Cholera, a waterborne, life-threatening, gastrointestinal infection, is caused by the Gram-negative bacterial pathogen Vibrio cholerae and affects millions of individuals annually with an estimate of 21,000 to 143,000 deaths recorded worldwide (Zuckerman et al., 2007; Ali et al., 2015; Cho J. Y. et al., 2021). It is characterized by watery diarrhea and vomiting which leads to rapid dehydration, hypovolemic shock, acidosis and death (Ali et al., 2015; Hsiao and Zhu, 2020). V. cholerae colonizes the epithelium of the small intestine where it can form a resistant biofilm, produces cholera toxin and toxin-coregulated pili and employs a type 6 secretion system to kill competing native gut bacteria (Faruque et al., 1998; Zhu and Mekalanos, 2003; Logan et al., 2018).

Cholera toxins are responsible for altering the hosts cellular signaling pathways causing cellular damage and the watery diarrhea characteristic of cholera; while the toxin-coregulated pili aids in colonization of the gut epithelium (Herrington et al., 1988; Cho J. Y. et al., 2021). Current treatments for cholera are centered around antibiotic therapy and while these have dramatically reduced case fatality, resistance to antibiotics has been demonstrated in cholera endemic and epidemic countries creating a need for new antimicrobial strategies (Clemens et al., 2017). Some natural probiotic microbes can aid in V. cholerae infections. For example, oral administration of Lactobacillus acidophilus and Bfidobacterium spp. in mouse models that were initially infected with cholera has shown to have enhanced mucosal and systemic immune response to the cholera toxin (Tejada-Simon et al., 1999).

Aside from antibiotic treatment, vaccination is considered key to controlling the spread of cholera (World Health Organization, 2018). Beyond the potential of their native therapeutic properties, probiotic microbial cells offer a potential vehicle for vaccine delivery to the gastric mucosa (Gardlik et al., 2012; Iqbal et al., 2021). The budding yeast Saccharomyces boulardii has properties including acid tolerance and good growth at 37°C that allow it to be employed as a probiotic (Sazawal et al., 2006; Czerucka et al., 2007; Kelesidis and Pothoulakis, 2012). This yeast has been engineered to express the V. cholerae toxin coregulated pilin gene tcpA, with the objective of inducing a pre-emptive immune response to inhibit gut colonization by V. cholerae (Awad et al., 2020). Although no data are available on whether ingestion of the probiotic can induce a strong mucosal immune response to V. cholerae, it is relevant that S. boulardii can be engineered to either secrete foreign antigens or display them on the cell surface (Wang et al., 2014; Li et al., 2021). Oral vaccination using recombinant probiotics as the delivery system could thus be an effective means to activate both the innate and adaptive immune responses of the host to reduce the risk of transmission of foodborne infections (Mathipa and Thantsha, 2017).

Many bacterial pathogens including V. cholerae employ a quorum sensing system (QS) to induce colonization, biofilm formation, virulence and broad dissemination in the host’s gastrointestinal tract (Hsiao and Zhu, 2020). QS is a signaling mechanism that bacteria use to respond to chemical hormone-like molecules called autoinducers, secreted by bacteria of the same species (Reading and Sperandio, 2006). The soluble QS molecule is of necessity unique to the pathogen and thus offers a target to allow detection of the pathogen. The V. cholerae quorum sensing system has been transplanted into E. coli and linked to a pathway to induce expression of a guide RNA that targeted dCAS9 to inhibit expression of a GFP reporter gene. In the absence of the V. cholerae quorum sensing molecule CA1, the GFP reporter was silenced. Upon binding of CA1 the expression of the gRNA was repressed leading to activation of the GFP reporter signal. This biosensor was extremely sensitive and allowed detection of V. cholerae (Holowko et al., 2016). The same group further engineered the E. coli biosensor and constructed a sense-and-kill suicide microbe (Jayaraman et al., 2017). Upon detection of the V. cholerae quorum sensing molecule a synthetic circuit was activated that induced expression of a synthetic endolysin that was exported to the periplasm leading to lysis of the biosensor and releasing the endolysin, leading to killing of the V. cholerae cells. A similar synthetic biosensor circuit was constructed in Lactococcus lactis which permitted effective detection of V. cholerae in infected mice (Mao N. et al., 2018). L. lactis has a natural capability to inhibit V. cholerae through secretion of lactic acid to acidify the local environment and consumption of a native strain is sufficient to reduce cholera-induced morbidity in mice (Mao N. et al., 2018). In this case the biosensor circuit was not linked to any mechanisms to kill or inhibit V. cholerae but this is a direction that would likely yield a powerful therapeutic tool (Mao N. et al., 2018). The power of such a synthetic circuit lies in the specificity of response, allowing the sense-and-kill strain to reside in the gut and only respond upon detection of the pathogen.

