1. Sample preparation for proteomics (advanced) [9]
1.1. Recapitulation
1.1.A. Bottom-up Proteomics
❥ Proteins are enzymatically digested into peptides (usually with trypsin).
❥ Peptides are separated by liquid chromatography (LC) and analysed by tandem MS (MS/MS).
❥ Protein identity and quantity are inferred from detected peptides.
❥ Advantages:
❣ High sensitivity and throughput
❣ Compatible with complex samples
❥ Disadvantages:
❣ Loss of information on intact Proteoforms
❣ PTM combinations may be missed
1.1.B. Top-down Proteomics
❥ Intact proteins are analysed directly by LC-MS/MS.
❥ Allows direct identification of proteoforms (isoforms, PTM combinations, truncations).
❥ Less common due to:
❣ Technical complexity
❣ Lower throughput
❣ More demanding instrumentation and data analysis
❥ Emerging approach with increasing relevance for PTM-focused studies.
1.1.C. Shotgun Proteomics
❥ Unbiased, global analysis aiming to identify as many peptides/proteins as possible in a sample.
❥ Usually performed in bottom-up mode.
❥ Strategy: acquire as many MS/MS fragmentation spectra as possible.
❥ Does not require prior knowledge of sample composition.
1.1.D. Data-Dependent vs Data-Independent Acquisition (DDA/DIA)
1. Data-Dependent Acquisition (DDA)
❣ Basic principle
⤷ First, a full MS1 spectrum is acquired with the quadrupole in transmission mode (no precursor
isolation).
⤷ The instrument software: Detects all peaks and ranks them by intensity
⤷ MS/MS spectra are acquired sequentially from the most intense precursors.
❣ Top-N workflow
⤷ MS/MS is acquired for:
⧙ The most intense peak in MS1
⧙ Then the 2nd most intense, 3rd, etc.
⤷ This continues until the Top-N precursors are fragmented.
⤷ After reaching N, the instrument acquires a new MS1 spectrum.
, ❣ Choice of N
⤷ N too low → important peptides are missed (no MS/MS data).
⤷ N too high → precursor ions may have eluted or fragmented before analysis.
⤷ N is chosen based on: LC peak width, instrument speed and sample complexity
❣ Refinements in DDA
⤷ Exclusion criteria to improve coverage:
⧙ Exclude singly charged ions (z = 1)
⧙ Exclude known contaminants
⧙ Dynamic exclusion (prevent repeated fragmentation of same precursor)
⤷ Can be extended to MSⁿ (top-N, top-M, etc.).
❣ General characteristics of DDA
⤷ Most widely used method in proteomics.
⤷ Simple and well-established data analysis.
⤷ Limitation: Lower-abundance peptides may never be selected for MS/MS.
2. Data-Independent Acquisition (DIA)
❣ Basic principle
⤷ The instrument does not isolate a single precursor ion.
⤷ Instead, it fragments all ions within a defined m/z window.
⤷ This ensures that all fragmentation information is present in the dataset.
❣ Advantages of DIA
⤷ No loss of information due to precursor selection bias.
⤷ Improved reproducibility across runs.
⤷ Low-abundance peptides are not excluded.
❣ Comparison with DDA
⤷ In DDA: A “top-5” experiment permanently loses information on the 6th most intense ion.
⤷ In DIA: Fragmentation data for all ions are recorded simultaneously.
❣ Disadvantages of DIA
⤷ MS/MS spectra contain fragments from multiple precursors.
⤷ Linking fragments to the correct precursor becomes computationally complex.
⤷ Requires advanced data analysis algorithms and often spectral libraries.
3. DIA on timsTOF: PASEF
❣ timsTOF instrument
⤷ Combines Trapped Ion Mobility Spectrometry (TIMS) with Time-of-Flight (TOF) MS.
⤷ Adds Arlington separation based on ion mobility, improving resolution.
❣ Parallel Accumulation – Serial Fragmentation (PASEF)
⤷ Ions are accumulated in parallel while previous ions are being fragmented.
⤷ Fragmentation occurs serially, increasing speed and sensitivity.
⤷ Benefits: Higher sequencing speed, improved sensitivity and better utilisation of ion beam
❣ DDA vs DIA on timsTOF
⤷ DDA-PASEF improves classical DDA efficiency.
