Neuropeptide Characterization Workflow from Sampling to Data-Independent Acquisition Mass Spectrometry

Abstract

Endogenous neuropeptides are key modulators of brain function, playing critical roles in behavior, stress, pain, and homeostatic regulation, yet their analysis remains difficult. Biologically, they are low in abundance, rapidly degraded, and processed variably from precursor proteins, with expression limited to small, localized cell populations. Technically, their detection is complicated by a wide dynamic range, diverse post-translational modifications, and sparse signals in mass spectrometry datasets. This protocol outlines a comprehensive workflow for neuropeptide analysis in Rattus norvegicus brain tissue using both data-dependent acquisition (DDA) and data-independent acquisition (DIA) mass spectrometry (MS) on a timsTOF platform. Following optimized brain sample preparation, including dissection, peptide extraction and clean-up, nano liquid chromatography (LC)-MS is performed with ion mobility gas-phase fractionation to improve detection sensitivity and accuracy. The DDA-generated spectral library supports DIA-based quantification in Skyline, enabling high-confidence MS2-level measurements. This integrated workflow increases neuropeptide coverage and enhances quantitative reproducibility, providing a robust platform for studying neuropeptides in complex brain tissue.

SUMMARY:

This protocol outlines a comprehensive neuropeptidomics workflow in rat brain tissue that combines data-dependent acquisition-parallel accumulation–serial fragmentation (DDA-PASEF) and data-independent acquisition (DIA) mass spectrometry.

INTRODUCTION:

Neuropeptides are a functionally diverse class of endogenous signaling molecules that serve as critical modulators of neuronal communication, nervous, and neuroendocrine system functions. Acting as neuromodulators, hormones, or co-transmitters, they influence a wide range of physiological processes, including pain modulation, stress response, circadian regulation, and appetite control. Unlike classical neurotransmitters, which are synthesized as small molecules, neuropeptides are encoded within large precursor proteins that are synthesized through ribosomal translation of mRNA. Precursor proteins, or prohormones, then undergo proteolytic processing to release the active peptides, which are packaged into dense-core vesicles (DCVs). Upon stimulation, neuropeptides are released to exert modulatory, typically slower and more prolonged effects through interactions with G-protein-coupled receptors. Their importance in maintaining central and peripheral nervous system function has made them an area of intense biological and clinical interest^1-5.

Despite their relevance, neuropeptides remain analytically challenging to characterize. Their highly heterogeneous structures—ranging from short to extended sequences, often bearing diverse post-translational modifications (PTMs) such as amidation, acetylation, phosphorylation, amino acid isomerization, or truncation—lead to variability in ionization and fragmentation behaviors in mass spectrometry (MS)^1,6-8. Furthermore, they are typically present at low concentrations relative to biological matrix components, easily degraded, and are localized in a spatially restricted manner within the complex tissue of the nervous system. These features complicate both identification and quantification, particularly when using traditional proteomics approaches that rely on predictable enzymatic digestion and uniform peptide properties^9-11.

Here, we present a comprehensive strategy for the identification and quantification of endogenous peptides in the brain of Rattus norvegicus, with a specific emphasis on neuropeptides. The workflow here begins with optimized brain sampling procedures, followed by extensive liquid chromatography (LC) separations, and concludes with high-resolution MS analysis using parallel accumulation–serial fragmentation (PASEF) on a trapped ion mobility spectrometry (TIMS) platform^12-16. By incorporating ion mobility gas-phase fractionation (IM-GPF) prior to precursor selection, the DDA-PASEF approach enables improved precursor separation and the enhanced detection of low-abundance neuropeptides, many of which would otherwise be missed in unidimensional DDA workflows¹⁷.

While data-dependent acquisition (DDA) remains the most commonly used strategy in peptide identification due to its high-quality MS/MS spectra, it is inherently limited by its reliance on precursor ion selection by intensity, which biases against low-abundance species^18-20. Moreover, its semi-stochastic sampling leads to missing values across replicates, hindering reproducible quantification. To overcome these limitations, we constructed a robust spectral library using our DDA-PASEF data, which can be subsequently employed for data-independent acquisition (DIA) analysis. In contrast to DDA, DIA samples all precursor ions within defined m/z windows, ensuring consistent sampling of both abundant and rare peptides, and when combined with the spectral library and targeted analysis in Skyline, allows for sensitive and reproducible MS2-level quantification^21-24.

