Optimizing ESI Parameters for Untargeted Metabolomics: A Comprehensive Guide to Enhance Sensitivity and Reproducibility

Abstract

This article provides a comprehensive guide for researchers and drug development professionals on optimizing Electrospray Ionization (ESI) parameters to maximize data quality in untargeted metabolomics. Covering foundational principles to advanced applications, we detail systematic optimization of spray voltage, ion source temperatures, gas flows, and needle positioning to improve metabolome coverage and signal stability. The guide incorporates robust quality control strategies using intrastudy quality control samples to correct instrumental drift and batch effects. Furthermore, we present comparative evaluations of data acquisition modes and column chemistries, alongside modern computational approaches for enhanced metabolite annotation. This end-to-end resource aims to empower scientists to generate highly reproducible and biologically meaningful metabolomic data for biomarker discovery and clinical research.

Understanding ESI Fundamentals and Their Impact on Metabolome Coverage

Electrospray Ionization (ESI) has become a cornerstone technique in liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics, enabling the detection of a vast array of metabolites from complex biological samples. As a "soft" ionization technique, ESI efficiently produces gas-phase ions from liquid samples without extensive fragmentation, making it ideal for detecting labile metabolites. However, the chemical diversity of metabolites presents a significant challenge, as no single set of ESI parameters can optimally ionize all compounds. Consequently, parameter optimization is not merely a technical refinement but a fundamental requirement for achieving comprehensive metabolome coverage, ensuring data quality, and generating biologically meaningful results. This application note details the critical ESI parameters requiring optimization and provides structured protocols to guide researchers in this essential process, framed within the broader context of advancing metabolomics research.

The Impact of ESI Parameters on Metabolomics Data Quality

The ionization process in ESI involves creating a fine aerosol of charged droplets from the LC eluent, followed by desolvation and ion formation. This complex process is influenced by numerous instrumental parameters that collectively determine which metabolites are detected and with what sensitivity. Suboptimal settings can lead to ion suppression, reduced sensitivity for critical metabolites, and the generation of artifactual adducts, ultimately compromising data integrity and leading to false biological conclusions.

Recent research demonstrates that systematic optimization of ESI conditions significantly enhances data quality. A comprehensive study evaluating ion source parameters for plasma untargeted metabolomics found that needle position, spray voltage, and gas temperatures notably affect signal stability and the number of compound detections [1]. The same study concluded that optimized parameters are essential for establishing appropriate system suitability and ensuring the acquisition of high-quality metabolomic data, which in turn eases comparison of data between laboratories [1] [2].

Key ESI Parameters and Their Effects

Table 1: Critical ESI Parameters and Their Impact on Metabolite Detection

The performance of these parameters is often interrelated, requiring a systematic approach to optimization rather than adjusting parameters in isolation. For instance, the fraction of organic solvent in the mobile phase interacts significantly with ESI parameters, affecting signal stability and metabolite annotations [1]. This interplay underscores the need for holistic optimization protocols that account for the complex relationships between LC conditions and ESI parameters.

Experimental Protocols for ESI Parameter Optimization

Protocol 1: Comprehensive Optimization of ESI Source Parameters

This protocol provides a systematic approach for optimizing ESI parameters to maximize metabolite coverage in untargeted analyses, adapted from recent methodology publications [1].

Materials and Reagents:

• Standard reference materials (e.g., NIST SRM 1950 Plasma)
• Quality control samples derived from pooled study samples
• Solvents: LC-MS grade water, acetonitrile, methanol, formic acid
• Appropriate analytical columns (e.g., C18 for reversed-phase, HILIC for polar metabolites)

Procedure:

• Sample Preparation:
  • Prepare extracts from standard reference materials using 4 volumes of ice-cold methanol for protein precipitation.
  • Centrifuge at 14,000 × g for 15 minutes at 4°C.
  • Transfer supernatant to LC-MS vials for analysis.

• Initial LC-MS Conditions:

  • Column: Appropriate for metabolite class (e.g., C18 for lipids, HILIC for polar metabolites)
  • Mobile Phase: A: 0.1% formic acid in water; B: 0.1% formic acid in acetonitrile
  • Flow Rate: 0.3 mL/min
  • Injection Volume: 5 μL
  • Gradient: Optimized for metabolite separation (e.g., 5-95% B over 15-30 minutes)

• ESI Parameter Optimization Sequence:

  • Begin with manufacturer's recommended settings as baseline.
  • Optimize needle position for maximum signal intensity using a standard metabolite mixture.
  • Systematically vary one parameter while holding others constant:
    • Spray voltage: Test in 0.5 kV increments for both positive and negative modes
    • Ion Transfer Tube (ITT) temperature: Evaluate from 250°C to 350°C in 25°C increments
    • Sheath gas: Test from 30 to 60 units in 5-unit increments
    • Auxiliary gas: Test from 5 to 20 units in 5-unit increments
    • Vaporizer temperature: Evaluate from 100°C to 400°C in 50°C increments

• Evaluation Metrics:

  • Total number of metabolic features detected
  • Signal intensity for a range of metabolite standards
  • Signal stability (relative standard deviation of internal standards)
  • Coverage across metabolite classes (assessed using authentic standards)

• Data Analysis:

  • Process data using untargeted analysis software (e.g., MS-DIAL, MZmine)
  • Use peak picking and alignment to compare feature counts across conditions
  • Apply statistical analysis to identify parameter sets that maximize coverage and signal quality

Table 2: Research Reagent Solutions for ESI Optimization

Protocol 2: Design of Experiments (DoE) Approach for DDA Parameter Optimization

This protocol employs a structured Design of Experiments (DoE) methodology to optimize Data-Dependent Acquisition (DDA) parameters, which directly impact ESI performance and metabolite fragmentation quality [3].

Procedure:

• Experimental Design:
  • Select critical parameters for optimization: collision energy and intensity threshold
  • Implement a Central Composite Design (CCD) with 10 experimental runs including repetitions
  • Test collision energy at 10, 20, and 30 eV with a fixed spread of 5 eV
  • Evaluate intensity thresholds at 1000, 2000, 3000, and 4000 counts

• LC-MS Analysis:

  • Use UHPLC system with appropriate column (e.g., Shim-pack Velox C18, 100 × 2.1 mm, 2.7 μm)
  • Maintain column temperature at 50°C
  • Employ binary gradient with 0.1% formic acid in water (A) and methanol (B)
  • Set flow rate to 0.3 mL/min with 5 μL injection volume

• MS Data Acquisition:

  • Operate in positive ESI mode with DDA
  • Set mass range to m/z 100-2000
  • Apply varying collision energies and intensity thresholds according to experimental design
  • Use calibration solution (e.g., sodium iodide) for mass accuracy

• Data Processing and Evaluation:

  • Convert raw data to mzML format
  • Process using molecular networking platforms (e.g., GNPS)
  • Set MS2 fragment ion tolerance to 0.02 Da
  • Create molecular networks with minimum cosine score of 0.7 and minimum of 6 matching peaks
  • Evaluate based on: number of network nodes, library identifications, and self-loops

• Statistical Analysis:

  • Use response surface methodology to model relationship between factors
  • Identify optimal parameter combinations for maximum metabolite coverage
  • Validate optimal settings with biological replicates

Workflow Visualization

ESI Optimization Workflow Diagram

Data Analysis and Technical Validation

Following parameter optimization, rigorous data analysis and validation are essential to confirm improved metabolome coverage. Modern approaches incorporate advanced computational and visualization strategies to assess data quality [4]. Effective data visualization facilitates method validation by enabling researchers to identify patterns, anomalies, and quality metrics that might be obscured in tabular data.

Network-based strategies have emerged as powerful tools for evaluating the outcomes of ESI optimization. The two-layer interactive networking topology implemented in tools like MetDNA3 enhances metabolite annotation by integrating data-driven and knowledge-driven networks [5]. This approach has demonstrated the capability to annotate over 1600 seed metabolites with chemical standards and more than 12,000 putatively annotated metabolites through network-based propagation, representing a significant advancement in annotation coverage [5].

When evaluating the success of ESI parameter optimization, consider these key metrics:

• Feature Quality: Assess using parameters like peak width and feature linearity, which have been shown to effectively distinguish true metabolites from noise [6].
• Annotation Confidence: Leverage computational tools like MSConvert, MZmine, and SIRIUS for systematic data analysis [7].
• Platform Integration: Combine multiple separation techniques (e.g., RPLC and HILIC) to expand metabolome coverage, as this approach can detect 60% new metabolic features compared to single-platform methods [1] [2].

Optimization of ESI parameters is a critical, non-negotiable component of rigorous untargeted metabolomics. The chemical diversity of the metabolome necessitates careful tuning of spray voltages, gas flows, and temperature parameters to achieve comprehensive coverage and reliable detection. The protocols presented herein provide a systematic framework for this optimization process, emphasizing the use of standard reference materials, structured experimental designs, and robust validation metrics. As the field advances toward greater reproducibility and inter-laboratory comparability, standardized approaches to ESI optimization will play an increasingly important role in generating high-quality, biologically meaningful metabolomics data.

Electrospray Ionization (ESI) is a cornerstone technique in liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics. It is a "soft" ionization method that produces ions from a liquid phase, allowing for the detection of a wide range of metabolites with minimal fragmentation [7]. The core principle involves applying a high voltage to a liquid to create an aerosol, which is then desolvated by a stream of hot gas to produce gas-phase ions for mass analysis [7]. The configuration and tuning of the ESI source are not trivial tasks; they profoundly determine the sensitivity, coverage, and reproducibility of the metabolomics experiment. As one study concludes, in untargeted LC/MS metabolomics, 'what you see is what you ionise', emphasizing that the ionisation efficiency directly defines the extent of metabolome coverage [8]. This application note details the function, optimization, and impact of key ESI parameters—spray voltage, gas flows, and temperatures—framed within the broader thesis that meticulous source optimization is indispensable for expanding and maximizing metabolome coverage in untargeted investigations.

Core ESI Parameters: Functions and Optimization Goals

The performance of the ESI source is governed by several interdependent electronic and pneumatic parameters. Understanding their individual functions is the first step toward systematic optimization.

• Spray Voltage: This high voltage (typically in kV) applied to the ESI capillary is responsible for charging the liquid surface and forming the "Taylor cone," from which a fine aerosol of charged droplets is emitted. The optimal voltage ensures stable electrospray operation and efficient droplet formation and disintegration, directly influencing the ionization efficiency of a wide range of metabolites [8] [1]. Its setting can vary between positive and negative ionization modes.

• Sheath and Auxiliary Gas Flows: These are inert gas flows (typically nitrogen) that serve distinct purposes. The sheath gas helps stabilize the spray and shape the plume of charged droplets for efficient entry into the mass spectrometer. The auxiliary (or drying) gas is a stream of heated gas that accelerates the desolvation of charged droplets, liberating gas-phase ions [7] [1]. Optimizing these flows is critical for achieving a stable signal and efficient desolvation without causing premature evaporation or scattering of droplets.

• Capillary and Vaporizer Temperatures: These heated elements facilitate the final stages of desolvation. The vaporizer temperature (or ion transfer tube temperature) controls the heat applied to the aerosol as it enters the source, driving off residual solvent. The capillary temperature pertains to the heating of the capillary that guides ions into the high-vacuum region of the mass spectrometer. Sufficient temperatures are required for complete desolvation, preventing solvent-related adducts and signal suppression, but excessive heat can degrade thermally labile metabolites [1].

The overarching goal of optimizing these parameters is to maximize the number and abundance of detected metabolite "features" (unique m/z-retention time pairs) while ensuring signal stability and reproducibility [8] [1]. This directly translates to a broader and deeper coverage of the metabolome.

Table 1: Core ESI Parameters, Their Functions, and Optimization Goals

Quantitative Parameter Recommendations from Experimental Data

While optimal settings are instrument- and sample-specific, recent systematic studies provide a quantitative baseline for method development. The following table summarizes experimental data from a study that optimized an Orbitrap mass spectrometer for plasma untargeted metabolomics, demonstrating the practical ranges and final optimized values for key parameters [1] [9] [10].

Table 2: Experimentally Optimized ESI Parameters for Untargeted Metabolomics on an Orbitrap MS

Another study investigating a ZSpray ESI source highlighted that the capillary voltage (1.5–3.0 kV) and sample cone voltage (10.0–40.0 V) significantly influenced not only the number and abundance of features but also the overall structure of the collected data, impacting the biological information extracted [8]. This underscores that parameter optimization is vital for unbiased hypothesis generation.

Detailed Experimental Protocol for ESI Parameter Optimization

This protocol provides a step-by-step guide for systematically optimizing ESI source parameters, based on methodologies used in recent publications [1] [9].

Materials and Reagents

• Mass Spectrometer System: Orbitrap-based or other high-resolution mass spectrometer with a tunable ESI source.
• Liquid Chromatography System: UHPLC system capable of delivering a stable gradient.
• Test Sample: Use a complex matrix representative of your study (e.g., NIST SRM 1950 human plasma [1] or a pooled quality control (QC) sample created from all study samples). A "master mix" sample can also be prepared for assessing LC and sensitivity [7].
• Solvents: LC-MS grade water, acetonitrile, and methanol. Additive: 0.1% formic acid is common.

Step-by-Step Optimization Procedure

• Sample Preparation: Extract the test sample using a standardized protocol relevant to your matrix (e.g., protein precipitation with cold acetonitrile for plasma). Reconstitute the final extract in a solvent compatible with your initial LC conditions.
• Initial LC-MS Conditions: Establish a representative, short LC gradient (e.g., 15-20 minutes). Set the mass spectrometer to acquire data in both positive and negative ionization modes over a suitable mass range (e.g., 60–1200 m/z). Use a moderate scan speed (e.g., 0.1–0.5 seconds) [7].
• System Suitability and Baseline: Inject the test sample with the manufacturer's default source settings. Use this data as a baseline for comparison. Monitor total ion chromatogram (TIC) stability and the number of metabolic features detected.
• Define the Experimental Design: A central composite design (response surface methodology) is highly effective for exploring multiple parameters and their interactions [8]. Alternatively, a one-factor-at-a-time (OFAT) approach can be used.
• Iterative Parameter Adjustment:
  • Begin by optimizing the ESI needle position relative to the inlet for maximum signal intensity [1] [10].
  • Systematically vary one parameter at a time (or according to your experimental design) while keeping others constant. The ranges in Table 2 serve as a starting point.
  • For each parameter set, inject replicates of the test sample to assess reproducibility.
• Data Acquisition and Processing: Acquire data in untargeted full-scan mode. Process all data files uniformly using software like MarkerLynx [8], MS-DIAL [11], or MZMine [7] to extract the number of metabolic features, their abundances, and signal stability metrics (e.g., %RSD for internal standards).
• Data Analysis and Selection of Optimal Conditions:
  • Analyze the processed data using chemometric approaches such as Principal Component Analysis (PCA) to visualize how different parameter settings affect the overall structure of the data [8].
  • The optimal set of parameters is the one that maximizes the number of reproducibly detected metabolic features and the overall stability of the TIC [1].

Workflow Visualization

The following diagram illustrates the logical workflow for the ESI parameter optimization protocol:

The Scientist's Toolkit: Essential Reagents and Materials

Successful optimization and application of an ESI-based metabolomics method rely on specific, high-quality reagents and materials. The following table details key solutions used in the featured experiments.

Table 3: Essential Research Reagent Solutions for ESI Metabolomics

Integrated Workflow and Concluding Remarks

Optimizing the ESI source is not an isolated activity; it is an integral part of a larger analytical workflow designed for comprehensive metabolome coverage. The most effective untargeted metabolomics strategies often combine optimized ESI settings with complementary chromatographic methods, such as Reversed-Phase (RP) and Hydrophilic Interaction Liquid Chromatography (HILIC), to capture both hydrophobic and hydrophilic metabolites [1] [12]. As demonstrated experimentally, this combined approach can detect 60% more metabolic features compared to using a RP column alone [9] [10].

In conclusion, the spray voltage, gas flows, and temperatures of the ESI source are critical levers that control the quality and scope of untargeted metabolomics data. A systematic, experimental approach to optimization, using complex, representative samples and chemometric analysis, is strongly recommended over relying solely on manufacturer defaults. By investing in this optimization process, researchers can significantly reduce analytical biases, improve reproducibility, and ensure that their studies achieve the deepest possible coverage of the metabolome, thereby maximizing the potential for robust biological discovery.

How Ion Source Conditions Influence Metabolite Detection and Annotation Confidence

Electrospray Ionization (ESI) is a pivotal soft ionization technique in liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics. The sensitivity, reproducibility, and breadth of metabolome coverage are profoundly influenced by the specific conditions of the ESI ion source. This application note details the critical impact of key ion source parameters—including needle position, spray voltage, gas flows, and temperatures—on data quality and annotation confidence. We provide optimized, experimentally-derived protocols for parameter tuning and sample preparation, supported by quantitative data. Adherence to these protocols ensures the generation of high-fidelity, reproducible metabolomic data, which is fundamental for robust biological interpretation and biomarker discovery in pharmaceutical and clinical research.

In untargeted metabolomics, the goal is to comprehensively profile all detectable small molecules in a biological sample without a priori knowledge of the compounds present [1]. The electrospray ionization (ESI) source is the critical interface where solution-phase analytes are converted into gas-phase ions, making the conditions within the ion source a primary determinant of the quality and quantity of the metabolic data generated [13] [14]. ESI uses electrical energy to generate a fine aerosol of charged droplets from a liquid stream, leading to the production of gas-phase ions through solvent evaporation and Coulomb fission [13]. As a "soft" ionization technique, ESI preserves molecular integrity, facilitating the observation of protonated or deprotonated quasimolecular ions like [M+H]+ or [M-H]- [14].

However, the ionization process is susceptible to numerous factors. Suboptimal ion source conditions can lead to poor ion yield, signal suppression from matrix effects, and reduced analytical reproducibility [1] [15]. Consequently, this directly compromises the confidence of metabolite annotation, a process defined as the inference of structures and functions for detected metabolites [16]. Therefore, a systematic approach to optimizing ESI parameters is not merely a technical exercise but a foundational requirement for any rigorous untargeted metabolomics study aiming to generate biologically meaningful and reliable results.

Key Ion Source Parameters and Their Optimization

The ionization efficiency of metabolites across diverse chemical classes is highly sensitive to the physical and electrical environment of the ESI source. The following parameters require careful optimization to expand metabolome coverage and improve data quality.

ESI Needle Position and Spray Voltage

The physical position of the ESI needle relative to the mass spectrometer's orifice and the voltage applied to the spray are fundamental to initiating a stable electrospray.

• Needle Position: The X, Y, Z coordinates of the ESI needle must be adjusted to maximize signal intensity and stability. Even minor misalignments can significantly degrade signal quality across a wide range of metabolites [1]. The optimal position ensures the spray is directly sampled into the inlet orifice without electrical discharge or turbulence.
• Spray Voltage: This high voltage (typically 2.5 – 6.0 kV) is applied to the capillary to disperse the liquid into a fine mist of charged droplets [13]. The optimal voltage is polarity-dependent. For positive ion mode, a voltage of 3.8 kV has been demonstrated to provide superior signal stability, whereas 3.2 kV is often optimal for negative ion mode [1]. Using excessively high voltage can lead to electrical discharge and increased background noise, while low voltage may cause an unstable spray and poor ionization.

