Mouse fatty liver proteomes in dda-PASEF and dia-PASEF from low input using the timsTOF SCP

Importance of the Topic

The global rise of non-alcoholic fatty liver disease (NAFLD) driven by western‐style diets creates an urgent need for sensitive biomarkers and translational tools to evaluate disease mechanisms and therapeutic interventions. High-resolution proteomic profiling of liver tissue from animal models delivers molecular insights but often requires large sample amounts. Developing workflows that generate deep proteome coverage from minimal input enables region-specific analysis, high-throughput screening and more refined studies of liver pathology.

Study Objectives and Overview

This work evaluates two PASEF-based acquisition strategies on the timsTOF SCP platform for low-input mouse liver proteomics. Key aims:

• Compare data-dependent (dda-PASEF) and data-independent (dia-PASEF) acquisition modes.
• Assess proteome depth from as little as 10 ng and 1 ng of peptide material.
• Differentiate proteomic signatures between wild-type controls and a fatty liver model induced by 16 weeks of high-fat diet with or without protein overexpression.

Methodology and Instrumentation

Mouse livers were flash-frozen, homogenized, reduced, alkylated and digested with trypsin. Peptides were cleaned using STRAP cartridges. Separation employed Bruker nanoElute UHPLC with an IonOpticks Aurora C18 column (25 cm × 75 µm, 400 nL/min, 90 min gradient). Mass spectrometry was performed on a Bruker timsTOF SCP in both dda-PASEF and dia-PASEF modes. Data processing:

• dda-PASEF results analyzed with PEAKS Online at 1% FDR.
• dia-PASEF data interrogated in Spectronaut 17 at 1% FDR using a project-specific spectral library.

Main Results and Discussion

dda-PASEF: Technical reproducibility was high across replicates. From 10 ng input of wild-type liver, an average of 3 658 protein groups and 41 353 peptides were identified per run. Fatty liver samples yielded ~4 415 protein groups and 49 000 peptides.

dia-PASEF: A bespoke library of ~5 996 protein groups and 55 000 peptides enabled deep coverage from 1 ng input. Wild-type runs averaged 4 500 protein groups, 40 000 peptides and >45 000 sequences; fatty liver achieved >5 500 protein groups, 50 000 peptides and ~60 000 sequences. Unsupervised clustering and volcano analysis clearly separated disease vs control profiles, highlighting early markers of lipid accumulation.

Benefits and Practical Applications

These optimized PASEF workflows deliver deep liver proteomes from nanogram-level inputs, allowing:

• Proteomic mapping of microdissected regions or small biopsy fragments.
• High-throughput screening of disease models or drug effects.
• Discovery of early NAFLD biomarkers and mechanistic targets.

Future Trends and Potential Applications

Advances may include integration with spatial proteomics to link molecular changes to histology, extension to single-cell or single-fibre analyses, and application in human clinical biopsies. Combining ultralow-input PASEF with machine learning could accelerate biomarker panels for personalised liver disease management.

Conclusion

On the timsTOF SCP platform, both dda-PASEF and dia-PASEF enable robust, deep proteome coverage from minute liver samples. DDA achieves >3 600 protein groups from 10 ng, while DIA yields >4 500 from 1 ng. This methodological advance paves the way for highly sensitive, region-specific and high-throughput liver proteomic analyses.

Reference

Gordon E., Willetts M., Assis D., Zhang L., Zhang Y. Mouse fatty liver proteomes in dda-PASEF and dia-PASEF from low input using the timsTOF SCP. ASMS 2023.