A central part of ZOE’s science has been an investigation of the microbiome and its links with an individual’s diet and metabolic health. Through the PREDICT studies, we've collected the world’s largest set of whole-metagenome sequenced gut microbiome data linked to in-depth metabolic responses and diet.
This enables us to investigate how the gut microbiome, diet, and metabolic health are interlinked, to provide personalized advice on how to improve both gut microbiome and health.
Our research suggests a complex interplay exists, whereby our gut microbes are associated with our diet and state of health, and those microbes may also change the level of dietary host effects after meals, thus impacting long-term health and weight (Figure 1). Our findings have been published in leading peer-reviewed scientific journals.
What's the relationship between the gut microbiome and human health?
Our gut contains a complex ecosystem of trillions of different microorganisms and their collective genomes, known as the gut microbiome. These microbes form a very complex ecological entity in our body that plays an important role in many aspects of nutrition and health. This includes:
The transformation and production of thousands of key metabolites, enzymes, and vitamins (many of which cannot be produced by humans and are not available directly through diet).
The extraction of nutrients from our food.
Our metabolic responses to food.
The structure and community of the gut microbiome have been associated with both health and disease states, including cardiovascular disease, type 2 diabetes, inflammatory bowel disease, cancer, and obesity(1).
The gut microbiota has been suggested to influence metabolic health through several different host-microbiome interactions mediated by diet both directly (through the availability of diet-dependent metabolites) and indirectly (through modulating the composition of the microbiome)(2).
Your gut microbiome is unique
Humans share 99% of the same DNA, however, the human gut microbiome is hugely variable from person to person. Through our research, we've found that even identical twins have very different gut microbiomes, with unrelated individuals sharing 30% and twins sharing 34% of the same gut microbes at species level.
By taking an even closer look at the gut microbiome using new, high-resolution shotgun sequencing technologies (“deep metagenomic sequencing”), it becomes clear that each of us has a unique set of genetic variants (strains) of each species present in each of our guts. This makes us, in fact, unique microbially.
Our research also suggests that host genetics only influences microbiome composition to a limited extent, making the microbiome a rich target for precision nutrition.
While you can’t change your genes, you can modify your microbiome. Changes to the gut microbiota can take place within days of dietary alteration(3,4) thus by eating the types of foods that are linked to “good” microbes, we can potentially change the suit of molecules that our gut microbes produce.
This does not imply your microbiome is completely modifiable, as the microbiome has several poorly understood mechanisms inducing colonization resistance of new strains, and your immune system is a complex machine that plays a critical role modulating the microbes in ways that are still not fully understood. However, your microbiome can be modified by what you eat, and this is a major area of ongoing research at ZOE.
The food we eat influences the gut microbiome
Diet is acknowledged to be a key modifiable factor in manipulating the microbial community(2), with habitual diet shaping the microbial ecology by curating its environment through continuous provision of dietary substrates(5).
The fact there is a relationship between diet, the human gut microbiome, and many metabolic diseases are well known, but until now we have lacked large-scale, high-resolution studies linking these outcomes to individual microbes or to the specific foods we eat(6). This is partly due to the complexity of our diets, the difficulty of measuring them accurately and at scale, disentangling them from other lifestyle and confounding factors, and the personalized nature of the microbiome(7,8,9).
Such analyses also require very large numbers of people with in-depth dietary and metabolic data, alongside new deep metagenomic sequencing techniques which can provide strain-level data for each individual at an affordable cost.
In the PREDICT study, for the first time, we've been able to study the gut microbiome and diet at a scale and complexity that has not been possible before. Using deep metagenomic sequencing, together with long-term dietary data and hundreds of fasting and postprandial blood marker measurements from each PREDICT 1 participant, we've been able to identify a clear set of microbial species that are strongly and consistently linked to dietary patterns, cardiometabolic and obesity-related biomarkers, and postprandial responses.
Figure 2 below illustrates the 15 “good” (shown at the bottom) and “bad” (shown at the top) gut microbes that correlate with diverse positive and negative health and diet indicators, respectively.
This analysis is now being extended by the further 1,000 individuals who participated in PREDICT 2 from a wide variety of ethnic and geographical background. Our understanding of these links will continue to improve based on data from individuals who participate in the ZOE program and join our ongoing PREDICT studies.