Gas Flows and Temperature Settings

Desolvation—the process of removing solvent molecules from charged droplets—is facilitated by heated gases and capillaries. Proper configuration is essential for efficient ion release.

• Sheath Gas: This gas stream (typically nitrogen) flows coaxially around the ESI needle to assist in nebulizing the liquid stream and shaping the spray. An optimal setting of 45 arbitrary units has been shown to support high-quality signals [1].
• Auxiliary Gas: Another stream of nitrogen (or other inert gas) that acts as a drying gas to assist in solvent evaporation from the charged droplets. A setting of 15 arbitrary units is often effective [1].
• Ion Transfer Tube (ITT) Temperature: This heated capillary serves as the entrance to the mass spectrometer. Its temperature must be high enough to ensure complete desolvation but not so high as to cause thermal degradation of labile metabolites. An optimized value of 300°C is recommended [1].

The Challenge of Signal Suppression

A significant phenomenon affecting quantification accuracy in ESI is signal suppression, where the ionization of an analyte is inhibited by co-eluting compounds [15]. This can occur due to charge competition or changes in the droplet surface properties.

• Mechanism: In a complex biological matrix, highly concentrated or more easily ionized compounds can "compete" for the limited charge available on the droplet surface, suppressing the signal of lower-abundance or less efficiently ionized metabolites [15].
• Experimental Evidence: A model study with metformin (MET) and glyburide (GLY) demonstrated that when co-eluted, the signal for GLY could be suppressed by up to 30% in the presence of high concentrations of MET. This suppression was dependent on the concentration of the interfering substance (MET) and not the analyte itself (GLY) [15].
• Mitigation Strategies:
  • Chromatographic Separation: Improving LC separation to resolve the interfering compounds is the most effective strategy.
  • Sample Dilution: Diluting the sample can reduce the absolute concentration of interferents, thereby alleviating suppression. However, this comes at the cost of reduced sensitivity for low-abundance metabolites [15].
  • Stable Isotope-Labeled Internal Standards (SIL-IS): Using SIL-IS for the analytes of interest can correct for the variable ion suppression, as the standard and analyte experience the same matrix effects, ensuring accurate quantification [15].

Table 1: Summary of Key Ion Source Parameters and Their Optimized Values

Protocols for Optimizing ESI Conditions

A systematic, step-by-step approach is required to transition from default settings to an optimized method tailored to your specific instrument and research question.

Protocol: Systematic Optimization of ESI Parameters

This protocol is designed for the iterative tuning of critical ion source settings to maximize metabolome coverage and signal quality [1].

• Initial Setup and Calibration: Begin with manufacturer-recommended settings. Acquire data for a pooled quality control (QC) sample—a mixture of all experimental samples—to establish a baseline.
• Needle Positioning: Using the QC sample, adjust the X, Y, and Z coordinates of the ESI needle in small increments while monitoring the total ion current (TIC) and the intensity of a set of representative internal standards. Lock the position that yields the highest and most stable signal.
• Spray Voltage Optimization: In the optimized needle position, test a range of spray voltages (e.g., from 2.8 kV to 4.2 kV in 0.2 kV steps) for both positive and negative ionization modes independently. The optimal voltage provides the highest number of detected metabolic features and the best overall signal stability.
• Gas and Temperature Optimization: Sequentially optimize the sheath gas, auxiliary gas, and ion transfer tube temperature. For each parameter, hold others constant and test a range of values. The goal is to maximize the signal for a broad range of metabolites without introducing excessive noise.
• Validation with Complex Mixtures: Validate the final optimized parameters by analyzing a standard reference material (e.g., NIST SRM 1950 plasma) and monitoring the number of detected features, signal stability, and intensity of key metabolite classes.

Protocol: Determining Optimal Sample Injection Concentration

Sample concentration injected into the LC-MS system is a critical, yet often overlooked, parameter. Overloading causes signal saturation and non-linearity, while under-loading results in poor detection of low-abundance metabolites [17].

• Sample Preparation: Perform a biphasic extraction on your target tissue (e.g., liver, adipose) to obtain pooled lipid and polar extracts [17].
• Serial Dilution: Reconstitute the dried pooled extracts in a series of different solvent volumes to create a dilution series covering a wide concentration range (e.g., from 3.91 mg/mL to 250 mg/mL tissue mass equivalent per mL of solvent) [17].
• LC-MS Analysis: Analyze the entire dilution series using your LC-MS method.
• Data Analysis:
  • Feature Detection: Process the data to extract the number of metabolic "features" (unique m/z-RT pairs) at each concentration.
  • Reproducibility Assessment: Calculate the coefficient of variation (CV) for features across technical replicates at each concentration.
  • Linearity Assessment: Evaluate the linearity of feature intensity across the dilution series. Features with a linear response (e.g., R² > 0.95) are within the dynamic range of the method.
• Concentration Selection: The optimal injection concentration is the one that maximizes the number of reproducible (low CV) and linear features. For example, one study found the optimum for pig liver lipidomics to be 125 mg/mL, which yielded 2811 linear features in ESI+ mode [17].

Experimental Design for Parameter Evaluation

To rigorously assess the impact of ion source conditions, a structured experimental design is essential.

Instrumentation and Reagents

• Mass Spectrometer: High-resolution mass spectrometer (e.g., Orbitrap, Q-TOF).
• Chromatography: Ultra-high-performance liquid chromatography (UHPLC) system.
• Columns: Utilize multiple column chemistries to expand metabolome coverage. This includes reversed-phase (RP-C18) columns for non-polar to mid-polar metabolites and hydrophilic interaction liquid chromatography (HILIC) columns for polar metabolites [1] [18].
• Mobile Phases:
  • Aqueous (A): 10 mM ammonium formate with 0.1% formic acid [18].
  • Organic (B): Acetonitrile with 0.1% formic acid [18].
• Standards and Samples: Use a mixture of authentic metabolite standards and standardized biofluids like NIST SRM 1950 Metabolites in Human Plasma to ensure benchmarking is reproducible and comparable across laboratories [1].

Data Processing and Annotation Confidence

The data acquired from optimized methods must be processed with stringent statistical controls to ensure annotation confidence.

• Feature Extraction: Use software (e.g., Compound Discoverer, XCMS) to pick chromatographic peaks and align features across samples.
• False Discovery Rate (FDR) Control: For confident annotation, employ target-decoy strategies adapted from proteomics. Tools like DIAMetAlyzer can generate decoy assays by re-rooting fragmentation trees, allowing for robust FDR estimation (e.g., at 1% or 5%) during metabolite identification [19].
• Annotation Grading: Implement a grading system for annotations [16]:
  • Grade A: Confirmed by authentic standard (accurate mass, retention time, and MS/MS spectrum).
  • Grade B: Putatively annotated by MS/MS spectral similarity to public libraries.
  • Grade C: Putatively characterized by accurate mass and predicted chemical class.

Table 2: The Scientist's Toolkit - Essential Reagents and Materials for ESI Metabolomics

Workflow Visualization

Diagram 1: ESI Parameter Optimization Pathway. This workflow outlines the sequential and iterative process of tuning ion source parameters to achieve a robust and sensitive LC-MS method for untargeted metabolomics.

Diagram 2: Untargeted Metabolomics Workflow. The ESI ion source is a critical gateway where physical parameters directly impact the quality and quantity of ions detected by the mass spectrometer, ultimately affecting downstream annotation confidence.

Electrospray Ionization (ESI) is a cornerstone technique in liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics, serving as the critical interface where liquid-phase analytes are converted into gas-phase ions. The performance of the ESI source is not merely a technical detail; it directly and profoundly determines the sensitivity, reproducibility, and overall coverage of the metabolomic investigation [2] [1]. Optimizing ESI parameters is therefore essential for minimizing analytical variance and bias, which is a prerequisite for generating high-quality, biologically meaningful data [1]. This Application Note delineates the quantitative impact of ESI performance on key data quality metrics and provides detailed protocols for its systematic optimization, framed within the broader objective of enhancing reliability in untargeted metabolomics research.

Impact of ESI and LC Configuration on Data Quality

The configuration of the ESI source and the liquid chromatography (LC) system collectively establishes the upper limit for data quality in an untargeted metabolomics experiment. Empirical evaluations demonstrate the measurable effects of different choices.

Comparative Performance of MS Acquisition Modes

The data acquisition mode is a fundamental selection that influences feature detection and reproducibility. A systematic comparison of three common acquisition modes highlights significant differences in their performance profiles [20].

Table 1: Performance Comparison of Data Acquisition Modes in Untargeted Metabolomics

Data-Independent Acquisition (DIA) demonstrated superior performance, detecting the highest number of metabolic features and exhibiting the best reproducibility and identification consistency [20]. This makes DIA a compelling choice for studies where data completeness and quantitative robustness are paramount.

Expanding Metabolome Coverage via Orthogonal Chromatography

A single chromatographic separation often inadequately captures the vast chemical diversity of the metabolome. Research confirms that combining reversed-phase (RP) and hydrophilic interaction liquid chromatography (HILIC) significantly expands metabolite coverage.

One study found that integrating an optimized RP-C18 method with a zwitterionic HILIC approach detected 60% new metabolic features compared to using RP chromatography alone for serum samples [2]. This dual-column strategy is crucial for comprehensive analysis, effectively reducing "analytical blind spots" for polar and non-polar metabolites [21].

Experimental Protocols for ESI and LC-MS Optimization

The following protocols provide a structured approach for optimizing the ESI source and evaluating LC methods to maximize data quality.

Protocol 1: Systematic Optimization of ESI Source Parameters

This protocol is designed to methodically tune ESI parameters for maximum signal intensity and stability across a broad metabolite range [2] [1].

1. Preparation of Standard and QC Solutions:

• Prepare a standard mixture of ~95 authentic metabolite standards from diverse chemical classes (e.g., amino acids, lipids, carbohydrates) at concentrations in the low µM range [1].
• Generate a Quality Control (QC) sample by pooling small aliquots of all experimental biological samples or using a commercial standard reference material like NIST SRM 1950 plasma [1].

2. Instrumental Setup:

• Use an Orbitrap-based or similar high-resolution mass spectrometer.
• Install a suitable RP or HILIC column (e.g., BEH C18 or BEH Z-HILIC) [2].

3. Iterative Parameter Tuning:

• Spray Voltage: Test both positive and negative ion modes. For Orbitrap systems, evaluate voltages in the range of 3.0 to 4.5 kV [1].
• Ion Transfer Tube (ITT) Temperature: Assess temperatures from 250°C to 350°C. Higher temperatures generally improve desolvation but may induce thermal degradation for labile compounds [1].
• Sheath and Auxiliary Gas: Optimize gas flow rates (e.g., 10-50 Arb for sheath gas). Higher flows can enhance spray stability in high aqueous mobile phases [1].
• Vaporizer Temperature: This parameter can further aid in desolvation; optimize based on the solvent flow and composition [1].
• Needle Position: Carefully adjust the sprayer position relative to the MS inlet for maximum ion signal. Even minor deviations can drastically impact sensitivity [2].

4. Data Acquisition and Evaluation:

• Inject the standard mixture and QC sample repeatedly after each parameter adjustment.
• Monitor the total ion chromatogram (TIC) for signal stability.
• Process data using software like Compound Discoverer or MZMine. The optimal condition is that which yields the highest number of annotated metabolites, the greatest total peak area, and the most stable signal (lowest %RSD) across replicate injections [1].

Protocol 2: Evaluating LC Columns for Metabolite Coverage

This protocol assesses different LC columns to maximize the breadth of metabolite detection [2] [1].

1. Column Selection:

• Select multiple column chemistries for evaluation. For RP, consider modern C18 columns like Premier CSH C-18, HSS T3 C-18, and BEH C-18. For HILIC, include a zwitterionic column such as BEH Z-HILIC [2].

2. Sample and Mobile Phase Preparation:

• Use the same standard mixture and QC sample from Protocol 1.
• Employ LC-MS Optima-grade solvents and additives (e.g., 0.1% formic acid) [1].

3. Chromatographic Method Development:

• Develop and run separate, optimized gradient elution methods for each column type.
• For RP, a gradient from aqueous to organic (e.g., acetonitrile) is standard.
• For HILIC, a gradient from high organic (e.g., 80-95% acetonitrile) to aqueous is used [2].

4. Data Analysis and Column Selection:

• Acquire data in full-scan MS mode, and if possible, data-dependent MS/MS.
• Process raw data to extract metabolic features (chromatographic peaks with associated m/z and retention time).
• Compare the total number of metabolic features, the number of annotated metabolites, and the diversity of chemical classes detected by each column.
• For comprehensive coverage, implement the orthogonal RP and HILIC methods on your sample set, as their combined coverage is vastly superior to either alone [2].

Diagram 1: ESI parameter optimization workflow. ITT: Ion Transfer Tube.

The Scientist's Toolkit: Essential Reagents and Materials

Successful method optimization relies on high-quality, specific materials. The following table lists key reagents and their critical functions based on the cited research.

Table 2: Essential Research Reagents and Materials for ESI-LC-MS Optimization

Concluding Remarks

The path to robust and comprehensive untargeted metabolomics data is paved with meticulous analytical optimization. As demonstrated, the conscious optimization of ESI source parameters and the strategic implementation of orthogonal chromatographic separations are not optional refinements but fundamental requirements. These protocols provide a clear roadmap for researchers to enhance the sensitivity, reproducibility, and coverage of their metabolomic studies, thereby strengthening the foundation for subsequent biological discovery and translational application.

Diagram 2: Relationship between optimized parameters and data quality outcomes. DIA: Data-Independent Acquisition.

Systematic Strategies for ESI Parameter Optimization and Method Development

Electrospray Ionization (ESI) is a pivotal technique in liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics, enabling the sensitive detection of a vast array of metabolites. The quality of data generated in these experiments is profoundly influenced by the optimization of ESI source parameters [7] [1]. Suboptimal settings can lead to reduced sensitivity, poor metabolite coverage, and the formation of adducts that complicate spectral interpretation [22]. This application note provides a detailed, step-by-step protocol for the systematic optimization of key ESI parameters—including needle positioning, spray voltage, and gas flow rates—within the context of untargeted metabolomics research. The procedures are designed to help researchers, scientists, and drug development professionals maximize signal stability, enhance metabolome coverage, and improve the reproducibility of their analyses.

Background and Significance

In untargeted metabolomics, the goal is to comprehensively profile all detectable small molecules (typically ≤ 2000 Da) in a complex biological sample without prior knowledge of the compounds present [7]. ESI is a "soft" ionization technique that produces gas-phase ions from a liquid solution with minimal fragmentation, making it ideal for detecting labile metabolites. It is particularly useful for biological samples and complex mixtures because it is relatively gentle and can ionize a wide range of molecules [7].

The ionization efficiency in ESI is highly dependent on the physicochemical properties of the analyte and the specific conditions within the ion source [23]. Major differences in ionization efficiency can result in widely varying responses, complicating quantitative analysis in non-targeted studies [23]. Therefore, careful optimization of the ESI source is not a trivial exercise but a fundamental requirement for generating high-quality, reliable metabolomic data. Parameters such as the sprayer position, spray voltage, and gas temperatures must be tailored to the specific experimental conditions, including the chromatographic method and the sample matrix, to ensure optimal performance [1] [22].

Experimental Protocol for ESI Optimization

Safety and Preparatory Procedures

• Personal Protective Equipment (PPE): Wear safety glasses, gloves, and a lab coat throughout the procedure.
• Solvent Handling: Use LC-MS grade solvents and handle them in a well-ventilated fume hood.
• Electrical Safety: Be aware that high voltages are used in the ESI source. Ensure the instrument is powered down and properly locked out before performing any mechanical adjustments not described in the software-controlled methods.
• System Setup: Prior to optimization, ensure the LC-MS system is properly calibrated and maintained. For infusion experiments, use a syringe pump and a stable, purified standard solution relevant to your metabolomics study (e.g., a metabolite standard mix dissolved in a solvent resembling the starting mobile phase conditions).

Step 1: Needle Position Optimization

The position of the ESI probe relative to the sample cone is critical for maximizing ion transmission and signal intensity [24] [22].

Detailed Procedure:

• Prepare a Tune File: Using the mass spectrometer's software, open a sample-infusion tune file [24].
• Configure Fluidics: Set the sample flow path to "Combined" to allow for simultaneous infusion of the standard and LC flow during optimization [24].
• Set Initial Parameters: Establish initial source temperatures and gas flows based on the expected LC flow rate. Refer to guidelines such as those in [24]:

• Start Flow and Stabilize: Infuse your standard solution at a typical flow rate (e.g., 10 µL/min) and start the external LC flow. Allow the source temperatures to stabilize [24].
• Optimize Position: Monitor the intensity of a dominant ion from your standard. Using the Vernier probe adjuster on the probe's mounting flange, carefully change the probe's position—typically its distance from the sampling cone—to maximize the signal intensity [24].
  • Expert Tip: The optimal position can be analyte-dependent. More polar analytes often benefit from the sprayer being farther from the cone, while larger hydrophobic analytes may respond better when the sprayer is closer [22].
  • Critical Warning: Positioning the probe too close to the sample cone can lead to nonlinear data, signal suppression, and require frequent source cleaning [24].
• Verify Electrical Grounding: Set the capillary voltage to zero. If the signal intensity does not drop to less than 10% of its optimum value, the probe is likely too close and should be retracted slightly. Restore the voltage and re-check the signal [24].

The following workflow summarizes the key steps for ESI source optimization:

Step 2: Spray Voltage Tuning

The spray voltage (or capillary voltage) is responsible for charging the liquid effluent and forming the Taylor cone. Its optimization is crucial for a stable electrospray.

Detailed Procedure:

• Initial Setting: With the needle position optimized, start with a moderate spray voltage (e.g., 3.0 kV for positive mode).
• Systematic Adjustment: Gradually adjust the voltage in small increments (e.g., 0.1 - 0.2 kV) while monitoring the total ion count and the signal of your specific analyte(s).
• Identify Optimal Range: Find the voltage that provides a stable, maximum signal intensity. Avoid voltages that are too high, as they can cause electrical discharge (evidenced by erratic signals and solvent cluster ions in positive mode) and unwanted redox reactions [22].
• Consider Mobile Phase: The optimal voltage is dependent on the mobile phase composition. More aqueous eluents generally require higher voltages, while organic-rich eluents require lower voltages. As a rule of thumb, use the lowest stable voltage that provides maximum signal [22]. The table below provides example threshold voltages for different solvents [22]:

• Gradient Methods: For LC methods with gradients, it is advisable to infuse your standard dissolved in the eluent composition at which the analyte elutes to find the optimal voltage for that specific time window [22].

Step 3: Optimization of Gas Flow Rates and Temperatures

The desolvation (drying) gas and nebulizer gas, along with their associated temperatures, are essential for efficiently evaporating the solvent from the charged droplets and liberating gas-phase ions.

