Professional information for education only. Not consumer advertising or product promotion. If you are a consumer, please visit G-Health for general health and lifestyle content. Probiotics are food supplements, not medical treatments. If you have a medical condition or are seeking medical advice, please consult your doctor for personalised guidance.
Introduction
The gut microbiome is the complex community of microorganisms living within the gastrointestinal tract. In practical terms, it encompasses the distribution and relative abundance of beneficial commensal bacteria, potentially harmful or opportunistic microorganisms, and the many other microbial species that coexist within this ecosystem.
Rather than remaining passive inhabitants of the gut, these microorganisms interact with the intestinal barrier, metabolise dietary components and produce compounds capable of influencing local and systemic immune signalling. When this community becomes significantly disrupted—a condition commonly described as gut dysbiosis—beneficial bacteria may be depleted while potentially harmful microorganisms become relatively more prominent.
COVID-19 provided a particularly instructive example. During the early stages of the pandemic, studies of hospitalised patients found that COVID-19 was accompanied by marked alterations in the gut microbiome, including the depletion of several beneficial commensal bacteria and the enrichment of opportunistic microorganisms. Among these patients, the degree and pattern of microbiome disruption were associated with disease severity, inflammatory cytokines and other blood markers of inflammation.¹, ²
These observational findings did not show that a favourable gut microbiome could prevent SARS-CoV-2 infection, or that gut dysbiosis caused COVID-19. They did, however, reveal a clinically relevant pattern: among patients who had already developed the disease, greater disruption of the microbial community was associated with more severe illness and heightened inflammatory responses.
This raised an important question for microbiome research. If a particular condition is associated with the depletion of specific beneficial bacteria, should intervention begin with a generic collection of popular probiotics—or with a careful investigation of the underlying microbial imbalance?
When an Unfavourable Gut Microbiome Starts to Look Like the New Normal
This question may be relevant beyond people experiencing an acute illness.
As discussed in our previous Professional Hub article, How the Gut Microbiome Influences Immune Response to Infection, findings from the analysis of stool samples from 1,000 individuals suggested that approximately 83% of this urban cohort exhibited varying degrees of gut dysbiosis. Nearly 40% showed moderate-to-severe dysbiosis—a degree of imbalance reported to be comparable with that observed in patients with COVID-19.³, ⁴
The Chinese University of Hong Kong raised concern that the significant dysbiosis observed in almost 40% of the cohort could indicate potentially impaired immunity.⁴ Although findings from one urban population should not automatically be generalised to every city or country, they suggest that an unfavourable gut microbiome may no longer be an uncommon exception among modern urban populations.
The scientific challenge, therefore, is not simply to add more bacteria. It is to identify which microbial changes matter, understand how those changes relate to the health problem being investigated, and determine how a carefully selected intervention might support a more favourable microbial environment.
COVID-19 Revealed a Persistent Microbial Imbalance
In an early longitudinal study, researchers conducted shotgun metagenomic sequencing of serial stool samples from 15 hospitalised patients with COVID-19. Their gut microbiomes were compared with those of healthy individuals and patients with community-acquired pneumonia.¹
The patients with COVID-19 showed enrichment of several opportunistic pathogens and depletion of beneficial commensal bacteria, including Faecalibacterium prausnitzii, Eubacterium rectale and members of Roseburia. Lower abundance of F. prausnitzii, a bacterium associated with anti-inflammatory activity, was also correlated with greater disease severity.¹
A subsequent two-hospital study analysed samples from 100 patients with laboratory-confirmed COVID-19 and 78 non-COVID-19 individuals. The researchers found that microbiome composition differed significantly between the groups, while the pattern of disruption was associated with disease severity, inflammatory cytokines and blood markers such as C-reactive protein.²
Importantly, the imbalance did not necessarily disappear when the acute infection resolved. In the first study, depletion of beneficial symbionts persisted in some patients even after respiratory symptoms had resolved and SARS-CoV-2 was no longer detected in throat samples.¹ In the larger study, several beneficial commensals remained underrepresented in samples collected up to 30 days after viral clearance.²
A separate six-month follow-up study subsequently found that microbiome richness in some recovered patients had still not returned to the level observed in non-COVID-19 controls.⁵
These studies were observational and could not prove that microbiome disruption caused severe disease or delayed recovery. Nevertheless, they identified a reproducible pattern of clinically associated dysbiosis. They also suggested that recovery from an infection might involve more than clearing the pathogen: restoration of the microbial ecosystem could represent a separate biological process.