Signalling molecules of the quorum sensing pathway are critical to regulating the expression of toxins and other virulence genes by V. cholerae. These signaling molecules provide not only a biomarker to allow detection of the pathogen but provide a potential means to disrupt induction of virulence factors. An engineered E. coli strain expressing and secreting the V. cholerae autoinducer CA1 was able to disrupt toxin expression in mice inoculated with V. cholerae, leading to a significant increase in survival (Duan and March, 2010). An additional strategy to reduce the harm done by cholerae toxin has been to engineer a probiotic E. coli to express a surface receptor for cholera toxin. This probiotic could significantly reduce the amount of cholera toxin binding to the gut wall and improve recovery in a mouse model of V. cholerae infection (Focareta et al., 2006).

The concept of recombinant probiotics improving the inherent functions of natural probiotic strains, adding a surveillance attribute, being used as a drug or vaccine delivery system, or being “weaponized” to neutralize pathogen virulence factors, is a promising direction for the development of new therapies intervening with the pathogenic action of V. cholerae.

Staphylococcus aureus are a species of potentially pathogenic bacteria that are widespread in the environment. They commonly infect the blood, respiratory and most notably, the gastrointestinal tract of affected hosts. Staphylococcus infection is also one of the primary causes of community and hospital acquired infection, often resulting in severe and sometimes life-threatening conditions (Díaz et al., 2021). It is estimated that staphylococcal food poisoning causes approximately 241,188 illnesses and 1,064 hospitalizations in the United States alone (Scallan et al., 2011). S. aureus poses a serious public health hazard owing to the ability to secrete heat resistant enterotoxins that trigger inflammatory responses and the ability to form highly resistant biofilms that resist the host defenses, decrease the efficiency of antimicrobial factors, and provide resistance to antibiotics (Otto, 2013; Bintsis, 2017; Paik et al., 2019; Díaz et al., 2021). This pathogen infects a host and employs a quorum sensing signaling pathway to induce expression a collection of virulence factors including toxins, enzymes and secreted biofilm components that allow it to colonize and become established in the host.

The rapid appearance of antibiotic resistance in this microbe presents a challenge for developing antimicrobial therapies to treat and prevent infections. S. aureus acquires mutations at a rate that is about 1,000 times faster than for E. coli (Pipiya et al., 2020). This has allowed for the emergence of the notorious methicillin-resistant (MRSA) strains that are resistant to the entire β-lactam antibiotic class and created the need for improved therapies for detection and clearance of S. aureus. This is where bioengineered probiotics may have a pivotal role.

Several probiotic organisms including lactobacillus species, Bacillus subtilis and S. boulardii have demonstrated inhibitory activities toward S. aurerus though competitive exclusion, inhibiting growth by metabolite secretion, inhibition of biofilm formation and disruption of quorum sensing (Saidi et al., 2019; Nataraj and Mallappa, 2021). These antimicrobial activities coupled to the more general probiotic properties of reducing inflammation and improving epithelial barrier function make these organisms excellent candidates for engineering to improve upon their existing capabilities and to add more specific antimicrobial functionalities (Mathipa and Thantsha, 2017).

One way to improve the function of favorable probiotic strains is to make them more competitive within the gut environment. While these strains are all naturally tolerant of the acid conditions of the stomach and the routine 37°C temperature, one place that an advantage can be given is through bestowing resistance to the natural population of bacteriophage that populate the gut. Resistance to phage populations might permit a desirable probiotic strain to more effectively compete with pathogens and opportunistic pathogens that invade the human gut. Nagarajan et al. employed mutagenesis and a selective resistance procedure to isolate a probiotic bacterium, Lactobacillus plantarum, that was resistant to a phage population. In vitro competition tests between S. aureus and the selected L. plantarum strain demonstrated that in the presence of a phage population the probiotic was able to out compete S. aureus thereby reducing the ability of the pathogen to adhere to and colonize the epithelial cell layer (Nagarajan et al., 2019). This proof of principle experiment clearly has limited utility in that the probiotic was selected for resistance to the phage population from a specific environment and would likely be susceptible to phage from other environments. Additionally, given the high rate of mutagenesis in bacteriophage it is likely that phage capable of lysing the probiotic would readily emerge. None-the-less improving the competitiveness of favorable probiotic strains is a notion worthy of further investigation. Bacterial species with fully engineered and recoded genomes have been produced to eliminate redundant codons (Fredens et al., 2019). These are fully resistant to phage infection as they lack tRNA required for phage gene expression (Robertson et al., 2021). Application of this technology to a probiotic strain could give that strain a competitive advantage, allowing it to easily outgrow opportunistic pathogens in the gut.

