By Dr. Mary Ellen Sanders PhD, Mary Ellen Sanders LLC, Colorado, USA, Dr. Hania Szajewska MD, Medical University of Warsaw, Poland, Prof. Karen Scott PhD, Rowett Institute, University of Aberdeen, Scotland, UK

From the time scientists began to understand how antibiotics disrupt the gut microbiota, questions have been raised about whether probiotics could help recovery from this disruption by providing bacteria beneficial to the gut environment or disrupting the bloom of harmful ones. The topic was brought to popular attention after a prominent 2018 scientific paper, widely covered in the media, demonstrated that a certain 11-strain probiotic formulation impaired gut microbiota recovery after antibiotic treatment, sparking debate about possible harms of probiotics.

A microbiota that supports health should be both resilient – meaning it can return to its original state after disruption – and adaptive, but can probiotics facilitate these important features of the microbiota? Under the auspices of International Scientific Association for Probiotics and Prebiotics, we met as part of a group of 8 experts in clinical medicine, biostatistics, basic microbiome science and probiotic microbiology to consider whether probiotics could help restore an antibiotic-perturbed gut microbiota in humans.

Different antibiotics vary in the extent to which they disrupt the microbiota, but in general antibiotics can reduce microbial diversity and alter how the microbiome functions. The therapeutic value of antibiotics is undisputed; however, their clinical use is associated with some harms, including an increase of antibiotic-resistant pathogens, Clostridioides difficile infection, and antibiotic-associated diarrhea. Although specific probiotics have been shown to mitigate some of these clinical harms associated with antibiotic use, how they accomplish this is not clear.

At the outset, we recognized that the human response to both antibiotics and probiotics was multi-factorial. Factors such as microbiota composition and host genetics can vary substantially among individuals. Variation is also observed among the different probiotic strains and doses used in intervention trials. Further, the studies that have been conducted to assess microbiota composition and recovery employed different methods and are therefore difficult to compare.

After conducting a systematic review of human studies within the scope of this topic, and reviewing available clinical and microbiome data (composition and function of the microbiota), we concluded that current evidence does not support the conclusion that probiotics studied so far help restore the gut microbiota to its pre-antibiotic-challenged state. Studies adequately designed to directly address this question are rare. We found that only a few probiotic strains, such as Bifidobacterium lactis subsp. animalis BB-12, have been tested in randomized, controlled trials for their ability to mitigate the impact of antibiotics on gut microbiota.

Given that nature of the interaction of probiotics with the microbiota may be specific at the strain level, strain-specific research is essential to uncover targeted benefits, rather than making broad claims about probiotics as a whole. It is unlikely that a conclusion on the effect of probiotics in general on an antibiotic-perturbed microbiota composition or function can be reached. Instead, future conclusions should be strain-specific, based on research on particular probiotic preparations using meaningful microbiome measures.

Multiple microbiota readouts, including strain abundance possibly adjusted for host-related parameters, should be used to assess microbiota recovery. We also recognized that higher microbiota diversity, although frequently assessed, has not been causally linked to good health in humans and may be insufficient as a measure of a more ideal microbiota.

Our conclusions do not detract from evidence that some probiotics can improve certain clinical outcomes related to antibiotic use, such as reducing the risk of antibiotic-associated diarrhea or C. difficile infections. However, these clinical effects have not been shown to be causally linked to a probiotic-mediated impact on the function or composition of the microbiota. An important avenue for future research is study of the mechanism(s) through which probiotic bacteria can provide these clinical benefits.

The panel was composed of Prof. Hania Szajewska MD, Prof. Karen P. Scott PhD, Prof. Tim de Meij MD, Dr. Sofia K. Forslund-Startceva PhD, Prof. Rob Knight PhD, Prof. Omry Koren PhD, Prof. Paul Little MD, Prof. Bradley C. Johnston PhD, Dr. Jan Łukasik MD, Dr. Jotham Suez PhD, Prof. Daniel J. Tancredi PhD, and Dr. Mary Ellen Sanders PhD. This review was published in Nature Reviews Gastroenterology and Hepatology.