Once this pattern had been identified, the next question was how to intervene. A generic answer: simply adding any probiotic would overlook the complexity of the ecosystem being investigated.
Why Simply Combining Any "Good Bacteria" Is Not Enough
Probiotics are often discussed as though all beneficial bacteria perform broadly interchangeable roles. In reality, different bacterial species possess different metabolic capabilities, utilise different nutrients and interact with the host and neighbouring microorganisms in different ways.⁶
Different probiotic species have not necessarily been studied for the same health conditions. For example, clinical studies involving preparations containing Lactobacillus salivarius have reported significant improvements in measures of atopic dermatitis severity or quality of life among adult and paediatric participants.⁷, ⁸ By contrast, a distinct Bifidobacteria-based synbiotic preparation, combining three different Bifidobacterium species with prebiotics, was evaluated for post-acute COVID-19 syndrome, commonly known as Long COVID—a condition in which symptoms persist after the acute phase of infection has resolved. In a large-scale, randomised, double-blind, placebo-controlled trial, a significantly higher proportion of participants receiving this bifidobacteria-based synbiotic preparation reported clinically meaningful improvement in their overall symptom burden after six months compared with those receiving placebo.⁹
These examples illustrate why probiotic selection should begin with the health problem being investigated and the evidence relevant to that particular context. A bacterial species or combination studied for one condition should not automatically be assumed to provide the same effects in another.
Some microorganisms support one another through cross-feeding. In this process, a compound produced when one bacterium metabolises a nutrient becomes a substrate that another bacterium can use. Laboratory research has, for example, demonstrated metabolic cross-feeding between Bifidobacterium adolescentis and the butyrate-producing commensal Faecalibacterium prausnitzii.¹²
At the same time, microorganisms can compete for fermentable carbohydrates, other nutrients and ecological niches. Experimental research has shown that the presence of competing species can influence which gut bacteria benefit from particular fibre-derived glycans.¹⁰ Community composition can therefore alter the outcome of an intervention that was intended to favour a particular microorganism.
Human research has similarly found that the ability of an orally administered Bifidobacterium longum to persist in the gut depended partly on the individual’s resident microbiome and the availability of an unoccupied ecological niche. Stable persistence occurred in only a proportion of participants.¹¹
These findings do not demonstrate that every combination of probiotic species will compete, nor do they prove that the selected bacteria within a particular finished preparation work against one another. They establish a broader ecological principle: both cooperation and competition can influence the response to supplementation.
Consequently, adding more bacterial species—or simply pursuing the highest possible viable count—does not automatically create a more targeted formula. A scientifically reasoned preparation should instead consider the health problem being investigated, the microbial changes associated with that problem, the evidence supporting the selected species and the ecological environment into which those bacteria will be introduced.
From Disease–Healthy Differences to Targeted Bacterial Selection
This is where a precision approach begins.
In this context, precision formulation does not mean that a separate preparation was individually prescribed for every patient. It describes a research-led process in which a defined health problem is studied at the microbiome level, a relevant pattern of imbalance is identified, and bacterial species are selected to target that pattern more deliberately.