Relatively simple genetic circuits have been engineered into probiotic strains to detect and in some cases respond to S. aureus infections. The probiotic bacteria strain Lactobacillus reuteri is effective in reducing S. aureus viability in a host through the mechanism of competitive exclusion, biosurfactant secretion and production of lactic acid (Sikorska and Smoragiewicz, 2013; Wang et al., 2015). An L. reuteri strain that expressed the S. aureus AgrA and AgrC genes to provide a receptor and response pathway for detection of the AIP1 signaling molecule was constructed such that activation of AgrA. would repress expression of a β-glucuronidase gene to allow colorimetric detection (Lubkowicz et al., 2018). This mimic system can detect the presence of the autoinducer at very low concentrations and could be useful for high-throughput quantitative and qualitative detection of S. aureus, especially in hospital settings (Lubkowicz et al., 2018). This method is also cost-effective in comparison to current mass spectrophotometry and DNA sequencing approaches. The probiotic system can also be fine-tuned as a chassis for the eradication of S. aureus.

Several probiotic organisms described above are capable of competing with S. aureus in the gut environment. In an effort to further improve the utility of probiotics to impeded S. aureus infections several groups have sought to “weaponize” probiotic strains through engineering them with genes encoding antimicrobial factors with specificity for the target pathogen. Bacteriocins are proteinaceous compounds produced by bacteria to facilitate anti-microbial activity against other bacteria (Bastos et al., 2010). Lysostaphin is a metallo-enzyme that is relatively specific for the cell wall structures of Staphylococcal species (Kokai-Kun et al., 2007; Bastos et al., 2010). P. pastoris strains that have been engineered to produce lysostaphin displayed efficient eradication of S. aureus during in vitro co-culture experiments (Pipiya et al., 2020). Despite the effectiveness of lysostaphin to kill S. aureus in vitro infections in the human gut pose a greater challenge owing to biofilm formation by the colonizing S. aureus.

A probiotic strategy to deal with S. aureus infections that has emerged is the use of engineered Mycoplasma pneumoniae. This organism has a small genome, is genetically manipulable, has an unusual codon usage and limited recombination that limits horizontal gene transfer (Osawa et al., 1990; Himmelreich et al., 1996; Krishnakumar et al., 2010; Sluijter et al., 2010). Additionally, it lacks a cell wall and so has a limited ability to trigger an inflammatory response in mammalian hosts (Sukhithasri et al., 2013). An M. pneumoniae with reduced virulence owing to deletion of the mpn133 and mpn372 genes was engineered to secrete dispersin B, a glycosyl hydrolase that can efficiently dissolve biofilms formed by S. aureus, and the bacteriocin lysostaphin (Garrido et al., 2021). The engineered strain was effective in disrupting and reducing S. aureus biofilm development in a catheterized mouse model as well as eliminating biofilms formed ex-vivo (Garrido et al., 2021).

Staphylococcus aureus remains a challenging pathogen for medicine to control owing to its ability to rapidly acquire drug resistance, the tenacious nature of its protective biofilms, ability to evade the immune system and ability to invade a variety of tissues. Clever genetic engineering of favorable host organisms to act as biosensors and “hunter-killers” that can disrupt biofilms may become a key tool that can be combined with other therapies including the use of engineered bacteriophage to combat this common microbial pathogen.

Clostridia species are anaerobic organisms prevalent in soil. Several of these are well known human pathogens including C. tetani, C. difficile and C. perfringens and can induce disease in humans and agricultural animals (Popoff and Bouvet, 2009; Onderdonk and Garrett, 2020). The ability of these organisms to form spores that are resistant to heat and many common cleaning agents makes them challenging to control (Augustyn et al., 2022). C. perfringens is common within ecosystems and frequents the gastrointestinal tracts of livestock and poultry (Abudabos et al., 2019; Bai et al., 2020). It is responsible for a range of severe conditions, including food poisoning, gas gangrene, respiratory infections and necrotizing enteritis (NE) in both humans and animals (Lindström et al., 2011; Gervasi et al., 2014a). C. perfringens ranks second among the causes of foodborne illness and is responsible for greater than one million illnesses each year in the United States alone (Bintsis, 2017).

