当前位置: X-MOL 学术Environ. Sci. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Advancing Critical Applications of High Resolution Mass Spectrometry for DOM Assessments: Re-Engaging with Mass Spectral Principles, Limitations, and Data Analysis.
Environmental Science & Technology ( IF 11.4 ) Pub Date : 2020-09-17 , DOI: 10.1021/acs.est.0c04557
Juliana D'Andrilli 1 , Sarah J Fischer 2 , Fernando L Rosario-Ortiz 2
Affiliation  

The development of high resolution mass spectrometry (HRMS) has been immensely valuable to the environmental chemistry community. HRMS has been used to identify organic compounds in environmental matrices(1) and to characterize complex mixtures of organic compounds present in dissolved organic matter (DOM).(2,3) In this work, we refer to “compounds” as the elemental compositions determined from the exact masses detected by HRMS. Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) using a high magnetic field has opened the door for unparalleled mass resolving power and mass accuracy necessary for nontarget, molecular-level identification of organic compounds in atmospheric, aquatic (freshwater and marine), and terrestrial DOM mixtures, in both carbon-poor and carbon-rich environments. The application of HRMS in the identification of organic compounds from complex mixtures such as DOM has surged over recent decades. Initial publications focused on identifying molecular formula of DOM compounds and characterizing substances in environmental isolates.(2,3) Quite rapidly, studies employing HRMS shifted to assessing DOM compound properties, identifying changes in molecular formula to specific biological and chemical processes (e.g., microbial biodegradation, natural photochemical oxidation). However, sample preparation and HRMS biases can confound results and subsequent interpretations. For example, sample preparation procedures on DOM samples prior to HRMS (e.g., XAD isolation, solid phase extraction, reverse osmosis, ultrafiltration), can bias the results to the fraction recovered. HRMS instruments (e.g., FT-ICR MS, Orbitrap) can introduce bias, for example, relative ionization efficiencies versus sample type.(4) Another common example of a potential issue with HRMS for DOM analysis is the focus on mass spectral peak intensity correlations with peak intensity data from non-HRMS-type analytical instruments targeting other various DOM properties, e.g., fluorescence. Given that HRMS peak intensities are a product of sample, ionization efficiency, and instrument setting biases, correlations without bias context or further analyses cannot describe absolute linkages. Lastly, HRMS detects ions of lower molecular weights (<1000 Da) and therefore represent ionizable subsets of the total DOM pool. The wealth of HRMS data offers many opportunities to conduct diverse analyses, where new correlations may drive forward knowledge on DOM. Recognizing the inherent biases and employing caution is recommended, as such biases and technical choices can lead to different perspectives on the chemical understanding of the material.(4) These limitations warrant discussion when interpreting HRMS results for DOM assessments and connection to physical or biogeochemical processes, or the role of DOM in ecosystem function. In many cases, HRMS data has been used to make inferences about DOM properties in conjunction with complementary data, but these possible correlations (e.g., using peak intensities) often lack proper causal evaluation and/or discussion of bias (e.g., ionization efficiencies). As we move forward with DOM HRMS applications for identification and characterization purposes, targeted approaches will advance this field of research. We propose that the use of HRMS is approached with a well-defined hypothesis (Figure 1). If a study investigates biological, photochemical, and/or other ecosystem processes of DOM production and transformation, an intent to collect specific causal molecular information from HRMS will improve mechanistic understanding. For example, DOM photooxidation creates identifiable shifts in molecular weights, functional groups, and in oxygen and heteroatom content.