Protein Structure

Day 12 part 2

Quiz ChatGPT conversation

🧬 Cross-Linking Mass Spectrometry (XL-MS)

🔗 What is a linker?

A linker (crosslinker) is a bifunctional molecule:

Has two reactive ends
Can connect two amino acids

👉 This means:

It links two regions of a protein OR two different proteins
Only works if those regions are close in space

✔️ So yes, you are correct:

It gives distance information between regions
Not just sequence proximity, but 3D proximity

📌 Key idea:

XL-MS gives distance constraints, not exact structure

🧪 Complex samples vs earlier methods

You noted:

“last methods before → individual protein?”

✔️ Correct.

Earlier methods (like basic bottom-up proteomics) → focus on single proteins
XL-MS → works well on:
- Protein complexes
- Cells
- Whole systems

👉 So XL-MS is more systems-level biology

🔗 Types of crosslinks

✔️ You are correct:

Intra-protein → within the same protein
Inter-protein → between different proteins

📌 Interpretation:

Inter = proteins interact
Intra = protein folding information

⚙️ Workflow (important concept)

Crosslink proteins
Digest into peptides
LC-MS/MS analysis
Identify crosslinked peptides
Extract distance constraints

📦 Enrichment – what does it mean?

You asked:

“Do enrichment → LC-MS/MS and side/site something?”

✔️ Explanation:

After digestion, mixture is VERY complex:

Normal peptides
Crosslinked peptides (rare!)

👉 Enrichment = increase proportion of crosslinked peptides

Methods:

Size-based → crosslinked peptides are bigger
Chemical enrichment → linker has handle/tag
Protein-level enrichment (e.g. His-tag)

📌 Why?

Improves detection → deeper analysis

⚛️ Zero-length crosslinkers

Example: EDC

✔️ Meaning:

No spacer → 0 Å distance
Direct bond between residues

👉 Interpretation:

The residues are in direct contact

📏 Spacer length & resolution

You asked:

“Link the spacer → increase data resolution?”

✔️ Yes — but carefully:

Short linker → strict distance → high resolution
Long linker → more flexibility → lower resolution

📌 Typical upper limit:

~30 Å (~3 nm) between Cα atoms

🧪 Amine-specific (Lysine, N-terminus)

✔️ Correct idea.

Many crosslinkers target amines
Found in:
- Lysine side chain
- N-terminus

👉 Meaning:

Crosslinks only occur at specific residues

🧪 DMTMM

✔️ Correction:

DMTMM is NOT the linker itself
It is a coupling reagent

👉 It helps:

Attach linker molecules (e.g. dihydrazine)

🧬 Crosslinked peptides (A + B)

You asked:

“Combination of peptide A and B?”

✔️ Yes.

After digestion:

You get:
- Peptide A
- Peptide B
- Crosslinked A–B peptides

👉 These are harder to analyze because:

Mass = A + B + linker

⚖️ Heavy vs Light linker (MS2)

✔️ Correct idea.

Use isotopic labeling:
- Light linker
- Heavy linker

👉 Result:

Same peptide appears with mass difference

📌 Purpose:

Identify true crosslinked peptides (filter noise)

✂️ Short vs Long fragments

You asked:

“Short fragment → peptide A & linker?”

✔️ Explanation:

When linker is cleavable:

It breaks in MS
Produces:
- Short fragment
- Long fragment

👉 These appear as doublets in spectra

📌 Key idea:

Detecting this pair confirms presence of linker

🔬 DSBU / DSSO and MS2/MS3

These are cleavable crosslinkers

MS2:

Fragment peptides + linker
Complex spectrum

MS3:

Fragment selected pieces again

✔️ Result:

Easier interpretation and sequencing

🔁 Multiple fragmentation rounds

✔️ Meaning:

MS1 → intact mass
MS2 → fragmentation
MS3 → further fragmentation

👉 Each step reduces complexity

⚗️ Homobifunctional crosslinkers

✔️ Means:

Both ends react with same group (e.g. amines)

⚖️ Linker length & delta mass

✔️ Correct:

Different linkers → different mass shifts
Helps identify:
- Which linker
- Where it is

🎯 Can we find the crosslink site?