The research programme that ultimately produced SIM01 began with metagenomic datasets derived from patients with COVID-19 and healthy individuals.¹³ Instead of starting with a generic list of popular probiotics, researchers first compared the overall microbial profiles of the two populations and identified bacterial species that were relatively enriched or depleted in patients.¹, ²
According to the published development methodology, three focal Bifidobacterium species—B. adolescentis, B. bifidum and B. longum—were selected on the basis of their positive correlations with bacterial species found to be depleted in patients with COVID-19.¹³
This distinction is important. The process was not simply:
Three bacteria were missing, so the same three bacteria were added back
Rather, the researchers identified a wider network of beneficial bacteria affected in COVID-19 and selected three Bifidobacterium species whose abundance was positively associated with members of that depleted network. The intended approach was therefore to support a more favourable microbial community rather than merely fill three isolated microbial “gaps”.
The authors reported that these relationships were then examined in two independent healthy cohorts comprising approximately 1,400 individuals. The abundance of the selected bacteria showed strong positive correlations with beneficial bacteria in these healthy microbiomes.¹³
Formulation extended beyond deciding which species to include. The ratio between the three bacteria was reportedly derived from their average naturally occurring proportions in a reference population of more than 3,000 healthy individuals.¹³ In other words, population-scale healthy microbiome data were used not only to inform bacterial selection, but also to guide how the selected species were combined.
This process can be understood as a sequence:
- Define the clinically relevant health problem.
- Compare disease-associated and healthy microbiome profiles.
- Identify microbial species and networks that differ between the groups.
- Select focal bacteria associated with the depleted beneficial network.
- Cross-check these relationships against independent healthy cohorts.
- Use population data to inform the bacterial ratio.
- Combine the selected bacteria with prebiotic substrates intended to support favourable microbial growth and activity.
It demonstrates that the preparation was developed from identifiable microbiome differences and large reference datasets, rather than assembled through guesswork or the indiscriminate inclusion of popular probiotic species.
How Metagenomics Made a Community-Wide Comparison Possible
The precision of this approach depended on the ability to examine the microbial community broadly. This is where shotgun metagenomics became important.
A targeted test, such as PCR designed around a preselected organism or gene, can answer a highly specific question: is the chosen genetic signal present, and how much of it can be detected? This can be extremely useful, but the test only looks for targets that researchers have decided upon in advance.
Shotgun metagenomic sequencing takes a more community-wide approach. Rather than examining only a handful of predetermined signals, it sequences large quantities of recoverable DNA extracted from a stool sample. This provides a broad survey of detectable microbial genetic material and supports more detailed identification at the species level than many targeted or 16S-based approaches.¹⁴
A stool sample contains a complex mixture of genetic material. In addition to DNA from numerous microorganisms, it may contain human DNA, low-quality sequence fragments and technical sequences introduced during laboratory processing. Advanced sequencing is therefore only the beginning of the analysis.
In the COVID-19 microbiome studies, DNA was extracted from stool and prepared for high-throughput sequencing using Illumina systems. The process generated millions of short DNA fragments known as sequence reads. One study produced an average of approximately 47 million reads per sample,² while the later pilot study generated an average of approximately 93.5 million raw reads per sample.¹³ Across a study containing many samples, the total analysis can therefore involve billions of sequence reads.
These raw data must then pass through several computational stages:
- Sequencing adapters and other technical sequences are removed.
- Low-quality bases and excessively short reads are filtered out.
- Sequences aligning with the human reference genome are removed.
- The remaining microbial reads are compared with curated microbial marker-gene databases.
- Bioinformatics software converts the resulting matches into a species-level relative-abundance profile.
- Statistical analyses then compare microbial patterns between clinical and reference groups.
The studies used MetaPhlAn2, a computational profiling system that assigns sequence reads to microbial groups using clade-specific marker genes.¹⁵ The output is not a literal inventory of every microorganism or every gene in the entire gastrointestinal tract. It is a broad, data-rich snapshot of the microbial DNA detectable in the stool sample under the study’s collection, extraction, sequencing and analytical conditions.
