e-HoMIniCS - elucidation of Host Microbiome Interactions in Cosmetic Skin

Lead Research Organisation: King's College London
Department Name: Dental Institute


The recent advances in high through-put data generation for DNA/RNA, proteins and metabolites has resulted in a paradigm shift in how we seek to answer some of the fundamental questions of biology. Over the past decade, significant amounts of these large data sets encompassing resident microbial communities (microbiome), specific host responses and environmental conditions have been generated. To date the integration and exploitation of these complex datasets in a structured way has been highly problematic. However recent advancements in in-silico methodologies can for the first time help to unlock the full potential of these data, facilitating improved understanding of and discovery of novel interventions for host-microbiome interactions. With the advent of these technologies it has become apparent that interactions between environmental, host and microbial factors give rise to the various changes in skin homeostasis that result in cosmetic conditions such as dry skin and dandruff. Dandruff and dry skin are widespread conditions impacting over 50% of the world's population affecting quality of life including self/body confidence and their treatment is the basis of a sector worth over 10bn Euros annually.

In this study, in collaboration with our industrial partners, Unilever, we will investigate the physiological changes of normal, dry skin and dandruff through unique integration of computational biology and modelling with microbiology. We will develop a computational and experimental platform for skin host-microbiome interactions to reveal the microbial mechanisms involved in different skin states. Using this approach, we will identify and evaluate new therapeutic targets as well as reveal the underlying physiological events in skin homeostasis.
Using a combination of skin samples collected by tape strips from normal, dry skin and dandruff, as well as data generated from reconstituted skin models and keratinocyte monolayers, we will generate data that accurately describes skin-microbe interactions. we will also identify the key species and strains of Malassezia, Staphylococcus and Cutibacterium associated with different skin states. In parallel by using the available multi-omics data from Unilever and the public domain, we will generate computational models for microbes and host skin tissue and cells. Having both in-silico and in-vitro set ups, we will investigate the impact of key metabolites and anti-metabolites on the relationship between the skin and key microbes and microbial communities. Finally, we will explore the impact of key host factors, such as cytokines (e.g. IL-36, IL-1, IL-17, IL-20 family) and antimicrobial peptides (e.g. beta-defensins, S100, LL-37) on the resident microbial communities. We will then categorize these therapies based on their mode of action on skin-microbiomes interactions. The new therapeutic targets generated and validated through this combination of both computational and experimental techniques can then be tested for host toxicity and efficacy. This cutting-edge integrative platform could be easily extended to identify new targets or drugs for different microbial constituents in human body, their association with a range of hosts and pathologies. As such it will delineate an entirely novel approach to investigating host-microbiome interactions that will have broad applicability across a wide range of sectors, including medical, veterinary, cosmetic and agricultural.

Technical Summary

Current approaches to investigating host-microbiome interactions are somewhat piecemeal. We will take a fully integrated approach, combining state-of-the-art computational modelling approaches to predict events and targets for modulation, with conventional microbiological and cell biology techniques to generate omics data and validate target efficacy in an iterative fashion. We will generate high-quality multi-cellular genome-scale metabolic models (GEMs) of Skin host-microbes interaction for dandruff, dry and healthy skin alongside in vitro models. These models will be derived from metagenomic, transcriptomic and metabolomic data generated from skin and in vitro models cocultured with unique multi-kingdom microbial. We will then integrate the computational and experimental outputs to predict and refine these communities. We will next investigate the impact of known metabolites and drugs on host-microbiome interactions, targeting metabolites through anti-metabolite analysis, and enzyme inhibition or gene silencing for gene/enzyme targeting. The in silico anti-metabolite and gene silencing analysis will be utilized to identify novel modulation targets for skin state and microbiome while toxicity of targets will be tested in host cells to check the side effects. A select panel of computational candidate targets will be validated experimentally before any further analysis. Additionally, this project develops a directed bipartite graph-based algorithm to model metabolic, regulatory and signaling networks simultaneously, predicting the effect of cytokines in maintaining host-microbiome homeostasis at the skin. Thus, we propose a pipeline for generating skin host-microbiome interactions to elucidate the physiology of different states, as well as identifying new therapeutic targets for their modulation. This approach can be extended to predict targets or drugs for host-microbiome interactions of other sites, including medical, veterinary and agricultural.

Planned Impact

Skin is the body's largest organ covering an area of over 2 square metres and contributing about 15% of total body weight. Dandruff and dry skin are widespread skin conditions impacting over 50% of the worlds' population, impacting significantly on quality of life and self/body confidence. Their treatment is the basis of a sector worth over 10bn euros annually, with the NHS alone spends more than £4.5 million each year on anti-dandruff medicated shampoos. Despite such high prevalence, their aetiology is not well understood. Treatment for these conditions often consists of antifungal shampoos. However, this is not always successful, and has the potential to increase antimicrobial drug resistance. Whilst there have been several recent studies generating significant amounts of data regarding the processes involved in microbiome-host interactions in healthy and dandruff skin, little effort has been made to effectively integrate the different data sets. By integrating different data sets, we will generate a novel approach that combines in silico modelling with in vitro and in vivo investigations to delineate the key mechanisms and events in skin homeostasis. By doing so, it will be possible to formulate new treatment modalities that will be less aggressive, focussing more on altering host-microbiome interactions through changing the environment, rather than killing of or removing microbes from otherwise healthy skin.

Academic researchers: Results from this work and the tools generated will provide an invaluable resource set that can be applied to understanding basic interactions between microbial and their host in both a healthy and unbalanced microbiota. These tools will help to answer key questions surrounding our current knowledge of microbiome, including understanding how imbalances of a microbial community can lead to changes in host tissue homeostasis and concomitant pathologies. Further, it will also have an impact on how we approach investigating host-microbiome interactions. As well as investigations of skin-microbiome, these tools will be equally applicable to investigating other tissue-microbiome interactions and will pave the way for a truly integrative approach to investigating microbial communities.

Industry: The approach and tools validated through this proposal will provide a rapid screening methodology that will enable industrial research to comprehensively screen a large panel of drugs, metabolites and anti-metabolite treatments to identify promising candidates for further investigation. This will reduce the amount of time and money spent following false leads and thereby streamline the drug/target discovery pipeline. This will have an impact on pharmaceutical, personal care and food producers.

Public: Through public dissemination plans, this work will provide the public with a greater understanding of our microbiomes, and how we can manipulate these communities to our advantage in many avenues. Moreover, completion of this work will allow the identification of novel compounds and treatment modalities for these common but potentially debilitating conditions, improving the quality of life for millions.
Governmental and NGOs: Results from this study will guide future studies investigating the microbiome and host-microbiome interactions, as well as methods to manipulate these communities. Further, by reducing the need for antimicrobial use, results of this study will have a significant impact on antimicrobial drug resistance development. This will reduce the healthcare burden for many modern-day conditions, both in the UK and in developing countries, where long-term treatments are not a viable option for many.


10 25 50