This integrative approach—spanning brain peptide extraction, chromatographic separation, ion mobility-enhanced MS acquisition, and hybrid DDA/DIA-based quantification—was developed with the complexity of neuropeptides in mind. It maximizes peptide coverage while minimizing data loss, addressing key limitations of a simpler DDA-driven peptidomics. As such, we present a flexible and powerful platform for exploring the neuropeptide landscape in the mammalian brain, which can be readily adapted to a wide range of experimental questions and tissue samples.

PROTOCOL:

All animal experiments in this study were done in accordance with the animal use protocol approved by the Illinois Institutional Animal Care and Use Committee (23228) with strict adherence to both national and ARRIVE standards for the ethical treatment and care of animals.

1. Animal dissection and brain isolation

1.1. Perform cold saline perfusion immediately followed by tissue snap freezing to preserve neuropeptide integrity, minimize degradation artifacts, and enhance data quality for downstream peptidomic analysis.

1.2. Euthanize the rat using the CO₂ method following institutional animal care guidelines. Confirm death by absence of heartbeat and lack of reflexes.

1.3. Place the rat supine; make an incision to expose the thoracic cavity.

1.4. Make a small incision in the left ventricle of the heart.

1.5. Make a small incision in the right atrium to allow blood and perfusion solution outflow.

1.6. Insert a cut-to-size plastic pipette tip attached to a 60 mL syringe filled with 50 mL of ice-cold physiological solution (e.g., 0.9% saline or mGBSS) into the left ventricle.

1.7. Perfuse the animal body with the solution. Decapitate the rat using sharp scissors or a guillotine.

1.8. Remove skin and skull overlying the brain with rongeurs or fine scissors.

1.9. Carefully lift the brain from the cranial cavity, cutting cranial nerves as needed.

1.10. Immediately freeze the brain on dry ice or snap-freeze in isopentane chilled on dry ice.

1.11. Transfer to labeled cryovials or foil, then store at −80 °C.

2. Peptide extraction from rat brain tissue

2.1. For whole brain samples, weigh each frozen tissue quickly using an analytical balance. Homogenize the tissue using LC-MS grade solution containing 10% glacial acetic acid and 1% water in methanol, using a 10:1 (v/w) solvent-to-tissue ratio.

2.2. Incubate the homogenized samples on ice for 20 min.

2.3. Centrifuge the samples at 16,000 × g for 20 min at 4 °C. Carefully collect the supernatant without disturbing the pellet.

2.4. Resuspend the pellet in LC-MS grade water, using a 10:1 (v/w) solvent-to-tissue ratio, and repeat the incubation on ice for 20 min.

2.5. Centrifuge again under the same conditions (16,000 × g, 20 min, 4 °C), and collect the second supernatant.

2.6. Combine both supernatants and transfer 1/10 of the total volume to a new tube for downstream processing. Dry this aliquot using a vacuum concentrator until completely dry. Store the remaining extract at −80 °C for future use.

3. Solid phase extraction (SPE) using C18 spin columns

NOTE: All solvents should be LC-MS grade. Unless otherwise noted, centrifugation steps are performed at 1,500 × g at room temperature using a benchtop centrifuge.