⤷ DIA-PASEF combines full-coverage fragmentation with ion mobility separation.
,1.2. Shotgun Proteomics vs Genome Sequencing
❥ Increasing biological complexity from DNA → RNA → proteoforms
❣ One gene can give rise to:
⤷ multiple mRNA isoforms (splicing),
⤷ multiple protein isoforms,
⤷ multiple proteoforms due to PTMs, truncations, and processing.
❣ Proteomics therefore measures a much more complex and dynamic system than genomics.
❥ No amplification step in proteomics
❣ Unlike DNA (PCR) or RNA (RT-PCR), proteins cannot be amplified.
❣ Dynamic range of MS is therefore limited:
⤷ high-abundance proteins dominate,
⤷ low-abundance proteins may not be detected.
❥ Incomplete sequence coverage
❣ It is very rare to identify peptides covering the entire protein sequence.
❣ Some peptides are:
⤷ too small or too large,
⤷ poorly ionized,
⤷ lost during sample preparation.
❥ Protein inference problem
❣ Identified peptides are not always unique to one protein.
❣ Shared peptides can belong to multiple protein isoforms or homologs.
❣ Leads to ambiguity in determining:
⤷ which proteins are truly present,
⤷ which isoform is expressed.
1.3. Example Questions Addressed by Proteomics
1. How do protein expression profiles differ between healthy and diseased states?
2. What is the absolute or relative concentration of a protein (biomarker discovery)?
3. Are PTM profiles (e.g. phosphorylation, glycosylation) altered in disease?
4. Which proteins interact with each other in the cell (interactomics)?
5. Where are proteins localized within tissues or subcellular compartments?
6. What is the effect of a single nucleotide polymorphism (SNP) on proteoforms?
, 1.4. Sample Preparation
1.4.A. General Steps
❥ Define the biological question and experimental design
❣ Sample prep depends on the goal: global proteomics, phosphoproteomics, interactomics,
membrane proteomics, etc.
❥ Literature research: Similar problems may already have established protocols, avoid reinventing
protocols that are known to work.
❥ Sample collection: Proper handling and storage to prevent degradation.
❥ Protein extraction / solubilisation: Efficient lysis is crucial, especially for membrane proteins.
❥ Protein separation or purification (optional): Used to reduce complexity or enrich specific protein
classes.
❥ Reduction and alkylation: Reduce disulfide bonds and alkylate cysteines to prevent re-oxidation.
❥ Peptide purification / selection: Removal of salts, detergents, and other MS-incompatible compounds.
❥ Prefractionation (optional): Reduces sample complexity and increases proteome depth.
1.4.B. General Recommendations
❥ Use high-quality samples and prevent degradation: protease inhibitors and phosphatase inhibitors.
❥ Avoid contamination: MS is extremely sensitive to contaminants and plasticizers (e.g. polyethylene
glycol, PEG) reduce LC-MS sensitivity.
❥ Always prepare fresh buffers: Degraded buffers introduce variability and artefacts.
❥ Use LC-MS grade solvents and reagents only.
❥ Never pipette directly from stock bottles: Pour into a clean beaker first to avoid contamination.
❥ Small samples are high-risk: Surface binding and handling losses become significant.
❥ Clear trend toward: automation, standardisation, miniaturisation and robotic workflows.
1.4.C. Protocol 1: Filter-Aided Sample Preparation (FASP)
❥ Miniaturisation and simplification: Reduces sample loss and handling steps.
❥ Highly compatible with SDS
❣ Allows use of high SDS concentrations for efficient lysis.
❣ Advantage: good extraction of membrane proteins.
❥ Workflow
❣ Lyse sample in: 4% SDS, 0.1 M DTT, 0.1 M Tris-HCl (pH 7.6).
❣ Sonication to disrupt cells.
❣ Add 8 M urea to dissociate SDS from proteins.
❣ Use a filter unit: proteins are retained but SDS and
small molecules pass through.
❣ Repeated washing with urea to remove SDS.
❣ Proteins remain on filter for downstream
digestion.
❥ Disadvantage
❣ Urea can decompose into cyanic acid.
❣ Cyanic acid reacts with: lysine, arginine and N-terminal amino groups.