Detailed Procedure:

• Nebulizer Gas: This gas is responsible for pneumatically assisting the formation of the spray and controlling initial droplet size. Optimize its flow rate to achieve a stable signal and fine droplet mist for the given LC flow rate [22].
• Desolvation Gas and Temperature: This gas stream, typically heated, helps to evaporate the solvent from the droplets.
  • Start with manufacturer-recommended settings (e.g., 350-500 °C for desolvation temperature and 800-1000 L/h for gas flow, depending on flow rate [24]).
  • Systematically adjust the temperature and flow, finding the balance where the signal is maximized. Excessive temperatures can lead to thermal degradation of labile metabolites, while insufficient temperatures result in poor desolvation and reduced sensitivity.
• Cone Gas (if applicable): Some sources use a cone gas (often N₂) directed at the sample cone to help break up solvent clusters and prevent non-volatile contaminants from accumulating. Start from a low value (e.g., 0 L/h) and increase in increments of 50 L/h, allowing stabilization after each adjustment. Use the highest flow that does not significantly reduce analyte signal [24].

Step 4: Cone Voltage / Declustering Potential Optimization

This parameter (known as Fragmentor, Cone Voltage, or Declustering Potential depending on the instrument vendor) controls the ion's energy as it moves from the atmospheric pressure source into the high-vacuum region of the mass spectrometer.

Detailed Procedure:

• Set Purpose: Determine the goal: a lower voltage (e.g., 10-30 V) is typically used to gently decluster ions and observe the intact [M+H]⁺ or [M-H]⁻ pseudomolecular ion. A higher voltage (e.g., 30-100 V) can be used to induce controlled in-source fragmentation (CID) for structural information [22].
• Infuse Standard: Infuse a metabolite standard and observe the spectrum.
• Find Balance: For untargeted metabolomics where the goal is broad detection of intact precursors, optimize for a voltage that maximizes the intensity of the pseudomolecular ion while minimizing in-source fragmentation. The voltage should be high enough to decluster solvent adducts but low enough to avoid breaking the molecule apart [22].

Key Research Reagents and Materials

The following table lists essential materials and their functions for conducting ESI optimization in metabolomics.

Results and Data Interpretation

A successfully optimized ESI source should yield the following outcomes, as demonstrated in optimization studies [1]:

• Enhanced Signal Stability and Quality: A stable total ion chromatogram (TIC) and base peak chromatogram (BPC) with minimal noise and fluctuation.
• Increased Metabolite Coverage: A higher number of detected metabolic features (unique m/z-RT pairs) in a complex sample like NIST SRM 1950 plasma.
• Improved Signal Intensity: The intensity of detected features should be maximized, improving the detection of low-abundance metabolites.
• Minimized Adduct Formation: Reduced prevalence of sodium and potassium adducts, leading to cleaner spectra and simplifying data interpretation [22].

The optimized parameters are interdependent. For instance, a study optimizing an Orbitrap MS for plasma metabolomics found that the ion transfer tube (ITT) temperature, vaporizer temperature, and sheath gas pressure were among the most critical parameters influencing overall signal quality and metabolome coverage [1]. The table below summarizes the key parameters and their typical effects.

Systematic optimization of the ESI source is a fundamental prerequisite for obtaining high-quality, reproducible data in untargeted metabolomics. This protocol provides a detailed, actionable guide for tuning critical parameters such as needle position, spray voltage, and gas settings. By following these steps, researchers can significantly enhance the sensitivity and coverage of their methods, leading to more confident metabolite annotation and more reliable biological insights. The optimized method should be validated using a complex, biologically relevant matrix to ensure its robustness before application to actual study samples.

Employing Statistical Design of Experiments (DOE) for Efficient Parameter Screening

In untargeted metabolomics, the electrospray ionization (ESI) source is a critical gateway for transferring analytes from the liquid chromatographic system to the mass spectrometer. Its performance directly influences sensitivity, reproducibility, and the overall coverage of the metabolome detected. The optimization of this source is notoriously complex, as multiple interdependent parameters—including gas flows, temperatures, and voltages—can exert significant and often non-linear effects on ionization efficiency. The traditional "One-Variable-At-a-Time" (OVAT) approach to optimization is inefficient, fails to capture interactions between factors, and risks identifying false optima [25] [26].

A systematic Design of Experiments (DOE) strategy overcomes these limitations by enabling the concurrent evaluation of multiple factors and their interactions in a minimal number of experimental runs [26]. This document outlines a robust, statistically-driven protocol for the efficient screening of ESI parameters, providing a solid foundation for sensitive and reliable untargeted metabolomics research.

Experimental Design and Workflow

The following workflow provides a step-by-step guide for implementing a DOE strategy to screen and optimize ESI parameters.

Pre-Experimental Planning

• Define the Objective: The primary goal is to identify ESI parameters that maximize the MS response (e.g., peak area or height) for a wide range of metabolites in an untargeted context, while ensuring stability.
• Select a Representative Metabolite Standard Mix: Prepare a mixture of chemically diverse standard metabolites that reflect the expected chemical space of your samples. This ensures the optimized conditions are not biased towards a single compound class [27] [28]. For a more targeted screening of poorly ionizing compounds, specific metabolites like 7-methylguanine (positive mode) and glucuronic acid (negative mode) can be used [25].
• Establish Chromatographic Conditions: The DOE must be performed using the final chromatographic method (column, mobile phase, and gradient) to account for the influence of the eluting solvent on ionization [28].
• Identify Critical Factors and Ranges: Based on literature and instrument capabilities, select the key adjustable ESI parameters and their practical ranges for investigation. A typical screening might include the factors listed in the table below.

Table 1: Key ESI Parameters for Initial Screening

The Screening Phase: A Fractional Factorial Design

With multiple factors to investigate, a full factorial design would be prohibitively time-consuming. A two-level Fractional Factorial Design (FFD) is the most efficient tool for initial screening. An FFD, such as a 2^(f-v) design, allows for the examination of f factors at two levels (high and low) in a fraction of the runs required for a full factorial design, successfully identifying the factors with the most significant impact on the response [25] [26].

• Experimental Execution:

  • Use statistical software (e.g., JMP, R, Modde Pro) to generate the randomized run table for the FFD.
  • Inject the standard metabolite mix for each unique combination of factor levels as per the experimental design.
  • Ensure system equilibration between runs where parameter changes are significant.

• Data Acquisition and Processing:

  • For each run, extract the MS signal (peak area or height) for every metabolite in the standard mix.
  • Compile the data into a response matrix for statistical analysis.

• Statistical Analysis:

  • Perform multivariate statistical analysis, such as Partial Least Squares (PLS) regression, to model the relationship between the ESI parameters (factors) and the MS responses [27].
  • Analyze the model to identify which factors have a statistically significant (p < 0.05) effect on the signal intensity.
  • Use this analysis to narrow down the list of critical factors for the subsequent optimization phase.

The Optimization Phase: Response Surface Methodology

Once the most influential factors are identified, the next step is to find their optimal levels. This is achieved using Response Surface Methodology (RSM).

• Select an RSM Design: For 2-4 critical factors, a Face-Centered Central Composite Design (CCD) or a Box-Behnken Design (BBD) is appropriate. These designs efficiently explore a three-level experimental space (low, middle, high) to model curvature and identify a true optimum [25] [30].
• Execution and Modeling:
  • Perform the experiments as per the RSM design.
  • Fit the data to a quadratic model and generate contour or 3D surface plots for the responses.
  • These plots visually represent the relationship between the factors and the response, allowing for the identification of a "sweet spot" that maximizes signal intensity [25] [28].

Robustness Testing

Finally, verify the robustness of the identified optimal settings by performing a small series of experiments around the optimum, for example, using a D-optimal design. This confirms that small, unavoidable variations in parameter settings during routine operation will not critically impact method performance [28].

Diagram 1: AThree-Phase DOE Workflow for ESI Optimization.

A Practical Example: Optimizing for Polar Metabolomics

The COLMeD (Comprehensive Optimization of LC-MS Metabolomics methods using DoE) workflow provides a proven framework for method development [27]. In one application, a screening DoE was used to optimize an ESI+ method for polar metabolites on a QqQ instrument. The factors investigated included mobile phase composition, additive concentration, column temperature, and flow rate. The response was the peak area for a panel of polar metabolites. The outcome was a method that yielded a median response increase of 161.5% and a 13.3% increase in metabolite coverage compared to the initial conditions [27].

Table 2: Example Reagent Solutions for LC-MS Metabolomics

Employing a structured DOE approach for ESI parameter screening is no longer an advanced tactic but a fundamental component of robust metabolomics method development. By moving beyond the inefficient OVAT method, researchers can efficiently account for complex factor interactions, maximize instrument sensitivity and metabolite coverage, and build a rigorous, statistically-defensible foundation for their untargeted discovery research. The protocols outlined herein provide a clear roadmap for implementing this powerful strategy.

In untargeted metabolomics, achieving comprehensive metabolome coverage is critically dependent on the stability and sensitivity of the electrospray ionization (ESI) source in liquid chromatography-mass spectrometry (LC-MS). The formation of a stable electrospray is fundamentally governed by the surface tension of the mobile phase, which directly impacts the initial droplet formation and subsequent ion liberation into the gas phase. This application note details the principles and practical protocols for selecting mobile phases and solvents to minimize surface tension, thereby promoting a stable Taylor Cone and enhancing overall MS signal quality for a broader detection of metabolites. Framed within the broader objective of optimizing ESI parameters, this guidance is essential for improving data quality in hypothesis-generating untargeted studies.

Theoretical Background: Surface Tension in the ESI Process

The electrospray ionization process begins with the application of a high voltage to a liquid emerging from a capillary, forming a Taylor Cone from which a fine plume of charged droplets is emitted [31]. The stability of this cone and the initial droplet size are paramount for efficient ion production.

Surface tension is the elastic tendency of a fluid surface that makes it acquire the least surface area possible. In the context of ESI, it is a force that pulls the liquid back into a spherical shape, resisting the electrical forces that pull the liquid into a cone and disperse it into droplets. A lower surface tension allows for a more stable Taylor Cone and the production of smaller primary droplets at a lower voltage, which subsequently undergo Coulombic fissions to yield gas-phase ions more efficiently [32] [31]. Solvents with high surface tension, such as pure water, require higher voltages to overcome this cohesive force, increasing the risk of electrical discharge and an unstable spray.

Solvent and Mobile Phase Selection Guidelines

The core strategy for minimizing surface tension involves the careful selection of organic modifiers and volatile additives. The following guidelines are curated to assist in mobile phase preparation.

Selection of Organic Solvents

The primary organic solvents used in reversed-phase LC-MS exhibit different surface tension properties and should be chosen appropriately.

Table 1: Properties of Common LC-MS Organic Solvents

As a general rule, methanol and acetonitrile are the preferred organic modifiers for ESI-MS. While isopropanol has the lowest surface tension, its use can result in poorer chromatographic performance and ESI response, making it less suitable as a primary solvent [33].

Mobile Phase Additives and Buffer Selection

Additives are essential for controlling chromatographic separation but must be selected for ESI compatibility.

Table 2: Common Mobile Phase Additives for ESI-MS

Best Practice: Mobile phases based on methanol or acetonitrile with formic acid and ammonium acetate are consistently the best-performing generic solvents for a wide range of analytes [33]. Phosphate buffers, non-volatile salts, and acids like trifluoroacetic acid (TFA) should be avoided as they can suppress ionization and cause intense source contamination.

Protocol: Optimizing Mobile Phase for Stable Spray

Objective: To prepare a mobile phase system that minimizes surface tension and ensures a stable ESI spray for untargeted metabolomics.

Materials:

• LC-MS grade Water
• LC-MS grade Acetonitrile
• LC-MS grade Methanol
• LC-MS grade Formic Acid
• LC-MS grade Ammonium Acetate
• Volumetric flasks, pipettes, and solvent bottles

Procedure:

• For a Generic Reversed-Phase Gradient:
  • Mobile Phase A: Water supplemented with 0.1% formic acid and 5-10 mM ammonium acetate.
  • Mobile Phase B: Acetonitrile (or Methanol) supplemented with 0.1% formic acid and 5-10 mM ammonium acetate.
• For Highly Aqueous Methods: If the initial mobile phase composition exceeds 95% water, add 1-2% (v/v) of methanol or isopropanol to the aqueous phase to significantly lower its surface tension and stabilize the spray [32].
• Mixing and Filtration: Prepare all mobile phases using high-purity LC-MS solvents. Filter through a 0.22 µm membrane filter to remove particulates.
• Degassing: Degas the mobile phases by sparging with helium or using an in-line degasser to prevent bubble formation in the LC system and ESI source.

Integrated ESI Source Optimization Workflow

Mobile phase selection is one component of a holistic ESI optimization strategy. The following workflow integrates solvent selection with other critical source parameters.

Diagram: Integrated ESI parameter optimization workflow, highlighting the foundational role of mobile phase (green) and its interaction with other source settings.

Concomitant Source Parameter Tuning

Once the mobile phase is optimized, fine-tune the following parameters for maximum sensitivity:

• Sprayer Voltage: Start with a lower voltage (~2.5-3.5 kV) and incrementally increase while monitoring the total ion chromatogram (TIC) baseline for stability. High voltages can lead to corona discharge, manifested by solvent cluster ions in the background [32].
• Nebulizing and Drying Gas: Optimize the nebulizing gas (sheath gas) flow rate to produce a stable spray at the LC flow rate used. The drying gas (auxiliary gas) flow and temperature should be set to ensure complete solvent evaporation from droplets without causing premature analyte thermal degradation [32] [1].
• Sprayer Position: The lateral and vertical position of the ESI probe relative to the MS inlet can significantly impact sensitivity. Typically, smaller polar analytes benefit from the sprayer being farther from the cone, while larger hydrophobic analytes benefit from a closer position [32].

The Scientist's Toolkit: Essential Research Reagents

Table 3: Key Reagent Solutions for ESI-Optimized Metabolomics

In untargeted metabolomics, the goal of achieving extensive metabolome coverage is significantly hampered by the vast dynamic range and wide chemical diversity of metabolites, including polarity. No single liquid chromatography (LC) mode can adequately retain and separate the entire metabolic complement of a biological sample. Reversed-Phase Liquid Chromatography (RPLC) excels at separating mid- to non-polar metabolites but often fails to retain highly polar compounds, which are eluted with the void volume. Conversely, Hydrophilic Interaction Liquid Chromatography (HILIC) effectively retains these polar metabolites that are poorly captured by RPLC [34] [1]. Consequently, integrating RPLC and HILIC has become a cornerstone strategy for comprehensive analysis, ensuring that a broader spectrum of metabolites, from lipids to amino acids, is captured within a single analytical workflow [34]. This integrated approach is fundamental to generating high-quality data in untargeted metabolomics, which is critically dependent on optimized Electrospray Ionization (ESI) parameters to maximize sensitivity and metabolite detection [1]. This application note provides detailed protocols and data for implementing a combined RPLC-HILIC strategy, framed within the context of optimizing ESI parameters for robust untargeted metabolomics.

Experimental Design and Rationale

The Orthogonality of RPLC and HILIC

The power of combining RPLC and HILIC stems from their orthogonal retention mechanisms. RPLC separates metabolites based on hydrophobicity, using a non-polar stationary phase and a polar mobile phase (typically water and acetonitrile). Retention increases with analyte hydrophobicity. In contrast, HILIC employs a polar stationary phase (e.g., bare silica or amide) and a mobile phase high in organic solvent (typically acetonitrile), where retention is primarily governed by hydrophilicity and often involves partitioning into a water-enriched layer on the stationary surface [35]. This fundamental difference in mechanism means that the elution order of metabolites in one mode is unrelated to the order in the other, thereby dramatically increasing the probability of resolving complex mixtures.

Addressing the Solvent Incompatibility Challenge

A significant challenge in directly coupling HILIC and RPLC, especially in two-dimensional LC (2D-LC), is solvent strength mismatch. The HILIC mobile phase is typically rich in acetonitrile, which is a strong eluent in RPLC. Transferring a large volume of ACN-rich effluent from the first dimension (e.g., HILIC) to the second dimension (RPLC) can cause severe peak broadening, distortion, or even analyte breakthrough in the second dimension column [36]. Several strategies have been developed to mitigate this issue:

• Active Solvent Modulation (ASM): This valve-based approach dilutes the effluent from the first dimension with a weak solvent (from the 2D pump) before it is transferred to the second dimension column, thereby reducing the elution strength of the transferred fraction [36].
• On-line Dilution with Trapping: The 1D effluent is diluted and focused on a short trap column before being eluted onto the analytical column for the second dimension separation. This approach was successfully implemented in an in-house 2D-LC setup to analyze pharmaceuticals with a wide polarity range (Log D between -5.75 and 4.22), achieving satisfactory recoveries between 82% and 99% [36].
• In-line Mixing Modulation (ILMM): A commercially available in-line mixer is installed before the 2D-column to homogenize and modify the composition of the mobile phase entering the second dimension [36].

Materials and Methods

Research Reagent Solutions

Table 1: Essential materials and reagents for integrated RPLC-HILIC metabolomics.

ESI Source Optimization Protocol

Optimal ESI conditions are paramount for sensitive and reproducible detection of a wide range of metabolites. The following protocol, adapted from Assress et al., outlines a systematic approach for optimizing ESI parameters on an Orbitrap mass spectrometer [1].

• Basic Protocol: ESI Parameter Optimization
  • Needle Position: Manually adjust the ESI needle position relative to the ion inlet. The optimal position is typically close to the orifice and slightly off-axis. Monitor the total ion chromatogram (TIC) and signal intensity of standard metabolites to find the position that maximizes signal and stability.
  • Spray Voltage: Test a range of voltages (e.g., 2.5 kV to 4.0 kV) in both positive and negative ionization modes. The optimal voltage provides the highest signal intensity for a broad set of metabolites without introducing excessive noise.
  • Ion Transfer Tube (ITT) Temperature: Evaluate temperatures between 250°C and 350°C. Higher temperatures can improve desolvation but may lead to thermal degradation of labile metabolites.
  • Vaporizer Temperature: This parameter assists in droplet desolvation. Test a range (e.g., 100°C to 400°C) to find the ideal setting for your specific application.
  • Sheath Gas and Auxiliary Gas: Optimize the flow rates of both sheath and auxiliary gas (nitrogen is common). These gases aid in nebulization and desolvation. Higher flows generally improve signal for higher flow rates but must be balanced to prevent premature evaporation or cooling effects.
  • Data Analysis: Use the number of detected metabolic features, the sum of peak areas, and signal stability (e.g., %RSD of QC samples) as key metrics to determine the optimal parameter set.

Integrated RPLC-HILIC Analysis Protocol

This protocol describes a parallel method using two separate analytical runs (RPLC-MS and HILIC-MS) to achieve comprehensive coverage, which is more accessible than online 2D-LC for most laboratories.

• Basic Protocol 1: Sample Preparation for LC-MS/MS

  • Prepare samples in a solvent compatible with both RPLC and HILIC injections. For broad coverage, reconstituting in the starting mobile phase of the HILIC method (e.g., >90% ACN) is often effective, as a small injection volume will not significantly disrupt the RPLC column [7].
  • Use protein precipitation with cold ACN or MeOH for plasma/serum samples. Centrifuge and collect the supernatant for analysis [1].
  • Prepare a "master mix" quality control (QC) sample by pooling a small aliquot of all samples. This QC is used to condition the system and is run at regular intervals throughout the batch to monitor system stability [7].