3.1. Reagent preparation

3.1.1. Prepare activation solution (A): 50% acetonitrile/50% 0.1% formic acid in water.

3.1.2. Prepare equilibration and wash solution (B): 0.1% formic acid in water.

3.1.3. Prepare elution solution (C): 50% acetonitrile/50% 0.1% formic acid in water.

3.2. Sample preparation

3.2.1. Reconstitute the dried peptide extract in 200 μL of solution (A).

3.2.2. Centrifuge the sample at 16,000 × g for 10 min at 4 °C.

3.2.3. Carefully collect the supernatant for SPE.

3.3. Column conditioning and equilibration

3.3.1. Add 150 μL of solution (A) to the C18 spin column.

3.3.2. Centrifuge for 1 min at 1,500 × g.

3.3.3. Repeat steps 3.3.1 and 3.3.2 two times to ensure complete activation.

3.3.4. Add 150 μL of solution (B).

3.3.5. Centrifuge for 1 min.

3.3.6. Repeat steps 3.3.4 and 3.3.5 two times.

3.4. Sample loading and washing

3.4.1. Load 200 μL of the prepared peptide solution onto the column.

3.4.2. Centrifuge for 1 min.

3.4.3. Reload the flowthrough back onto the column three times to maximize peptide retention.

3.4.4. Add 150 μL of solution (B).

3.4.5. Centrifuge for 1 min.

3.4.6. Repeat steps 3.4.4 and 3.4.5 three times to remove non-retained contaminants.

3.5. Peptide elution

3.5.1. Add 150 μL of elution solution (C) to the column.

3.5.2. Centrifuge for 1 min.

3.5.3. Repeat elution steps 3.5.1–3.5.2 once more.

3.5.4. Combine both eluates and store on ice or proceed to drying under vacuum.

4. LC-MS/MS Analysis

4.1. Prepare solution (A): 0.1% formic acid in water, solution (B): 0.1% formic acid in acetonitrile, solution (C): 25 fmol/μL retention time standard (iRT) 0.1% formic acid in water.

4.2. Resuspend the samples in 20 μL solution (C) prior to LC-MS injection.

NOTE: Two biological samples were analyzed, and each was injected five times for both DDA and DIA runs, resulting in five technical replicates per method per sample.

4.3. Perform the peptide separation using a nano-high-performance nanoflow liquid chromatography (nano-HPLC) system, with mobile phase A consisting of solution (A) and mobile phase B consisting of solution (B).

4.4. Load 1 μL of samples onto a twenty-five C18 analytical column (150 μm × 250 mm, 1.9 μm particles, 120 Å pore size) and maintain at 40 °C throughout the run.

4.5. Carry out the LC separation at a flow rate of 600 nL/min using the following gradient profile: 0–5 min: 2–10% B; 5–85 min: 10–45% B; 85–87 min: 45–50% B; followed by high organic wash and re-equilibration.

NOTE: The LC was directly coupled to a timsTOF mass spectrometer (MS) via an ion source.

4.6. For DDA analysis, use the default application method DDA PASEF-standard_1.1sec_cycletime on timsTOF MS.

4.6.1. Briefly, operate the mass spectrometer in PASEF mode with dynamic exclusion enabled. Set each cycle to consist of 10 PASEF MS/MS scans over a 1.1 s duty cycle.

4.6.2. Set the mass range to m/z 100–1700, and the ion mobility range from 0.60 to 1.60 V·s/cm², using a ramp time of 100.0 ms. Ramp the collision energy from 20.0 to 59.0 eV as a function of ion mobility.

4.7. Ion mobility gas phase fractionation (IM-GPF)

4.7.1. Segment the ion mobility range of 0.6–1.6 V·s/cm² into three overlapping windows: 0.6–1.0, 0.9–1.3, and 1.2–1.6 V·s/cm².

4.7.2. To do this, change the mobility start and end values in the TIMS setting tap of timscontrol or timscontrol editor to the corresponding window values.

4.7.3. Acquire data for each window. For all ion mobility acquisition windows, use a constant accumulation and ramp time of 100 ms.

4.7.4. Acquire data in DDA-PASEF mode using the same parameters detailed in step 4.6.

4.8. For DIA analysis, employ an isolation window width of 25 m/z across the mass range of 400–1200 m/z. Employ the same ion mobility experimental (IMEX) settings used in the DDA method.

4.8.1. Additional DIA settings include one MS1 ramp followed by 16 MS/MS ramps per cycle, yielding a total cycle time of 1.8 s.

4.8.2. Use the default proteomics DIA-PASEF method as a template for DIA method development. After creating the DIA method, load it into the method tab of Hystar. No activation of DIA mode is required.

NOTE: The DIA method will automatically be registered once successfully loaded.