❣ Results in carbamylation, an unwanted modification.
1.1. Recapitulation
1.1.A. Bottom-up Proteomics
❥ Proteins are enzymatically digested into peptides (usually with trypsin).
❥ Peptides are separated by liquid chromatography (LC) and analysed by tandem MS (MS/MS).
❥ Protein identity and quantity are inferred from detected peptides.
❥ Advantages:
❣ High sensitivity and throughput
❣ Compatible with complex samples
❥ Disadvantages:
❣ Loss of information on intact Proteoforms
❣ PTM combinations may be missed
1.1.B. Top-down Proteomics
❥ Intact proteins are analysed directly by LC-MS/MS.
❥ Allows direct identification of proteoforms (isoforms, PTM combinations, truncations).
❥ Less common due to:
❣ Technical complexity
❣ Lower throughput
❣ More demanding instrumentation and data analysis
❥ Emerging approach with increasing relevance for PTM-focused studies.
1.1.C. Shotgun Proteomics
❥ Unbiased, global analysis aiming to identify as many peptides/proteins as possible in a sample.
❥ Usually performed in bottom-up mode.
❥ Strategy: acquire as many MS/MS fragmentation spectra as possible.
❥ Does not require prior knowledge of sample composition.
1.1.D. Data-Dependent vs Data-Independent Acquisition (DDA/DIA)
1. Data-Dependent Acquisition (DDA)
❣ Basic principle
⤷ First, a full MS1 spectrum is acquired with the quadrupole in transmission mode (no precursor
isolation).
⤷ The instrument software: Detects all peaks and ranks them by intensity
⤷ MS/MS spectra are acquired sequentially from the most intense precursors.
❣ Top-N workflow
⤷ MS/MS is acquired for:
⧙ The most intense peak in MS1
⧙ Then the 2nd most intense, 3rd, etc.
⤷ This continues until the Top-N precursors are fragmented.
⤷ After reaching N, the instrument acquires a new MS1 spectrum.
, ❣ Choice of N
⤷ N too low → important peptides are missed (no MS/MS data).
⤷ N too high → precursor ions may have eluted or fragmented before analysis.
⤷ N is chosen based on: LC peak width, instrument speed and sample complexity
❣ Refinements in DDA
⤷ Exclusion criteria to improve coverage:
⧙ Exclude singly charged ions (z = 1)
⧙ Exclude known contaminants
⧙ Dynamic exclusion (prevent repeated fragmentation of same precursor)
⤷ Can be extended to MSⁿ (top-N, top-M, etc.).
❣ General characteristics of DDA
⤷ Most widely used method in proteomics.
⤷ Simple and well-established data analysis.
⤷ Limitation: Lower-abundance peptides may never be selected for MS/MS.
2. Data-Independent Acquisition (DIA)
❣ Basic principle
⤷ The instrument does not isolate a single precursor ion.
⤷ Instead, it fragments all ions within a defined m/z window.
⤷ This ensures that all fragmentation information is present in the dataset.
❣ Advantages of DIA
⤷ No loss of information due to precursor selection bias.
⤷ Improved reproducibility across runs.
⤷ Low-abundance peptides are not excluded.
❣ Comparison with DDA
⤷ In DDA: A “top-5” experiment permanently loses information on the 6th most intense ion.
⤷ In DIA: Fragmentation data for all ions are recorded simultaneously.
❣ Disadvantages of DIA
⤷ MS/MS spectra contain fragments from multiple precursors.
⤷ Linking fragments to the correct precursor becomes computationally complex.
⤷ Requires advanced data analysis algorithms and often spectral libraries.
3. DIA on timsTOF: PASEF
❣ timsTOF instrument
⤷ Combines Trapped Ion Mobility Spectrometry (TIMS) with Time-of-Flight (TOF) MS.
⤷ Adds Arlington separation based on ion mobility, improving resolution.
❣ Parallel Accumulation – Serial Fragmentation (PASEF)
⤷ Ions are accumulated in parallel while previous ions are being fragmented.
⤷ Fragmentation occurs serially, increasing speed and sensitivity.
⤷ Benefits: Higher sequencing speed, improved sensitivity and better utilisation of ion beam
❣ DDA vs DIA on timsTOF
⤷ DDA-PASEF improves classical DDA efficiency.
