The gut-kidney-heart axis as a driver of cardiovascular disease progression

Even a moderate decrease in how well our kidneys work (by 20-30%) at levels that would not refer a patient to the kidney clinic, can double the risk of future heart disease. For example, in the UK, more than 5million people are predicted to have diabetes by 2030. Of those, approximately 40% (2 million) will develop kidney complications and many will require dialysis as their disease progresses. Patients with diabetes whose kidneys do not work well are up to 10 times more likely to die from strokes and heart attacks in the next 10 years than those who do not. It is therefore important to find new ways to protect the kidney.

The millions of bugs that live in our digestive system, especially gut; known as the gut microbiota, affect how we use our food. One way that these bugs in our guts can communicate with their host is by producing chemicals (called metabolites) that I and other researchers have shown that can also affect how our kidneys work.

In this research, I want to find new ways to protect the kidney by reducing harmful metabolites produced by the bacteria that live in our gut.
In pilot studies, using machine learning and data from 2200 patients from Germany, France and Denmark, I discovered that a chemical made by bugs from the amino acid phenylalanine called phenylacetylglutamine can be harmful for the kidney, while another chemical that can also be made from phenylalanine called 3-phenylpropionate is protective.

Now, working together with leading scientists and medical doctors from Germany, France, Denmark and Canada I will use machine learning and big data analyses to see if the balance of these chemicals in the blood can be used as an early warning sign of future serious kidney and heart complications in humans. In addition to the original study (MetaCardis, N=2200) I generated my pilot data in, I will also use information from two human studies that followed and collected data from healthy people (Longitudinal Canadian Study of Aging with 9,500 participants) or people at early stages of chronic kidney disease (German CKD with 5,000 participants) for up to 6 years. For this part of my work, I will use a wealth of information from 16,700 individuals from three independent human studies that has costed tens of millions of pounds to be collected.

Then, in the second part of my study, in London and Oxford, together with a Research Assistant I will hire as part of this project and in collaborations with scientist from the Imperial College National Heart and Lung Institute and MRC Harwell, I will treat cells and mice that have diabetes and high blood pressure with these chemicals to study how they change the way the heart and kidney works. In this way, I hope to better understand their mode of action and how to better protect the heart and the kidney by harnessing the microbiome.

Finally, in London and Oxford at the Imperial College NHLI and MRC-Harwell and in collaboration with the leading expert of bacterial phenylalanine metabolism from the University of Stamford in the USA, I together with the Research Assistant will test modified bacteria, existing drugs and probiotics in mice to see if I can restore the balance between these chemicals and protect these mice from heart and kidney disease. By re-purposing interventions already safe for use in patients I hope to be able to "hack" the microbiome to produce a beneficial chemical instead of a harmful one.

In summary, if successful, my research will introduce a new risk factor (microbiota phenylalanine metabolism) that can help medical doctors better predict which patient is more at risk of heart attack or kidney failure and prioritise treatment. Additionally, my work will generate new information to improve our understanding of how the bacteria in our gut change the way our kidneys and heart works. Finally, by finding ways to re-program microbiome phenylalanine metabolism, my work can directly lead to human clinical trials.

Even a modest decrease in kidney function (by 20-30%) can result in an up-to two-fold increased cardiovascular disease (CVD) risk. The gut microbiome associates with deteriorating cardiorenal phenotypes, thus offering untapped opportunities for novel therapeutic interventions.

I discovered in the MetaCardis study (N=2,200) that imbalance in microbial phenylalanine catabolism is a hallmark of cardiorenal deterioration. Phenylacetylglutamine (PAG), originating from microbial oxidative phenylalanine catabolism, associated with worsening cardiorenal phenotypes whilst 3-phenylpropionate (3-PP), that can be produced by reductive microbiota phenylalanine catabolism had contrasting effects. Circulating PAG increased as disease progressed and was associated with aberrant microbiota. Moreover, in MetaCardis specific bacterial species correlated with improved 3-PP/PAG ratio and phenylalanine catabolism balance could be restored by clinically-actionable interventions.

I will now build on my pilot data and use three independent human populations with a total of 16,700 participants with available omics data and follow-up clinical information for up to 10 years and an animal model of progressive CVD mirroring the risk factors in MetaCardis (obesity and hypertension to demonstrate that: (1) Distinct human gut metagenomic and phenotypic signatures associate with phenylalanine catabolism imbalance and are predictive of future adverse cardiorenal events. (2) PAG adversely impacts the kidney, whilst 3-PP is protective. (3) Restoring 3-PP/PAG balance protects the kidney and alleviates associated CV risk.

My proposal could lead to novel diagnostic tests better predicting cardiorenal disease risk and clinically-actionable interventions to restore phenylalanine catabolism balance; thereby impeding CVD progression.