✔️ Yes — using MS data:

Fragment patterns
Linker-specific signals
Known chemistry (e.g. Lys-specific)

👉 Not trivial, but doable

⚠️ XL-MS limitation

✔️ Important:

Provides upper distance limit
Not exact position

📌 Example:

If crosslinked → distance ≤ ~30 Å

🧩 Structural insight

✔️ Key uses:

Protein folding
Protein interactions
Complex topology

👉 Can combine with:

SAXS
Cryo-EM
X-ray

🧠 De novo modeling

✔️ Meaning:

Build structure from scratch
Using constraints (like crosslinks)

🔄 Conformational changes

✔️ You asked:

“crosslinking reduced?”

✔️ Yes:

Less crosslinking = residues moved apart
More crosslinking = closer

👉 Used to study:

Apo vs holo
Drug binding

🔪 LiP-MS (Limited Proteolysis MS)

🧬 Core idea

Mild protease digestion BEFORE trypsin

👉 Cuts only accessible regions

🔍 What does it tell?

You asked:

“peptides disappear or emerge?”

✔️ Exactly.

Interpretation:

Disappearing peptide → region became accessible
New peptides → new cleavage sites

👉 Indicates:

Structural/conformational change

⚙️ Why semi-tryptic analysis?

Pre-digestion creates non-tryptic ends
So:
- One end = tryptic
- One end = random

👉 Search space increases

🔄 Condition comparison

Example:

Condition 1 → folded → protected
Condition 2 → unfolded → exposed

👉 Result:

Different peptide patterns

📊 Output

Fold changes
Volcano plots
Mapping on sequence

🔗 XL-MS + LiP-MS together

✔️ Complementary:

XL-MS → distance constraints
LiP-MS → accessibility/dynamics

👉 Together:

Much stronger structural insight

🧠 Final Takeaway

XL-MS = who is close to whom (distance)
LiP-MS = what becomes exposed (dynamics)

Neither works alone: 👉 Combined with other methods → realistic protein models

🧬 Additional XL-MS Concepts You Didn’t Explicitly Mention

🧪 XL-MS as a “surface probing” method

XL-MS is related to surface labeling techniques
But instead of modifying ONE site, it:
- Connects two sites simultaneously

📌 Insight:

Only accessible and spatially close residues get crosslinked → gives information about surface exposure + proximity

🧠 Distance constraints → not exact structures

A key conceptual point often missed:

XL-MS does not give a structure directly
It gives:
- “Residue A is within X Å of residue B”

👉 This is a constraint, not a coordinate

📌 Think:

Like solving a puzzle with distance rules

🧬 Why XL-MS rarely stands alone

Important conceptual point from the lecture:

XL-MS alone → insufficient for full structure
Must be combined with:
- SAXS
- Cryo-EM
- X-ray
- HDX-MS
- LiP-MS

📌 Reason:

Each method gives partial information

🧪 Crosslinker design variability (big concept)

Crosslinkers are highly tunable:

They differ in:

Reactive groups (what residues they bind)
Spacer length (distance constraint)
Cleavability (MS-friendly or not)
Isotope labeling (heavy/light)
Enrichment handles

📌 Insight:

Choice of linker = defines what biological question you can answer

🔬 Dead-end crosslinks

You briefly mentioned it, but conceptually:

Only one side reacts
Other side is “quenched”

👉 Result:

Single peptide with linker attached

📌 Use:

Helps identify accessible residues

🔄 Ambiguity in crosslinks (important limitation)

A tricky concept:

Same mass spectrum could represent:

Intra-protein crosslink
Inter-protein crosslink
Dead-end crosslink

👉 Interpretation is not always straightforward

📌 Requires:

Database searching
Additional constraints

⚠️ Why XL-MS data is more complex than normal MS

Compared to bottom-up proteomics:

You now have:

Linear peptides
Crosslinked peptides (A–B)
Dead-end peptides
Loop links (within same peptide)

👉 Huge increase in complexity

📌 Consequence:

Heavy reliance on computational tools

🔍 Database dependence

For non-cleavable linkers:

You cannot directly interpret spectra
Must match against:
- Protein sequence databases

📌 Reason:

Fragments are mixtures of A + B → ambiguous

🧪 MS-cleavable linkers: why they matter

Major conceptual advantage:

Break inside the linker
Separate peptide A and B signals

👉 Makes identification:

More targeted
More reliable

📌 Key idea:

