Food webs - 2021

Sampling was conducted in Kematen in Tirol, Austria, where spring barley (Hordeum vulgare L.) was grown in six fields that have had organic management for the past ten years, three in 2020 and 2021, respectively. Each field was split in half, one side being fertilized with manure and the other remaining unfertilized, as a negative control. The manure treatment was independently applied by each field’s respective owner, at a standard rate of 1.5 metric tons (1 500kg) per hectare, using manure spreaders, after tilling and pressing the soil. Four 5x5m plots per treatment were delimited, to carry out all sampling. These plots had no barriers or fences surrounding them that would impede or limit movement between the inside of the plots and the rest of the field area. Additionally, the plots were at least 5m from the field edge, to avoid edge effects. They were at least 10m way from each other along the field’s width, and between 15 to 20m along the length of the field. To avoid any spillover of treatment effects, plots were set at least 25m away from the border between the two treatments within the same field.Sampling was conducted every two weeks, with one week of sampling and one of rest. The sampling period started on the 21st of April and ended on the 14th of July in 2020, while it started on the 3rd of May and ended on the 16th of July in 2021, to encompass almost the entire crop season. This resulted in seven sampling time points for the first year and six for the second.A total of 2404 ground beetles, 913 rove beetles and 567 spiders were captured in 2020, and 1977 ground beetles, 891 rove beetles and 250 spiders in 2021. The beetles’ gut content and the spiders’ full bodies were extracted with a BioSprint 96 DNA Blood Kit (Qiagen, Hilden, Germany) on a QIAGEN Biosprint96® workstation for automated DNA extraction, following the manufacturer’s recommendations. After extraction, three different multiplex-PCR assays were run per sample. The first assay targeted several herbivore and detritivore taxa (assay in Rennstam Rubbmark et al. 2019), the second targeted generalist and specialist predators, such as spiders and ladybeetles, as intra- and extraguild predation (primers from Sint et al. 2014 and Staudacher et al. 2016). Lastly the third assay identified the genus of beetles consumed, also as intraguild predation, from a selected set of common taxa consisting of Bembidion spp., Harpalus spp., Poecilus spp., Pterostichus spp., Philonthus carbonarius and Philonthus cognatus. The intraguild interactions, and implications on competition, have been analysed in greater detail elsewhere (Leote et al. 2024).<br>Following Rennstam Rubbmark et al. (2019) a diagnostic multiplex-PCR was chosen over metabarcoding, as our regurgitate samples contain not only different amounts of DNA from several food sources, but also a lot of consumer DNA. This drastically lowers the detection of food DNA, as the primers targeting consumed taxa also bind to the taxonomically closely related consumer. Therefore, to ensure a more reliable and robust consumed DNA detection, we employed the three PCR assays described above and in the supplementary tables.After the multiplex-PCRs, the samples were screened using capillary electrophoresis on an QIAxcel Advanced system and the ScreenGel software, with a DNA Screening Kit (2400) using a 15-3k base pair alignment marker, following the manufacturer’s recommendations. The screening profile on ScreenGel was set to the standard AM320 method, with the relative fluorescence units (RFU) detection threshold defined at 0.07, as opposed to the default 0.1, to account for the fact the samples were extracted from gut content and thus partially digested.

本研究的采样工作于奥地利蒂罗尔州克马滕（Kematen in Tirol）开展，供试作物为春大麦（*Hordeum vulgare* L.），共设置6块实施有机管理达10年的试验田，其中2020年与2021年各设3块。每块试验田均被均分为两个区域：一侧施加粪肥处理，另一侧作为阴性对照不施加任何肥料。粪肥处理由各田块的所有者独立实施，采用粪肥撒播机，在翻耕并压实土壤后，按照每公顷1.5公吨（1500千克）的标准用量施加。每个处理组设置4个5×5米的样地以完成全部采样工作，样地未设置任何阻碍或限制样地内部与田块其余区域间移动的障碍物或围栏。此外，所有样地与田块边缘的间距至少为5米，以规避边缘效应；沿田块宽度方向，样地间的最小间距为10米，沿田块长度方向则介于15至20米之间。为避免不同处理间的效应溢出干扰，同一块田内两个处理区之间的样地至少相距25米。

采样工作每两周开展一次，即采样一周、休整一周。2020年的采样时段为2020年4月21日至7月14日，2021年为2021年5月3日至7月16日，覆盖了几乎整个作物生长季。由此第一年共获得7个采样时间点，第二年为6个。2020年共捕获步甲科昆虫（ground beetles）2404头、隐翅虫科昆虫（rove beetles）913头以及蜘蛛（spiders）567只；2021年则捕获步甲科昆虫1977头、隐翅虫科昆虫891头、蜘蛛250只。

所有样本的核酸提取工作均采用BioSprint 96 DNA血液提取试剂盒（Qiagen，德国希尔德），在QIAGEN Biosprint96®自动化核酸提取工作站上，严格按照制造商的操作指南完成：其中甲虫样本仅提取肠道内容物，蜘蛛样本则提取全虫体基因组DNA。提取完成后，对每个样本开展3组多重聚合酶链式反应（multiplex-PCR）检测：第一组检测靶向多种植食性和腐食性类群（检测方法参照Rennstam Rubbmark等，2019）；第二组靶向广食性与专食性捕食者（如蜘蛛和瓢虫），用于检测集团内和集团外捕食关系（引物序列取自Sint等，2014与Staudacher等，2016）；第三组则用于鉴定被捕食甲虫的属级分类单元，同样针对集团内捕食关系，所选类群均为常见类群，包括*Bembidion* spp.、*Harpalus* spp.、*Poecilus* spp.、*Pterostichus* spp.、*Philonthus carbonarius*及*Philonthus cognatus*。有关集团内捕食互作及其对种间竞争的影响，已在另一项研究中进行了详细分析（Leote等，2024）。

参照Rennstam Rubbmark等（2019）的研究方案，本研究选择诊断性多重PCR而非元条形码（metabarcoding）技术，原因在于本研究的反流物样本不仅包含来自多种食物源的不同含量的DNA，还含有大量消费者自身的DNA。这会大幅降低食物DNA的检出率，因为靶向被捕食物种的引物同时也会与分类学上亲缘关系较近的消费者DNA结合。因此，为确保被捕食物DNA检测的可靠性与稳定性，我们采用了前文所述及补充表格中的3组PCR检测方案。

多重PCR扩增完成后，使用QIAxcel Advanced毛细管电泳系统与ScreenGel软件，并结合DNA筛选试剂盒（2400），以15-3kb的碱基对校准标记物，严格按照制造商的操作指南对样本进行电泳筛选。ScreenGel软件的筛选程序设置为标准AM320方法，将相对荧光单位（relative fluorescence units, RFU）的检测阈值设定为0.07，而非默认的0.1，以适配从肠道内容物中提取的样本——此类样本的DNA已发生部分降解。