Travel costs of experts to London UK and local meeting expenses were paid by the International Scientific Association for Probiotics and Prebiotics.

See here for ISAPP’s clinical resource on probiotic use alongside antibiotics

By Prof. Maria Marco PhD and Prof. Kevin Whelan PhD

In studies on probiotics, prebiotics, and other biotics, the demonstrated health effect may depend on the personal characteristics and habits of the study participants. Take the following scenario: Person A and Person B consume the same prebiotic supplement. Person A typically eats a Mediterranean-style diet, meeting daily fiber requirements by consuming various fruits, leafy greens, nuts, and seeds daily. Person B eats mostly convenience foods that are low in fiber – pastries, fried foods, and frozen dinners. Is the prebiotic likely to have the same effect in both individuals?

The possibility that a prebiotic may have different effects in these two individuals seems intuitive. A plausible scenario is that the gut microbiome of Person A is enriched for bacterial species that feed well on the the prebiotic substrate and therefore is “primed” for more robust responses to prebiotic intake; Person B, on the other hand, doesn’t have the target bacterial species and therefore receives limited benefit from the prebiotic.

Both dietary patterns and specific foods are known to affect the digestive tract environment, either directly or indirectly (for example, by shaping the gut microbiome), and are therefore of interest as potential factors in the response of an individual to a biotic intervention. Surprisingly in the scientific literature, however, diet has been almost completely overlooked as a potential contributor to the effectiveness of probiotics and prebiotics.

In 2022 we gathered an international group of experts to discuss whether diet is potentially an important factor contributing to the efficacy of biotics. The result of our discussions was recently published as a Perspective in Nature Microbiology.

Evidence for how diet affects biotic efficacy

When our group reviewed the literature, we found surprisingly few examples of biotic clinical trials in which the researchers had gathered data on the background diet of the participants. Yet from the studies we did find, dietary influences on biotic efficacy are promising for future study. In one trial of a probiotic in participants with metabolic syndrome, dietary characteristics differentiated responders and non-responders. And in a prebiotic trial, participants who consumed more fiber at baseline had a more marked gut microbiota response and reported feeling fuller than those with lower fiber intake.

Our group members pointed out that, although the gut microbiome itself has increasingly been measured in biotics studies, microbiome features don’t capture the full picture of why someone responds to a biotic intervention. Diet may affect the efficacy of a biotic intervention through microbiome-mediated mechanisms or via other effects on the host, such as alterations in digestive function.

Recommendations for biotics trials

Our group agreed that including dietary data as a matter of course in biotics trials is a crucial way to move the field forward and continue to understand how biotics work in different individuals. From our extensive discussions and follow-up, we came up with ten recommendations for future studies on biotics to account for the effects of diet on biotic efficacy:

• Consider whether and how to harmonize the diet of your study population before intervention.
• Ideally, participants should not be involved in a biotic trial over religious festivals or holidays during which dietary intake can be dramatically altered.
• The composition of the prebiotic (structure, degree of polymerization, source and purity) or probiotic (for example, strain), as well as the structure and composition of the delivery matrix, should be reported.
• Details of prebiotic or probiotic intake (number of doses, timing, whether taken with or between meals and consumed with which foods) should be reported.
• Appropriate microbiome analyses should be performed to disentangle the specific effect of the intervention versus background diet.
• Dietary assessment should be performed during screening (if it is a requirement of inclusion), at baseline and at the end of an intervention.
• Dietary assessment should record intakes of microbiome-relevant dietary exposures where possible and feasible.
• Dietary assessment should be relevant to the hierarchical ontology of the dietary exposure of interest (for example, nutrient, food, food group or dietary pattern).
• The method and duration of dietary assessment should depend on the research question and the exposure of interest.
• A dietitian or nutritionist with expertise in microbiome research and dietary assessment should be included in the research team.

We acknowledge that in some cases implementing these recommendations will result in additional costs for a biotic trial, but we think these costs will be well justified given the potential to show stronger effect sizes, reduce unnecessary trials, and zero in on the populations that will benefit from a given biotic intervention.