Adds an extra layer of information

🔁 Probability of fragmentation

Important subtle point:

Linker can break on either side
Fragmentation is not symmetric

👉 You won’t get equal intensities

📌 Interpretation:

Presence matters more than intensity

🧬 Trifunctional crosslinkers (advanced concept)

You didn’t explicitly mention the implication:

Can link 3 sites
Or:
- 2 proteins + dead-end
- 3 proteins together

👉 Enables:

Direct probing of protein complexes

📌 Tradeoff:

Much harder data analysis

🧠 Protein interaction networks

XL-MS can scale to:

Identify who interacts with whom
Build interaction networks

📌 Especially useful for:

Whole proteome studies
In vivo experiments

🧬 In vivo XL-MS (very important concept)

Crosslinking can be done:
- In cells
- In tissues

👉 Captures:

Native interactions in real biological context

🔄 Structural topology mapping

Using crosslinks:

Map:
- Domains close in 3D
- Subunit organization

📌 Example insight:

Regions far apart in sequence → close in space

🔄 Sequence vs structure distinction

Critical concept:

Sequence distance ≠ spatial distance

👉 XL-MS reveals:

Long-range contacts

📌 Essential for:

Understanding protein folding

🧬 Circle plots (topology visualization)

Visualize crosslinks across sequence
Arcs represent connections

👉 Helps identify:

Local vs long-range interactions

🔄 Dynamics vs static structure

XL-MS captures:

Probabilities of interactions
Not fixed positions

👉 Important:

Proteins are dynamic, not rigid

🧪 Quantitative XL-MS (conformational changes)

You saw:

Heavy vs light comparison

👉 Used to detect:

Structural shifts between conditions

🔪 Additional LiP-MS Concepts You Didn’t Mention

🧪 Why use broad-specificity proteases

Example: Proteinase K

✔️ Cuts:

Many amino acids

👉 Needed because:

Want to probe structure, not sequence rules

⚠️ Controlling digestion level

Important:

Too much digestion → complete degradation
Too little → no information

👉 Must be limited proteolysis

🧠 Structural accessibility principle

Core idea:

Folded regions → protected
Unfolded regions → accessible

👉 Protease acts as:

A “structure sensor”

📊 Fold-change interpretation

Fold change ~1 → no change
High/low fold change → structural difference

🌋 Volcano plots (conceptual meaning)

Combine:
- Fold change
- Statistical significance

👉 Highlight:

Regions affected by condition change

🔄 Multiplexing (important advantage)

Using labeling (e.g. TMT):

Analyze many conditions at once

👉 Enables:

High-throughput structural screening

⏱️ Time-resolved LiP-MS

Study changes over time
Track folding/unfolding dynamics

🔗 Complementarity with HDX-MS

Even though not your focus:

HDX-MS → backbone dynamics
LiP-MS → protease accessibility
XL-MS → spatial constraints

👉 Together:

Multi-layer structural understanding

🧠 Final Conceptual Layer (Most Important)

All methods in your file share one principle:

❗ They do NOT directly give structure

They give:

Constraints
Accessibility
Dynamics

👉 Structure is inferred via:

Integration
Modeling
Computational fitting

📌 One-line summaries

XL-MS → “Which residues are close in 3D?”
LiP-MS → “Which regions become exposed?”
HDX-MS → “Which regions are flexible/dynamic?”

Quiz

Score: 0/41 (0%)

Q0. In XL-MS, what is the main reason a bifunctional crosslinker provides structural information?

It measures the exact atomic coordinates of all residues in the protein.

It links two reactive sites only when they are within a restricted spatial distance.

It unfolds the protein and reveals the primary sequence directly.

It labels every amino acid equally, regardless of accessibility or chemistry.

Q1. Which statement best distinguishes XL-MS from many simpler bottom-up proteomics workflows?

XL-MS only works on purified single proteins, while bottom-up proteomics works only on tissues.

XL-MS can provide spatial proximity and interaction information, while basic bottom-up proteomics mainly identifies and quantifies peptides or proteins.

XL-MS does not require digestion, while bottom-up proteomics always requires digestion.

XL-MS directly gives protein crystal structures, while bottom-up proteomics gives peptide masses.

Q2. What is the difference between intra-protein and inter-protein crosslinks?