C. perfringens infection has been associated with the development of necrotizing enterocolitis (NEC), a gastrointestinal disorder affecting pre-term infants with a mortality rate of about 20% (Duchon et al., 2021). This pathogenic species is also the leading cause of necrotic enteritis (NE) in farm animals, costing the global poultry industry over two billion dollars annually, mainly due to the high prices of antibiotic treatments and therapeutic feed supplementation (Van Immerseel et al., 2009; Paiva et al., 2014; Smith et al., 2014). C. perfringens, like S. aureus, employs a quorum sensing system and secretes a battery of adhesins, proteolytic enzymes and toxins allowing it to colonize and invade the host (Pruteanu and Shanahan, 2013; Kiu and Hall, 2018; Navarro et al., 2018; Bai et al., 2020). They are also able to secrete bacteriocins that inhibit some competing bacteria to increase availability of nutrients and clear space to colonize (Mora et al., 2020).

Antibiotic drugs are the most common means to treat and prevent C. perfringens infection. However, the growing number of antibiotic resistant pathogenic strains found in isolated livestock and agricultural sources suggest their use may not be beneficial in the long term (Slavić et al., 2011). Indeed, up to 33% of isolates from some meat samples harbor C. perfringens with resistance to the common antibiotics: ampicillin, tetracyclin, amoxicillin, ciprofloxiacin and chloramphenicol. Further multidrug resistance was found in 38% of those isolates and multidrug resistant C. perfringens has been isolated from livestock (Zhang et al., 2021; Hassani et al., 2022). Vaccination processes that have been successful against C. tetani have also been developed for C. perfringens but have yet to become widely and successfully established (Alizadeh et al., 2021). The growing health and economic threat posed by resistant forms of C. perfringens is creating an urgent need to develop new approaches to controlling infection by this pathogen that are effective, affordable and can be widely disseminated.

The requirements for an effective chassis for a live cell therapeutic would include the ability to thrive in the gut environment, being genetically manipulable, controllable, exhibit limited horizontal gene transfer and have intrinsic properties that make it competitive to the pathogen. A variety of probiotic organisms including Lactobacillus, Bifidobacterium, S. cerevisiae and S. boulardii, have been demonstrated to compete with or inhibit the growth of C. perfringens in vitro and in vivo (Schoster et al., 2013; Guo et al., 2017; Chuang et al., 2019; Gong et al., 2020; Khalique et al., 2020; Wang et al., 2020). A limited number of probiotic species have been demonstrated to secrete bacteriocins that inhibit C. perfringens (O'Shea et al., 2009; Heo et al., 2018; Golneshin et al., 2020; Lee et al., 2021). These organisms and their bacteriocins may be exploited in the future as tools to inhibit opportunistic C. perfringens infections.

Extensive effort has been expended on engineering probiotic organisms to express C. perfringens virulence factors with the aim of generating effective live cell vaccines. Delivering toxins to humans or animals as a vaccine has inherent risks even if those toxins have been inactivated with heat or formaldehyde. In contrast, toxoids mimic toxins released by the pathogens but are not associated with virulence. A live cell expressing target antigens has the potential benefit of inducing immunity at the level of the gut mucosa and provides a mechanism that allows extended exposure to the toxoid antigen rather than single doses. Probiotic Bacillus subtilis has been engineered to produce, in cells, or display on spore walls, a fragment of C. perfringens toxin-α fused to gluthatione S-transferase. Mice immunized with toxoid displaying spores developed immunity and were protected against lethal doses of the toxin (Hoang et al., 2008). This approach provided immunity to only a single toxin but it could clearly be extended to cover other toxins. Additionally, the use of spores as a display and delivery vehicle provides a stable vaccine that could easily be produced in large volumes.