Diet quality and the gut microbiome
While nutrients and foods are not consumed in isolation, but rather as part of a broad and varied dietary pattern, there are a few key dietary components that are known to impact our gut microbiome to a greater extent (e.g., dietary fiber, polyphenols, animal products, artificial sweeteners, and dietary fats).
Data from the PREDICT 1 study has found strong associations between microbes and specific nutrients, individual foods, food groups, and overall dietary patterns(10). As discussed, we've identified sets of “good” and “bad” gut bugs, creating microbial signatures that consistently correlate with “healthy” and “unhealthy” foods and dietary patterns(11,12,13).
These microbial clusters were also strongly linked to indicators of obesity and general health and fasting circulating metabolites connected with cardiometabolic risk, as seen in Figure 2.
For example, we found that the same group of beneficial microbes seen in people with diets rich in "healthy" plant foods was also seen in those eating "healthy" animal foods containing good fats, such as oily fish. Similarly, the cluster of "unhealthy" microbes was observed in those eating "unhealthy" plant- and "unhealthy" animal-based foods.
We are the first group to look at dietary intake data in this much detail to disentangle the impact that food sources and quality has on the gut microbiome and health outcomes. This clear segregation of "good" and "bad" microbes according to the heterogeneity of the food source (healthy or unhealthy animal or plant), quality (processed or unprocessed), and overall dietary patterns, highlights the importance of looking beyond just nutrients, macronutrients and single foods in diet-microbiome research.
Linking individual foods to the gut microbiome
Another novel outcome of our program of research is that we have identified specific foods that correlate with individual microbes (shown in Figure 3). Unsurprisingly, most foods that support the growth of “good” microbes are high in fiber. Dietary fiber is used to describe a variety of plant-derived compounds that are not broken down by human enzymes. Without microbes, humans are unable to make full use of both soluble and insoluble dietary fibers. Different microbes are specialized at breaking down different types of fiber, including prebiotic fibers, which are fermented by gut microbiota in the colon. As a result, a wide variety of fiber is important to support a healthy gut microbiome.
By eating specific high-fiber foods, one can likely increase the population of particular microbes. The naturally occurring fiber found in the context of a food matrix is highly complex, which we believe is part of the reason why individuals eating unprocessed foods have a more diverse set of gut microbes and better health outcomes than those eating highly-processed foods. As a result of this research, ZOE scores differentiate between different types of fiber and do not give much credit to highly-processed sources of fiber.
The gut microbiome influences cardiometabolic health and our responses to foods
Data from the PREDICT 1 study suggests that an individual’s gut microbiome is predictive of cardiometabolic markers and personalized responses to food(16). For example, “healthy” and “unhealthy” microbial signatures were associated with several traditional fasting cardiometabolic measures biomarkers (e.g. blood pressure, fasting lipids, fasting glucose, glycosylated hemoglobin (HbA1c), inflammatory markers), as well as postprandial cardiometabolic and glycemic biomarkers. Notably, we observed these clusters under both fasting and postprandial conditions.
Whilst it is known that the gut microbiome plays a role in glucose metabolism and therefore postprandial glycemia(14,15), a novel finding from our research is that that the gut microbiome plays an even more important role in postprandial lipemic responses after a mixed nutrient meal than glycemic responses(10). This is particularly relevant given that both postprandial lipemia and glycemia are independent risk factors for CVD and initiate a cascade of negative effects.
The gut microbiome is an important target for precision nutrition
PREDICT 1 is the first study to identify a panel of gut microbes consistently linked with both dietary intake and metabolic health. The scale and depth of our data also show much stronger links between the food that we eat, the gut microbiome, and cardiometabolic health than previously believed. PREDICT 1 is also the first study to identify both individual components of the gut microbiome and overall microbial signature associated with dietary intake and cardiometabolic health measures.
These findings strongly support the interactions between the food we eat, the microbes they support, and chronic disease outcomes, and highlight the gut microbiome as both a target for personalized dietary advice and valuable input for predicting personalized responses to food.