Intra-protein crosslinks occur within one protein, while inter-protein crosslinks occur between different proteins.

Intra-protein crosslinks occur only in cells, while inter-protein crosslinks occur only in purified systems.

Intra-protein crosslinks are always false positives, while inter-protein crosslinks are always true interactions.

Intra-protein crosslinks involve RNA, while inter-protein crosslinks involve DNA.

Q3. Why is enrichment often useful before LC-MS/MS in XL-MS?

Because crosslinked peptides are usually rare compared with linear unmodified peptides.

Because enrichment converts all peptides into crosslinked peptides.

Because enrichment removes the need for database searching.

Because enrichment guarantees that all detected crosslinks are inter-protein crosslinks.

Q4. What does a zero-length crosslinker such as EDC imply about the linked residues?

They are separated by at least 30 Å.

They are probably in very close contact because no spacer is added.

They must be on different proteins.

They must be in a flexible disordered region.

Q5. Why can changing the spacer length of a crosslinker change the structural information obtained?

Because spacer length changes the genetic sequence of the protein.

Because different spacer lengths impose different maximum distance constraints.

Because longer spacers always produce higher-resolution data.

Because spacer length determines whether trypsin can digest the protein.

Q6. Many common XL-MS crosslinkers are amine-specific. Which protein groups are commonly targeted by these linkers?

Lysine side chains and protein N-termini

Methionine sulfur atoms and cysteine disulfides only

Backbone carbonyl oxygens only

Phosphorylated serine and threonine only

Q7. In the lecture transcript, what is the role of DMTMM in certain crosslinking strategies?

It is mainly a fluorescent dye for imaging crosslinked proteins.

It is a coupling reagent that helps attach linkers such as dihydrazines.

It is a protease used in LiP-MS.

It is a database search algorithm for identifying crosslinked peptides.

Q8. Why are non-cleavable crosslinked peptides difficult to interpret by MS/MS?

Because they cannot ionize under any condition.

Because fragments may come from peptide alpha, peptide beta, or combinations involving the linker.

Because they always produce only one fragment ion.

Because they are always smaller than linear peptides.

Q9. What is the purpose of using heavy and light isotope-labeled crosslinkers in XL-MS?

To make crosslinked peptides invisible in MS1.

To create predictable mass differences that help identify linker-containing fragments.

To prevent digestion by trypsin.

To convert inter-protein crosslinks into intra-protein crosslinks.

Q10. What is a key advantage of MS-cleavable crosslinkers such as DSSO or DSBU?

They remove the need for mass spectrometry.

They can generate diagnostic fragment patterns, such as doublets, that help identify crosslinked peptides.

They only react with DNA and RNA.

They make all peptide fragments exactly the same mass.

Q11. Why can MS3 make XL-MS data easier to interpret?

MS3 removes all protein-protein interactions before analysis.

MS3 further fragments selected MS2 products, producing simpler spectra for sequence and site localization.

MS3 replaces the need for digestion.

MS3 measures protein folding directly by microscopy.

Q12. What does it mean that many crosslinkers are homobifunctional?

Both reactive ends have the same chemical specificity.

They contain three reactive groups.

They only form dead-end modifications.

They can only be used in homogeneous cell cultures.

Q13. What does XL-MS usually provide as a distance constraint?

An exact distance between two atoms with no error.

An upper limit for the allowed distance between linked residues.

Only the total length of the protein sequence.

Only the molecular weight of the intact protein.

Q14. What does a long-range intra-protein crosslink often indicate?

Two residues far apart in sequence are close together in 3D structure.

The protein has no folded structure.

The protein was not digested.

The crosslink must be between two different proteins.

Q15. How can XL-MS reveal conformational changes between two conditions?

By showing changes in the abundance of specific crosslinks between the conditions.

By sequencing the genome under each condition.

By measuring the pH of the LC solvent only.

By converting all proteins into antibodies.

Q16. What is the theoretical meaning of de novo modeling in the context of XL-MS?

Building a draft structural model using experimental constraints without relying only on a pre-existing structure.

Repeating LC-MS/MS until no peptides remain.

Removing all crosslinks from the protein before analysis.

Identifying only linear tryptic peptides.

Q17. What is the purpose of limited proteolysis in LiP-MS?

To completely digest proteins into single amino acids.