The probiotic bacteria Lactobacilli casei has been engineered to express a C. perfringens α-toxoid mimicking the α-toxin. Mice vaccinated with the probiotic developed anti-α toxin antibodies and survived challenge with a lethal dose α-toxin while the control animals rapidly succumbed to the effects of the toxin (Alimolaei et al., 2018; Song et al., 2018; Gao et al., 2019). Oral administration of an ε-toxoid expressing L. casei strain or a β-toxoid expressing L. casei strain was also able to induce an effective antibody response to the toxin (Alimolaei et al., 2016; Alimolaei et al., 2017). Similar results were found by Bai et al. and Zhao et al. who also engineered L. casei to express toxoids mimicking the C. perfringens α, ε, β1, and β2 toxins and vaccinated rabbits with the recombinant probiotic. Challenge experiments demonstrated that this strategy yielded an 80% protection rate, even more effective than vaccinating with inactivated, whole C. perfringens cells (Zhao et al., 2017; Bai et al., 2020). An attenuated Salmonella typhimurium expressing C. perfringens antigens NetB, α-toxoid and Fba administered to chickens induced both serum and mucosal antibody responses and provided partial protection from necrotizing enteritis intestinal lesions (Wilde et al., 2019). A similar investigation using a self-lysing S. typhimurium vector expressing α-toxoid and NetB administered to broiler chickens yielded induction of IgA, IgY, and IgM antibodies against the toxins and provided significant protection from a subsequent C. perfringens infection (Jiang et al., 2015). The yeast S. boulardii was engineered to secrete a fusion protein including the C. perfringens enterotoxin CPE. Oral administration of this live cell vaccine to mice induced both IgG and IgA antibody responses suggesting that S. boulardii may be an effective vector for application as a live cell vaccine (Bagherpour et al., 2018). Additionally, a “disarmed” C. perfringens strain lacking the NetB toxin, when orally administered to vaccinate broiler chickens, produced a significant protective effect and reduced the incidence of necrotizing enteritis lesions upon challenge with virulent C. perfringens (Mishra and Smyth, 2017). Recently, a native microbial strain isolated from poultry, Ligilactobacillus agilis La3, has been engineered to effectively produce C. perfringens NetB (Vezina et al., 2021). Although its effectiveness has yet to be characterized, it is likely that a native member of the poultry gut microbiota will be an effective delivery vehicle.

Metabolic engineering of probiotics has also been directed toward generating live cell therapeutics capable of not just competing with C. perfringens but able to attack and lyse C. perfringens to promote their clearance. This work has borrowed from the highly effective endolysin enzymes produced by bacteriophage to penetrate the cell walls of infected host cells to allow phage release (Schmelcher et al., 2012). Bacteriophage derived endolysins specific for binding to and degrading the cell walls of C. perfringens and have been identified and demonstrated to be effective for killing those organisms in vitro (Gervasi et al., 2014a). Additionally, protein engineering has been applied to improve activity and increase specificity of C. perfringens killing (Swift et al., 2015). Clostridia and other pathogenic bacteria can acquire resistance to antibiotic drugs through modification of or degradation of the drug or the use of efflux pumps. Endolysins specifically bind to and cleave highly conserved peptidoglycan structures that are essential to the integrity of the cell wall (Fischetti, 2010). Acquiring resistance to an endolysin would require the cells to develop changes in critical conserved cell wall structures that would likely reduce the integrity of that structure and so is expected to be a very rare event (Fischetti, 2006).

Synthetic biology techniques have been applied to engineer the probiotic bacteria, Lactobacillus johnsonii, secrete an endolysin against effective against C. perfringens. They found that this led to an improved ability of the probiotic bacteria to lyse C. perfringens cells and increased clearance of the pathogen in plate assays (Gervasi et al., 2014a; Gervasi et al., 2014b; Cho J.-H. et al., 2021). S. cerevisiae have been engineered to effectively display active C. perfringens-specific endolysins on their cell walls (Ritter and Hackel, 2019). These investigators used the yeast display system to engineer the endolysin for improved stability but they did not test them for C. perfringens killing. Such a system could be potentially employed with a probiotic yeast for application in vivo. While engineered C. perfringens killing probiotics have yet to be proven in vivo they display the potential to reduce pathogen load and aid in control of serious infections. Overall, these models demonstrate that probiotics can be manipulated to produce anti-microbial factors and directly promote the clearance of the pathogenic bacteria C. perfringens.