• Basic Protocol 2: RPLC Data Collection

  • Column: C18 column (e.g., 150 x 2.1 mm, 1.7-1.8 µm).
  • Mobile Phase: A: 0.1% FA in H₂O; B: 0.1% FA in ACN.
  • Gradient: Start at 1% B, ramp to 99% B over 10-20 minutes, hold for 2-3 minutes, then re-equilibrate at 1% B for 5-7 minutes.
  • Flow Rate: 0.3-0.4 mL/min.
  • Column Temperature: 40-55°C.
  • Injection Volume: 1-5 µL.

• Basic Protocol 3: HILIC Data Collection

  • Column: Bare silica or amide column (e.g., 150 x 2.1 mm, 1.7-1.8 µm).
  • Mobile Phase: A: 95% ACN with 10 mM ammonium acetate (pH neutral); B: 50% ACN with 10 mM ammonium acetate.
  • Gradient: Start at 100% A, ramp to 100% B over 10-15 minutes, hold for 2-3 minutes, then re-equilibrate at 100% A for 8-10 minutes.
  • Flow Rate: 0.3-0.5 mL/min.
  • Column Temperature: 30-45°C.
  • Injection Volume: 1-5 µL.

Results and Data Analysis

Performance Comparison of LC Columns

A critical step in method development is the selection of the appropriate chromatographic column. The following table summarizes data from a systematic evaluation of different RPLC and HILIC columns for untargeted metabolomics, as performed by Assress et al. [1].

Table 2: Evaluation of RPLC and HILIC columns for plasma untargeted metabolomics. Performance was rated based on the number of detected metabolic features, peak shape, and retention. [1]

Advanced 2D-LC: Reversed HILIC (revHILIC) as a Novel Tool

An innovative approach to overcoming solvent incompatibility in 2D-LC is the use of reversed HILIC (revHILIC). This technique uses a polar stationary phase (like HILIC) but with a gradient that starts with a high-water composition and increases the ACN content—the reverse of a conventional HILIC gradient [35]. This places revHILIC in an intermediate position between RPLC and HILIC in terms of selectivity.

• Application in 2D-LC: When revHILIC is used in the first dimension and RPLC in the second (revHILIC × RPLC), the solvent mismatch problem is alleviated because the 1D effluent transferred to the 2D is already rich in water, which is a weak solvent for RPLC. This setup has been shown to provide impressive peak capacity, sensitivity, and complementary selectivity compared to both RPLC × RPLC and HILIC × RPLC [35].
• Retention Behavior: The retention mechanism in revHILIC (at high water content >60%) is believed to involve interactions with hydrophobic siloxane groups and charged silanols on the silica surface, which is distinct from the partitioning mechanism dominant in traditional HILIC [35].

The workflow below illustrates the decision process for implementing these integrated LC strategies.

Discussion

The integration of RPLC and HILIC is a powerful and necessary strategy for expanding metabolome coverage in untargeted studies. The choice between implementing parallel one-dimensional runs versus an online 2D-LC system depends on the specific research goals, sample complexity, and available resources.

Parallel RPLC- and HILIC-MS analyses are more straightforward to implement and are sufficient for many biomarker discovery applications. However, for extremely complex samples where maximum peak capacity is required, online comprehensive 2D-LC (e.g., HILIC × RPLC or revHILIC × RPLC) is the superior approach [35]. The recent development of revHILIC is particularly promising, as it offers a practical solution to the longstanding problem of mobile-phase incompatibility in 2D-LC, while also providing unique and complementary selectivity [35].

Throughout this workflow, the stability and sensitivity of the ESI source are critical. As demonstrated, systematic optimization of ESI parameters is not a trivial step but a foundational one that directly impacts the number of metabolites detected and the quality of the resulting data [1]. The optimized parameters should be consistently applied and monitored using QC samples throughout the analytical batch.

This application note provides a detailed framework for integrating RPLC and HILIC to achieve comprehensive metabolite coverage in untargeted metabolomics. By providing structured protocols for ESI optimization, column selection, and chromatographic method development, along with data-driven performance comparisons, this guide serves as a practical resource for researchers and scientists in drug development and biomedical research. The adoption of these integrated strategies, particularly with attention to emerging techniques like revHILIC, will significantly enhance the depth and reliability of metabolomics data, thereby strengthening subsequent biological interpretation.

Electrospray Ionization (ESI) is a cornerstone technique in untargeted metabolomics, serving as the critical interface where liquid-phase analytes are transferred into the gas phase for mass spectrometric detection. The sensitivity and coverage of an untargeted analysis are profoundly influenced by the performance of the ESI source [37]. Suboptimal ESI parameters can lead to reduced metabolome coverage, poor signal stability, and increased analytical bias, ultimately compromising biological interpretation. While many researchers recognize the need for ESI optimization, a significant gap often exists between establishing optimal parameters in controlled experiments and robustly implementing them in day-to-day routine analyses. This application note addresses this gap by providing a detailed, practical workflow for translating optimized ESI conditions into a reliable, routine untargeted metabolomics pipeline, framed within the broader context of a thesis focused on maximizing data quality in metabolomic research.

The Scientist's Toolkit: Essential Reagents and Materials

The following table details key reagents and materials essential for implementing the protocols described in this application note.

Table 1: Key Research Reagent Solutions and Materials

ESI Parameter Optimization Strategy: A DoE Approach

Implementing a robust ESI method begins with a systematic optimization of key source parameters. A one-factor-at-a-time (OFAT) approach is inefficient and fails to reveal synergistic interactions between parameters. A Design of Experiments (DoE) strategy is a far more powerful and informative alternative [38] [28].

Core Protocol: Factorial Design for ESI Optimization

The following protocol, adapted from recent studies, outlines a generalizable DoE strategy for ESI optimization on Q-TOF or Orbitrap mass spectrometers [38] [37] [28].

Objective: To systematically optimize ESI source parameters for maximum feature detection in untargeted metabolomics. Materials: Standard metabolite mixture dissolved in a representative matrix (e.g., solvent-matched to the chromatographic starting conditions); LC-MS system with tunable ESI source. Software: Statistical software capable of generating and analyzing experimental designs (e.g., Modde Pro, JMP, R).

• Factor Selection: Identify the key ESI parameters to be optimized. The table below lists common critical factors.
• Define Ranges: Establish a realistic range for each factor based on instrument manufacturer recommendations and preliminary scouting experiments.
• Experimental Design: Select an appropriate experimental design, such as a Box-Behnken or a D-optimal design. These designs efficiently explore the multi-dimensional parameter space with a reduced number of experiments.
• Randomization and Execution: Randomize the order of experiments to minimize the effects of instrument drift. For each experimental run, inject the standard metabolite mixture and acquire data in both positive and negative ionization modes.
• Response Measurement: The primary response is the number of high-quality molecular features detected. Additional responses can include the total ion chromatogram (TIC) area, signal intensity for key metabolites, and signal-to-noise ratios. Reproducibility, measured by the relative standard deviation (RSD%) of features in replicate injections, is a critical quality metric [38].
• Data Analysis and Model Building: Use Response Surface Methodology (RSM) to build a statistical model linking the factors to the responses. Analyze the model to identify significant factors and their interactions.
• Identify Robust Set Point: Determine the parameter settings that maximize the desired responses (e.g., number of features, signal intensity). The goal is to find a "robust" region in the design space where performance is optimal and less sensitive to minor fluctuations in parameter settings.

Table 2: Key ESI Parameters for Optimization in Untargeted Metabolomics

Integrated Workflow for Routine Implementation

Successfully transferring optimized parameters from a DoE study to a stable routine method requires a structured workflow encompassing quality control, sample preparation, and data preprocessing. The following diagram and subsequent sections detail this integrated process.

Diagram 1: Integrated workflow for routine ESI-MS analysis.

Protocol: Standardized Sample Preparation for Biofluids

Consistent sample preparation is non-negotiable for robust routine analysis. This protocol for biofluids (e.g., plasma, urine) ensures minimal introduction of pre-analytical variation [18].

• Thawing: Thaw frozen biofluid samples slowly on ice or in a refrigerator at 4°C.
• Aliquoting: Aliquot a precise volume (e.g., 50-100 µL) of the sample into a clean microcentrifuge tube.
• Protein Precipitation: Add a chilled organic extraction solvent (e.g., Acetonitrile:Methanol:Formic acid, 74.9:24.9:0.2, v/v/v) containing stable isotope-labeled internal standards (IS). A typical sample-to-solvent ratio is 1:3 (v/v) [18].
• Vortexing and Centrifugation: Vortex the mixture vigorously for 1-2 minutes, then centrifuge at >14,000 × g for 10-15 minutes at 4°C to pellet precipitated proteins.
• Supernatant Collection: Carefully collect the supernatant and transfer it to a clean LC-MS vial with insert for analysis. If not analyzing immediately, store the extracts at -80°C.

Protocol: Implementing Quality Control (QC) in the Sequence

QC measures are the backbone of a reliable routine workflow, ensuring that the optimized ESI parameters are performing consistently over time [18] [39].

• System Suitability Test: At the beginning of each sequence, inject a pooled QC sample (a mixture of a small amount of all study samples) or a standard metabolite mixture multiple times (e.g., 5-10 injections) until stable signal intensity and retention time are achieved for key metabolites and IS.
• Pooled QC Samples: Throughout the analytical run, inject the pooled QC sample repeatedly (e.g., after every 5-10 experimental samples). This monitors instrumental performance and is used for data normalization.
• Blank Samples: Inject solvent blanks periodically to identify and monitor carryover and background ions.

Data Preprocessing and Quality Assessment

The complex data files generated require robust preprocessing before statistical analysis [39].

• Peak Picking and Alignment: Use software (e.g., XCMS, MZmine, Compound Discoverer) to detect chromatographic peaks, align them across samples based on retention time and m/z, and group them into molecular features.
• Componentization: Combine features representing the same metabolite (e.g., adducts, isotopes) into a single compound.
• Missing Value Imputation: Apply strategies to handle missing values, which may be caused by low abundance falling below the detection limit. Common approaches include replacing them with a small value (e.g., half of the minimum positive value for that variable) or using k-nearest neighbor (KNN) imputation [39].
• Normalization: Correct for systematic drift using the data from the pooled QC samples. Common methods include probabilistic quotient normalization (PQN) or regression-based normalization using the IS [39].
• Data Transformation and Scaling: Apply log or power transformation to make the data more Gaussian-like. Follow this with scaling. Autoscaling (mean-centering followed by division by the standard deviation of each variable) is often effective as it gives all metabolites equal weight, preventing highly abundant compounds from dominating the statistical model [40].

Optimizing ESI parameters via a systematic DoE approach and embedding them within a rigorously controlled analytical workflow is paramount for generating high-quality, reproducible data in untargeted metabolomics. This application note provides a detailed, practical pipeline—from initial parameter selection and optimization to final data preprocessing—that ensures the analytical platform remains stable and sensitive for routine analysis. By adhering to this workflow, researchers can minimize technical variance, maximize biological discovery, and build a robust foundation for their metabolomics research, particularly within the demanding context of drug development and biomarker discovery.

Solving Common ESI Challenges: Signal Instability, Adduct Formation, and Ion Suppression

Electrospray Ionization (ESI) is the cornerstone of liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics, directly impacting the sensitivity, robustness, and chemical comprehensiveness of the analysis [41]. Signal loss or variation represents a critical failure point that can compromise data quality and lead to erroneous biological conclusions. Within the broader thesis of optimizing ESI parameters, this guide provides a systematic framework for diagnosing and remedying signal loss, ensuring data acquisition with high fidelity for researchers and drug development professionals.

Fundamentals of ESI Signal Generation and Common Pitfalls

The ESI process creates charged droplets from a liquid sample, leading to gas-phase ions for mass spectrometry detection. Optimal signal depends on a stable electrospray, efficient desolvation, and effective ion transmission. Signal loss can occur at any of these stages and is often manifested as a sudden drop in overall intensity, increased variability, or a complete absence of signal for specific analytes.

A significant, yet often overlooked, source of signal variation is ionization interference or signal suppression. This occurs when the presence of one compound affects the ionization efficiency of another within the ESI source [42]. In untargeted metabolomics, where samples contain thousands of compounds, co-eluting drugs and their metabolites can mutually interfere, leading to nonlinear calibration curves and systematic quantitative errors [42]. This effect is concentration-dependent and may not be evident in standard method validation protocols.

Systematic Diagnosis of Signal Loss

A structured approach to diagnosis is essential for efficiently identifying the root cause of signal loss.

Preliminary Checks and Instrumental Diagnostics

Before delving into complex diagnostics, rule out simple causes:

• Solvent and Mobile Phase: Verify correct composition, purity, and absence of contaminants or salts that could precipitate.
• Sample Integrity: Ensure samples are properly prepared, free of particulates, and compatible with the LC solvent.
• LC System: Check for leaks, pressure fluctuations, or blockages in the chromatographic system.
• MS Instrument Status: Review system pressure gauges, foreline pressures, and detector voltages against established operational ranges.

Profiling Signal Loss with a Dilution Series

A defined nontargeted test using a sample dilution series is an efficient strategy to profile instrumental performance and diagnose issues related to ionization suppression or in-source fragmentation [41].

Table 1: Key Diagnostic Analyses and Their Interpretation

The following diagram illustrates the logical workflow for diagnosing the root cause of signal loss.

The Scientist's Toolkit: Essential Research Reagents and Materials

Table 2: Key Research Reagent Solutions for Signal Loss Investigation

Experimental Protocols for Diagnosis and Remediation

Protocol: Dilution Series for Ionization Performance Evaluation

This protocol assesses ESI source performance and identifies nonlinear signal responses [41].

• Sample Preparation: Prepare a pooled QC sample from a subset of your biospecimens [43]. Create a serial dilution series in a suitable solvent (e.g., 1:1, 1:4, 1:16, ..., 1:16,384). A minimum of eight dilution levels is recommended.
• LC-MS Analysis: Inject the dilution series in randomized order. Use standard LC-MS methods for untargeted profiling.
• Data Processing: Use software like XCMS or MZmine for feature detection (peak picking) on each file [34] [43]. Generate a table of features with m/z, retention time, and intensity.
• Data Analysis:
  • For each feature, plot the log-intensity against the log-dilution factor.
  • Visually inspect curves for nonlinearity.
  • Calculate robust fold-changes across dilution levels to identify features with aberrant responses.

Protocol: Assessing and Mitigating Ionization Interference

This protocol diagnoses and resolves signal suppression between co-eluting compounds [42].

• Experimental Setup: Prepare solutions containing suspected interfering compounds (e.g., a drug and its metabolite) both individually and in mixtures at varying concentration ratios.
• LC-MS Analysis: Analyze all solutions with the relevant LC-ESI-MS method.
• Data Analysis: Quantify the signal intensity for each analyte in individual and mixed solutions. A significant reduction in signal in the mixture compared to the individual solution indicates ionization interference.
• Remediation Strategies:
  • Chromatographic Separation: Optimize the LC method to increase the retention time difference between the interfering compounds.
  • Sample Dilution: Dilute the sample to reduce the absolute concentration of interferents [42].
  • Stable Isotope Internal Standards: Use a stable labeled isotope internal standard for each analyte to correct for suppression effects [42].

Remediation Strategies for Signal Loss

Optimizing ESI Source Parameters

Based on diagnostic outcomes, adjust key parameters:

• For General Sensitivity Loss: Optimize source temperature and gas flows (nebulizer, drying gas) for improved desolvation.
• For Increased In-Source Fragmentation: Reduce source fragmentor voltage or capillary temperature to preserve the molecular ion.
• To Minimize Interference: Ensure robust spray stability by properly positioning the ESI needle and optimizing voltages.

Computational and Informatic Corrections

For issues that cannot be fully resolved instrumentally, computational methods provide a solution.

• Batch Effect Correction: In large-scale studies, signal drift is common. A random forest-based approach has been shown to outperform other methods for batch correction [43].
• Data Normalization: Systematic normalization of metabolomics data is required before statistical analysis. Techniques include log transformation and variance stabilizing normalization to reduce technical variation [44] [45].

The following workflow integrates both analytical and computational remediation strategies into a cohesive untargeted metabolomics pipeline.

Signal loss in ESI-based untargeted metabolomics is a multifactorial challenge that demands a systematic diagnostic approach. By integrating practical experimental protocols, such as dilution series and interference tests, with robust remediation strategies, including both instrumental optimization and computational corrections, researchers can significantly enhance data quality. This systematic guide, framed within the overarching goal of ESI parameter optimization, empowers scientists to proactively diagnose issues, implement effective solutions, and generate reliable, high-fidelity metabolomic data essential for advanced research and drug development.

Minimizing Metal Adduct Formation and Contamination from Samples and Solvents

In untargeted metabolomics, the comprehensive profiling of small molecules is often compromised by the formation of metal adducts during electrospray ionization-mass spectrometry (ESI-MS) analysis. These adducts, primarily with sodium (Na+), potassium (K+), and other metal ions originating from samples and solvents, complicate mass spectra, suppress target analyte signals, and reduce the reliability of metabolite annotation and quantification [46]. The challenge is particularly pronounced in hydrophilic interaction liquid chromatography (HILIC), where inorganic ions are retained and can co-elute with analytes of interest, leading to unpredictable ion suppression and adduct formation that varies with the concentration of metal ions in the sample [46]. For research aimed at biomarker discovery and pathological understanding, such analytical biases are unacceptable. This application note provides detailed protocols for minimizing metal adduct formation, framed within the broader objective of optimizing ESI parameters for robust untargeted metabolomics.

Core Principles and Challenges

Metal adduct formation occurs when analyte molecules (M) form non-covalent complexes with metal cations (e.g., Na+, K+) instead of the typical protons, resulting in ions such as [M+Na]+ or [M+K]+ in positive ionization mode [46]. While sometimes exploited to enhance sensitivity for specific compounds, this phenomenon is generally detrimental in untargeted profiling for several reasons:

• Spectral Complexity: A single metabolite can generate multiple signals (e.g., [M+H]+, [M+Na]+, [M+K]+), complicating spectral interpretation and inflating the number of molecular features [46].
• Reduced Sensitivity: The total ion current for an analyte is distributed across several adduct species, which can lead to a significant loss in the sensitivity of the primary [M+H]+ or [M-H]- ion [46].
• Poor Reproducibility: Adduct formation is highly sensitive to the metal ion concentration in the sample. Variations in sample matrix between individuals or batches can lead to inconsistent adduct patterns, compromising reproducibility and quantitative accuracy [46].

The primary sources of metal ion contamination include biological matrices themselves (e.g., plasma, urine), impurities in solvents and chemicals, and even LC system components [46]. Therefore, a multi-pronged strategy addressing sample preparation, mobile phase composition, and instrument parameters is essential for effective mitigation.

Experimental Protocols

Protocol 1: Optimized Sample Preparation for Plasma Metabolomics

This protocol is designed to efficiently extract polar metabolites while incorporating steps to manage metal ion interference.