(5) Collecting and interpreting HRMS data with clear expectations of chemical indicators will avoid perfunctory, misidentified, or retroactive correlations. Even though it is possible that a given DOM property statistically correlates with data obtained by HRMS, it may not confirm a linked causal effect. We find that in many cases, there is a limited scientific basis for the proposed correlations and simply having HRMS data does not guarantee that a clear conclusion can be reached. Instead, using an approach that begins with a well-defined hypothesis, has clear research questions, and pursues targeted analysis of the DOM chemistry will produce sound molecular-level interpretations and understanding (Figure 1A). Figure 1. High resolution mass spectrometry (HRMS) dissolved organic matter (DOM) assessment steps showing (A) a robust workflow, (B) mass spectra for Suwannee River (top) and Pony Lake (bottom) fulvic acids with their molecular formula at m/z 311 (insets),(4) and (C) mass spacings for formula assignment (Dalton; Da) and targeted research approaches. HRMS researchers begin molecular DOM interpretations directly on the formula information based on individual masses and elemental content (Figure 1B), before visualizing larger composition data on van Krevelen diagrams and performing supplementary analyses. This is an essential step, beginning with the data generated first, prior to higher-level assessments. HRMS researchers rely on the principles governing mass spectra generation and the strength of highly resolved ion peaks in determining unambiguous assignments of molecular formula.(2−5) Without careful analysis of fundamental mass spectral spacing patterns of highly resolved peaks (e.g., confirming monoisotopic peaks having 12Cn versus 12Cn–113C1 used to confirm singly charged ions; Figure 1C), assigning individual peaks to DOM molecular formula would not be possible within a reasonable error margin (e.g., <1 ppm).(2−5) Thus, there is a reliance on the repeating spectral patterns to determine and confirm DOM composition. Mass spectral spacing patterns continue to be undeniably powerful for DOM molecular-level understanding as no manual assignments or algorithm can produce unambiguous molecular formula without them. The utility of mass spectral spacing patterns, for example, between two peaks (monoisotopic spacings) and among multiple peaks (e.g., methylene homologous series, CH4 versus 16O)(2,3) are reliable resources to first understand and then analyze data (Figure 1C). We also assert that the utility of mass spectral spacing patterns is not limited to these examples and can continue to evolve to target molecular information on specific processes, for example, photooxidation. Therefore, we propose a refocusing on the utility of mass spectral spacing patterns because they not only set the foundation of confirming charged ions (e.g., singly versus doubly or multiply) and formula assignments (and adducts) for DOM mixtures, but also are essential resources for targeted inquiries and biogeochemical relationships. Ultimately, HRMS allows DOM researchers to investigate complex organic