Understanding the eco-evolutionary drivers of emerging antifungal resistance

环境中的微生物暴露在不断变化的条件下，这些条件会选择最适合的变体。这种持续的适应过程导致种群的遗传组成随着最适突变的变化而在空间和时间上发生变化。不幸的是，当被设计用来杀死微生物的化学物质施加选择时，那些具有基因抗性的物质出现的频率就会上升；这导致了在环境中不断演变的抗菌素耐药性这一全球性问题。虽然细菌中出现的抗菌素耐药性得到了广泛的认可，但对抗真菌化学物质具有耐药性的真菌的出现却没有得到充分的重视，这损害了我们种植无枯萎病作物和治疗严重的人类真菌疾病的能力——因此出现了一个经典的“同一个健康”困境。我们项目的核心重点是曲霉，这是一种常见的环境霉菌，由于它们在空气中无处不在，所有人类都暴露在这种霉菌中。值得注意的是，烟曲霉影响着全世界数百万易感个体（包括COVID-19患者），并且越来越多地引起对用于治疗该病的一线抗真菌药物具有耐药性的疾病。至关重要的是，这类化学物质与农民用作杀真菌剂的化学物质是同一类的，这导致了抗唑烟曲霉数量的激增，因为这种霉菌在自然环境中受到这些化学物质的选择。然而，我们目前对导致杀菌剂化学残留物积累到选择和增强霉菌抗性的浓度的景观尺度途径知之甚少。对于不同杀菌剂组合使用对耐药性产生的影响，或者与更广泛的微生物群落的相互作用如何阻碍（或帮助）耐药性的产生，我们了解得更少。我们的项目将研究嵌套的人为驱动因素——农业实践和绿色废物回收——目的是了解它们如何为抗真菌耐药性病原体创造进化热点。我们将重点关注的霉菌嵌入复杂的微生物生态系统中，我们将确定规模的影响，从全国范围内的真菌分布，通过在富含杀菌剂的中生态环境中看到的生态演代，一直到个体模型微观模型。为此，我们将结合基于贝叶斯的统计方法进行实地和实验室研究，在空间明确的框架内考虑到进化和生态复杂性。通过这样做，我们将能够识别、理解并将导致抗杀菌剂霉菌形成热点的关键因素联系起来。我们测量的变量——土地利用、杀菌剂、真菌遗传学和微生物群落生态学——将被整合到一个系统网络分析中，该分析将杀菌剂在环境中的使用与选择耐药性的生态环境联系起来。这些“贝叶斯概率网络”是一个强大的工具，它将使我们能够预测真菌耐药性的热点，并允许我们建模方法，通过减少我们确定的特定“关键点”的杀菌剂投入来减轻这种风险。最终，通过剖析杀菌剂使用的延伸（无意）后果，因为这些化学物质推动了真菌抗菌素耐药性的演变，我们的项目将在更大的“同一个健康”背景下解决这一问题。迫切需要我们的方法来开发知识库，以了解当前的风险，并减轻环境中真菌抗微生物药物耐药性未来出现的选择压力。

Microbes in their environment are exposed to changing conditions, which select for the most fit variants. This continual process of adaptation leads to the genetic composition of populations shifting in space and time as the fittest mutations track change. Unfortunately, when selection is imposed by chemicals that are designed to kill microbes, then those that are genetically resistant rise in frequency; this results in the global problem of antimicrobial resistance evolving in the environment.While emerging antimicrobial resistance is widely recognised in bacteria, the emergence of fungi that are resistant to antifungal chemicals is underappreciated yet is compromising our ability to grow blight-free crops and to treat serious human fungal diseases -therefore presenting a classic One Health dilemma. The core focus of our project is Aspergillus species, common environmental moulds to which all humans are exposed due to their ubiquitous presence in the air. Of note, A. fumigatus affects millions of susceptible individuals worldwide (including those with COVID-19) and is increasingly causing disease that is resistant to the frontline azole antifungal drugs that are used to treat it. Crucially, this is the same class of chemicals that are used by farmers as fungicides, which is driving a surge in azole-resistant A. fumigatus as this mould comes under selection by these chemicals in its natural environment. However, we currently have very little understanding of the landscape-scale pathways that lead to fungicide chemical residues accumulating to the concentrations that select for, and amplify, resistance in moulds. We understand even less about the consequences combinations of different fungicides on the emergence of resistance, or how interactions with the wider microbial community that may hinder (or help) the emergence of resistance.Our project will examine the nested anthropogenic drivers - agricultural practices and green-waste recycling - with the aim of understanding how they create hotspots of evolution for antifungal resistant pathogens. The moulds on which we will focus are embedded in complex microbial ecosystems and we will determine the impact of scale from country-wide distributions of the fungus, through the ecological succession seen in fungicide-rich mesocosm environments, and down to individual model microcosm models. To do this, we will couple field and laboratory studies with Bayesian-based statistical methods that take into account both evolutionary and ecological complexity within a spatially-explicit framework. In doing so, we will be able to identify, understand and link the key factors that lead to hotspots of fungicide-resistant moulds forming. The variables that we measure - landuse, fungicides, fungal genetics and microbial community ecology - will be integrated into a systems network analysis that links the usage of fungicides in the environment to ecological settings where resistance is selected for. These 'Bayesian probabilistic networks' are a powerful tool which will allow us predict hotspots for fungal drug-resistance, as well as allowing us to model methods to mitigate against this risk by reducing fungicide-inputs into specific 'pinch-points' that we identify. Ultimately, by dissecting the extended (unintentional) consequence of fungicide use as these chemicals drive the evolution of fungal antimicrobial resistance, our project will address this problem within its greater 'One Health' context. Our approach is urgently needed to develop the knowledge-base that is needed to understand the current risk as well as to mitigate the selection-pressure driving future emergence of fungal antimicrobial resistance in the environment.