To mildly cut structurally accessible regions before standard bottom-up analysis.

To crosslink residues that are close in space.

To label lysine residues with isotope-coded linkers.

Q18. Why is semi-tryptic database searching important in LiP-MS?

Because the limited protease can create peptide ends that are not generated by trypsin.

Because trypsin always cuts every amino acid randomly.

Because LiP-MS does not produce peptides.

Because semi-tryptic searching removes all false positives automatically.

Q19. In LiP-MS, what can it mean if a tryptic peptide disappears and new semi-tryptic peptides appear under condition 2?

The protein region became more accessible to limited proteolysis under condition 2.

The protein sequence mutated during LC-MS/MS.

The crosslinker became longer under condition 2.

The peptide was converted into RNA.

Q20. Why can LiP-MS be useful for studying drug-protein interactions?

Drug binding may alter local structure or accessibility, changing the limited proteolysis pattern.

Drug binding always prevents all digestion completely.

LiP-MS directly visualizes drugs under an electron microscope.

LiP-MS only works on purified drugs, not proteins.

Q21. What is the benefit of combining LiP-MS with TMT multiplexing?

Many conditions or time points can be analyzed together in one pooled dataset.

TMT converts LiP-MS into X-ray crystallography.

TMT removes the need for statistical analysis.

TMT prevents any protein conformational change.

Q22. Which comparison best describes XL-MS, HDX-MS, and LiP-MS together?

XL-MS gives distance constraints, HDX-MS reports backbone exchange/dynamics, and LiP-MS reports protease accessibility.

XL-MS gives DNA sequence, HDX-MS gives RNA abundance, and LiP-MS gives lipid composition.

All three directly solve atomic-resolution structures without modeling.

All three measure only protein concentration and nothing about structure.

Q23. Why do XL-MS, HDX-MS, and LiP-MS usually need computational modeling or integration with other methods?

They provide partial structural constraints rather than full atomic structures by themselves.

They cannot detect peptides.

They only work if no protein structure exists.

They always produce identical results regardless of protein state.

Q24. What is the main theoretical advantage of performing XL-MS in vivo or in situ?

It can capture protein structures and interactions in a more native cellular or tissue context.

It guarantees that every protein in the organism is fully sequenced.

It prevents all nonspecific crosslinking.

It removes the need to extract or analyze proteins.

Q25. True or false: XL-MS directly determines a complete protein structure at atomic resolution without needing modeling or additional methods.

True

False

Q26. True or false: A crosslink between two residues means those residues were close enough in 3D space for the linker chemistry and spacer length to connect them.

True

False

Q27. True or false: Inter-protein crosslinks can provide evidence that two proteins are near each other or interacting.

True

False

Q28. True or false: Enrichment is often used in XL-MS because crosslinked peptides are usually abundant enough to dominate the peptide mixture naturally.

True

False

Q29. True or false: A zero-length crosslinker adds a long flexible spacer between two residues.

True

False

Q30. True or false: NHS ester crosslinkers commonly react with primary amines such as lysine side chains and protein N-termini.

True

False

Q31. True or false: In a crosslinked peptide pair, the observed mass can include peptide alpha, peptide beta, and the crosslinker.

True

False

Q32. True or false: Heavy and light isotope-labeled crosslinkers can help identify crosslinked species because they produce predictable mass shifts.

True

False

Q33. True or false: MS-cleavable crosslinkers are useful because they can produce diagnostic doublet patterns after fragmentation.

True

False

Q34. True or false: MS3 is always mandatory in every XL-MS experiment.

True

False

Q35. True or false: A homobifunctional crosslinker has two reactive ends with the same chemical specificity.

True

False

Q36. True or false: XL-MS distance constraints are usually interpreted as exact distances with no tolerance.

True

False

Q37. True or false: If the abundance of a specific crosslink decreases between two conditions, one possible interpretation is that the linked residues moved farther apart or became less favorably positioned.

True

False

Q38. True or false: LiP-MS uses limited proteolysis to probe structural accessibility and conformational changes.

True

False

Q39. True or false: In LiP-MS, semi-tryptic peptides can appear because the limited protease creates cleavage sites that are not produced by trypsin.

True

False

Q40. True or false: LiP-MS and XL-MS provide identical structural information and are therefore redundant.

True

False