ZOE’s at-home test kit uses the same sample collection methods and deep metagenomic sequencing techniques as used to analyze the gut microbiomes of participants in our PREDICT studies. This allows us to identify the presence of these “good” or “bad” microbes in any individual and calculate a personalized overall index of microbiome health based on the relative abundance of these microbes.
All ZOE product users are provided with a personalized gut health report, which includes the identification of particular foods that are associated with and thus can potentially support the growth of “good” bugs that are under-represented in their gut. As more individuals join the ZOE studies and can share their microbiome and we validate causal links, their dietary intake and metabolic responses, our ability to link individual microbes to foods will continue to expand significantly. As a result, dietary guidance will further improve in the coming months and years as we better understand the fine details of foods linked to particular microbes.
Singh RK, Chang H-W, Yan D et al. Influence of diet on the gut microbiome and implications for human health. J Transl Med. 2017;15:73. doi:10.1186/s12967-017-1175-y
Tilg H, Moschen AR. Microbiota and diabetes: an evolving relationship. Gut. 2014 Sep;63(9):1513-21. doi:10.1136/gutjnl-2014-306928
Valdes A, Walter J, Segal E et al. Role of the gut microbiota in nutrition and health. BMJ. Jun 2018;361:k2178. doi:10.1136/bmj.k2179
David LA, Maurice CF, Carmody RN et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014 Jan 23;505(7484):559-63. doi:10.1038/nature12820
Leeming ER, Johnson AJ, Spector T. Effect of Diet on the Gut Microbiota: Rethinking Intervention Duration. Nutrients. 2019 Dec;11(12):2862. doi:10.3390/nu11122862
Menni C, Zhu J, Le Roy CI, et al. Serum metabolites reflecting gut microbiome alpha diversity predict type 2 diabetes. Gut Microbes. 2020 Nov;11(6):1632-42. doi:10.1080/19490976.2020.1778261
Sze MA, Schloss PD. Looking for a Signal in the Noise: Revisiting Obesity and the Microbiome. mBio. Aug 2016;7(4):e01018-16. doi:10.1128/mBio.01018-16
Pasolli E, Asnicar F, Manara S et al. Extensive Unexplored Human Microbiome Diversity Revealed by Over 150,000 Genomes from Metagenomes Spanning Age, Geography, and Lifestyle. Cell. 2019 Jan;176(3):649-62. doi:10.1016/j.cell.2019.01.001
Gilbert JA, Blaser MJ, Caporaso JG et al. Current understanding of the human microbiome. Nat Med. 2018 Apr;24:392-400. doi:10.1038/nm.4517
Spector T, Asnicar F, Berry S et al. Microbiome Signatures of Nutrients, Foods and Dietary Patterns: Potential for Personalized Nutrition from The PREDICT 1 Study. Curr Dev Nutr. 2020 Jun;4(Suppl 2):1587. doi:10.1093/cdn/nzaa062_044
Vadiveloo M, Dixon LB, Mijanovich T et al. Development and evaluation of the US Healthy Food Diversity index. Br J Nutr. 2014 Nov;112(9):1562-74. doi:10.1017/S0007114514002049
Guenther PM, O Casavale K, Reedy J et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet. 2013 Apr;113(4):569-80. doi:10.1016/j.jand.2012.12.016
Fung TT, McCullough ML, Newby PK et al. Diet-quality scores and plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr. 2005 Jul;82(1):163-73. doi:10.1093/ajcn/82.1.163
Allin KH, Tremaroli V, Caesar R et al. Aberrant intestinal microbiota in individuals with prediabetes. Diabetologia. 2018 Apr;61(4):810-20. doi:10.1007/s00125-018-4550-1
Kootte RS, Levin Evgeni, Salojarvi J et al. Improvement of Insulin Sensitivity after Lean Donor Feces in Metabolic Syndrome Is Driven by Baseline Intestinal Microbiota Composition. Cell Metab. 2017 Oct;26(4):611-619.e6. doi:10.1016/j.cmet.2017.09.008
Berry S, Ordovas J, Asnicar F et al. Influence of Gut Microbial Communities on Fasting and Postprandial Lipids and Circulating Metabolites: The PREDICT 1 Study. Curr Dev Nutr. 2020 Jun;4(s2):1547. doi:10.1093/cdn/nzaa062_004