1. Materials and Reagents

• Internal Standard Extraction Solution: Acetonitrile/Methanol/Formic acid (74.9:24.9:0.2, v/v/v) containing stable isotope-labeled internal standards (e.g., L-Phenylalanine-d8 at 0.1 µg/mL and L-Valine-d8 at 0.2 µg/mL) [18].
• Solvents: LC-MS grade water, acetonitrile, and methanol.
• Equipment: Microcentrifuge, vortex mixer, nitrogen evaporator, analytical balance.

2. Procedure

• Aliquot: Transfer 100 µL of plasma sample into a 1.5 mL microcentrifuge tube [47].
• Precipitate and Extract: Add 395 µL of ice-cold methanol and 5 µL of a stable isotope-labeled internal standard mix to the sample. The internal standards are crucial for monitoring extraction efficiency and compensating for ion suppression [47].
• Vortex and Incubate: Vortex the mixture vigorously for 30-60 seconds and incubate on ice for 30 minutes to ensure complete protein precipitation and metabolite extraction [47].
• Centrifuge: Centrifuge at 18,000×g at 4°C for 15 minutes to pellet precipitated proteins [47].
• Collect Supernatant: Carefully transfer 320 µL of the supernatant (containing the extracted metabolites) to a new microcentrifuge tube.
• Dry and Reconstitute: Evaporate the supernatant to dryness under a gentle stream of nitrogen gas at room temperature. Reconstitute the dried metabolite extract in 50 µL methanol and 50 µL LC-MS grade water. Vortex thoroughly to ensure complete dissolution before LC-MS analysis [47].

Protocol 2: Configuration of Mobile Phase to Control Adducts

The choice of mobile phase modifiers is one of the most effective levers for controlling adduct formation.

1. Materials and Reagents

• Mobile Phase A (Aqueous): 10 mM ammonium formate with 0.125% formic acid in LC-MS grade water. This combination has been shown to provide an optimal balance for signal intensity and isomer separation in HILIC-MS [48].
• Mobile Phase B (Organic): 0.1% formic acid in LC-MS grade acetonitrile [18].
• HILIC Column: ACQUITY UPLC BEH Amide column (50 mm × 2.1 mm i.d., 1.7 µm) or equivalent [48].

2. Procedure

• Preparation: Prepare fresh mobile phases daily using high-purity solvents and additives. Filter through 0.2 µm membranes and degas by sonication.
• LC Configuration:
  • Column Temperature: 40°C
  • Flow Rate: 0.4 mL/min
  • Injection Volume: 2-5 µL
  • Gradient: Initiate with a high organic phase (e.g., 95% B) and ramp to a higher aqueous phase over 8.5-15 minutes for fast analysis [48].
• Principle of Action: The ammonium ions (NH4+) from the mobile phase modifier compete with Na+ and K+ for adduct formation sites on the analytes. Because ammonium adducts are often less stable and can be easily promoted to yield [M+H]+ in the source, this strategy effectively reduces the prevalence of persistent sodium and potassium adducts, leading to simpler spectra and more reproducible ionization [46] [48].

The Scientist's Toolkit

Table 1: Essential Research Reagents for Minimizing Metal Adducts

Data Presentation and Optimization Guidelines

Impact of Mobile Phase Modifiers

Systematic evaluation of mobile phase modifiers is critical for platform optimization. The data below demonstrate how modifier choice directly impacts analytical performance.

Table 2: Performance of Different Mobile Phase Modifiers in HILIC-ESI(+)-MS for Polar Metabolites [48]

Optimization of ESI Source Parameters

Beyond chromatography, the ESI source itself must be tuned to stabilize the spray and maximize the efficiency of protonated ion formation.

Table 3: Key ESI Source Parameters for Minimizing Adduct Formation [10]

Workflow and Strategic Application

The following diagram synthesizes the strategic workflow for minimizing metal adducts, integrating sample preparation, LC-MS configuration, and data processing considerations.

Concluding Remarks

Minimizing metal adduct formation is not a single-step fix but a holistic strategy that spans the entire analytical workflow. By implementing the protocols outlined herein—rigorous sample preparation with quality controls, the use of ammonium-based mobile phase modifiers, and systematic optimization of ESI source parameters—researchers can significantly reduce analytical variability and enhance the quality of their untargeted metabolomics data. This approach ensures that the resulting biological insights, particularly in drug development and clinical research, are built upon a robust and reliable analytical foundation.

Strategies to Overcome Ion Suppression in Complex Biological Matrices

Ion suppression represents a significant challenge in mass spectrometry (MS)-based untargeted metabolomics, dramatically decreasing measurement accuracy, precision, and sensitivity [49]. This phenomenon occurs when the ionization of target analytes is interfered with by co-eluting matrix components, leading to suppressed or enhanced signals and ultimately compromising data quality and reproducibility [49]. Within the broader context of optimizing electrospray ionization (ESI) parameters for untargeted metabolomics research, overcoming ion suppression is paramount for obtaining biologically relevant results. This application note details practical strategies and protocols to identify, quantify, and correct for ion suppression effects, enabling more robust metabolomic profiling in complex biological matrices such as plasma, urine, and cell cultures.

Understanding Ion Suppression and Its Impact

Ion suppression stems from matrix effects where co-eluting compounds compete for charge or disrupt droplet formation during the ESI process [49]. The mechanisms are influenced by multiple factors including ionization source type, mobile phase composition, gas temperature, and physicochemical properties of analytes and matrix components [49]. The consequences can be severe, with studies reporting ion suppression ranging from 1% to >90% for detected metabolites, with coefficients of variation ranging from 1% to 20% [49].

The extent of ion suppression varies significantly across different analytical conditions. As illustrated in Table 1, the degree of suppression is influenced by chromatographic system, ionization mode, and source cleanliness [49].

Table 1: Magnitude of Ion Suppression Under Different Analytical Conditions

Strategic Approaches to Overcome Ion Suppression

Sample Preparation and Cleanup

Effective sample preparation is the first line of defense against ion suppression. The key objectives are to remove interfering matrix components while maintaining the integrity of the metabolome.

Table 2: Sample Preparation Methods for Reducing Ion Suppression

Protocol: Optimized Sample Preparation for Multiple Biofluids [51]

• Urine Processing:

  • Combine urine with cold methanol in a 1:8 ratio (urine:methanol)
  • Vortex for 30 seconds followed by incubation at -20°C for 1 hour
  • Centrifuge at 12,000-16,000 × g for 15 minutes at 4°C
  • Collect supernatant for analysis

• Plasma/Serum Processing:

  • Utilize monophasic extraction with methanol (1:2 or 1:8 sample:methanol ratios)
  • As an alternative, employ biphasic extraction with methanol/chloroform (1:1:1 or 1:1.5:2.5 sample:methanol:chloroform)
  • Vortex thoroughly and centrifuge to separate phases
  • Collect the appropriate phase for analysis

• Cell Culture Processing:

  • Remove 25 μL of spent media from each well
  • Add 45 μL of cold acetonitrile to 15 μL of spent media
  • Centrifuge for 10 minutes at 3,000 × g
  • Transfer supernatant for LC-MS analysis

The IROA TruQuant Workflow for Ion Suppression Correction

The Isotopic Ratio Outlier Analysis (IROA) TruQuant Workflow represents a groundbreaking approach to directly measure and correct for ion suppression effects using stable isotope-labeled internal standards [49].

IROA Workflow for Ion Suppression Correction

Protocol: IROA TruQuant Implementation [49]

• Internal Standard Preparation:

  • Spike all samples with the IROA Internal Standard (IROA-IS) library at constant concentrations
  • Prepare IROA Long-Term Reference Standard (IROA-LTRS) as a 1:1 mixture of chemically equivalent standards at 95% ( ^{13}C ) and 5% ( ^{13}C )

• LC-MS Analysis:

  • Analyze samples using your preferred chromatographic system (IC, RPLC, or HILIC) in both positive and negative ionization modes
  • Maintain consistent ESI source conditions throughout the analysis

• Data Processing with ClusterFinder Software:

  • Import raw LC-MS data into ClusterFinder software (version 4.2.21 or higher)
  • Identify metabolites based on their unique IROA isotopolog ladder pattern
  • Apply the suppression correction algorithm using Equation 1:

  [ \text{AUC-12C}{\text{suppression-corrected}} = \frac{\text{AUC-12C}{\text{observed}} \times \text{AUC-13C}{\text{expected}}}{\text{AUC-13C}{\text{observed}}} ]

  • Perform Dual MSTUS normalization for final data output

• Quality Assessment:

  • Verify the signature IROA peak pattern for each metabolite
  • Ensure proper distinction between biological signals and artifacts
  • Validate linearity of suppression-corrected signals across sample dilutions

ESI Source Parameter Optimization

Optimizing ESI parameters is crucial for minimizing ion suppression effects. Systematic evaluation of source conditions can significantly improve ionization efficiency and reduce matrix effects.

Protocol: ESI Source Optimization for Orbitrap Mass Spectrometers [1]

• Needle Position Optimization:

  • Adjust the ESI needle position in three-dimensional space
  • Maximize signal intensity for a standard metabolite mixture
  • Find the optimal position that provides stable spray and maximum sensitivity

• Source Parameter Calibration:

  • Optimize spray voltage: Typically 3.5-3.8 kV for positive mode and 2.8-3.2 kV for negative mode
  • Set ion transfer tube temperature: 300-350°C
  • Adjust vaporizer temperature: 250-400°C
  • Optimize sheath gas (25-50 arb) and auxiliary gas (5-25 arb) settings

• Comprehensive Evaluation:

  • Test the optimized parameters across different metabolite classes
  • Validate using quality control samples and reference materials (e.g., NIST SRM 1950)
  • Verify performance across multiple chromatographic systems (RPLC and HILIC)

Chromatographic Method Selection

The choice of chromatographic method significantly influences ion suppression effects. Employing complementary separation techniques can reduce co-elution of interfering compounds.

Protocol: Multi-Platform Chromatographic Analysis [51]

• Reversed-Phase Liquid Chromatography (RPLC):

  • Use C18 columns (e.g., Thermo Hypersil GOLD aq C18, 100 × 2.1 mm, 1.9 µm)
  • Employ gradient elution with water (0.1% formic acid) and acetonitrile (0.1% formic acid)
  • Program: 0-10 min (10%-100% B), 10-17 min (100% B), 17-19 min (100-10% B), 19-20 min (10% B)
  • Flow rate: 0.3 mL/min, column temperature: ambient

• Hydrophilic Interaction Liquid Chromatography (HILIC):

  • Use zwitterionic stationary phases (e.g., ACQUITY UPLC HSS T3, 100 × 2.1 mm × 1.8 µm)
  • Employ gradient elution with aqueous solvent (0.05% formic acid) and acetonitrile
  • Program: 0-1 min (95% A), 1-12 min (95-5% A), 12-13.5 min (5% A), 13.5-13.6 min (5-95% A), 13.6-16 min (95% A)
  • Flow rate: 0.3 mL/min, column temperature: 40°C

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Reagents and Materials for Ion Suppression Management

Effective management of ion suppression is essential for generating high-quality, reproducible data in untargeted metabolomics. The strategies outlined in this application note—comprehensive sample preparation, implementation of the IROA TruQuant Workflow, systematic ESI parameter optimization, and strategic chromatographic method selection—provide researchers with a robust framework to overcome matrix effects. By adopting these protocols, scientists can significantly improve measurement accuracy, precision, and sensitivity, thereby enhancing the reliability of their metabolomic findings in drug development and clinical research applications.

Correcting for Instrumental Drift and Batch Effects Using Quality Control Samples

In untargeted metabolomics, which comprehensively analyzes small molecules in biological systems, liquid chromatography-mass spectrometry (LC-MS) has become a preeminent analytical technique due to its sensitivity and wide dynamic range [7] [1]. However, the reliability of data from large-scale studies is compromised by technical variations, broadly categorized as instrumental drift and batch effects [52] [53]. Instrumental drift refers to the gradual shift in instrument response over time within a single batch, manifesting as fluctuations in retention time (RT) and signal intensity [52]. Batch effects are systematic differences in measured signals when a study is divided into multiple analytical batches, caused by factors such as instrument maintenance, column replacement, or mobile phase preparation [52] [54].

These technical variations can introduce unwanted bias that obscures true biological signals, leading to reduced statistical power and potentially false discoveries [52] [53]. The implementation of a rigorous quality control (QC) strategy using intrastudy QC samples is, therefore, critical to monitor and correct for these non-biological variations [52] [55]. This protocol details the application of QC samples for effective drift and batch effect correction, framed within a research project focused on optimizing electrospray ionization (ESI) parameters for untargeted metabolomics.

Materials and Methods

Research Reagent Solutions and Essential Materials

The following table lists key reagents and materials essential for implementing a robust QC protocol in untargeted metabolomics.

Table 1: Essential Research Reagents and Materials for QC-Based Metabolomics

Experimental Workflow for QC-Based Correction

The following diagram illustrates the comprehensive workflow for correcting instrumental drift and batch effects using QC samples, integrating steps from sample preparation to normalized data output.

Diagram 1: Experimental workflow for QC-based correction.

Protocol for Quality Control Sample Preparation and Analysis

Preparation of Intrastudy Pooled QC Samples

The most effective QC samples are intrastudy pooled QCs, created by combining equal-volume aliquots from all biological samples in the study [52] [55]. This ensures the QC sample closely mirrors the overall metabolite composition and matrix of the entire sample set.

• Aliquot Pooling: After processing all individual study samples (e.g., through protein precipitation), take a small, equal volume (e.g., 10-20 µL) from each sample extract and combine them into a single vial.
• Homogenization: Vortex the pooled mixture thoroughly to ensure homogeneity.
• Aliquoting: Dispense the homogenized pooled QC into multiple low-volume injection vials to avoid repeated freeze-thaw cycles.
• Storage: Store the QC aliquots at -80°C until analysis [52].

LC-MS Acquisition Sequence Design

A properly designed acquisition sequence is vital for tracking technical variation over time. The sequence should start with a system conditioning and equilibration phase, followed by the analytical run.

• System Conditioning: Inject the pooled QC sample 4-8 times at the beginning of the sequence. The data from these initial injections are often unstable and are typically discarded, serving solely to condition the column and equilibrate the LC-MS system [52] [53].
• Sample Analysis Order: Randomize the injection order of biological samples to avoid confounding technical drift with biological group differences [52].
• QC Spacing: Analyze a pooled QC sample regularly throughout the sequence. A common practice is to inject a QC after every 5-10 study samples [53]. This frequency provides sufficient data points to model the temporal drift of metabolite features.

Table 2: Quantitative Evaluation Metrics for Data Quality Assessment using QC Samples

Results and Data Interpretation

Visualizing and Diagnosing Technical Variation

The first step after data pre-processing is to diagnose the presence and extent of technical variation. This is effectively done using unsupervised multivariate statistics.

• PCA of Raw Data: Perform PCA on the raw data matrix, including both study samples and QC samples, coloring the samples by injection order or batch.
• Interpretation: In a stable system, all QC samples should cluster tightly together in the PCA scores plot. A trajectory or spread of QCs along a principal component (often PC1) is a clear indicator of instrumental drift over time [53]. Similarly, separation of samples by batch in the PCA plot indicates a strong batch effect that requires correction.

Correcting Signal Intensity Drift

Several algorithms can correct for signal intensity drift modeled from the pooled QC samples. The following diagram outlines the core logic of this correction process.

Diagram 2: Logic of signal intensity drift correction.

Table 3: Comparison of Batch-Effect Correction Methods Based on QC Samples

Addressing Retention Time Drift

While intensity drift is a primary concern, retention time (RT) shifts can also occur, complicating peak alignment across samples [52].

• Detection: Plot the RT of identified features in QC samples against injection order. A consistent drift indicates an RT shift.
• Correction: Most modern data pre-processing software (e.g., MZmine, XCMS) include algorithms for RT alignment. These algorithms typically use a set of "landmark" features detected across all samples to warpthe RT axis and align peaks. The consistent pooled QC samples provide an ideal basis for this alignment [52] [54].

Application Notes and Protocol Integration

Integration with ESI Parameter Optimization

The stability of the ESI source is paramount for minimizing baseline technical variation. Optimization of parameters like spray voltage, gas temperatures, and gas flows should be performed as a foundational step prior to large-scale sample analysis [1] [57]. A well-tuned source reduces the magnitude of drift, making subsequent computational corrections more effective and reliable. The QC protocols described here serve a dual purpose: they are essential for correcting drift in a production run, and they are also a powerful diagnostic tool during method development to assess the real-world stability of different ESI source settings.

Troubleshooting and Best Practices

• QC Clustering Failure: If QC samples do not cluster in PCA after correction, investigate potential causes such as sample carry-over, source contamination, or degradation of the QC pool itself [53].
• Limited Sample Volume: If creating a comprehensive pooled QC from all study samples is not feasible, a surrogate QC from a representative subset of samples or a commercially available reference material (e.g., NIST SRM 1950) can be used, though with the understanding that it may not perfectly represent the study's metabolic matrix [52] [55].
• Data Filtering: After correction, apply data filtering. Typically, metabolic features with an RSD% in the pooled QCs greater than 20-30% are considered noisy and removed from further analysis [56] [53].

Instrumental drift and batch effects are inevitable challenges in untargeted metabolomics. The systematic use of intrastudy pooled quality control samples, integrated within a carefully planned analytical workflow, provides a robust framework to monitor, diagnose, and correct these technical variations. By implementing the detailed protocols for QC preparation, acquisition sequencing, and application of correction algorithms like TIGER or QC-RSC, researchers can significantly enhance the quality and reliability of their data, ensuring that biological insights are derived from true metabolic phenotypes rather than analytical artifacts.

Electrospray ionization (ESI) stands as a cornerstone technique in untargeted metabolomics, enabling the detection of thousands of metabolites. However, the accuracy and depth of metabolome coverage are heavily influenced by multiple factors, including matrix effects (MEs) and the efficiency of ion generation. MEs, caused by co-eluting compounds that suppress or enhance analyte ionization, represent a significant challenge, particularly in complex biological samples like plasma and serum [58]. Simultaneously, traditional ESI sources are often tuned to minimize in-source fragmentation (ISF), potentially overlooking valuable fragmentary data that can aid in metabolite annotation [59]. This application note details two advanced strategies—Post-column Infusion (PCI) for combating MEs and Enhanced In-Source Fragmentation (EISF) for improving molecular identification—framed within the critical context of optimizing ESI parameters for robust untargeted metabolomics.

Post-column Infusion for Matrix Effect Compensation

Principle and Workflow

Matrix effects can severely compromise quantification accuracy in liquid chromatography–tandem mass spectrometry (LC–MS/MS). Post-column infusion serves as a powerful diagnostic tool to identify chromatographic regions affected by ion suppression or enhancement. More recently, it has been innovatively adapted as a quantification technique itself, particularly when stable isotope-labeled internal standards (SIL-IS) are unavailable or cost-prohibitive [60].

The fundamental principle of the diagnostic PCI involves the continuous infusion of a target analyte, or a labeled analog, into the LC eluent post-column via a T-piece. As a blank matrix extract is injected and separated, the constant signal of the infused analyte is monitored. A dip in the signal indicates ion suppression, while a signal increase signifies ion enhancement at that specific retention time [58]. This provides a qualitative map of MEs across the chromatographic run.

