Watching immune cells fight malaria
Watching immune cells fight malaria
- Scientists reveal for the first time how immature mouse immune cells take on different roles to fight malaria infection.
- Researchers combined single cell genomics and machine learning methods to model and unpick complex biological processes.
- Study identifies new genes that may be involved in controlling antibody production during infection, and could be targets for drugs that aim to boost immunity to malaria.
Researchers from the European Bioinformatics Institute (EMBL-EBI), the Wellcome Trust Sanger Institute and QIMR Berghofer Medical Research Institute, Australia have used single-cell genomics and machine learning methods to track individual mouse T cells during infection with malaria parasites. Their findings, published in Science Immunology, provide valuable insights into T-cell specialisation.
The study also discovered genes that may be involved in controlling antibody production during malaria infection. One of these, Galectin 1, encouraged development of a particular type of T cell when active. These genes are possible targets for drugs that aim to boost immunity to malaria and other infections.
The immune system is extremely complex and responds to pathogens by developing specific types of immune cells. To fight infection, immature T cells develop into specific subtypes – T helper1 (Th1) and T follicular helper (Tfh) cells – that carry out different functions. Th1 cells help remove parasites from the bloodstream, and are needed early on in an infection. Tfh cells have a bigger role to play in longer-term immunity.
In this study, the researchers discovered that more Th1 cells were produced when a gene encoding Galectin 1 was active.
“Using single-cell genomics, we uncovered the inter-cellular conversation that is taking place between immune cells such as monocytes and Th1 cells – this has not been seen before,” says Dr Oliver Stegle, Research Group Leader at EMBL-EBI and a senior author on the paper. “Our data have allowed us to uncover tens or hundreds of new genes that may be involved in controlling the production of antibodies.”
Stegle explains that activity in these genes may either help the body, for example in curing an infection, or work against it by allowing cancerous cells to flourish.
“The principles and computational methods we have developed here could be applied to future studies to explore these questions,” he adds.
“This is the first time that Galectin 1 acting inside T cells has been seen to influence Th1 fate, and has shown that Galectin 1 is a possible therapeutic target for malaria,” explains Dr Ashraful Haque, joint lead author from QIMR Berghofer. “An important next step will be to test many of the new gene targets identified by our studies, to see if they can be targeted by drugs to boost immunity to malaria.”
The exact molecules that encourage the T cells to develop into one form or the other are not well understood.
In this study, the researchers used single-cell RNA sequencing to take high-resolution ‘snapshots’ of the active genes produced by each individual T cell after in a mouse infected with the malaria parasite. The team put the snapshots together and reconstructed a T cell’s journey from immature to fully specialised.
“This is the first high resolution time-course of cells using a pathogen in mice, where we have used cutting edge genomics coupled with computational methods to reconstruct how cells evolve and develop over infection,” concludes Dr Sarah Teichmann, Head of Cellular Genetics at the Sanger Institute and senior author on the paper. “Building on powerful methods from machine learning, we have simplified really complex biological processes into something we can understand. This approach could also be applied to resolve other biological developmental process.”
Modelling fate with machine learning
The team developed a new computational modelling method called GPfates, which allowed them to see how all the cells related to each other. GPfates builds on approaches from spatio-temporal statistics to characterise how cells differentiate. In this case, the team used it to show which genes are switched on in each of the two distinct cell states (Th1 and Tfh).
The software framework underlying this machine-learning method for single cell RNA data analysis was first developed by computer scientists in Sheffield, UK to enable the flexible implementation of a variety of computational models.
GPfates and the PlasmoTH database for TH1/TFH fate commitment discovery are freely available.
Lönnberg T, et al. (2017) Single-cell RNA-seq and computational analysis using temporal mixture modeling resolves TH1/TFH fate bifurcation in malaria. Science Immunology. Published online 3 March; DOI: 10.1126/sciimmunol.aal2192
This news story is based on a press release, first published on the Wellcome Trust Sanger Institute website.