Key Points

Introduction

“Medicine used to be simple, ineffective and relatively safe. Now it is complex, effective and potentially dangerous” —Cyril Chantler, Lancet 1999

First, the safety record of probiotic strains in current use does not necessarily apply to new strains in development and each needs assessment on a case-by-case basis. Second, probiotic strains are highly varied without a uniform mechanism of action and, therefore, unlikely to have the same adverse effects in all situations. Third, there is no such thing as zero risk, whether for drugs, probiotics, or even therapeutic nihilism. Fourth, there is poor public understanding of risk in general and risk/benefit analysis in particular, which needs to be addressed. Fifth, because some probiotic products are marketed to those seeking alternative medicine in health-food stores or are available from sources under dubious regulatory constraints, the quality of the product in terms of potential contaminants may be more important than concerns regarding the specific properties of the probiotic constituent. Finally, although probiotics have commonly been selected from the nonpathogenic components of the commensal microbiota and generally regarded as safe, the relationship between commensals and pathogens is not one of mutual opposites, but rather they are at different positions on a spectrum of low to high pathogenic potential.

Probiotics are usually defined as “live microorganisms, which when administered in adequate amounts confer a health benefit on the host.” The limitations of the restrictive nature of this definition have been commented on elsewhere because it excludes dead organisms, probiotic fragments, and metabolites, such as bioactive polysaccharides, nucleotides, and proteins. Like the fate of the original definition of antibiotics, which excluded sulphonamides and synthetic antimicrobials, the definition of probiotics may have outlived its usefulness and is likely to undergo refinements that will be informed by science. At present, yeast ( Saccharomyces cervisiae ) and bacteria comply with the current definition of probiotics with the latter including species of Lactobacillus , Streptococcus , Enterococcus , Bifidobacterium , Propionibacterium , Bacillus , and Escherichia coli . In addition, the organisms may be either naturally occurring or genetically modified. This degree of heterogeneity underscores the problem of using a collective generic name. In the same way that the safety and efficacy of drugs would never be addressed in collective, generic terms, so probiotic bacteria need consideration individually on a case-by-case basis, particularly those still in development.

When friendly bacteria behave poorly

Because consumption of a probiotic represents, in essence, an intention to mimic, supplement, or otherwise harness the commensal microbiota, the distinction between commensals and pathogens becomes critical. Organisms with a propensity to cross biologic boundaries, like the mucosal barrier, are obviously pathogens; but the distinction from commensals is not always readily evident, particularly when the host has a particular vulnerability, such as an acquired barrier defect or a genetic susceptibility. How does the host distinguish pathogens from commensals or consumed probiotics? This process is complex and cannot be based solely on the recognition of microbe-associated molecular patterns because the patterns involved in the recognition of pathogens are also expressed by nonpathogens, including probiotics.

The problem is solved for some commensals by the production of symbiosis-associated molecular patterns. This point is exemplified by Bacteroides fragilis , a prominent member of the commensal microbiota, which produces an immunomodulatory polysaccharide that signals through Toll-like receptor 2 on regulatory T cells within the host to suppress T _H 17 effectors, thereby avoiding an adverse immune response. The degree to which this mechanism is deployed by other commensals/probiotics is unclear, but the host has a backup surveillance system for the detection of danger signals from the microbiota and for modifying the composition of the microenvironment. This detection is achieved by the inflammasomes, which are intracellular multi-protein complexes that sense exogenous and endogenous stress or damage. Experimental defects in inflammasome function have shown their importance within the epithelium and mucosal phagocytes, not only in detecting pathogenic components within the microbiota but also in initiating a cascade of immunologic responses to restore compositional equilibrium within the microbiota.

Of course, some organisms may, depending on the context and/or depending on the host susceptibility, behave either as a commensal or as a pathogen. The clearest example of commensals in the wrong place at the wrong time is that of the baby born too soon, born before maturation of the mucosal and blood-brain barriers, and before full development of immunity. In this setting, colonization with what would otherwise be considered harmless commensals poses a threat. The case for using probiotics in premature infants has been well made, albeit still controversial. Regardless, the use of probiotics in premature infants is not without risk and should be considered as an attempt to substitute organisms of low pathogenic risk for those that may represent greater risk to the vulnerable neonate.

The importance of context in relation to risk and benefit is also illustrated by the case of Helicobacter pylori . Many clinicians have adopted the unidimensional view that the only good H. pylori is a dead H. pylori , but the outcome of the host-helicobacter interaction may be favorable, unfavorable, or both for the host, depending on the bacterial strain and on a variety of factors, including the age and susceptibility of the host. The organism is usually acquired in childhood but only after a variable period of apparent health causes peptic ulceration in adulthood in a minority of those infected. At a later age, the organism may cause gastric cancer. The potential benefits from the same organism also depend on the age of the host. In early life, H pylori may protect against asthma and infections; but later, it protects against reflux-associated complications, such as metaplasia and neoplasia at the gastroesophageal junction.

