 and lipases 88. Sequence based metagenomics can involve the complete sequencing of clones containing phylogenetic anchors that indicate the taxanomic group that is the probable source of the DNA fragment or it can involve the random sequencing of clones to identify a gene of interest and subsequent the search for phylogenetic anchors in the flanking DNA, which can provide a link of phylogeny with the functional gene 89. Sequence based analysis has been used to identify a bacterial rhodopsin gene; the first evidence that rhodopsins are not exclusive to Archaea 90. Sequenced based approaches have also been used to identify novel oxidative coupling enzymes 91, chitinases 92 and a novel fibrinolytic metalloprotease 93 to name a few.

Research in our laboratory has focused on the human gut metagenome. The human distal gut is the highest density natural bacterial ecosystem known and the number of bacterial cells on or in our bodies is estimated to be 10 times greater than the number of human cells 94. Furthermore, the number of genes in the representative species probably exceeds the number of human genes by 100-fold 95. In this respect it has been suggested that we as humans should be considered a superorganism in symbiosis with our vast microflora 96. This microbiome is largely untapped and contains a virtually limitless supply of novel genes to be exploited for use in medicine, science and industry. Exploiting or 'mining' the human gut metagenome for the development of novel therapeutics and designer probiotics, is essentially an extension of the patho-biotechnology concept. Considering these genes would be isolated from commensal species they are in essence 'self' genes and may alleviate some concerns regarding the use of genetic elements from pathogenic species. 'Mining' the intestinal flora to discover novel antimicrobial peptides, immunoregulatory molecules and stress tolerance genes for therapeutic purposes, is a promising and exciting area of research and a logical extension of the patho-biotechnology concept 97. We thus coin the term 'meta-biotechnology' to describe the use of metagenomics as a robust tool to identify novel genes for the use in biotechnology, specifically building on the patho-biotechnology concept, by increasing the technological and physiological robustness and clinical efficacy of probiotic bacteria.

As mentioned above our research has focused on the human gut metagenome and more specifically the identification of novel osmotolerance loci used by bacteria in this environment. A metagenomic library of the human gut microbiome was screened to identify clones with increased osmotolerance compared to host strain. Representative clones were subsequently subjected to transposon mutagenesis in an attempt to disrupt osmotolerance loci. We have identified a number of such genes and current work is focused on testing their potential to increase osmotolerance in other bacterial strains. The ability to cope with fluctuations in environmental osmolarity is key to the survival and viability of all microorganisms and especially so to both bacteria transiently moving through the human gastrointestinal tract and to those who permanently colonize and proliferate in that environment. Mechanisms of osmotolerance have been well characterized for the pathogenic bacterium L. monocytogenes (for a review see 98) and recently, osmolyte transporters have been shown to increase bile resistance 99, an important factor in gastrointestinal persistence. We hope that the identification of novel osmotolerance loci from our host microflora using a functional metagenomic approach will enable us to create probiotic strains with an increased physiological and technological robustness, as has been demonstrated previously with genes from the pathogen L. monocytogenes 717879.

The major disadvantage with designer probiotics is that they are genetically modified organisms (GMOs) and as such their use in the treatment of humans would essentially constitute the deliberate release of such a GMO into the environment. Therefore the safety of such strains needs to be guaranteed and stringently monitored so that they do not; (i) possess antibiotic selection markers, (ii) have the ability to accumulate in the environment and (iii) transfer the genetic modification to other bacteria through lateral dissemination. Biological containment can be divided into active and passive forms. Active containment is conditionally controlled by the production of a compound that is toxic to the cells. Gene expression is tightly regulated and controlled by an environmentally responsive element. Passive containment meanwhile, is dependent on complementation of an auxotrophy by supplementation with either an intact gene or the essential metabolite (for a recent review see; 100. Perhaps the most elegant method developed to date, which combines the advantages of both active and passive containment, is that of the thymidylate synthase gene (thyA) 101. The thyA gene form L. lactis, which is essential for growth, was replaced with the expression cassette for human interleukin-10 (hIL-10) by double crossover using a non-replicative plasmid (pORI19). An indigenous suicide system is induced by the activation of the SOS repair system and subsequent DNA fragmentation resulting from thymine and thymidine auxotrophy. Because thymine or thymidine is essential for growth, the thyA-deficient strain cannot accumulate in the environment in the absence of the essential growth factors. This approach deals completely with the biosafety issues mentioned above, in that no resistance marker is required to guarantee stable inheritance of the transgene, thus removing any concerns regarding the spread of antibiotic resistance. Also, the risk of the GMO accumulating in the environment is negligible due to rapid death in the absence of thymine or thymidine. Finally, in the unlikely event that a functional thyA gene is acquired by homologous recombination from a closely related bacterium, the transgene would be lost 101. One of the strains developed using this method was approved in The Netherlands for the use in a human clinical trial for the treatment of IBD. This was the first clinical trial in which live genetically modified bacteria were used as bio-therapeutics in humans 50.