5. Data processing

NOTE: Following data acquisition, peptide identification should be performed using bioinformatics tools such as PEAKS Online, PEAKS Studio, MSFragger, Maxquant, or similar platforms. This study used Peaks Online, capable of working with raw data in .d format.

5.1. Database search

5.1.1. Use an appropriate reference database for peptide searches, such as a curated rat signaling peptide database, a rat proteome, or the human proteome, or a curated signaling peptide database based on the sample source.

NOTE: These databases can be obtained from reliable sources such as NCBI or UniProt (https://www.uniprot.org/). Note only protein or peptides in the database will be identified. Rat signal data was curated by filtering the rat proteome with "signal peptide". In the Uniprot terms, filtering by "Signal peptide" means narrowing down the proteome to proteins involved in secretion, membrane localization, or compartmental trafficking. This custom database contains 5630 proteins. Limiting the size of the protein database from 50,000 to 5,000 offers several advantages. First, it enhances identification confidence by reducing the likelihood of random peptide-spectrum matches, thereby increasing the accuracy of peptide assignments. Second, a smaller database improves computational efficiency, allowing for faster processing times and reduced demand for computational resources. Third, a more concise database can lead to better control of the false discovery rate (FDR), resulting in more reliable and statistically robust protein identifications. However, the database needs to be relatively large to avoid artificial inflation of confidence scores and inaccurate FDR estimation.

5.1.2. Configure the database search in the program GUI with the following recommended parameters: Precursor mass error tolerance: 10–25 ppm; fragment mass error tolerance: 0.01–0.05 Da; enzyme specificity: none (unspecific digestion); peptide length range: 4–45 amino acids.

5.1.3. Define the set of post-translational modifications (PTM) to be included in the search.

5.1.4. After completing the database search, export the pepxml and mzXML (or MGF) files corresponding to each ion mobility window, including the standard full mobility range (0.6–1.6 V·s/cm²).

5.2. Skyline library build-up

5.2.1. Open Skyline and create a new document, ensure the Proteomics interface is selected, and use the default settings unless otherwise specified.

5.2.2. Navigate to File > Import > Peptide Search.

5.2.3. In the Import Peptide Search Wizard, select Build and add the relevant pepxml files (ensure the corresponding mzXML or MGF files are in the same directory).

5.2.4. Set iRT standard peptides to Automatic, choose DIA as the workflow, then click Next.

5.2.5. After library construction, choose approximately 10–15 peptides as iRT standards for alignment across replicates. If prompted to recalibrate iRT, click OK to proceed.

5.2.6. In the Extract Chromatograms page, click Browse to select and import all DIA raw files, then click continue.

5.2.7. In the Transition Settings window, set precursor charges range to 1–5, product charges range to 1–3, Ion types to ,p,y,b and define product ion selection From ion 2 To last ion. Other recommended settings: ion match tolerance: 0.05 Da, Peaks to pick: 5.

5.2.8. In the Peptide Settings window, set modifications that match the Peaks search setting described in 4.1.3.

5.2.9. On the Isolation Scheme page, enter a name for the isolation and scheme and select Prespecified Isolation Windows. Click on import and load one of the DIA files to automatically import the experiment's isolation window configuration.

5.2.10. Set Full scan acquisition settings with the following parameters. Acquisition method: DIA; Product mass analyzer: Centroided; Isolation scheme: Use the one defined above; Mass accuracy: 10–20 ppm; Mobility resolving power: 30–50; Enable Use only scans within 5 min of predicted RT; Optionally, modify MS1 Filtering if importing MS1 chromatograms is needed. Select an appropriate mass analyzer and resolution for optimum MS1 chromatogram extraction.

5.2.11. Import FASTA and Generate Decoys. Set Decoy generation method: Shuffle Sequence.

5.2.12. Select enzyme: digestion is unspecific in endogenous peptide analysis. Edit the enzyme by selecting add from the drop-down enzyme menu. Set the name of the enzyme, Type to both, cleave c-terminal to KR and cleave the N-terminal to KR, allow semi-cleavage, and click OK. Set missed cleavages to 3.

5.2.13. Import the target FASTA file, then click Finish. Confirm the Associated Proteins Page if prompted.

5.2.14. Ensure that the spectral library has been successfully created and that DIA data has been imported. Manually inspect the list of identified peptides and proceed with downstream quantitative and qualitative analysis.