⤷ DIA-PASEF combines full-coverage fragmentation with ion mobility separation.
,1.2. Shotgun Proteomics vs Genome Sequencing
❥ Increasing biological complexity from DNA → RNA → proteoforms
❣ One gene can give rise to:
⤷ multiple mRNA isoforms (splicing),
⤷ multiple protein isoforms,
⤷ multiple proteoforms due to PTMs, truncations, and processing.
❣ Proteomics therefore measures a much more complex and dynamic system than genomics.
❥ No amplification step in proteomics
❣ Unlike DNA (PCR) or RNA (RT-PCR), proteins cannot be amplified.
❣ Dynamic range of MS is therefore limited:
⤷ high-abundance proteins dominate,
⤷ low-abundance proteins may not be detected.
❥ Incomplete sequence coverage
❣ It is very rare to identify peptides covering the entire protein sequence.
❣ Some peptides are:
⤷ too small or too large,
⤷ poorly ionized,
⤷ lost during sample preparation.
❥ Protein inference problem
❣ Identified peptides are not always unique to one protein.
❣ Shared peptides can belong to multiple protein isoforms or homologs.
❣ Leads to ambiguity in determining:
⤷ which proteins are truly present,
⤷ which isoform is expressed.
1.3. Example Questions Addressed by Proteomics
1. How do protein expression profiles differ between healthy and diseased states?
2. What is the absolute or relative concentration of a protein (biomarker discovery)?
3. Are PTM profiles (e.g. phosphorylation, glycosylation) altered in disease?
4. Which proteins interact with each other in the cell (interactomics)?
5. Where are proteins localized within tissues or subcellular compartments?
6. What is the effect of a single nucleotide polymorphism (SNP) on proteoforms?
, 1.4. Sample Preparation
1.4.A. General Steps
❥ Define the biological question and experimental design
❣ Sample prep depends on the goal: global proteomics, phosphoproteomics, interactomics,
membrane proteomics, etc.
❥ Literature research: Similar problems may already have established protocols, avoid reinventing
protocols that are known to work.
❥ Sample collection: Proper handling and storage to prevent degradation.
❥ Protein extraction / solubilisation: Efficient lysis is crucial, especially for membrane proteins.
❥ Protein separation or purification (optional): Used to reduce complexity or enrich specific protein
classes.
❥ Reduction and alkylation: Reduce disulfide bonds and alkylate cysteines to prevent re-oxidation.
❥ Peptide purification / selection: Removal of salts, detergents, and other MS-incompatible compounds.
❥ Prefractionation (optional): Reduces sample complexity and increases proteome depth.
1.4.B. General Recommendations
❥ Use high-quality samples and prevent degradation: protease inhibitors and phosphatase inhibitors.
❥ Avoid contamination: MS is extremely sensitive to contaminants and plasticizers (e.g. polyethylene
glycol, PEG) reduce LC-MS sensitivity.
❥ Always prepare fresh buffers: Degraded buffers introduce variability and artefacts.
❥ Use LC-MS grade solvents and reagents only.
❥ Never pipette directly from stock bottles: Pour into a clean beaker first to avoid contamination.
❥ Small samples are high-risk: Surface binding and handling losses become significant.
❥ Clear trend toward: automation, standardisation, miniaturisation and robotic workflows.
1.4.C. Protocol 1: Filter-Aided Sample Preparation (FASP)
❥ Miniaturisation and simplification: Reduces sample loss and handling steps.
❥ Highly compatible with SDS
❣ Allows use of high SDS concentrations for efficient lysis.
❣ Advantage: good extraction of membrane proteins.
❥ Workflow
❣ Lyse sample in: 4% SDS, 0.1 M DTT, 0.1 M Tris-HCl (pH 7.6).
❣ Sonication to disrupt cells.
❣ Add 8 M urea to dissociate SDS from proteins.
❣ Use a filter unit: proteins are retained but SDS and
small molecules pass through.
❣ Repeated washing with urea to remove SDS.
❣ Proteins remain on filter for downstream
digestion.
❥ Disadvantage
❣ Urea can decompose into cyanic acid.
❣ Cyanic acid reacts with: lysine, arginine and N-terminal amino groups.
❣ Results in carbamylation, an unwanted modification.
