Griffin Eric A, Traw M Brian, Morin Peter J, Pruitt Jonathan N, Wright S Joseph, Carson Walter P
Department of Biological Sciences, University of Pittsburgh, A234 Langley Hall, 4249 Fifth Avenue, Pittsburgh, Pennsylvania, 15260, USA.
Department of Ecology, Evolution, and Natural Resources, Rutgers University, 14 College Farm Road, New Brunswick, New Jersey, 08901, USA.
Ecology. 2016 Nov;97(11):2998-3008. doi: 10.1002/ecy.1537.
The phyllosphere (comprising the leaf surface and interior) is one of the world's largest microbial habitats and is host to an abundant and diverse array of bacteria. Nonetheless, the degree to which bacterial communities are benign, harmful, or beneficial to plants in situ is unknown. We tested the hypothesis that the net effect of reducing bacterial abundance and diversity would vary substantially among host species (from harmful to beneficial) and this would be strongly mediated by soil resource availability. To test this, we monitored tree seedling growth responses to commercial antibiotics among replicated resource supply treatments (N, P, K) in a tropical forest in Panama for 29 months. We applied either antibiotics or control water to replicated seedlings of five common tree species (Alseis blackiana, Desmopsis panamensis, Heisteria concinna, Sorocea affinis, and Tetragastris panamensis). These antibiotic treatments significantly reduced both the abundance and diversity of bacteria epiphytically as well as endophytically. Overall, the effect of antibiotics on performance was highly host specific. Applying antibiotics increased growth for three species by as much as 49% (Alseis, Heisteria, and Tetragastris), decreased growth for a fourth species by nearly 20% (Sorocea), and had no impact on a fifth species (Desmopsis). Perhaps more importantly, the degree to which foliar bacteria were harmful or not varied with soil resource supply. Specifically, applying antibiotics had no effect when potassium was added but increased growth rate by almost 40% in the absence of potassium. Alternatively, phosphorus enrichment caused the effect of bacteria to switch from being primarily beneficial to harmful or vice versa, but this depended entirely on the presence or absence of nitrogen enrichment (i.e., important and significant interactions). Our results are the first to demonstrate that the net effect of reducing the abundance and diversity of bacteria can have very strong positive and negative effects on seedling performance. Moreover, these effects were clearly mediated by soil resource availability. Though speculative, we suggest that foliar bacteria may interact with soil fertility to comprise an important, yet cryptic dimension of niche differentiation, which can have important implications for species coexistence.
叶际（包括叶片表面和内部）是世界上最大的微生物栖息地之一，栖息着大量多样的细菌。然而，细菌群落对原位植物是无害、有害还是有益，目前尚不清楚。我们检验了这样一个假设：减少细菌丰度和多样性的净效应在宿主物种间会有很大差异（从有害到有益），并且这将受到土壤资源可利用性的强烈影响。为了验证这一点，我们在巴拿马的一片热带森林中，对重复资源供应处理（氮、磷、钾）下的树木幼苗进行了为期29个月的监测，观察它们对商业抗生素的生长反应。我们对五种常见树种（黑叶阿尔赛木、巴拿马德斯莫普斯木、秀丽海斯特木、近缘索罗木和巴拿马四蕊木）的重复幼苗施加抗生素或对照水。这些抗生素处理显著降低了附生细菌和内生细菌的丰度和多样性。总体而言，抗生素对植物生长的影响具有高度的宿主特异性。施用抗生素使三种树种的生长增加了49%（阿尔赛木、海斯特木和四蕊木），使第四种树种的生长减少了近20%（索罗木），而对第五种树种（德斯莫普斯木）没有影响。也许更重要的是，叶际细菌有害程度随土壤资源供应情况而变化。具体而言，添加钾时施用抗生素没有效果，但在不添加钾的情况下，生长速率提高了近40%。或者，磷的富集导致细菌的影响从主要有益转变为有害，反之亦然，但这完全取决于氮的富集情况（即重要且显著的相互作用）。我们的结果首次表明，减少细菌丰度和多样性的净效应可能对幼苗生长产生非常强烈的正负影响。此外，这些影响明显受到土壤资源可利用性的调节。尽管只是推测，但我们认为叶际细菌可能与土壤肥力相互作用，构成生态位分化的一个重要但隐秘的维度，这可能对物种共存具有重要意义。