Clostridioides difficile, like C. perfringens, is an anaerobic Gram positive bacteria that is widely disseminated in the environment. It is unclear how prevalent C. difficile is within the gut microbiome of human populations but it can be isolated from the gut of new born and healthy children (Tullus et al., 1989; Cui et al., 2021). C. difficile infection leads to diarrhea and colitis that can be fatal and is responsible for greater than 29,000 deaths each year in the United States (Lessa et al., 2015a; Lessa et al., 2015b). Indeed, C. difficile infection is among the top ten forms of health care associated infection and is a continuing problem in hospital environments (Suetens et al., 2018). C. difficile, like C. perfringens, secretes a set of toxins tcdA (toxin A), tcdB (toxin B), and CDT that are responsible for the tissue damage and pathology associated with infections (Kuehne et al., 2014; Chandrasekaran and Lacy, 2017). The pathology of C. difficile infection in humans and animals has been extensively reviewed (Abt et al., 2016; Moono et al., 2016).

Pathogenic infections with C. difficile are associated with conditions that yield dysbiosis including antibiotic treatments or severe stress that provide the opportunity for C. difficile to colonize the gut (Hooks and O'Malley, 2017; Vasilescu et al., 2021). The use of a subset of antibiotic drugs is associated with an increased risk of inducing C. difficile infections likely owing the reduction of competing microbial species in the gut caused by the treatment (Vardakas et al., 2016). Infections with C. difficile are notoriously persistent. Most antibiotic treatments are not successful and even following extensive antibiotic regimens infections are recurrent in 20%–40% of patients (Pepin et al., 2005; Vardakas et al., 2016). The high rate of recurrence may be due to the ability of the cells to sporulate and embed spores in the gut wall (Castro-Cordova et al., 2021). Owing to the persistent nature of C. difficile infections the antibiotics currently used in treatment are vancomycin, fidaxomicin and metronidazole (McDonald et al., 2018). An ominous trend is that these antibiotic treatments now display reduced effectiveness in treatment of C. difficile. The effectiveness of metronidazole has reduced from 95% to 75% since 2010. Even vancomycin has displayed a reduced effectiveness from 95% to 85% (McDonald et al., 2018). The increased incidence of recurrent C. difficile infections reflects a failure of the currently employed antibiotic treatments and reveals the need for more effective drugs or alternative therapeutic approaches (Marra et al., 2020; Sholeh et al., 2020).

There has long been evidence that the normal diversity and composition of the gut microbiome is able to suppress outgrowth of C. difficile and limit pathological colonization of the gut (Freter, 1955). Administration of probiotic microbes including Bifidobacterium breve and Lacticaseibacillus casei, S. boulardii, and L. rhamnosus GG, either alone or in combination with antibiotic regimens have seen some success in inhibiting or resolving C. difficile induced gut pathologies in animal models (Gorbach et al., 1987; McFarland et al., 1994; Esposito et al., 2021; Panpetch et al., 2021; Yang and Yang, 2021; Yang et al., 2022). Treatment with defined combinations of probiotic strains that compete with C. difficile for mucin as a nutrient source have shown some success (Pereira et al., 2020). More complex combinations of probiotic strains have reached stage III clinical trials for treatment of C. difficile induced pathology and the application of fecal transplantation is gaining acceptance as a treatment despite concerns over the need to characterize the donor microbiota (Khoruts et al., 2021; McGovern et al., 2021). The mechanism by which probiotic administration aids in resolving C. difficile infection is likely a combination of competition for resources coupled with the ability of a subset of microbiota to secrete secondary bile acids, bacteriocins, and other compounds inhibitory to C. difficile vegetative cells, germination of C. difficile spores and potentially disruption of C. difficile quorum sensing (Spinler et al., 2017; Thanissery et al., 2017; Lv et al., 2020; Gunaratnam et al., 2021).