5.3. Qualitative and quantitative measurements

NOTE: The qualitative and quantitative analyses are done in Skyline. For details on the analysis of DIA data in Skyline, visit https://skyline.ms/wiki/home/software/Skyline/page.view?name=tutorial_dia_swath.

5.3.1. If DDA data is loaded, select Settings > Transition Setting from the menu, choose DDA in Acquisition Method; in Settings > Peptide Setting, choose MS 1 level for quantification.

5.3.2. If DIA data is loaded, Settings > Transition Setting, choose DIA in Acquisition Method; in Settings > Peptide Setting, choose MS 2 level for quantification.

5.3.3. In View > Document Grid, select Peptide Quantification, then Export the spreadsheet. Combine the spreadsheets from DDA results and DIA results.

NOTE: The spreadsheet summarizes the peak areas for all the peptides in the library across all samples. In this experiment, peptide peak areas were normalized to the retention time standard peptide GISNEGQNASIK, which is part of the Pierce retention time mixture. Normalization was performed by calculating the ratio of each peptide's peak area to that of the standard.

REPRESENTATIVE RESULTS:

A representative experimental scheme is shown in Figure 1. Analysis of the DDA-PASEF data (Figure 2) acquired using both the standard ion mobility range and IM-GPF revealed a substantially higher peptide coverage for the IM-GPF approach, which revealed significantly greater peptide coverage with IM-GPF (Figure 2D). Notably, most peptides identified using the standard DDA-PASEF method overlapped with those found in the IM-GPF dataset. To maximize peptide identification coverage, a comprehensive spectral library was generated by integrating identifications from both approaches.

Figure 1: Overview of the peptidomics workflow for identification and quantification of endogenous peptides from rat brain tissue.

The workflow includes tissue dissection, peptide extraction and purification, LC-ion mobility-MS (LC-IM-MS) data acquisition using gas-phase fractionation (IM-GPF), and data processing to generate spectral libraries and perform quantitative analysis. Portions of the Figure were created in BioRender. Tan, Y. (2025) https://BioRender.com/vrejrsk.

Figure 2: Database Peptide and Protein Identification.

(A) Bar chart comparing the total number of precursors, peptide-spectrum matches (PSMs), and unique peptides identified across different ion mobility windows. (B) Pie chart illustrating the proportion of peptides commonly identified across all mobility windows and those uniquely detected within individual windows. (C,D) Heatmaps depicting the distribution of ion signals across the m/z and ion mobility dimensions for the full mobility range (0.6–1.6 V·s/cm²) (C) and the restricted mobility window (0.6–1.0 V·s/cm²) (D). (E) Upset plot showing the number of proteins identified in datasets acquired using DDA and DIA methods, highlighting shared and unique protein identifications. (F) Heatmap representing peptides derived from the proenkephalin precursor identified in DIA-acquired replicates (top panel) and DDA-acquired replicates (bottom panel). Leu-enkephalin was not detected in two of the four DDA replicates, indicating potential missing values in DDA-based acquisition.

Further database analysis of replicate datasets acquired via DDA- and DIA-PASEF (Figure 3), using a curated rat signaling peptide database, revealed 217 proteins common to both methods. Additionally, 115 proteins were uniquely identified by DIA-PASEF, while 69 were exclusive to DDA-PASEF. Exploratory analyses utilizing spectral libraries derived from DDA and IM-GPF data demonstrated that DIA-PASEF enabled the confident detection of neuropeptides such as NPAFLFQPQRF (neuropeptide SF) and YGGFMRRVGRPEWWMDYQ (derived from proenkephalin). These neuropeptides were not reliably detected in the corresponding DDA-PASEF datasets due to poor peak assignment (Figure 3B,C).

Figure 3: Comparative and exploratory analysis of DDA-PASEF and DIA-PASEF data using spectral libraries.

(A) Retention time alignment of neuropeptide SF across DIA and DDA-PASEF replicates. (B) EIC of PENK (212–229) fragments ions from DIA (top) and DDA (bottom) datasets. (C) Fragment ion chromatograms of neuropeptide SF from DIA replicates demonstrating high co-elution and spectral quality.

