Within the plant kingdom, there is an exceptional amount of chemical diversity that has yet to be annotated. It is for this reason that non-targeted analysis is of interest for those working in novel natural products. To increase the number and diversity of compounds observable in root exudate extracts, several workflows which differ at three key stages were compared: 1) sample extraction, 2) chromatography, and 3) data preprocessing. Plants were grown in Hoagland's solution for two weeks, and exudates were initially extracted with water, followed by a 24-h regeneration period with subsequent extraction using methanol. Utilizing the second extraction showed improved results with less ion suppression and reduced retention time shifting compared to the first extraction. A single column method, utilizing a pentafluorophenyl column, paired with high-resolution mass spectrometry ionized and correctly identified 34 mock root exudate compounds, while the dual column method, incorporating a pentafluorophenyl column and a porous graphitic carbon column, retained and identified 43 compounds. In a pooled quality control sample of exudate extracts, the single column method detected 1,444 compounds. While the dual method detected fewer compounds overall (1,050), it revealed a larger number of small polar compounds. Three preprocessing methods (targeted, proprietary, and open source) successfully identified 43, 31, and 38 mock root exudate compounds to confidence level 1, respectively. Enhancing signal strength and analytical method stability involves removing the high ionic strength nutrient solution before sampling root exudate extracts. Despite signal intensity loss, a dual column method enhances compound coverage, particularly for small polar metabolites. Open-source software proves a viable alternative for non-targeted analysis, even surpassing proprietary software in peak picking.

中文翻译：

---

双柱色谱结合高分辨率质谱提高了植物根系分泌物非靶向分析的覆盖范围


在植物界中，存在大量尚未注释的化学多样性。正是由于这个原因，非靶向分析引起了从事新型天然产品研究的人员的兴趣。为了增加根分泌物提取物中可观察到的化合物的数量和多样性，比较了在三个关键阶段不同的几种工作流程：1) 样品提取，2) 色谱法，3) 数据预处理。植物在霍格兰溶液中生长两周，首先用水提取渗出物，然后进行 24 小时的再生期，随后使用甲醇提取。与第一次萃取相比，第二次萃取显示出更好的结果，离子抑制更少，保留时间漂移也更小。单柱方法利用五氟苯基柱，搭配高分辨率质谱电离并正确鉴定了 34 种模拟根分泌物化合物，而双柱方法结合了五氟苯基柱和多孔石墨碳柱，保留并鉴定了 43 种化合物。在渗出物提取物的混合质量控制样品中，单柱方法检测到 1,444 种化合物。虽然双重方法检测到的化合物总体较少 (1,050)，但它揭示了大量的小极性化合物。三种预处理方法（靶向、专有和开源）分别成功识别出 43、31 和 38 种模拟根分泌物化合物，置信度为 1。增强信号强度和分析方法稳定性包括在对根分泌物提取物进行采样之前去除高离子强度营养液。尽管信号强度损失，双柱方法增强了化合物覆盖率，特别是对于小极性代谢物。 事实证明，开源软件是非目标分析的可行替代方案，甚至在峰值选取方面超过了专有软件。