Turning that snapshot into useful evidence requires high-throughput sequencing instruments, carefully controlled laboratory procedures, substantial data storage and computing capacity, curated genomic databases, specialist bioinformatics pipelines and statistical analysis. It is the combination of these technologies—not sequencing alone—that allows researchers to move from millions of fragments of raw genetic information to a structured comparison between disease-associated and healthy microbiomes.
Metagenomics is therefore not the central purpose of precision formulation. It is one of the advanced tools that makes evidence-led precision possible.
Why Pair Selected Bacteria with Prebiotics?
Selecting bacteria is only one part of a synbiotic strategy. The ecological environment that supports—or restricts—their activity must also be considered.
A prebiotic is scientifically defined as a substrate that is selectively utilised by host microorganisms and confers a health benefit.¹⁶ Many recognised prebiotics are carbohydrates that resist complete digestion in the upper gastrointestinal tract. They can therefore reach the large intestine, where microorganisms possessing the appropriate carbohydrate-utilisation enzymes may ferment them.
This fermentation provides energy and growth substrates to responsive microorganisms. It can also generate metabolites such as acetate and lactate, which may subsequently be used by other commensal bacteria through cross-feeding. Prebiotics may therefore influence not only a single organism, but also relationships across a wider microbial network.¹²
Not every prebiotic feeds every probiotic equally. Microorganisms differ in the enzymes and metabolic pathways they possess, while resident bacteria may compete for the same substrate. The outcome can consequently depend on the type of prebiotic, the bacteria present and the original composition of the host microbiome.¹⁰, ¹⁷
The selected Bifidobacterium species were therefore paired with a three-substrate prebiotic blend, such as xylo-oligosaccharides (XOS) and resistant dextrin.¹³
Experimental research has demonstrated that XOS can be utilised and fermented by B. adolescentis, supporting its growth under controlled conditions.¹⁸ Human research involving resistant dextrin has also shown that it can modify the abundance and carbohydrate-metabolism activity of particular resident gut bacteria. The response to a prebiotic can be selective and may depend on the carbohydrate-utilisation genes present within an individual’s resident microorganisms.¹⁹
The purpose of incorporating prebiotics was therefore not merely to increase the ingredient count. It was to provide an ecological support system intended to stimulate favourable microbial growth or activity, while recognising that the eventual response would also be shaped by the wider resident microbiome.
Conclusion: Precision Is About Relevance, Not Simply More
Microbiome research increasingly shows that a sophisticated probiotic preparation is defined by the relevance of its bacterial selection—not simply by the number of species it contains or the pursuit of a higher viable bacterial count.
Different bacteria possess different biological capabilities and bodies of evidence. Once introduced into the gut, they become part of an established microbial ecosystem in which microorganisms may support one another through cross-feeding, compete for nutrients and occupy overlapping ecological niches. Their growth and activity are also influenced by the availability of suitable nutritional substrates.
A precision synbiotic approach therefore begins with a clearly defined health problem. Scientific evidence is used to identify the associated pattern of microbiome imbalance; advanced technologies enable detailed comparisons between disease-associated and healthy microbiomes; population-scale data inform bacterial selection and proportion; and carefully chosen prebiotics provide ecological support for favourable microbial growth and activity.
The development of the SIM01 Synbiotic Preparation illustrates this evidence-led pathway—from disease-associated microbiome discovery and large-scale data analysis to targeted bacterial selection and synbiotic formulation.¹³ It also reinforces a broader principle for microbiome intervention: more is not necessarily better; precision lies in selecting relevant bacteria, combining them in a reasoned proportion and supporting them with appropriate prebiotic substrates.
How this formulation was subsequently evaluated in patients recovering from COVID-19 will be explored separately in our review of the Recovery Study.
References
Zuo T, Zhang F, Lui GCY, et al. Alterations in gut microbiota of patients with COVID-19 during time of hospitalisation. Gastroenterology. 2020;159(3):944–955.e8.