The future of biotics

At this stage in the field of biotics, the literature indicates that demonstrating the efficacy of a biotic intervention through a clinical trial must take into account the complexity of individual participants – not just demographic factors, but also lifestyle factors such as diet. We believe that background diet may be a critical factor in determining biotic effectiveness, and thus should be included in as many future biotic studies as possible. By accounting for the influence of diet, scientists studying biotics may be able to show stronger and more targeted health benefits, allowing more personalized and effective interventions.

By Prof. Maria Marco PhD, University of California, Davis

“Kills 99.9% of bacteria and viruses.” This percentage and others like it are frequently found on disinfectant labels.

Ideally, the microbicidal effect of the product is sufficient to kill more than the numbers of the target pathogen or pathogens expected to be in the environment where the disinfectant is used. However, if more of the target pathogen is present, 99.9% can be misleading or even result in illness. For example, for a surface with one million pathogenic E. coli cells, a 99.9% reduction would mean that 1000 living and infectious E. coli cells remain. That amount of E. coli could be sufficient to cause life threatening illness.

Declaring a microbe dead

The issue of quantity is only one of the conundrums when relaying information about microbial death. Another issue is how do we even know if a microbe is really dead? For viruses, they are not cells and so the terms “live” and “dead” do not apply. Viral inactivation is inferred by a loss in the ability to infect and multiply in host cells. For bacteria, the answer is not so simple. Bacteria can form dormant states, whereby their metabolic activity is minimal, but then, when conditions are right, they grow and divide again. Endospores formed by some bacteria are a great example of dormancy. But even bacteria that do not form endospores can be viable (alive) but dormant for long periods of time and may be inferred to be “dead”. Sometimes these dormant states are desirable, such as when drying bacterial strains for use as probiotics and retaining their capacity to reactivate when the conditions are right. 

Measuring dead microbes

Another problem with microbial death is how we measure it. Colony forming unit (CFU) enumeration on laboratory culture media is the gold-standard for quantifying viable bacterial cell numbers. However, it is now well known that only a very small fraction of all bacteria on Earth have been grown or “cultured” in the laboratory. Many microbes are “unculturable” mainly because the nutrients and environmental conditions needed for their enrichment in the laboratory have yet to be found. So, it is not possible to quantify viability of “unculturable” microorganisms using the gold standard approach.

Even for microbes that grow well in the laboratory (for example E. coli), they too may not always be culturable. After exposure to certain conditions, such as nutrient limitations, some bacteria can form Viable But Non Culturable (VBNC) states. Those VBNC microbes will not grow using routine culture methods, but they are viable and may return to a metabolically active, reproductive state later, such as when nutrients or environmental conditions change. Testing via CFU therefore may underestimate levels of viable bacteria – with important implications for measuring and monitoring both beneficial and pathogenic organisms.

Dead bacterial cells can best be described as having different states. Completely lysed bacterial cells, wherein the cell membrane is destroyed, and intracellular contents are released, are obviously dead. Yet, this state may constitute only a minor fraction of dead bacteria in a population. Bacteria can also be dead, but still have an intact cell membrane. Several methods have been developed to assess viability of cells in that state, including measurements of cellular enzymatic activity (MTT conversion assay) and uptake of fluorescent dyes impenetrable to intact cell membranes (SYTO9-propidium iodide staining).  While this question of evaluating cell viability is far from resolved, an intriguing recent recommendation was to use multiple tests, assessing both metabolic activity and reproductive capacity (1).

Implications for biotics

How does all this relate to biotics? Probiotics should be alive (viable) at the time of administration. Postbiotics are preparations of dead (inanimate) microbes. Both must deliver a health benefit. Decisions on how viable and dead microbes are enumerated in biotic preparations should address the fact that there are different bacterial viability states. The use of a single method such as CFU enumeration can lead to underestimating numbers of viable cells and will not be helpful for quantifying dead cells. Although we may never be able to say that a microbial population is either absolutely 100% alive or dead, such viability states may affect how well either a probiotic or a postbiotic performs for its intended purpose of conferring a health benefit.