A novel quantification approach using PCI was demonstrated for the immunosuppressant tacrolimus in whole blood. In this method, the analyte itself (tacrolimus) is continuously infused, acting as its own internal standard. A second multiple reaction monitoring (MRM) transition for the same analyte, with a slightly different mass, is monitored to represent the infused "IS" signal. The actual area of the infused standard is calculated by subtracting the peak area of the tacrolimus originating from the injected sample from the total area of the "IS" channel over a fixed elution window. The response, derived from the ratio of the sample peak area to the calculated infused standard area, is then used to build a calibration curve [60].

The following diagram illustrates the core logical relationship and workflow of the PCI technique:

Experimental Protocol: PCI for Quantification

The following protocol is adapted from a proof-of-concept study quantifying tacrolimus in whole blood using PCI [60].

1. LC-MS/MS Setup:

• Instrumentation: An Agilent InfinityLab LC-MSD system or equivalent LC-MS/MS system.
• Chromatography: Utilize a suitable C18 column (e.g., ZORBAX 300 SB-C18, 0.5 × 150 mm, 5 μm) maintained at 60°C. The mobile phase consists of (A) 2 mM ammonium acetate/0.1% formic acid in water and (B) 2 mM ammonium acetate/0.1% formic acid in methanol. Employ a gradient elution from 30% B to 95% B over 5 minutes.
• Mass Spectrometry: Operate in positive ESI mode with multiple reaction monitoring (MRM). For tacrolimus, monitor two transitions: one for the native analyte (821.7000 > 768.7000) and a second, slightly offset transition for the infused "IS" (821.7001 > 768.7001).

2. Post-column Infusion Setup:

• Connect a syringe pump capable of delivering a constant flow (e.g., 10 μL/min) to a T-piece installed between the column outlet and the MS inlet.
• Prepare a solution of the pure analyte (tacrolimus) in a suitable solvent (e.g., methanol/water mixture) and load it into the syringe. The concentration should be adjusted to produce a stable baseline signal.

3. Data Acquisition and Calculation:

• Inject the calibration standards, quality controls, and unknown samples.
• For each run, integrate the peak area for the native tacrolimus (grey area).
• Manually integrate the signal for the "tacrolimus-IS" channel over a fixed elution time window (e.g., 0.9 to 2.0 min, red hatched area).
• Calculate the actual area of the infused IS (light red area) using the formula: Area IS = Area tacrolimus-IS (total) - Area tacrolimus (from sample)
• Calculate the response for each calibrator and sample: Response = Area tacrolimus (from sample) / Area IS (infused)
• Construct a calibration curve by plotting the Response against the nominal concentration of the calibrators.

4. Validation:

• The method should be validated according to relevant guidelines (e.g., European Medicines Agency guideline on bioanalytical method validation). The cited study achieved imprecisions and inaccuracies with coefficient of variation and relative bias below 15%, and a strong correlation (r = 0.9532) with conventional IS quantification [60].

Research Reagent Solutions for PCI

Table 1: Key Reagents and Materials for Post-column Infusion Quantification

Enhanced In-Source Fragmentation and Ion Source Optimization

Principle of Enhanced In-Source Fragmentation

Traditionally, ESI source parameters are configured to minimize in-source fragmentation (ISF) to preserve the molecular ion. However, recent research demonstrates that strategically enhancing ISF can generate pseudo-MS/MS spectra within a single MS¹ full-scan experiment. This approach, termed Enhanced In-Source Fragmentation Annotation (eISA), involves tuning source-induced dissociation parameters to produce fragment patterns comparable to higher-energy collision-induced dissociation (CID) spectra found in libraries like METLIN, without critically compromising precursor ion intensity [59].

Optimization of the ion source is critical for this technique and for untargeted metabolomics in general. Key parameters include the ESI needle position, spray voltage, vaporization and ion transfer tube temperatures, and sheath and auxiliary gas flows. For instance, one study found that positioning the ESI needle at the farthest point on the Z-axis and the closest tested position on the Y-axis relative to the MS inlet yielded the best signal reproducibility and number of metabolite annotations. Optimal values included a spray voltage between 2.5 and 3.5 kV (positive) and 2.5–3.0 kV (negative), vaporization and ion transfer tube temperatures between 250 and 350 °C, 30–50 arbitrary units of sheath gas, and at least 10 units of auxiliary gas [57] [1].

The following diagram illustrates the strategic decision-making process and outcomes when optimizing the ion source, weighing the trade-off between minimizing and enhancing in-source fragmentation.

Experimental Protocol: eISA Method Development

This protocol outlines the steps for developing an Enhanced In-Source Fragmentation Annotation (eISA) method on a QTOF mass spectrometer, based on the validation study of 50 endogenous metabolites [59].

1. Standard and Sample Preparation:

• Prepare a mixture of known metabolite standards (e.g., 30 μM) covering a range of chemical classes.
• For biological application, extract metabolites from your sample of interest (e.g., macrophage cell pellet extracted with 1 mL of ice-cold acetonitrile).

2. LC-MS Analysis with Parameter Ramping:

• Chromatography: Employ both reversed-phase (e.g., ZORBAX 300 SB-C18 column with positive ESI) and HILIC (e.g., Luna NH2 column with negative ESI) methods to cover a broad metabolome.
• Mass Spectrometry: Use a high-resolution MS like a Bruker impact II QTOF. In the source, identify the parameter controlling collision energy in the interface (e.g., transfer isCID energy). While keeping other parameters (e.g., spray voltage, temperatures) at their previously optimized values, ramp this fragmentation energy (e.g., from 0 eV to 100 eV in 10 eV increments). Acquire data in full-scan mode.

3. Data Analysis and Optimal Condition Selection:

• For each metabolite standard in the mixture, analyze the data across the energy ramp. Focus on three factors:
  • Precursor Ion Intensity: Monitor the decrease in the precursor ion's intensity.
  • Fragment Match: Identify the number of in-source fragments whose m/z matches fragments in a reference MS/MS library (e.g., METLIN 20 eV spectra).
  • Spectral Similarity: Assess the relative intensity correlation between the experimental in-source fragments and the library spectra (e.g., using a match factor).
• Select the optimal energy that generates a fragmentation pattern consistent with the library for >90% of molecules, while keeping the median precursor ion intensity loss to ≤10%. In the cited study, this was 30 eV for positive mode and 40 eV for negative mode [59].

4. Method Validation and Application:

• Compare the fragmentation patterns and precursor sensitivity of eISA against conventional data-dependent acquisition (DDA) and data-independent acquisition (DIA). eISA has been shown to provide comparable fragmentation patterns with significantly higher precursor ion peak intensities [59].
• Apply the optimized eISA conditions to the analysis of biological samples. Putative identifications can be made by matching the in-source fragments in the MS¹ data to low-energy CID spectra in libraries.

Quantitative Comparison of Fragmentation Techniques

Table 2: Performance Comparison of eISA versus Traditional MS/MS Acquisition Modes

Integrated Application in Metabolomics Workflow

The true power of these advanced techniques is realized when they are integrated into a cohesive metabolomics workflow. An optimized ESI source is the foundation, ensuring stable spray, efficient desolvation, and controlled fragmentation, which in turn enhances the reliability of both PCI and eISA methods. For instance, comprehensive metabolomic profiling, such as the study that identified novel metabolite associations in Parkinson's disease, often relies on a dual-column (HILIC and C18) LC-HRMS setup with carefully optimized ESI conditions to maximize metabolome coverage [61].

Furthermore, the combination of different data streams is a growing trend. As demonstrated in studies that integrate direct-infusion ESI-MS (DI-ESI-MS) with NMR data, using multiblock multivariate statistics like multiblock PCA (MB-PCA) and PLS (MB-PLS), the correlation structures between different analytical blocks can be leveraged to provide a more robust biological interpretation [62]. Similarly, the pseudo-MS/MS data from eISA and the quantitative rigor from PCI can be combined with conventional MS¹ data to build a more complete and confident picture of the metabolome.

Optimizing ESI parameters extends beyond achieving a stable spray; it involves strategically manipulating the ion source to address specific analytical challenges. Post-column infusion has evolved from a diagnostic tool for matrix effects into a novel quantification strategy, offering a viable path forward when internal standards are lacking. Conversely, enhanced in-source fragmentation challenges the conventional wisdom of minimizing ISF, instead leveraging it to acquire informative fragmentation data within a simple MS¹ experiment, thereby boosting molecular identification confidence. By adopting and integrating these advanced techniques, researchers can significantly improve the quality, depth, and reliability of their untargeted metabolomics analyses.

Evaluating ESI Performance: QC Strategies, Data Acquisition Modes, and Annotation Tools

In the context of optimizing Electrospray Ionization (ESI) parameters for untargeted metabolomics, implementing robust quality control (QC) is not merely an optional step but a fundamental requirement for generating credible, reproducible data. Untargeted metabolomics serves as a powerful, discovery-oriented approach for identifying unknown small molecules (approximately ≤2000 Da) from highly complex biological mixtures, a situation frequently encountered in biomedical and environmental research [7]. The analytical challenge is substantial; unlike nucleic acid or protein sequencing, metabolomics deals with structurally diverse molecules lacking common building blocks, making confident identification dependent on correlating fragmentation data with retention time [7]. Within this framework, quality assurance (QA) and quality control (QC) are two integral quality management processes critical for success in any high-throughput analytical chemistry laboratory [63]. QA encompasses all planned and systematic activities implemented before sample collection to provide confidence that the analytical process will fulfill predetermined quality requirements. In contrast, QC describes the operational techniques and activities used to measure and report these quality requirements during and after data acquisition [63]. For researchers fine-tuning ESI source parameters, a robust QC system provides the essential feedback loop to distinguish between genuine metabolic changes and analytical artifacts introduced by instrument instability.

Core Concepts: System Suitability and Intrastudy QC Samples

The QC framework for untargeted metabolomics relies on two complementary sets of samples: those used to validate instrument performance before data acquisition (system suitability) and those used to monitor stability throughout the analytical batch (intrastudy QC) [63].

System suitability testing is performed prior to the analysis of any biological samples to minimize the loss of potentially irreplaceable biological material due to analytical failure. This process qualifies the instrument as "fit-for-purpose" and involves the analysis of a "blank" gradient to check for solvent or column contamination, followed by a solution containing a small number of authentic chemical standards (typically five to ten analytes) dissolved in a chromatographically suitable diluent [63]. These analytes should be distributed as fully as possible across the mass-to-charge (m/z) and retention time ranges to assess the entire analytical window. Acceptance criteria must be pre-defined; an example includes a mass-to-charge (m/z) error of ≤ 5 ppm compared to the theoretical mass, a retention time error of < 2%, a peak area within ± 10% of a predefined acceptable value, and a symmetrical peak shape with no evidence of splitting [63].

Intrastudy QC Samples are analyzed intermittently throughout the sequence of biological samples. Key types include:

• Isotopically-labelled Internal Standards: Added to each sample to assess system stability for every individual analysis [63].
• Pooled QC Samples: Created by combining small aliquots of all study samples, these are used to condition the analytical platform, perform intra-study reproducibility measurements, and mathematically correct for systematic errors [63].
• Long-Term Reference (LTR) QC Samples and Standard Reference Materials (SRMs): Applied for inter-study and inter-laboratory assessment of data quality, providing a benchmark for performance over time and across different locations [63].

Table 1: Types and Functions of Quality Control Samples in Untargeted Metabolomics

Experimental Protocols

Basic Protocol 1: Preparing System Suitability Test Samples

This protocol ensures the LC-ESI-MS/MS system is analytically stable and sensitive before running valuable study samples.

Materials:

• Authentic chemical standards (5-10 compounds covering a range of m/z and polarities)
• Appropriate solvents (e.g., 0.1% formic acid in H₂O, acetonitrile)
• Volumetric flasks and pipettes
• LC vials

Procedure:

• Select Standards: Choose a mixture of 5 to 10 metabolite standards that are chemically diverse and expected to elute across the entire chromatographic gradient. Common choices include compounds like caffeine, reserpine, or MRM standards for positive and negative ionization modes.
• Prepare Stock Solutions: Dissolve each standard in a compatible solvent to create concentrated stock solutions.
• Create Master Mix: Combine appropriate volumes of each stock solution into a single volumetric flask. Dilute to the mark with the starting mobile phase (e.g., 0.1% formic acid in H₂O) to create a "master mix." The final concentration of each analyte should be high enough for easy detection (typically 1-10 µM) but not so high as to overload the column or mass spectrometer [7].
• Storage: Aliquot the master mix into LC vials and store at -80°C if not used immediately to prevent degradation.

Basic Protocol 2: Implementing Intrastudy QC Samples

This protocol outlines the creation and use of QC samples within an analytical sequence.

Materials:

• Aliquots of all biological samples in the study
• Isotopically-labelled internal standard mixture
• Solvents for reconstitution

Procedure:

• Prepare Pooled QC Sample: Take a small, equal-volume aliquot (e.g., 10-20 µL) from each prepared biological sample and combine them into a single tube. Mix thoroughly. This pooled sample represents the average composition of the entire study cohort [63].
• Prepare Internal Standard Solution: Create a solution of isotopically-labelled internal standards covering various metabolite classes. This solution should be added to every sample—including blanks, pooled QCs, and study samples—during the reconstitution step prior to LC-MS analysis.
• Sequence Integration: Design the analytical sequence with the following structure:
  • Inject several (e.g., 5-10) pooled QC samples at the beginning to "condition" the analytical system.
  • Analyze the system suitability sample to confirm initial performance.
  • Analyze study samples in a randomized order.
  • Inject a pooled QC sample after every 4-10 study samples to monitor stability.
  • Include process blanks periodically to track contamination.
  • Analyze the system suitability sample again at the end of the batch to check for performance drift.

Support Protocol: Data Quality Assessment and Acceptance Criteria

After data acquisition, the QC data must be evaluated to determine if the entire batch is acceptable.

Assessment Metrics:

• System Suitability: Check that the pre-defined acceptance criteria (e.g., m/z error < 5 ppm, retention time drift < 0.2 min) are met for the initial and final system suitability tests.
• Retention Time Stability: Calculate the relative standard deviation (RSD%) of the retention time for key internal standards across all pooled QC injections. A RSD of < 2% is typically desirable.
• Peak Area Stability: Calculate the RSD% of the peak areas for internal standards and a large number of endogenous features detected in the pooled QC samples. In a well-controlled system, >70% of metabolic features should have an RSD < 20-30% in the pooled QCs.
• Principal Component Analysis (PCA): Perform PCA on the entire data set. The pooled QC samples should cluster tightly in the scores plot, indicating stable analytical performance. Any significant drift or outlier in the QCs suggests a problem with the batch.

Workflow Visualization: A Comprehensive QC Strategy for Untargeted Metabolomics

The following diagram illustrates the integrated workflow for implementing robust quality control, from initial sample preparation to final data quality assessment, within an untargeted metabolomics study.

The Scientist's Toolkit: Essential Reagents and Materials

A successful untargeted metabolomics study relying on robust QC requires careful selection of reagents and materials. The following table details key research reagent solutions essential for implementing the protocols described in this application note.

Table 2: Essential Research Reagents and Materials for Metabolomics QC

The optimization of ESI parameters in untargeted metabolomics must be underpinned by a rigorous and comprehensive quality control strategy. The integrated use of system suitability tests and intrastudy QC samples, as detailed in these protocols, provides the necessary framework to ensure that the data generated is of high quality, reproducible, and biologically meaningful. By adhering to these practices, researchers can confidently distinguish true metabolic phenotypes from analytical variation, thereby maximizing the return on investment from their valuable samples and advancing the reliability of metabolomics science in drug development and biomedical research.

In liquid chromatography-mass spectrometry (LC-MS) based untargeted metabolomics, the choice of data acquisition mode is a critical determinant for success. This analysis focuses on two primary strategies: Data-Dependent Acquisition (DDA) and Data-Independent Acquisition (DIA). The performance of these modes is particularly relevant within the broader scope of optimizing electrospray ionization (ESI) parameters, as stable and efficient ionization is foundational for generating high-quality data in either acquisition mode [9].

Performance Metrics: A Quantitative Comparison

Extensive empirical studies have directly compared DDA and DIA, revealing distinct performance profiles. The table below summarizes key quantitative metrics from recent research.

Table 1: Quantitative Performance Comparison of DDA and DIA Modes

Experimental Protocols for Performance Evaluation

To obtain the performance metrics outlined above, robust and standardized experimental protocols are essential. The following section details the core methodologies for sample preparation, LC-MS analysis, and data processing.

Sample Preparation for Complex Matrices

Principle: Consistent and reproducible sample preparation is crucial to minimize analytical bias and ensure that observed differences are due to the acquisition mode and not preparation artifacts [7] [9].

Protocol: Metabolite Extraction from Plasma/Serum [7] [67]

• Protein Precipitation: Mix 100 µL of plasma with 400 µL of ice-cold methanol (LC-MS grade) to precipitate proteins.
• Solubilization and Shaking: Vortex the mixture thoroughly for 30-60 seconds to ensure complete mixing.
• Incubation: Place the sample on ice for a slow precipitation for 90 minutes.
• Centrifugation: Centrifuge at 20,000 × g for 15 minutes at 4°C to pellet the protein.
• Collection and Drying: Carefully transfer the supernatant (containing metabolites) to a new tube. Evaporate the solvent to dryness under a gentle stream of nitrogen gas.
• Reconstitution: Reconstitute the dried metabolite pellet in 100 µL of a compatible solvent (e.g., water:acetonitrile, 95:5, with 0.1% formic acid). Vortex well to ensure complete solubilization before LC-MS analysis.

Liquid Chromatography and Mass Spectrometry Configuration

Principle: High-resolution separation coupled with high-resolution mass spectrometry is the standard for untargeted metabolomics. The protocols below can be adapted for both DDA and DIA modes.

Protocol A: Reversed-Phase Liquid Chromatography (RPLC) Method [64] [67]

• Column: C18 core-shell column (e.g., 150 mm x 2.1 mm, 1.9 µm).
• Mobile Phase: A: 0.1% Formic Acid in Water; B: 0.1% Formic Acid in Acetonitrile.
• Gradient:
  • 0-2 min: Hold at 5% B.
  • 2-13 min: Ramp linearly from 5% to 100% B.
  • 13-25.5 min: Hold at 100% B.
  • 25.5-30 min: Re-equilibrate at 5% B.
• Flow Rate: 0.5 mL/min.
• Column Temperature: 40°C.
• Injection Volume: 5 µL.

Protocol B: Data-Dependent Acquisition (DDA) Method [66] [67]

• Full MS1 Scan: Acquire a full scan at a high resolution (e.g., 120,000) over the desired m/z range (e.g., 70-1050 m/z).
• Precursor Selection: Isolate the top N (e.g., N=10) most intense ions from the MS1 scan using a narrow isolation window (e.g., 1.0-1.5 m/z).
• Fragmentation: Fragment the selected precursors using normalized collision energy (e.g., 25-35% for HCD).
• MS2 Scan: Acquire the fragment ion spectra at a high resolution (e.g., 30,000).
• Dynamic Exclusion: Enable dynamic exclusion (e.g., for 15-20 seconds) to prevent repeated fragmentation of the same abundant ions.