From a quantitative perspective, the scatter plot in Figure 4A reveals a moderate positive correlation between DDA precursor intensity and DIA fragment ion intensity, as expected given that both approaches quantify the same peptides but capture different ion types. The box plots for selected peptides shown in Figure 4B further illustrate the reproducibility of the quantitative measurements obtained by both approaches. The box plots show the relative peak areas for each peptide measured by DIA and DDA, with error bars representing measurement variability across technical replicates. The selected peptides are derived from opioid prohormones, including proenkephalin (PENK) and prodynorphin (PDYN), both of which play central roles in modulating pain and analgesia.

Figure 4: Comparison of quantitative measurements between DIA-PASEF (MS2-level) and DDA-PASEF (MS1-level).

(A) This scatter plot compares peptide signal intensities obtained using IDA and DDA modes. Specifically, the x-axis represents the log₁₀-transformed precursor peak area from DDA, while the y-axis shows the log₁₀-transformed fragment peak area from DIA for each peptide. (B) Box plots represent the relative peak areas of four representative peptides, PENK 107–133, PENK 188–195, PDYN 221–233, and PDYN 235–248. Error bars represent variability across technical replicates.

DISCUSSION:

In peptidomics, particularly when analyzing endogenous peptides, preserving the native peptide profile is essential for data integrity and biological interpretation. Cold saline perfusion not only removes blood but also chills the brain, which significantly reduces postmortem proteolytic degradation occurring rapidly due to the activity of endogenous proteases. By lowering the brain temperature, enzymatic activity is slowed and circulating proteases are flushed from the vasculature, while also reducing postmortem metabolic changes^25,26. Perfusing with ice-cold saline also flushes out blood from the brain's vasculature. This step is especially important, as blood contains high-abundance proteins (such as albumin and globulins) and circulating proteases that can interfere with neuropeptide detection. Postmortem degradation of abundant blood proteins can increase the complexity of neuropeptide extracts and mask low-abundance signals, ultimately compromising both identification and quantification^25-27. Immediate freezing of the brain tissue, either on dry ice or via snap-freezing in chilled isopentane, halts enzymatic processes almost instantly. This preservation is especially important for detecting low-abundance and full-length mature neuropeptides and ensuring reliable quantification. Without these steps, peptide degradation can lead to increased variability and artifacts that compromise both identification and quantification. Thus, blood removal, cold perfusion, and rapid brain chilling are foundational techniques that directly impact the quality, sensitivity, and reproducibility of peptidomics data prior to measurements.

Analysis of peptides extracted from rat brain revealed the presence of two distinct ion mobility plumes (Figure 2C,D). This phenomenon may be attributed either to different molecular classes or to varying charge states of the peptides. If the plumes correspond to distinct molecular classes, coeluting non-peptide species may compete with peptides for selection and fragmentation. Also, if the separation arises from differences in charge states, the plume containing highly charged peptides would be expected to yield higher-quality MS/MS spectra, facilitating more confident peptide identification.

To investigate this, we implemented an ion mobility fractionation strategy, acquiring data across three discrete mobility windows: 0.6–1.0, 0.9–1.3, and 1.2–1.6 V·s/cm². These windows collectively span the full ion mobility range typically analyzed in standard DDA-PASEF experiments for peptides. As illustrated in Figure 2, although the total number of precursors detected was highest in the full scan range (0.6–1.6 V·s/cm²), a significantly greater number of peptide-spectrum matches (PSMs) were observed within the 0.6–1.0 V·s/cm² window. This finding suggests that the presence of two plumes may indeed be due to interference from other molecules.

Ion mobility–based precursor fractionation resulted in a 30% increase in peptide coverage compared to the conventional full-scan method. Approximately 82% of peptides identified in the full-scan dataset overlapped with peptides identified using fractionation (Figure 2B), highlighting the consistency and robustness of the fractionation strategy. Based on these observations, we employed the ion mobility fractionation and the full-scan data to construct spectral libraries for the quantitative analysis of neuroendocrine peptides using DIA-PASEF.