The effective deployment of probiotic microbes to resolve and prevent C. difficile infection has motivated efforts to engineer probiotics for improved specific functions. The pathologies and morbidity associated with C. difficile infection are induced by the toxins tcdA, tcdB, and CDT. Monoclonal antibodies directed against the toxins have proven effective in limiting the pathology of C. difficile infection (Wilcox et al., 2017). But thus far vaccination with toxoids has had limited success at inducing immunity or treating C. difficile infection (Dieterle et al., 2019; de Bruyn et al., 2021) However, engineered microbial cells expressing C. difficile toxins can induce effective protective immune response when orally administered in animal models. An attenuated V. cholerae strain secreting a toxin A fragment fusion protein induced anti-toxin A antibody production and a protective immune response in mice that survived a subsequent challenge with C. difficile infection (Ryan et al., 1997). Engineered attenuated Salmonella typhimurium and Bacillus subtilis have both been used as vectors for secretion of toxin A fragments to demonstrate the induction of protective immune response through mucosal delivery of the vaccinating antigen (Ward et al., 1999; Permpoonpattana et al., 2011). Oral vaccination with a “disarmed” C. difficile engineered to express toxin fragments provided substantial protective immunity to C. difficile challenge in animal models (Donald et al., 2013; Wang et al., 2018). Similarly, administration of an engineered Lactococcus lactis expressing a fragment of toxin A, either secreted from the cells or displayed on the cell wall induced a strong immune response and provided substantial protection to C. difficile challenge in mice (Yang et al., 2013). Engineering L. lactis to express toxin A and toxin B fragments in combination provided no greater protection than toxin A alone (Guo et al., 2015). Clostridia employ surface proteins to assist with adherence to the intestinal cell wall and promote colonization. L. casei engineered to express and display the C. difficile SlpA on its cell wall were able to effectively engraft in mouse intestines following inoculation and triggered a robust immune response that protected hamsters from death following infection with C. difficile (Vedantam et al., 2018). Additionally, C. difficile surface antigen CD0873 involved in attachment to the epithelial cell wall has emerged as an effective antigen for induction of a protective immune response and may be a good candidate for surface display on probiotics to induce immunity (Karyal et al., 2021). The apparent effectiveness of the live cell vaccines in animal models is encouraging and the relative ease with which a variety of toxin and surface antigens can be tested for the ability to induce protective immunity means that this approach may achieve further uptake.

One of the limitations of vaccination efforts against C. difficile is the relatively slow sero-conversion time as well as the inherent limitation of systemic antibodies directed against toxins in the intestinal lumen (Greenberg et al., 2012; de Bruyn et al., 2016). Microbial cells have been engineered to deliver therapeutic antibodies to neutralize C. difficile toxins during infection. L. paracasei strains were developed to express and secrete or display on their cell walls antibody chains capable of binding toxin B. Oral delivery of the recombinant probiotics provided protection against infection with a toxin B expressing C. difficile strain in a hamster model (Andersen et al., 2016). The yeast S. boulardii has been employed to express and secrete a multidomain neutralizing antibody directed against the major toxins of C. difficile (Chen et al., 2020). The engineered live cell therapeutic was effective in limiting disease caused by acute and recurrent infections by C. difficile in a mouse model. An advantage of a S. boulardii as a chassis for delivering antimicrobial activity is that it can be used in combination with antibiotic drugs that would render a bacterial probiotic ineffective. Additionally, this strategy has potential for further enhancement through improved expression and secretion of the therapeutic antibody. It could also be more broadly applied through expression of antibodies against other virulence factors.

A final consideration for deployment of live cell therapeutics to treat C. difficile infection is the application of lytic bacteriophage. Phage that specifically target C. difficile have been identified but treatment with phage has had limited success (Mayer et al., 2008; Meader et al., 2013; Nale et al., 2016; Whittle et al., 2022). The effectiveness of a C. difficile-specific phage was improved through engineering the phage to provide guide RNA to induce the endogenous C. difficile CRISPR system to target its own genome (Selle et al., 2020). Despite the improved effectiveness in killing C. difficile vegetative cells this phage will likely suffer the limitation of all phage therapies in that the target cells will rapidly evolve resistance.

Bacteriophage encoded endolysins with specificity for C. difficile cell walls have been identified (Mayer et al., 2008; Mondal et al., 2020). Oral application of purified endolysin has proven effective in controlling C. difficile infection in mice but owing to limitations of protein stability the therapeutic effect is modest (Peng et al., 2018). This limitation might be overcome through expression of the endolysin by a probiotic microbial strain allowing prolonged secretion into the gut and extending the time allowed for effective killing of colonizing C. difficile. While this is an intriguing idea it would be necessary to ensure that the endolysin did not have “off target” effects by killing other microbial species which might add to any dysbiosis rather than simply eradicate the pathogen.

The increasing frequency of drug resistant C. difficile infections and the severity of their pathology coupled with the high incidence of infections occurring in health care facilities creates urgency around the development of new strategies to control this pathogen. The research described above highlight the potential to use bioengineered probiotics as live cell vaccines and vehicles for delivering therapeutics to effectively control the damaging effects of opportunistic C. difficile outgrowth.