Protocol C: Data-Independent Acquisition (DIA) Method [64] [67]

• Full MS1 Scan: Acquire a full scan at a high resolution (e.g., 120,000) over the desired m/z range.
• Windowed MS2 Scans: Divide the total m/z range into consecutive, variable-width windows. Fragment all ions within each window simultaneously.
• Fragmentation and Detection: Fragment ions using stepped collision energy and acquire MS2 spectra for each window at high resolution.

Data Processing and Analysis

Principle: The processing workflow differs significantly between DDA and DIA due to the nature of the acquired data.

• DDA Data Processing: MS2 spectra are directly matched against experimental or in-silico spectral libraries (e.g., mzCloud, GNPS) for compound identification [7] [69].
• DIA Data Processing: Specialized software (e.g., MS-DIAL, Skyline, MetaboDIA) is required to deconvolute the complex, multiplexed MS2 spectra by leveraging project-specific or public spectral libraries [67].

Workflow Visualization

The logical relationship and key decision points for choosing and implementing DDA and DIA acquisition modes are summarized in the workflow below.

The Scientist's Toolkit: Essential Research Reagents and Materials

The following table lists key materials and solutions required for conducting a comparative analysis of DDA and DIA, with special consideration for ESI optimization in untargeted metabolomics.

Table 2: Essential Research Reagents and Materials for LC-MS Metabolomics

The choice between DDA and DIA is not a matter of one being universally superior, but rather a strategic decision based on research objectives. DIA demonstrates clear advantages for large-scale quantitative studies requiring high reproducibility and comprehensive coverage of the metabolome, including low-abundance species. Conversely, DDA remains a valuable tool for applications demanding high-quality, interpretable MS/MS spectra for confident identification of unknown compounds, particularly when building spectral libraries. Ultimately, the optimal acquisition mode depends on the specific balance required between coverage, reproducibility, and identification confidence for a given experimental context.

Within the broader context of optimizing electrospray ionization (ESI) parameters for untargeted metabolomics, the selection and implementation of chromatographic column chemistries present a fundamental challenge. The goal of comprehensive metabolite profiling is hindered by the vast physicochemical diversity of metabolites, which no single chromatographic mode can sufficiently retain and separate. Reversed-phase liquid chromatography (RPLC) excels for non-polar to mid-polar compounds but often fails to adequately retain highly polar metabolites. Hydrophilic interaction liquid chromatography (HILIC) has emerged as a powerful orthogonal technique that addresses this limitation by providing robust retention for polar and ionizable compounds. This application note provides detailed protocols and benchmarking data for leveraging RPLC and HILIC, both separately and in innovative coupled setups, to significantly expand metabolome coverage in untargeted metabolomics research.

Theoretical Background and Orthogonality of RPLC and HILIC

The orthogonality of RPLC and HILIC stems from their diametrically opposed retention mechanisms. RPLC separates compounds based on hydrophobic partitioning between a non-polar stationary phase (e.g., C18) and a polar mobile phase, typically starting with a high percentage of water. In contrast, HILIC employs a polar stationary phase (e.g., bare silica, amide, or zwitterionic materials) and a mobile phase rich in organic solvent (usually >70% acetonitrile). Retention in HILIC is primarily governed by hydrophilic partitioning of analytes into a water-enriched layer immobilized on the stationary phase, supplemented by hydrogen bonding and electrostatic interactions [71] [72].

This fundamental difference means that elution orders in RPLC and HILIC are often reversed: metabolites with strong retention in RPLC are typically poorly retained in HILIC, and vice versa [73]. The integration of these two techniques capitalizes on their complementary nature, enabling the analysis of a more chemically diverse set of metabolites from a single sample injection than either method could achieve alone. Studies have demonstrated that combining optimal RPLC and HILIC methods can increase the number of annotated metabolites by approximately 60% compared to using RPLC alone [1] [57].

Experimental Protocols

Protocol 1: Serially Coupled RPLC/HILIC for Single-Injection Analysis

This protocol describes a novel setup for serially coupling RPLC and HILIC columns, enabling the analysis of both polar and non-polar metabolites in a single injection [74].

• Instrumentation: UHPLC system capable of handling multiple solvents and a T-piece connector; High-resolution mass spectrometer (e.g., Orbitrap) with HESI II interface.
• Columns:
  • First Dimension: Hypersil GOLD RPLC column (100 mm × 1.0 mm, 1.9 µm).
  • Second Dimension: SeQuant ZIC-pHILIC HILIC column (150 mm × 4.6 mm, 5 µm).
• Connection: The two columns are connected in series through a T-piece. The third port of the T-piece is connected to the second pump delivering the HILIC-compatible mobile phase [74].
• Mobile Phases:
  • RPLC Mobile Phase: Solvent A: LC-MS grade water; Solvent B: Acetonitrile.
  • HILIC Mobile Phase: Solvent A: LC-MS grade water + 20 mM ammonium carbonate, pH 9; Solvent B: Acetonitrile.
• Gradients:
  • RPLC Gradient: Flow rate: 65 µL/min. Initial 5% B held for 2 min, increased to 95% B over 15 min, held for 5 min, then re-equilibrated at 5% B for 12 min.
  • HILIC Gradient: Flow rate: 350 µL/min. Initial 90% B held for 5 min, decreased to 20% B in 15 min, then to 5% B and held for 5 min, followed by reconstitution to 90% B in 0.1 min and re-equilibration for 12 min.
• MS Detection: Full scan mass range 70-1400 m/z in both positive and negative ionization switching mode. Capillary temperature: 275°C.

This configuration allows non-polar metabolites to be retained and separated on the RPLC column, while polar metabolites that are not retained by the RPLC column are subsequently trapped and separated on the downstream HILIC column [74]. The results from this study validate this simple yet powerful metabolomics approach for complex samples like beer extracts, providing good retention time reproducibility (RSD < 5% for tested organic acid standards) [74].

Protocol 2: Sequential Analysis Using Optimized RPLC and HILIC Methods

For laboratories where serial coupling is not feasible, sequential analysis using separately optimized RPLC and HILIC methods provides an excellent alternative for expanded coverage [1] [57] [73].

• Sample Preparation: For plasma samples, use protein precipitation with cold acetonitrile (2:1 solvent-to-sample ratio). For plant extracts (e.g., Hypericum perforatum), use ultrasound-assisted extraction with methanol/water or ethanol/water mixtures (80:20, v/v) [73].
• Optimized ESI Source Parameters (Orbitrap-based): The following parameters were identified as optimal for broad metabolite coverage in untargeted metabolomics [1] [57]:
  • Needle Position: Farthest position in the Z-direction and closest tested position in the Y-direction with respect to the MS inlet.
  • Spray Voltage: 2.5–3.5 kV (positive mode); 2.5–3.0 kV (negative mode).
  • Vaporizer and Ion Transfer Tube (ITT) Temperature: 250–350 °C.
  • Sheath Gas: 30–50 arbitrary units.
  • Auxiliary Gas: ≥10 arbitrary units.

RPLC Method for Untargeted Profiling

• Column: C18 column, e.g., Accucore C18 (100 mm × 2.1 mm, 1.9 µm) or similar.
• Mobile Phase: Solvent A: Water + 0.1% Formic Acid; Solvent B: Acetonitrile + 0.1% Formic Acid.
• Gradient: Flow rate: 0.4 mL/min. Initial 1% B held for 1 min, increased to 99% B over 10-15 min, held for 2-3 min, then returned to 1% B for re-equilibration.
• Column Temperature: 40–50°C.
• Injection Volume: 2–5 µL.

HILIC Method for Untargeted Profiling

• Column Selection: Zwitterionic sulfobetaine-based columns (e.g., ZIC-HILIC) have demonstrated superior performance for a wide range of polar metabolites compared to amide columns [1] [57] [73].
• Mobile Phase: Solvent A: Water + 10-20 mM Ammonium Acetate or Formate (pH ~3-4 for positive mode; ~9 for negative mode); Solvent B: Acetonitrile.
• Gradient: Flow rate: 0.4-0.5 mL/min. Initial 95-90% B held for 1-2 min, decreased to 40-60% B over 10-15 min, held for 1-2 min, then returned to starting conditions for re-equilibration. Note: HILIC typically requires longer re-equilibration times than RPLC (e.g., 12-15 min).
• Column Temperature: 30–40°C.
• Injection Volume: 2–5 µL. Critical: The sample diluent should have a high organic content (≥80% ACN) to match the starting mobile phase and ensure sharp peak shapes [71].

Benchmarking Data and Column Performance

Performance Comparison of Different Stationary Phases

Evaluation of five different RPLC-C18 columns and two HILIC columns for plasma untargeted metabolomics revealed that while various modern C18 columns showed comparable performance in the number of metabolites annotated, the zwitterionic (ZIC-HILIC) column outperformed an amide-based HILIC column [1] [57]. A separate study on plant extracts (Hypericum perforatum) further compared a C18 column with three HILIC columns (silica, amide, and zwitterionic), confirming the utility of HILIC for resolving challenging isobaric compound pairs that are difficult to separate by RPLC alone [73].

Table 1: Benchmarking of Column Chemistries for Metabolite Annotation

Critical Consideration: HILIC Retention Time Reproducibility

A major challenge in HILIC has been poor retention time repeatability, often attributed to slow column equilibration. Recent research has identified that leaching of ions (sodium, potassium, borate) from standard borosilicate glass solvent bottles can alter the water layer on the HILIC stationary phase, causing significant retention time shifts [75] [76].

Solution: Replace borosilicate glass bottles with inert PFA (perfluoroalkoxy) polymer solvent bottles. This simple change dramatically improved retention time reproducibility in a tested metabolite mixture, reducing the average %RSD from 8.4% (with glass) to 0.14% (with PFA) over 30 injections [75] [76]. This is a critical recommendation for any HILIC-based untargeted metabolomics workflow.

The Scientist's Toolkit: Essential Research Reagents and Materials

Table 2: Key Reagents and Materials for RPLC/HILIC Metabolomics

Workflow Visualization

Workflow for Expanded Metabolite Coverage

The strategic combination of RPLC and HILIC chromatographies is indispensable for achieving comprehensive metabolite retention in untargeted metabolomics. This application note provides two robust experimental pathways—serial coupling for single-injection analysis and optimized sequential methods—enabling researchers to significantly expand metabolome coverage beyond the limitations of RPLC alone. The critical importance of using inert PFA solvent bottles for HILIC reproducibility and the superior performance of zwitterionic HILIC columns for broad polar metabolite coverage are highlighted as key practical considerations. When integrated with optimized ESI source parameters, these chromatographic protocols form a solid foundation for generating high-quality, reproducible metabolomic data, thereby enhancing the discovery of biologically significant metabolites in pharmaceutical and clinical research.

Leveraging High-Resolution MS and Two-Layer Networking for Confident Metabolite Annotation

A primary bottleneck in untargeted metabolomics is the transition from vast arrays of unknown mass spectral features to confident metabolite identifications. The immense structural diversity of metabolites, coupled with the limitation of existing spectral libraries, means a significant fraction of detected compounds remains uncharacterized [5] [77]. High-Resolution Accurate Mass Mass Spectrometry (HRAM-MS) provides the foundational data quality necessary for detailed analysis, but sophisticated computational strategies are required for interpretation. This protocol details the integration of a novel two-layer interactive networking topology with HRAM-MS data to significantly enhance the coverage, accuracy, and efficiency of metabolite annotation. This methodology is framed within the critical context of optimizing Electrospray Ionization (ESI) parameters, as comprehensive metabolome coverage is a prerequisite for successful downstream networking analysis [9].

Technical Foundation & Key Concepts

The Challenge of Metabolite Annotation

Untargeted metabolomics aims to profile endogenous metabolites unbiasedly, offering critical insights into cellular metabolism and disease mechanisms. Despite advancements in liquid chromatography–mass spectrometry (LC–MS), metabolite annotation remains challenging. Standard library-based spectral matching is limited to known metabolites with available reference spectra, creating a significant hurdle for annotating unknown compounds [5]. The confidence levels for metabolite identification, as defined by the Metabolomics Standard Initiative, range from Level 0 (complete 3D structure) to Level 4 (unknown compound). Achieving Level 1 (confirmed structure with standard) or Level 2 (probable structure via spectral library match) is a central goal [78].

The Role of High-Resolution Mass Spectrometry

HRAM-MS is one of the most widely used techniques due to its low limit of detection and wide range of detectable masses [7]. Instruments like quadrupole-time-of-flight (Q-TOF) and Orbitrap systems provide high mass accuracy (often < 0.001 Da) and high resolution, enabling the distinction of molecules with similar mass and the determination of elemental compositions [7] [79]. The data-dependent acquisition (DDA) mode is commonly used, though recent evaluations show Data-Independent Acquisition (DIA) can provide superior reproducibility and a higher number of metabolic feature identifications in complex matrices [20].

The Two-Layer Interactive Networking Strategy

The two-layer networking approach overcomes the limitations of standalone data-driven or knowledge-driven methods by creating a synergistic framework for annotation propagation.

Conceptual Framework

This strategy integrates two distinct network layers:

• Knowledge Layer (Metabolic Reaction Network - MRN): A comprehensive, curated network where nodes represent metabolites and edges define known or predicted biochemical reaction relationships (e.g., from KEGG, MetaCyc, HMDB) [5].
• Data Layer (Feature Network): A network derived from experimental LC-MS/MS data, where nodes represent MS1 features and edges denote spectral similarity or other data-derived relationships [5].

The power of the method lies in the interactive pre-mapping between these layers, which establishes direct metabolite-feature relationships and ensures consistent network topologies, thereby refining the annotation process.

Workflow Implementation

The workflow, as implemented in tools like MetDNA3, consists of two major steps [5]:

Step 1: Curation of Two-Layer Network Topology Experimental data (metabolic features) are pre-mapped onto the knowledge-based MRN through a sequential process:

• MS1 m/z Matching: Experimental features are matched to metabolites in the MRN based on accurate mass.
• Reaction Relationship Mapping: Biochemical relationships from the MS1-constrained MRN are mapped onto the data layer.
• MS2 Similarity Constraints: MS2 spectral similarity between features is calculated and used to filter and refine the network structure, eliminating unwanted nodes.

This process results in a "data-constrained MRN" and a "knowledge-constrained feature network," establishing coherent links within and between the two layers.

Step 2: Recursive Metabolite Annotation Propagation Once the topology is established, annotation propagation proceeds recursively. Metabolites annotated with high confidence (e.g., via library match) serve as "seed" annotations. Their annotations are then propagated to connected, unknown features in the data layer via the edges defined by the knowledge network, dramatically expanding annotation coverage.

Table 1: Performance Metrics of the Two-Layer Networking Approach (MetDNA3) in Biological Samples [5]

The following diagram illustrates the core workflow and interaction between the two network layers:

Figure 1: Two-Layer Interactive Networking Workflow. The process integrates a pre-curated knowledge base with experimental HRAM-MS data to enable recursive annotation.

Experimental Protocol: From Sample to Annotation

This section provides a detailed, step-by-step protocol for implementing the described strategy.

Sample Preparation and LC-MS Analysis

Materials:

• Solvents: LC-MS grade Water, Methanol, Acetonitrile, Formic Acid [7] [9].
• Columns: Reversed-Phase C18 (e.g., HSS T3, BEH C18) and/or HILIC (e.g., BEH Z-HILIC) for expanded coverage [9].
• Internal Standards: e.g., Methionine sulfone, for quality control and, in CE-MS, relative migration time calculation [78].

Procedure:

• Sample Preparation: Precipitate proteins from biofluids (e.g., serum/plasma) using cold methanol-water (80:20). Vortex and incubate on ice, then centrifuge to pellet precipitate. Transfer supernatant for analysis [80].
• ESI Source Optimization: Critically optimize ESI parameters for maximal signal intensity and metabolite coverage. This includes:
  • Needle position and spray voltage (positive/negative mode).
  • Sheath gas, auxiliary gas, and vaporizer temperature.
  • Ion transfer tube temperature.
  • Monitor signal stability and number of detected features across different solvent compositions [9].
• Chromatographic Separation:
  • Utilize a C18 column for reversed-phase separation with a water/acetonitrile gradient containing 0.1% formic acid.
  • For comprehensive coverage of polar metabolites, complement with a HILIC method. Combining RPLC and HILIC can reveal 60% new metabolic features compared to RPLC alone [9].
• HRAM-MS Data Acquisition:
  • Acquire data in DDA or DIA mode. DIA has demonstrated superior reproducibility (10% CV vs. 17% for DDA) and higher feature detection in complex matrices [20].
  • Set mass resolution to > 30,000 (higher is preferred, e.g., 60,000-120,000 on Orbitrap instruments).
  • Set mass accuracy to < 5 ppm.

Data Preprocessing and Network Construction

• Feature Detection: Process raw MS data using software (e.g., MZmine, MS-DIAL) to perform peak picking, alignment, and deconvolution. Extract MS1 and MS2 spectra for each feature [7] [81].
• Build the Knowledge Layer (MRN): Curate a metabolic reaction network by integrating databases (KEGG, MetaCyc, HMDB). Enhance coverage and connectivity using a Graph Neural Network (GNN) model to predict potential reaction relationships between metabolites, expanding it to include hypothetical metabolites generated by tools like BioTransformer [5].
• Construct the Two-Layer Network: Use a computational workflow (e.g., MetDNA3) to execute the interactive pre-mapping as described in Section 3.2. This links the experimental feature table with the curated MRN.

Annotation and Confidence Assessment

• Seed Annotation: Identify a set of high-confidence "seed" metabolites by matching experimental MS2 spectra to authentic spectral libraries (e.g., MoNA, HMDB).
• Propagation: Execute the recursive annotation algorithm to propagate identifications from seeds to network-connected unknowns.
• Validate and Score: Increase confidence in annotations using orthogonal strategies:
  • Relative Migration Time (RMT): In Capillary Electrophoresis-MS, use RMT normalized with internal standards to add a physicochemical identifier [78].
  • In-Source Fragmentation: Harness in-source fragment ions as pseudo-MS/MS spectra to support annotation [78].
  • COSMIC-like Scoring: For in silico annotations, employ a confidence score like that in the COSMIC workflow, which combines kernel density P-value estimation and a support vector machine (SVM) with enforced feature directionality to separate correct from incorrect hits [77].

Table 2: The Scientist's Toolkit: Essential Reagents and Software for Implementation [5] [7] [81]

The integration of optimized HRAM-MS platforms with a two-layer interactive networking strategy represents a significant advancement in untargeted metabolomics. This approach moves beyond the limitations of simple spectral matching by leveraging the power of biological context and data relationships. By following the detailed protocols outlined herein—from critical ESI parameter optimization to the computational construction of interactive networks—researchers can achieve a dramatic increase in both the coverage and confidence of metabolite annotations, thereby unlocking deeper biological insights from their metabolomic data.

The integration of advanced analytical techniques is revolutionizing the diagnosis of inherited metabolic disorders (IMDs). This application note delineates robust validation protocols for untargeted metabolomics workflows, specifically optimized for IMD diagnosis. We provide detailed methodologies for liquid chromatography-mass spectrometry (LC-MS) parameter optimization, experimental design for complex sample analysis, and computational data processing pipelines. Within the broader context of optimizing electrospray ionization (ESI) parameters for untargeted metabolomics, this document serves as a practical guide for clinical researchers and laboratory professionals seeking to implement validated, high-throughput diagnostic approaches for rare genetic diseases.