Database analysis of replicate datasets acquired via DDA and DIA revealed a higher number of peptide and protein identifications in the DIA data, including 115 proteins uniquely identified by DIA (Figure 2E). These included neuropeptides such as Pro-FMRFamide-related peptides FF and VF. In contrast, 69 proteins were exclusively identified by DDA, many of which are associated with cell adhesion and neuronal development. These findings suggest that DDA and DIA methodologies can be applied in a complementary manner to achieve broader peptide coverage.

Further analysis demonstrated that DIA acquisition effectively addresses some of the missing value problem frequently encountered in endogenous peptide analysis using DDA. For example, Leu-enkephalin was detected in only two out of four DDA replicates (Figure 2F, bottom panel), whereas it was consistently identified across all four DIA replicates (Figure 3F, top panel).

Skyline, a platform widely used for quantitative proteomics, was utilized to analyze both DDA and DIA-PASEF datasets²⁸. The software facilitates peptide identification through spectral library matching, fragment ion co-elution, and retention time alignment. To evaluate the ability of DIA-PASEF to reduce missing identification, we performed a qualitative assessment of peptides exclusively detected in the DIA dataset.

As shown in Figure 3D, neuropeptide SF was consistently detected across all DIA replicates within the same retention time window. Detection in the DDA replicates exhibited considerable variability, with retention times ranging from 25 min to 45 min (Figure 3A). The observed inconsistencies, including peak misassignments with low confidence scores, were likely due to limited or low-quality fragment ion coverage, as shown in Figure 3B. In contrast, Figure 3C shows that fragment ions corresponding to neuropeptide SF were well aligned across all DIA replicates, supporting confident automated identification.

A comparison of MS/MS spectral quality for the PENK (212–229) peptide, derived from proenkephalin, further demonstrated the advantages of DIA-PASEF. Spectra obtained from DIA exhibited higher intensity and improved fragment ion quality relative to DDA (Figure 3), which is particularly beneficial for detecting low-abundance endogenous peptides.

Quantitative analysis confirmed that DIA-PASEF yielded consistent and reproducible measurements across biological replicates, with performance comparable to DDA (Figure 4). Moreover, DIA enabled quantification at the MS2 level, offering fragment ion-level specificity and increased confidence in peptide identification relative to MS1-based DDA. These findings support the use of DIA-PASEF as a reliable and efficient approach for targeted neuropeptide quantification in complex biological matrices, particularly when data completeness and reproducibility are critical.

This protocol outlines a comprehensive strategy for neuropeptide analysis and quantification. The IM-GPF approach facilitates the construction of high-quality spectral libraries. While DIA-PASEF offers improved reproducibility and increased peptide coverage compared to DDA-PASEF, it also enhances MS2-based quantification. Combining both acquisition strategies increases overall peptide coverage, as each method uniquely contributes to the identification of distinct peptides (Figure 2E). The limitation of this protocol is the requirement for ample sample material to support both spectral library generation and data acquisition via DDA- and DIA-PASEF methods.

Name of Material/ Equipment	Company	Catalog Number	Comments/Description
0.1% Formic Acid Water	Fisher Scientific	LS118-500	LC/MS grade
Acetic Acid	Fisher Scientific	A11350	LC/MS grade
Acetonitrile	Fisher Scientific	AA47138K7	LC/MS grade
Formic Acid	Fisher Scientific	PI28905	LC/MS grade
Methanol	Fisher Scientific	AA47192K7	LC/MS grade
nanoELute2	Bruker
Pierce C18 spin columns	Fisher Scientific	AA47192K8
Pierce Peptide Retention Time Calibration Mixture	Thermo Fisher	A11351	LC/MS grade
SpeedVac vacuum concentrator	Genevac	https://scientificproducts.com/product_cat/benchtop-solvent-evaporators/	The specific model used in this study is no longer available from the manufacturer. A link to the current equivalent model is provided for reference
timsTOF Pro2	Bruker	https://www.bruker.com/en/products-and-solutions/mass-spectrometry/timstof/timstof-pro-2.html
Water	Fisher Scientific	AA47146M6	LC/MS grade