Introduction

“Medicine used to be simple, ineffective and relatively safe. Now it is complex, effective and potentially dangerous” —Cyril Chantler, Lancet 1999

First, the safety record of probiotic strains in current use does not necessarily apply to new strains in development and each needs assessment on a case-by-case basis. Second, probiotic strains are highly varied without a uniform mechanism of action and, therefore, unlikely to have the same adverse effects in all situations. Third, there is no such thing as zero risk, whether for drugs, probiotics, or even therapeutic nihilism. Fourth, there is poor public understanding of risk in general and risk/benefit analysis in particular, which needs to be addressed. Fifth, because some probiotic products are marketed to those seeking alternative medicine in health-food stores or are available from sources under dubious regulatory constraints, the quality of the product in terms of potential contaminants may be more important than concerns regarding the specific properties of the probiotic constituent. Finally, although probiotics have commonly been selected from the nonpathogenic components of the commensal microbiota and generally regarded as safe, the relationship between commensals and pathogens is not one of mutual opposites, but rather they are at different positions on a spectrum of low to high pathogenic potential.

Probiotics are usually defined as “live microorganisms, which when administered in adequate amounts confer a health benefit on the host.” The limitations of the restrictive nature of this definition have been commented on elsewhere because it excludes dead organisms, probiotic fragments, and metabolites, such as bioactive polysaccharides, nucleotides, and proteins. Like the fate of the original definition of antibiotics, which excluded sulphonamides and synthetic antimicrobials, the definition of probiotics may have outlived its usefulness and is likely to undergo refinements that will be informed by science. At present, yeast ( Saccharomyces cervisiae ) and bacteria comply with the current definition of probiotics with the latter including species of Lactobacillus , Streptococcus , Enterococcus , Bifidobacterium , Propionibacterium , Bacillus , and Escherichia coli . In addition, the organisms may be either naturally occurring or genetically modified. This degree of heterogeneity underscores the problem of using a collective generic name. In the same way that the safety and efficacy of drugs would never be addressed in collective, generic terms, so probiotic bacteria need consideration individually on a case-by-case basis, particularly those still in development.

When friendly bacteria behave poorly

Because consumption of a probiotic represents, in essence, an intention to mimic, supplement, or otherwise harness the commensal microbiota, the distinction between commensals and pathogens becomes critical. Organisms with a propensity to cross biologic boundaries, like the mucosal barrier, are obviously pathogens; but the distinction from commensals is not always readily evident, particularly when the host has a particular vulnerability, such as an acquired barrier defect or a genetic susceptibility. How does the host distinguish pathogens from commensals or consumed probiotics? This process is complex and cannot be based solely on the recognition of microbe-associated molecular patterns because the patterns involved in the recognition of pathogens are also expressed by nonpathogens, including probiotics.

The problem is solved for some commensals by the production of symbiosis-associated molecular patterns. This point is exemplified by Bacteroides fragilis , a prominent member of the commensal microbiota, which produces an immunomodulatory polysaccharide that signals through Toll-like receptor 2 on regulatory T cells within the host to suppress T _H 17 effectors, thereby avoiding an adverse immune response. The degree to which this mechanism is deployed by other commensals/probiotics is unclear, but the host has a backup surveillance system for the detection of danger signals from the microbiota and for modifying the composition of the microenvironment. This detection is achieved by the inflammasomes, which are intracellular multi-protein complexes that sense exogenous and endogenous stress or damage. Experimental defects in inflammasome function have shown their importance within the epithelium and mucosal phagocytes, not only in detecting pathogenic components within the microbiota but also in initiating a cascade of immunologic responses to restore compositional equilibrium within the microbiota.

Of course, some organisms may, depending on the context and/or depending on the host susceptibility, behave either as a commensal or as a pathogen. The clearest example of commensals in the wrong place at the wrong time is that of the baby born too soon, born before maturation of the mucosal and blood-brain barriers, and before full development of immunity. In this setting, colonization with what would otherwise be considered harmless commensals poses a threat. The case for using probiotics in premature infants has been well made, albeit still controversial. Regardless, the use of probiotics in premature infants is not without risk and should be considered as an attempt to substitute organisms of low pathogenic risk for those that may represent greater risk to the vulnerable neonate.

The importance of context in relation to risk and benefit is also illustrated by the case of Helicobacter pylori . Many clinicians have adopted the unidimensional view that the only good H. pylori is a dead H. pylori , but the outcome of the host-helicobacter interaction may be favorable, unfavorable, or both for the host, depending on the bacterial strain and on a variety of factors, including the age and susceptibility of the host. The organism is usually acquired in childhood but only after a variable period of apparent health causes peptic ulceration in adulthood in a minority of those infected. At a later age, the organism may cause gastric cancer. The potential benefits from the same organism also depend on the age of the host. In early life, H pylori may protect against asthma and infections; but later, it protects against reflux-associated complications, such as metaplasia and neoplasia at the gastroesophageal junction.