A genome-scale census of virulence factors in the major mould pathogen of human lungs

Infectious diseases caused by fungi are a worldwide problem causing at least as many deaths as malaria and tuberculosis. For infections caused by spore-forming moulds there is only one available class of oral drug, the azoles. Azole resistance amongst clinical mould isolates is increasing and one possible cause of this problem is the widespread use of azoles in the environment as agricultural fungicides. There is therefore a desperate need to develop new antifungal treatments.

Fungal particles (or spores), which are continually present in the air that we breathe, are able to survive in the human lung. Usually, this occurs when the immune system of the infected individual is not functioning optimally. Always, it involves the germination of the fungal spore (rather like a plant seed) and growth of long fungal cells called hyphae. Hyphae penetrate and dissolve lung tissues by secreting enzymes. The major fungal pathogen in the air we breathe is called Aspergillus fumigatus and people affected by cancer, or requiring organ transplants, are particularly at risk of fatal infections caused by it. Unfortunately our lack of understanding of A. fumigatus, and the technical difficulties encountered when working with it, has limited progress in this field. We do not fully understand why Aspergillus is able to survive inside the lung environment. We urgently need more basic information about this pathogen so we can begin to design future therapies.

When scientists try to figure out how cellular processes work, a first port of call is often transcription factors. Transcription factors control the activity of many genes simultaneously so they command great power over cellular processes. If a transcription factor is important for a particular process, one will then look at which genes it regulates and this will give clues to the biology underlying that process. If a transcription factor is not involved then it, and all of the genes it regulates, can be eliminated from the investigation. By asking which transcription factors control the ability of A. fumigatus to cause disease, we can achieve a global view of the regulatory network which drives infection. We intend to couple this approach to a state of the art DNA sequencing technology to gain maximal insight on A. fumigatus pathogenicity, with minimal usage of mice.

We have searched the A. fumigatus genome for transcription factor genes and created a collection of 401 A. fumigatus mutants, each one lacking a transcription factor. To ask which of the transcription factors is important for mammalian lung infection we will infect mice with the collection of mutants and see which are unable to survive inside the lung. We have devised a method which will allow us to test many mutants at the same time. This involves infecting all of the mutants into a single mouse, and after sacrifice, sequencing the DNA extracted from the lung to tell us which mutants have managed to survive. We will use this method to reveal all of the transcription factors driving pathogenicity (P-TFs) in A. fumigatus. Thus the lowest possible number of animals will be utilised. When we have identified all of the P-TFs we will then work to identify the genes which are regulated by each P-TF. In order to do this we will use two approaches. First we will chemically fix the transcription factor to A. fumigatus DNA to identify the binding sites of each transcription factor. Second we will look at gene expression in mutants lacking P-TFs and work out which genes become deregulated. The success of this investigation will 1) Redress the meagre progress made in recent decades in defining the molecular basis of pathogenicity of Aspergillus fumigatus 2) Lead us directly to the fungal processes we need to target with new drugs 3) Hugely reduce the number of animal infection studies performed by other scientists on A. fumigatus 4) Help us to optimise high throughput manipulation of A. fumigatus for future drug screens.

Spores of the common Aspergillus moulds are agents of widespread, often fatal, human and avian diseases, however, amongst a repertoire of more than 400 predicted transcription factors, fewer than ten have been characterised for their role in disease.

We have developed a 'first-in-field' collection of 401 transcription factor knock out (TFKO) mutants for Aspergillus fumigatus, the major mould pathogen of humans, and now propose to identify those which govern pathogenicity. We have also developed and optimised a high throughput parallel fitness screen with which to functionally annotate transcription factor activities. We will use this method to assess the fitness of each mutant under infection-relevant laboratory-imposed stress, and in two murine models of aspergillosis. This will enable the functional annotation of the entire genomic cohort of transcription factors and identification of those which support pathogenicity (P-TFs). We will construct isolates expressing epitope-tagged P-TFs to identify the regulatory network of gene products they govern. We will use chromatin immunoprecipitation to capture DNA bound by P-TFs and massively parallel DNA sequencing to identify them. We will then use RNA-Seq to verify the functional significance of promoter binding by P-TFs. This work will allow us to construct a map of the functional regulatory network which governs pathogenicity and elevate our understanding of this important pathogen.

1. Individuals suffering, or at risk from, Aspergillus-related disease: The health burden of diseases caused by fungi is considerable. Where severe invasive infections are concerned, deep-organ infection occurs as a negative corollary to otherwise successful, but nonetheless expensive, treatments for serious illness. In 2002 almost 10,000 patients in England were estimated to have suffered a deep-organ fungal infection. Almost half of them died from their fungal disease (HPA). The pipeline for new antifungal compounds is sparse and antifungal resistance is advancing. The use of agricultural azoles is prompting emergence of clinically significant resistant isolates. Three new triazoles (albaconazole, isavuconazole and ravuconazole) are in clinical trials however, as with extant triazoles, they target lanosterol 14-demethylase and are likely to be dogged by similar resistance issues. The glucan synthase inhibitor in development by Merck has been stalled in Phase I since 2009, as have the novel compounds corifungin (Acea biotech), T-2307 (Toyama) and the HDAc inhibitor of Methylgene. All of this indicates potential problems in development.

2. Healthcare Service Providers: The cost of invasive infections caused by Aspergillus sp. in the US alone was estimated at $633 million in 1996. This cost included the expense of failed chemotherapy, or bone marrow or organ transplantation [Dasbach EJ, Davies GM, Teutsch SM (2000) Burden of aspergillosis-related hospitalizations in the United States. Clin Infect Dis 31:1524-8].

3. Pharmaceutical companies seeking novel antifungal agents: The global market for clinical antifungals was estimated to be $9.4 billion US in 2010 (GBI research) and is anticipated to exceed $12 billion by 2016. The advancement of novel target-based strategies currently falls outside the remit of many large drug discovery companies and it will therefore fall to academic partners to progress research in this area. This deficiency will require pharmaceutical companies to interact with academics that have expertise in drug design and toxicology. The tools we will develop will have direct applicability to these efforts, and their availability will help to forge links that will provide tangible and valuable outputs far beyond the scope of this project.

4. Scientists and industries pursuing the advancement of systems biology in Aspergilli: Our study will lead to whole genome systems approaches in Aspergilli, which are a hugely significant fungal genus to man, not least to the human economy. Aspergilli are used in production of acidification agents (citric acid), hydrolytic enzymes (such as amylase used for hydrolysis of starch in bread and beer), invertase (confectionary), and pectinases (fruit juice and wine production). Aspergillus oryzae is widely used for the production of fermented foods and beverages in Japan. Over half all bread production in the USA utilises A. oryzae proteases to liberate amino acids required for yeast growth and respiration. A. niger is used as a producer of polysaccharide-degrading enzymes (amylases, pectinases, xylanases) and organic acids. The world market for such enzymes has an estimated worth of US$ 5 billion. High yield production strains of Trichoderma reesei and Aspergillus niger are used for the manufacture of enzymes worth in the region of $1 billion per annum. Aspergilli are important sources of natural products. Statins, whose use has had a dramatic impact upon incidence of coronary artery diseases, and prevention of stroke and peripheral vascular disease, were first derived as lovastatin from Aspergillus terreus. It is widely recognised that enzyme cost is the single biggest economic barrier to the mass conversion of lignocellose waste for bioethanol production. Our parallel fitness model could be employed in academic laboratories to assess in a high throughput manner, the fitness of multiple genetic variants in industrially relevant contexts.