Scientists at the Francis Crick Institute have generated a comprehensive database of mouse gene expression across 10 different infectious and inflammatory diseases, and are making the resource freely available to scientists around the world through a web app. The data, which define how the mouse immune system responds to different pathogens and a common allergen, show the activity of all of the 45000 or so mouse genes in lung tissue and blood for different infections. Scientists can use the resource to evaluate the activity of any gene, without having to generate experimental mice. Traditionally, they would have had to infect mice, extract tissue samples, and sequence the RNA to study the genes.
“By sequencing both lung tissue and blood, we can also see how the immune response in the blood reflects the local response in the lung, and vice versa,” said the Crick Institute study lead Anne O’Garra, PhD, who is also corresponding author on the team’s paper in Nature Communications. “This will help us to understand what we can learn from genetic signatures in the blood, since for most diseases doctors can’t realistically get lung samples from patients.”
The research, which was coordinated by Christine Graham, PhD, principal laboratory research scientist at the Crick Institute, included a team of collaborators in the U.K. and U.S. Their published paper is titled, “Transcriptional profiling unveils type I and II interferon networks in blood and tissues across diseases.”
In response to infection and inflammation animals, including mice and humans, mount complex immune responses, which may be driven by different groups of cytokines, the authors wrote. Understanding how different immune challenges elicit different responses is critical for diagnosing and deciphering immune regulation. However, “there are few transcriptional studies or data resources on the global immune responses spanning different experimental models of diseases across distinct types of immune responses.”
A number of prior studies have investigated gene transcription signatures in human blood in response to disease, the team continued. “… many published whole blood disease-specific signatures are dominated by IFN [interferon]-inducible signatures and those attributable to innate immune responses, as broadly described in both experimental models and human diseases.” However, despite these indicators it’s not really understood how immune responses in blood are reflected at the disease sites.
The researchers used a next-generation RNA sequencing technology, RNA-seq, to evaluate gene expression in the lungs of mice infected with one of a range of pathogens. “To determine the global changes in the host response to infection and allergens, we performed RNA-based next-generation sequencing (RNA-seq) on RNA isolated from both lung and blood, at the predetermined peak of the response of mice infected with T. gondii [Toxoplasma gondii] influenza A virus (influenza); respiratory syncytial virus (RSV); acute Burkholderia pseudomallei (B. pseudomallei); Candida albicans (C. albicans); or challenged with the allergen house dust mite (HDM) … blood transcriptional signatures were also investigated in a distinct set of mice infected with Plasmodium chabaudi chabaudi (P. chabaudi, malaria), murine cytomegalovirus (MCMV), Listeria monocytogenes (Listeria) and chronic B. pseudomallei.”
The authors said they chose the pathogens and an allergen to trigger a wide breadth of different types of immune response in the lung, representative of TH1, type I IFN, TH17, and TH2 responses, under the hypothesis that “distinct responses underlying the immune response in each model could be determined by the transcriptional signature of unseparated lung cells.”
The team then used sophisticated bioinformatics strategies, including Weighted Gene Co-expression Network Analysis, to group genes that were co-expressed across the lung and blood samples into different modules, which made it possible to decipher the global transcription responses in the lungs, and to determine to what extent each of the responses was retained in the blood. “Gene activity can show us how the body responds to infections and allergens,” O’Garra explained. “There are thousands of genes involved in any immune response, so Akul Singhania, a bioinformatics postdoc, in our lab used advanced bioinformatics approaches to cluster the genes into modules. These modules represent clusters of genes that are co-regulated and can often be annotated to determine their function and known physiological roles.”
The results confirmed that the different infections triggered specific immune responses in the lungs, which were preserved in the blood. Overall, type 1 IFN and IFN-γ signatures were common across all the diseases, except during C. albicans infection. “The overall spectrum of signatures demonstrated distinct immune response patterns in the lung, “ranging from very high IFN- γ expression, type I IFN-inducible gene expression, IL-17-induced neutrophil dominated signatures and expression of Il4, Il5, Il13, and mast cell-associated genes.”
For example, O’Garra continued, “of 38 lung modules there is a module associated with allergy, and seen only in the allergy model, containing over 100 genes and another module associated with T cells containing over 200 genes.”
Broadly, the results showed that the discrete modules were dominated by genes associated with type I or type II interferons, IL-17 or allergy type responses. Type I interferons are known to be released in response to viruses, while type II interferon (IFN- γ) activates phagocytes to kill intracellular pathogens, and IL-17 attracts neutrophils causing early inflammatory immune responses.
Interestingly, interferon gene signatures were present in blood modules similar to the lung, but IL-17 and allergy responses were not. Genes associated with type I interferon were in addition highly active in both the lungs and blood of mice infected with the Toxoplasma gondii parasite and also seen in response to Burkholderia pseudomallei infection although to a lesser degree. “IFN-inducible signatures were preserved in blood, with strongest co-expression during T. gondii infection,” they wrote. This finding challenges the notion that type I interferon-associated genes are indicative of viral infections.
“We found that mice without functioning interferon pathways were less able to fight off Toxoplasma infection,” O’Garra stated. “This was true for both type I and type II interferons, which have a complex relationship with each other. We found that both play a key role in protection against the parasite in part by controlling the neutrophils in the blood which in high numbers can cause damage to the host.”
The researchers suggest that their work represents “a comprehensive resource of modular transcriptional signatures from various infectious and inflammatory diseases to identify commonalities and differences in the immune response to specific infections or challenges to aid the discovery of pathways in disease … Our framework provides an ideal tool for comparative analyses of transcriptional signatures contributing to protection or pathogenesis in disease.”
The results are the culmination of a decade of research. Initiated in 2009, the project was originally focused on the use of then well-established microarray techniques to analyze gene activity in blood and lung samples. However, after an unforeseen hiccup, the researchers had to abandon the approach, and were subsequently presented with the opportunity to exploit newly launched RNA-Seq technology to re-analyze the samples. “Ten years since the project began, we now have an open-access resource of gene expression that anyone in the world can use to look up their favorite genes and also see if they are regulated by type I or type II interferon signaling,” said O’Garra. “Nobody said science was easy, but it’s certainly worthwhile.”
The open resource is available via the online webapp: https://ogarra.shinyapps.io/MouseModules/