We believe we have found a contributing factor – sugar, in particular glucose. Diabetes is defined by elevated blood glucose. 13 years ago, Prof Emma Baker and Prof Debbie Baines (at St George’s University of London) noticed that additionally, diabetics have increased airway glucose. Normally, the cells that line the airways pump any glucose that leaks into the lungs back into the blood. In diabetes, there is too much sugar in the blood and the pumps are overwhelmed, leading to a rise in airway glucose.They hypothesized that the increased level of sugar in the lungs would allow more bacteria to grow in the lungs – the biological equivalent of leaving a jam jar open!
In our latest paper in Scientific Reports (Increased airway glucose increases airway bacterial load in hyperglycaemia) we set out to test this hypothesis using a number of different techniques. First we looked in hospitalised patients to see if there was a link between glucose and bacterial infection, and there was, patients with high blood sugar were twice as likely to have a bacterial lung infection. We know this thanks to our collaborators, Dr Luke Moore and Professor Alison Holmes, who have been tracking bacterial infections in London hospitals. This kind of a study is called an association or correlation study, and these studies are very good at showing that one thing is linked to another, but do not tell whether the link is causal and if it is how (the mechanism in scientific parlance).
In order to understand the how, we investigated how bacteria use glucose in the lung. The way we do this is to delete individual bacterial genes and compare the function of these gene deleted mutant bacteria to bacteria with all their genes (wild type). We deleted four different genes that based on their shape and similarities to genes from other bacteria were predicted to be important for the bug to be able to use glucose. These studies were performed using a bacteria called Pseudomonas aeruginosa, which, unless you have cystic fibrosis, you’ve probably never heard of, but causes many cases of pneumonia each year, especially in hospitalised patients. The first step was to demonstrate that deleting the genes affected Pseudomonas ability to use glucose to grow. Great news, they do.
The final step was to link everything - high glucose, in the lungs and bacteria - together. We did this using mice with diabetes (yes they do exist). As seen in people with diabetes, diabetic mice get more severe bacterial lung infections, unless you infect them with bacteria that can’t use glucose. When these bacteria were used, there was no difference in the bacterial lung infection. Boom, job done.
But why stop there, understanding the factors that increase infection gives us new ways to fight infection. This is particularly important for bacterial infections because our arsenal of antibiotics is rapidly being depleted and we desperately need new treatments. If increased lung glucose increases infection it follows that drugs that reduce lung glucose should reduce infection. We tested the common anti-diabetic drug, metformin. Diabetic mice treated with metformin had lower lung glucose and less bacterial infection.
In conclusion, we have linked increased bacterial infection in people with diabetes to the level of glucose in the lungs, and used this finding to test new antibacterial treatments. If you want to read more details the paper is here.