Inherited metabolic disorders (IMDs) represent a group of over 1,450 rare genetic diseases characterized by disruptions in enzymatic activities or metabolic pathways, often leading to the accumulation of toxic compounds and deficient synthesis of essential metabolites [82]. The molecular diagnosis of IMDs is particularly challenging due to disease rarity and significant locus heterogeneity, necessitating comprehensive testing approaches [83]. Untargeted metabolomics has emerged as a powerful discovery tool for identifying novel metabolic perturbations and biomarkers associated with these conditions [7]. This approach enables the relatively fast and inexpensive identification of metabolites in situations where many chemical species are unknown before analysis begins, making it particularly valuable for investigating rare and novel IMDs [7]. The critical importance of methodological validation becomes apparent when considering that delays in IMD diagnosis can lead to irreversible neurological damage or fatal outcomes, especially in pediatric populations [82]. This document establishes validated protocols specifically for untargeted metabolomics applications in IMD diagnosis, with particular emphasis on ESI parameter optimization to maximize sensitivity, metabolite coverage, and analytical reproducibility.

Experimental Design and Validation Framework

Sample Preparation Protocol

Basic Protocol 1: Sample Preparation for LC-MS/MS Analysis

• Principle: Consistent sample preparation is crucial for minimizing analytical variability and ensuring reproducible results in untargeted metabolomics. This protocol outlines a standardized approach for processing biological samples (e.g., plasma, serum) for IMD investigation.
• Materials:
  • Biological samples (plasma/serum recommended for IMD screening)
  • Cold methanol, acetonitrile, or methanol:acetonitrile mixture (1:1 v/v) for protein precipitation
  • Centrifuge capable of maintaining 4°C
  • SpeedVac or lyophilizer for sample concentration
  • Reconstitution solution (appropriate LC mobile phase)
• Procedure:
  • Protein Precipitation: Thaw samples on ice. Aliquot 100 µL of plasma/serum into a microcentrifuge tube. Add 300-400 µL of cold precipitation solvent (e.g., methanol). Vortex vigorously for 30-60 seconds.
  • Incubation: Incubate samples at -20°C for 60 minutes to enhance protein precipitation.
  • Centrifugation: Centrifuge at 14,000-16,000 × g for 15 minutes at 4°C to pellet precipitated proteins.
  • Collection: Carefully transfer the supernatant (containing metabolites) to a new clean tube.
  • Concentration: Evaporate the supernatant to dryness using a SpeedVac or lyophilizer without heating.
  • Reconstitution: Reconstitute the dried metabolite extract in 100 µL of initial LC mobile phase (e.g., 0.1% formic acid in water) compatible with the chromatographic method.
  • Clarification: Centrifuge at 14,000 × g for 10 minutes at 4°C and transfer the clarified supernatant to an LC vial for analysis [7].

Research Reagent Solutions

Table 1: Essential Research Reagents for Untargeted Metabolomics in IMD Diagnosis

Optimized ESI and LC Parameters for Untargeted Metabolomics

Recent methodological investigations have demonstrated that ESI parameters and chromatographic conditions significantly influence metabolomics results, necessitating systematic optimization for comprehensive metabolite coverage [9].

ESI Source Optimization

Support Protocol 1: ESI Parameter Optimization for Metabolite Detection

• Objective: To empirically determine optimal ESI source parameters that maximize signal intensity and the number of detectable metabolic features for IMD-relevant metabolite classes.
• Experimental Approach: A design-of-experiment (DoE) approach is recommended to evaluate the interaction effects of multiple ESI parameters simultaneously.
• Parameters for Evaluation:
  • Spray Voltage: Optimize for both positive ion mode (+3.0 to +5.0 kV) and negative ion mode (-2.5 to -4.5 kV).
  • Sheath Gas Flow: Typically ranges from 30-60 arbitrary units.
  • Auxiliary Gas Flow: Typically ranges from 10-25 arbitrary units.
  • Vaporizer Temperature: Test range of 250-450°C.
  • Ion Transfer Tube Temperature: Test range of 250-350°C [9].
• Validation Metrics: The optimal parameter set should maximize the number of detected metabolic features, ensure signal stability (low coefficient of variation), and provide balanced detection across diverse metabolite classes (e.g., amino acids, organic acids, lipids, acyl-carnitines).

Table 2: Optimized ESI Source Parameters for Untargeted Metabolomics on Orbitrap Platforms

Liquid Chromatography Method Selection

Chromatographic separation is paramount for resolving the immense chemical diversity present in the metabolome. Combining different chromatographic modes dramatically expands metabolome coverage.

Basic Protocol 2: Comprehensive LC-MS/MS Data Collection for IMD Screening

• Instrumentation: Ultra-high-performance liquid chromatography (UHPLC) system coupled to a high-resolution mass spectrometer (e.g., Q-TOF, Orbitrap).
• Chromatographic Recommendations:
  • Reversed-Phase (RP) Chromatography:
    • Columns: C18 columns, specifically premier CSH C-18, HSS T3 C-18, and BEH C-18 have shown comparable performance.
    • Application: Ideal for separating medium to non-polar metabolites (lipids, acyl-carnitines, many organic acids).
  • Hydrophilic Interaction Liquid Chromatography (HILIC):
    • Columns: Zwitterionic BEH Z-HILIC columns demonstrate superior performance for polar metabolites.
    • Application: Essential for retaining and separating highly polar metabolites (amino acids, sugars, nucleotides) that elute poorly in RP [9].
• Critical Finding: Combining optimal RP and HILIC methods can detect approximately 60% more metabolic features compared to using RP chromatography alone, providing a more comprehensive view of the metabolic perturbations in IMDs [9].

Data Analysis Workflow for IMD Diagnosis

Basic Protocol 3: Data Analysis Pipeline for Untargeted Metabolomics

• Software Tools: MSConvert (for file conversion), MZMine (for feature detection and alignment), and SIRIUS (for metabolite annotation and fragmentation analysis) [7].
• Key Analysis Steps:
  • Raw Data Conversion: Convert vendor-specific files to open formats (e.g., mzML) using MSConvert.
  • Feature Detection: Identify chromatographic peaks corresponding to ions (features) from the MS data.
  • Retention Time Alignment & Gap Filling: Correct for minor retention time shifts across samples and fill in missing peaks.
  • Metabolite Annotation: Utilize accurate mass and tandem MS/MS fragmentation data to query metabolite databases (e.g., HMDB, KEGG). Tools like SIRIUS can assist in deciphering fragmentation patterns for confident annotation [7].
• Validation via Multimodal Data Integration: For IMD diagnosis, untargeted metabolomics findings should be correlated with clinical presentation and, where available, genetic testing to confirm diagnosis [82].

Diagram 1: Untargeted Metabolomics Workflow for IMD Diagnosis. The process begins with standardized sample preparation, proceeds through comprehensive LC-MS analysis and computational data processing, and concludes with validation against clinical and genetic findings.

Application in Inherited Metabolic Disorder Diagnosis

The clinical utility of validated untargeted metabolomics is demonstrated across the major classifications of IMDs.

Diagnostic Classification and Metabolomic Correlates

Table 3: IMD Classification and Associated Metabolomic Features Detectable by Untargeted LC-MS

Integrated Diagnostic Pathway

The validated untargeted metabolomics workflow fits into a broader multi-omics diagnostic pathway for IMDs, which is critical for confirming diagnosis and guiding treatment.

Diagram 2: Integrated Diagnostic Pathway for IMDs. Untargeted metabolomics acts as a pivotal hypothesis-generating tool, narrowing the focus for subsequent genetic confirmation.

Concluding Remarks

The application of rigorously validated untargeted metabolomics, underpinned by optimized ESI and chromatographic parameters, represents a transformative approach for the diagnosis and investigation of inherited metabolic disorders. The protocols and application notes detailed herein provide a framework for clinical researchers to implement these powerful techniques, enhancing diagnostic sensitivity and metabolome coverage. As public metabolite databases continue to expand and analytical technologies advance, the role of untargeted metabolomics in diagnosing and understanding rare diseases is poised to grow significantly, ultimately improving outcomes for patients with these complex conditions.

Conclusion

Optimizing ESI parameters is not a one-time task but a fundamental component of rigorous untargeted metabolomics that directly impacts data quality and biological interpretation. By systematically addressing parameters from spray voltage to source geometry, researchers can significantly enhance sensitivity and metabolome coverage. The integration of robust QC protocols is essential for monitoring instrumental performance and correcting technical variations, especially in large-scale studies. Furthermore, combining optimized ESI methods with advanced data acquisition strategies like DIA and computational tools such as two-layer networking for annotation creates a powerful pipeline for discovery. As the field advances, these optimized workflows will be crucial for unlocking the full potential of metabolomics in biomarker discovery, drug development, and precision medicine, ultimately leading to more accurate and clinically actionable insights.

References

• 1. Evaluation of ion source parameters and liquid ... [https://www.sciencedirect.com/science/article/abs/pii/S1570023225001163]
• 2. Publication : USDA ARS [https://www.ars.usda.gov/research/publications/publication/?seqNo115=426621]
• 3. Multiparametric Optimization of Data-Dependent Acquisition ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12303636/]
• 4. Effective data visualization strategies in untargeted ... [https://pubs.rsc.org/en/content/articlehtml/2025/np/d4np00039k]
• 5. Knowledge and data-driven two-layer networking for ... [https://www.nature.com/articles/s41467-025-63536-6]
• 6. Impact of three different peak picking software tools on the ... [https://www.sciencedirect.com/science/article/pii/S073170852400342X]
• 7. Design and Analysis of Untargeted Metabolomics ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12517316/]
• 8. Mass spectrometry in untargeted liquid chromatography ... [https://pubmed.ncbi.nlm.nih.gov/28990281/]
• 9. Publication : USDA ARS [https://www.ars.usda.gov/research/publications/publication/?seqNo115=407470]
• 10. Evaluation of Ion Source Parameters and Liquid ... [https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5042103]
• 11. Structure Annotation of All Mass Spectra in Untargeted ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11426395/]
• 12. An optimized LC-HRMS untargeted metabolomics workflow ... [https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0260354]
• 13. Electrospray Ionisation Mass Spectrometry: Principles and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC1853331/]
• 14. Electrospray ionization [https://en.wikipedia.org/wiki/Electrospray_ionization]
• 15. Signal Suppression in LC-ESI-MS/MS from Concomitant ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9862991/]
• 16. Metabolite annotations based on the integration of mass ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC2440531/]
• 17. Tissue-Specific Sample Dilution: An Important Parameter to ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6680868/]
• 18. A Protocol for Untargeted Metabolomic Analysis: From Sample ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9284939/]
• 19. DIAMetAlyzer allows automated false-discovery rate ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8924252/]
• 20. mass spectrometry based- untargeted metabolomics ... [https://www.sciencedirect.com/science/article/pii/S2001037025002090]
• 21. Advances in Dual-Column Chromatography for ... [https://www.sciencedirect.com/science/article/pii/S2772391725000647]
• 22. 10 Tips for Electrospray Ionisation LC-MS [https://www.elementlabsolutions.com/uk/chromatography-blog/post/10-tips-for-electrospray-ionisation-lc-ms]
• 23. Quantification Approaches in Non-Target LC/ESI/HRMS ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11483430/]
• 24. Re-optimizing the ESI probe position - Waters Knowledge Base [https://support.waters.com/Select/cIMS/User_Manuals_and_Video_Training/Troubleshoot_and_maintain_my_system/Re-optimizing_the_ESI_probe_position]
• 25. Optimization of the Electrospray Ionization Source with ... [https://www.spectroscopyonline.com/view/optimization-electrospray-ionization-esi-source-use-design-experiments-doe-approach-lc-ms-ms-determi]
• 26. Design of Experiments in metabolomics-related studies [https://www.sciencedirect.com/science/article/abs/pii/S0731708518317229]
• 27. Comprehensive Optimization of LC-MS Metabolomics ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC5364801/]
• 28. Design of Experiment Strategy for an Electrospray ... [https://www.chromatographyonline.com/view/design-experiment-strategy-electrospray-ionization-optimization-supercritical-fluid-chromatography-m]
• 29. JGI/LBNL Metabolomics - Standard LC-MS/MS ESI Method [https://www.protocols.io/view/jgi-lbnl-metabolomics-standard-lc-ms-ms-esi-method-dnhe5b3e.html]
• 30. Optimization of Electrospray Ionization by Statistical Design of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4972871/]
• 31. Principles of Electrospray Ionization - PMC - PubMed Central [https://pmc.ncbi.nlm.nih.gov/articles/PMC3134074/]
• 32. 10 Great Tips for Electrospray Ionization LC–MS [https://www.chromatographyonline.com/view/the-lcgc-blog-10-great-tips-for-electrospray-ionization-lc-ms]
• 33. Understanding mobile phase buffer composition and ... [https://pubmed.ncbi.nlm.nih.gov/37054638/]
• 34. Guide to Metabolomics Analysis: A Bioinformatics Workflow [https://pmc.ncbi.nlm.nih.gov/articles/PMC9032224/]
• 35. Reversed HILIC Gradient: A Powerful Strategy for On-Line ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10179806/]
• 36. Detailed comparison of in-house developed and ... [https://www.sciencedirect.com/science/article/abs/pii/S0021967323007902]
• 37. Ion Source Optimization and Lc Separation Method... :: SSRN [https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4563986]
• 38. Factorial design applied to LC- ESI -QTOF mass spectrometer... [https://pubs.rsc.org/en/content/articlelanding/2023/ay/d3ay00094j]
• 39. Pretreating and normalizing metabolomics data for statistical ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10827599/]
• 40. Centering, scaling, and transformations: improving the ... [https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-7-142]
• 41. Towards Unbiased Evaluation of Ionization Performance in ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9144264/]
• 42. Signal interference between drugs and metabolites in LC ... [https://www.sciencedirect.com/science/article/pii/S2095177924000510]
• 43. An Untargeted Metabolomics Workflow that Scales to ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11018270/]
• 44. Pretreating and normalizing metabolomics data for ... [https://www.sciencedirect.com/science/article/pii/S2352304223002246]
• 45. MetaboAnalyst [https://www.metaboanalyst.ca/]
• 46. Adduct formation in electrospray ionisation-mass ... [https://www.sciencedirect.com/science/article/abs/pii/S0021967319304339]
• 47. A diagnostic algorithm for inherited metabolic disorders ... [https://link.springer.com/article/10.1007/s11306-025-02302-7]
• 48. of Mobile Phase Modifiers for Fast LC-MS-Based... Optimization [https://pmc.ncbi.nlm.nih.gov/articles/PMC9916776/]
• 49. Ion suppression correction and normalization for non- ... [https://www.nature.com/articles/s41467-025-56646-8]
• 50. Advancing Untargeted Metabolomics Through FT‐ICR‐MS [https://pmc.ncbi.nlm.nih.gov/articles/PMC12451246/]
• 51. Optimization of Sample Preparation for Metabolomics ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8305469/]
• 52. Instrumental Drift in Untargeted Metabolomics [https://www.mdpi.com/2218-1989/13/5/665]
• 53. Evaluation of batch effect elimination using quality control ... [https://www.sciencedirect.com/science/article/abs/pii/S0003267018302812]
• 54. Alignment and Analysis of a Disparately Acquired Multibatch ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11701891/]
• 55. Effect of different pooled qc samples on data quality during ... [https://pubmed.ncbi.nlm.nih.gov/39557686/]
• 56. Untargeted metabolomics-based elucidation of metabolic ... [https://eurjmedres.biomedcentral.com/articles/10.1186/s40001-025-02924-6]
• 57. Evaluation of ion source parameters and liquid ... [https://pubmed.ncbi.nlm.nih.gov/40209549/]
• 58. Compensate for or Minimize Matrix Effects? Strategies ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7412464/]
• 59. Enhanced in-Source Fragmentation Annotation Enables ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8966047/]
• 60. Standardization via Post Column Infusion—A Novel and ... [https://www.mdpi.com/1420-3049/29/16/3829]
• 61. Untargeted serum metabolomics reveals novel metabolite ... [https://molecularneurodegeneration.biomedcentral.com/articles/10.1186/s13024-023-00694-5]
• 62. Combining DI-ESI–MS and NMR datasets for metabolic ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4354777/]
• 63. Guidelines and considerations for the use of system ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC5960010/]
• 64. High-resolution accurate mass- mass spectrometry based ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12173630/]
• 65. a comparative study of DIA and DDA mass spectrometry [https://www.sciencedirect.com/science/article/pii/S2667145X25000240]
• 66. Comparative Study of DDA and DIA Acquisition Modes ... [https://pubmed.ncbi.nlm.nih.gov/41115138/?utm_source=FeedFetcher&utm_medium=rss&utm_campaign=None&utm_content=14mzXgUgXxOhz72eR8oGB-fhcdG9zbPCC1041iU1Zt22GVtQAD&fc=None&ff=20251024010210&v=2.18.0.post22+67771e2]
• 67. Comparative Evaluation of Data Dependent and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7240956/]
• 68. Mass Spectrometry Acquisition Mode Showdown: DDA vs. ... [https://www.metwarebio.com/mass-spectrometry-acquisition-modes-dda-vs-dia-vs-mrm-vs-prm/]
• 69. Evaluation of significant features discovered from different ... [https://www.sciencedirect.com/science/article/abs/pii/S0003267020309120]
• 70. Characterization of Electrospray Ionization Complexity in ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11238156/]
• 71. Advances in Hydrophilic Interaction Liquid ... [https://www.chromatographyonline.com/view/advances-hydrophilic-interaction-liquid-chromatography-pharmaceutical-analysis]
• 72. Applications of Hydrophilic Interaction Chromatography in ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12430451/]
• 73. Comparative evaluation of reversed-phase and hydrophilic ... [https://link.springer.com/article/10.1007/s00216-025-06030-8]
• 74. Serially coupling hydrophobic interaction and reversed- ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4559102/]
• 75. Solving the retention time repeatability problem of ... [https://www.sciencedirect.com/science/article/pii/S0021967324004345]
• 76. HILIC Retention Time Issues Addressed with New Approach [https://www.chromatographyonline.com/view/hilic-retention-time-issues-addressed-with-new-approach]
• 77. High-confidence structural annotation of metabolites ... [https://www.nature.com/articles/s41587-021-01045-9]
• 78. Enhancing confidence of metabolite annotation in Capillary ... [https://www.sciencedirect.com/science/article/abs/pii/S0021967320310323]
• 79. An insight into high-resolution mass-spectrometry data - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC2697344/]
• 80. compMS2Miner: An Automatable Metabolite Identification ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6338221/]
• 81. Two-layer homolog network approach for PFAS nontarget ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11735632/]
• 82. Inherited metabolic disorders: presentation, clinical types ... [https://ojrd.biomedcentral.com/articles/10.1186/s13023-025-03979-8]
• 83. Design and validation of a metabolic disorder ... [https://pubmed.ncbi.nlm.nih.gov/20578233/]