Let the right ones in: Testing microeconomic models of screening in an ant-bacteria microbiome

美洲热带地区的一群蚂蚁，被称为“阿丁”或“切叶者”，至少在5000万年前进化出了农业。这些蚂蚁收集植物并将它们带回巢穴，在那里它们咀嚼植物以喂养一种特殊的真菌，这种真菌只能与切叶蚁一起生活。毫不奇怪，在新大陆的某些地方，切叶蚁是一种害虫，能够在一个晚上从整棵橙子树上剥下叶子，以喂养巨大的真菌花园，这些真菌花园填满了两层楼高的地下蚁巢（在伦敦，而不是拉斯维加斯）。作为住房和食物的回报，真菌产生富含脂肪和糖的结构，称为gongylidia，蚂蚁收获作为食物。科学家称这种相互依赖为互利共生，因为蚂蚁和真菌相互受益。蚂蚁还通过清除霉菌来保护它们宝贵的真菌花园，如果不加以控制，霉菌最终会吞噬花园。蚂蚁还使用抗生素来杀死外来的霉菌。它们从另一个共生者那里得到抗生素，共生者也和蚂蚁生活在一起。这些共生菌是一组特殊的细菌，称为放线菌，它们（在生物学家中）以制造多种抗生素而闻名，其中一些抗生素我们用作药物，如红霉素。因此，放线菌也是蚂蚁和真菌花园的互利共生者，因为细菌对抗疾病，作为回报，它们生活在蚂蚁的身体上，那里专门的腺体似乎为细菌提供食物。我们已经表明，许多放线菌物种生活在蚂蚁身上，并提供抗生素的混合物，可能是为了减缓入侵真菌花园的疾病的抗生素耐药性的进化。生物学家将生活在宿主生物体上的细菌群落称为微生物组。在蚂蚁的微生物组中，一组被称为假诺卡氏菌的放线菌被认为是世代相传的，适应了它的蚂蚁宿主。其他放线菌，大多数是链霉菌，似乎每一代都从土壤中重新获得。这是令人惊讶的，因为土壤中充满了细菌，其中大多数不是链霉菌，但不知何故，蚂蚁能够选择性地吸收有用的，产生寄生虫的细菌，而不是有害或无用的细菌。在概念层面上，这个问题与汽车救援公司所面临的问题是一样的，这些公司只想向拥有可靠汽车的客户出售保险。土壤细菌和潜在客户的特征是隐藏的，但每个人，无论好坏，可靠或不可靠，都希望生活在蚂蚁身上或在路上被拯救，经济学已经开发出一种解决隐藏特征问题的方法，即所谓的筛选。应用于蚂蚁，我们的假设是，蚂蚁提供了正确的资源组合，以促进细菌之间的战斗。赢家是那些能够释放抗生素的人，因为抗生素的真实的目的是让生产者杀死其他细菌。碰巧的是，产生抗生素的细菌也有基因，使它们对自己的抗生素产生抗药性（而且，因为细菌交换基因，对许多其他抗生素产生抗药性），否则，它们在制造抗生素时就会自杀。这场斗争产生了一个由产抗生素和耐药细菌主导的微生物组，这当然是预期的结果。我们的目标之一是了解促进细菌之间“正确战斗”的资源组合。现在看来，每种动物和每种植物都有一个微生物组，可以提供重要的益处，例如合成必需的营养素和抵御疾病。人类医学中一个新兴的想法是，管理我们自己的微生物组可以治愈一些更难治愈的疾病。Attine微生物组只是众多微生物组中的一个，但它的优点是我们可以用它做实验，这给了我们希望，我们可以制定出管理如何创建和管理微生物组的一般原则。

A group of ants in tropical America, known as the 'attines' or 'leafcutters,' evolved agriculture at least 50 million years ago. These ants collect plants and take them back to their nests, where they chew up the plants to feed a special fungus that is only able to live with leafcutter ants. Not surprisingly, in some parts of the New World, leafcutter ants are a pest, able to strip leaves off whole orange trees in one night, to feed enormous fungal gardens that fill underground ant nests as big as a two-story house (in London, not Las Vegas). In return for housing and food, the fungus produces fat- and sugar-rich structures, called gongylidia, that the ants harvest as food. Scientists call this co-dependence a mutualism because the ants and the fungus mutually benefit each other. The ants also protect their valuable fungal garden by weeding out moulds, which, if not controlled, would eventually consume the garden. The ants also apply antibiotics to kill the foreign moulds. They get the antibiotics from another mutualist, which also lives with the ants. These other mutualists are a special set of bacteria, called the actinomycetes, which are famous (amongst biologists) for making many kinds of antibiotics, some of which we use as medicine, like erythromycin. The actinomycetes are therefore also mutualists with the ant and the garden, because the bacteria fight disease, and in return, live on the ant bodies, where specialised glands appear to feed the bacteria.We have shown that many actinomycete species live on the ants and provide a mixture of antibiotics, probably to slow down the evolution of antibiotic resistance in the diseases that invade the fungus gardens. Biologists call the bacterial communities that live on a host organism its microbiome. In the attine microbiome, one group of actinomycetes, known as Pseudonocardia, are thought to have been handed down over generations, adapting to its ant hosts. Other actinomycetes, mostly in the group called Streptomyces, appear to be acquired anew from the soil in each generation. This is surprising, because the soil is full of bacteria, most of which are not Streptomyces, but somehow the ant is able to selectively take up useful, antibiotic-producing bacteria, and not harmful or useless bacteria. At a conceptual level, this problem is the same as the one faced by auto rescue companies wanting to sell coverage only to customers who own reliable cars. The characteristics of the soil bacteria and of potential customers are hidden, but everyone, good or bad, reliable or unreliable, wants to live on the ant or be rescued on the road.Economics has developed a solution to this problem of hidden characteristics, which is known as screening. Applied to ants, our hypothesis is that the ants provide the right mix of resources to promote fighting amongst bacteria. The winners are the ones that can release antibiotics, since the real purpose of antibiotics is to allow the producers to kill other bacteria. It happens that antibiotic-producing bacteria also have genes that make them resistant to their own antibiotics (and, because bacteria exchange genes, to many others), otherwise, they would commit suicide when they make antibiotics. The fighting produces a microbiome dominated by antibiotic-producing and -resistant bacteria, which, of course, is the desired outcome. One of our goals is to understand the mix of resources that promote the 'right kind of fighting' amongst bacteria. It appears now that every animal and every plant has a microbiome that provides important benefits, such as synthesising essential nutrients and defending against disease. An emerging idea in human medicine is that managing our own microbiome could cure some of our more recalcitrant diseases. The attine microbiome is just one of many, but its advantage is that we can do experiments with it, which gives us hope that we can work out general principles governing how to create and manage microbiomes.