Yeoh YK, Zuo T, Lui GCY, et al. Gut microbiota composition reflects disease severity and dysfunctional immune responses in patients with COVID-19. Gut. 2021;70(4):698–706.
G-NiiB Science Editor. How the Gut Microbiome Influences Immune Response to Infection: Insights from Clinical Research. G-NiiB Professional Hub.
The Chinese University of Hong Kong. 40% of Hong Kong people show gut dysbiosis comparable to that of COVID-19 patients.
Chen Y, Gu S, Chen Y, et al. Six-month follow-up of gut microbiota richness in patients with COVID-19. Gut. 2022;71(1):222–225.
Hu J, Zhang L, Lin W, et al. Review article: Probiotics, prebiotics and dietary approaches during COVID-19 pandemic. Trends in Food Science & Technology. 2021;108:187–196.
Drago L, Iemoli E, Rodighiero V, et al. Effects of Lactobacillus salivarius LS01 treatment on adult atopic dermatitis: a randomised placebo-controlled study. International Journal of Immunopathology and Pharmacology. 2011;24(4):1037–1048.
Niccoli AA, Artesi AL, Candio F, et al. Preliminary results on clinical effects of probiotic Lactobacillus salivarius LS01 in children affected by atopic dermatitis. Journal of Clinical Gastroenterology. 2014;48(Suppl 1):S34–S36.
Lau RI, Su Q, Lau ISF, et al. A synbiotic preparation (SIM01) for post-acute COVID-19 syndrome in Hong Kong (RECOVERY): a randomised, double-blind, placebo-controlled trial. The Lancet Infectious Diseases. 2024;24(3):256–265. doi:10.1016/S1473-3099(23)00685-0
Patnode ML, Beller ZW, Han ND, et al. Interspecies competition impacts targeted manipulation of human gut bacteria by fibre-derived glycans. Cell. 2019;179(1):59–73.e13.
Maldonado-Gómez MX, Martínez I, Bottacini F, et al. Stable engraftment of Bifidobacterium longum AH1206 in the human gut depends on individualised features of the resident microbiome. Cell Host & Microbe. 2016;20(4):515–526.
Rios-Covian D, Gueimonde M, Duncan SH, et al. Enhanced butyrate formation by cross-feeding between Bifidobacterium adolescentis and Faecalibacterium prausnitzii. FEMS Microbiology Letters. 2015;362(21):fnv176.
Zhang L, Xu Z, Mak JWY, et al. Gut microbiota-derived synbiotic formula (SIM01) as a novel adjuvant therapy for COVID-19: an open-label pilot study. Journal of Gastroenterology and Hepatology. 2022;37(5):823–831.
Ranjan R, Rani A, Metwally A, et al. Analysis of the microbiome: advantages of whole genome shotgun versus 16S amplicon sequencing. Biochemical and Biophysical Research Communications. 2016;469(4):967–977.
Segata N, Waldron L, Ballarini A, et al. Metagenomic microbial community profiling using unique clade-specific marker genes. Nature Methods. 2012;9:811–814.
Gibson GR, Hutkins R, Sanders ME, et al. The ISAPP consensus statement on the definition and scope of prebiotics. Nature Reviews Gastroenterology & Hepatology. 2017;14:491–502.
Swanson KS, Gibson GR, Hutkins R, et al. The ISAPP consensus statement on the definition and scope of synbiotics. Nature Reviews Gastroenterology & Hepatology. 2020;17:687–701.
Amaretti A, Bernardi T, Leonardi A, et al. Fermentation of xylo-oligosaccharides by Bifidobacterium adolescentis DSMZ 18350: kinetics, metabolism and β-xylosidase activities. Applied Microbiology and Biotechnology. 2013;97:3109–3117.
Thirion F, Da Silva K, Plaza Oñate F, et al. Diet supplementation with NUTRIOSE, a resistant dextrin, increases the abundance of Parabacteroides distasonis in the human gut. Molecular Nutrition & Food Research. 2022;66(11):e2101091.