mixtures. However, a clear understanding of the instrument’s principles and limitations is essential, especially when relating HRMS data to other complementary analytical techniques having their own principles and limitations. We assert that with the abundance and breadth of information gained by HRMS combined with its biases and limitations, the most informative use of HRMS data requires the following approach: targeted, hypothesis-driven research, intentional complementary assessments for other physicochemical information, and utilization of spectral patterns that center on molecular chemical responses of individual processes. Using this workflow, spurious correlations with complementary analyses will be reduced. The authors declare no competing financial interest. This work was supported by the National Science Foundation, Division of Environmental Engineering (Award Numbers: 1804704, 1804736/2027431), USA. This article references 5 other publications.

中文翻译:

推进用于DOM评估的高分辨率质谱的关键应用:重新参与质谱原理,限制和数据分析。

高分辨率质谱(HRMS)的发展对环境化学界具有巨大的价值。HRMS已用于鉴定环境基质中的有机化合物(1)并表征溶解有机物(DOM)中存在的有机化合物的复杂混合物。(2,3)在本文中,我们将“化合物”称为元素组成由HRMS检测到的确切质量确定。使用强磁场的傅里叶变换离子回旋共振质谱(FT-ICR MS)为无与伦比的质量分辨能力和质量准确性打开了大门,这对于大气,水生(淡水和海洋)中的有机化合物的非目标分子水平鉴定是必需的以及低碳和高碳环境中的陆地DOM混合物。近几十年来,HRMS在从复杂混合物(例如DOM)中鉴定有机化合物的应用激增。最初的出版物侧重于鉴定DOM化合物的分子式和表征环境分离物中的物质。(2,3)很快,采用HRMS的研究转向评估DOM化合物的性质,确定分子式在特定的生物和化学过程(例如微生物)中的变化。生物降解,自然光化学氧化)。但是,样品制备和HRMS偏差会混淆结果和后续解释。例如,在HRMS之前对DOM样品进行样品制备程序(例如XAD分离,固相萃取,反渗透,超滤)可能会使结果偏向回收的馏分。HRMS仪器(例如FT-ICR MS,Orbitrap)可能会引入偏差,例如相对电离效率与样品类型之间的关系。(4)用于DOM分析的HRMS潜在问题的另一个常见示例是关注质谱峰强度与非HRMS类型的峰强度数据的相关性针对其他各种DOM属性(例如荧光)的分析仪器。鉴于HRMS峰强度是样品,电离效率和仪器设置偏差的乘积,因此没有偏差背景或没有进一步分析的相关性无法描述绝对联系。最后,HRMS检测到较低分子量(<1000 Da)的离子,因此代表了整个DOM库中可电离的子集。丰富的HRMS数据为进行各种分析提供了许多机会,其中新的相关性可能推动对DOM的了解。建议认识到固有偏差并谨慎行事,因为这样的偏差和技术选择可能导致人们对材料的化学理解产生不同的观点。(4)在解释用于DOM评估的HRMS结果以及与物理或生物地球化学过程的联系时,这些局限性值得讨论或DOM在生态系统功能中的作用。在许多情况下,HRMS数据已与补充数据一起用于推断DOM属性,但这些可能的相关性(例如,使用峰强度)通常缺乏适当的因果评估和/或对偏差的讨论(例如,电离效率)。随着我们为识别和表征目的而使用DOM HRMS应用程序的发展,有针对性的方法将推动这一研究领域的发展。我们建议以明确定义的假设来接近HRMS的使用(图1)。如果一项研究调查DOM生产和转化的生物学,光化学和/或其他生态系统过程,那么从HRMS收集特定因果分子信息的意图将改善对机理的理解。例如,DOM光氧化会在分子量,官能团以及氧和杂原子含量方面产生明显的变化。(5)收集并解释对HRMS数据的化学指标有明确期望的方法将避免过分的,错误的识别或追溯相关性。即使给定的DOM属性可能与HRMS获得的数据在统计上相关联,也可能无法确认相关的因果关系。我们发现在许多情况下,所建议的相关性的科学依据有限,仅拥有HRMS数据并不能保证可以得出明确的结论。相反,使用以明确定义的假设开始,具有明确的研究问题并进行DOM化学的有针对性的分析的方法,将产生合理的分子水平的解释和理解(图1A)。图1.高分辨率质谱(HRMS)溶解有机物(DOM)评估步骤,显示了(A)稳健的工作流程,(B)Suwannee River(上)和Pony Lake(下)黄腐酸的质谱,其分子式为 并针对DOM化学进行有针对性的分析,将产生合理的分子水平解释和理解(图1A)。图1.高分辨率质谱(HRMS)溶解有机物(DOM)评估步骤,显示(A)稳健的工作流程,(B)Suwannee River(上)和Pony Lake(下)黄腐酸的质谱,其分子式为 并针对DOM化学进行有针对性的分析,将产生合理的分子水平解释和理解(图1A)。图1.高分辨率质谱(HRMS)溶解有机物(DOM)评估步骤,显示了(A)稳健的工作流程,(B)Suwannee River(上)和Pony Lake(下)黄腐酸的质谱,其分子式为m / z 311(插入),(4)和(C)用于公式分配的质量间距(道尔顿; Da)和有针对性的研究方法。HRMS研究人员在基于van Krevelen图可视化较大的成分数据并进行补充分析之前,直接基于单个质量和元素含量(图1B)在分子式信息上直接进行分子DOM解释。这是必不可少的步骤,首先要先生成数据,然后再进行更高级别的评估。HRMS研究人员在确定分子式明确分配时依赖于控制质谱生成和高分辨离子峰强度的原理。(2-5)无需仔细分析高分辨峰的基本质谱间距模式(例如,确定单同位素峰)有12 C n12 C n –113 C 1用于确认单电荷离子;图1C),不可能在合理的误差范围内(例如<1 ppm)将单个峰分配给DOM分子式。(2-5)因此,依赖于重复的光谱图来确定和确认DOM组成。质谱间距模式对于DOM分子水平的理解仍然具有不可否认的强大功能,因为没有它们,任何手动分配或算法都无法产生明确的分子式。质谱间距模式的用途,例如,在两个峰之间(单同位素间距)和多个峰之间(例如,亚甲基同源系列,CH 416O)(2,3)是首先了解然后分析数据的可靠资源(图1C)。我们还断言,质谱间距模式的用途不限于这些示例,并且可以继续发展以针对特定过程(例如光氧化)的分子信息。因此,我们建议重新关注质谱图间距模式的实用性,因为它们不仅为确认带电离子(例如单,双或乘或乘)和DOM混合物的公式分配(和加合物)奠定了基础,而且还是必不可少的资源进行有针对性的查询和生物地球化学关系。最终,HRMS使DOM研究人员能够研究复杂的有机混合物。但是,清楚了解该工具的原理和局限性是至关重要的,特别是在将HRMS数据与其他具有自身原理和局限性的补充分析技术联系起来时。我们断言,由于HRMS所获得的信息丰富且广度,加上其偏见和局限性,对HRMS数据的最有意义的使用需要以下方法:有针对性的,假设驱动的研究,对其他理化信息的有意补充评估以及对HRMS的利用。以单个过程的分子化学响应为中心的光谱模式。使用此工作流程,将减少与补充分析的虚假关联。作者宣称没有竞争性的经济利益。这项工作得到了美国国家科学基金会环境工程司的资助(授予号:1804704,1804736/2027431)。
更新日期:2020-10-06
down
wechat
bug