🧬 Heteronuclear NMR Assignment of Proteins (15N / 13C labeled)

This video explains how we assign protein backbone and sidechain resonances using multidimensional heteronuclear NMR — when the protein is isotopically labeled with:

• ¹⁵N
• ¹³C

High labeling efficiency is required so that we can observe scalar (J) couplings between these nuclei.

🔗 1. The Backbone Coupling Network

A protein backbone repeats:

N – Cα – C' – N – Cα – C' – ...

With ¹⁵N and ¹³C labeling, we can observe couplings between:

Key Coupling Constants

Important Concept

• Nitrogen couples strongly to:
  • Its own Cα (1-bond)
  • Previous residue Cα (2-bond but still visible!)
  • Previous residue carbonyl (1-bond)

These couplings are the basis for sequential assignment.

🧪 2. Triple Resonance Experiments

These experiments transfer magnetization stepwise along bonded atoms.

Their names describe the magnetization pathway.

🧩 HNCA

Pathway:

H → N → Cα

Result:

• Shows two Cα peaks per residue:
  • Cα (same residue)
  • Cα (previous residue)

Why two? Because both ¹J and ²J couplings are strong enough.

🧩 HNCOCA

Pathway:

H → N → C' → Cα

Important:

• Carbonyl (C') is in parentheses → shift NOT recorded
• Magnetization is relayed through it

Result:

• Shows only Cα of previous residue
• Same-residue pathway too weak

🔎 Why Compare HNCA and HNCOCA?

If a Cα peak appears in both: → It belongs to the previous residue

If it appears only in HNCA: → It belongs to the same residue

This comparison lets you determine sequential order.

🧱 3. HNCO and HNCACO

HNCO

H → N → C'

• Shows one peak per residue
• Carbonyl shift of previous residue

HNCACO

H → N → C' → C'

• Shows carbonyl of:
  • Previous
  • Same residue

Again → compare the two to identify which is which.

📊 4. Chemical Shift Ranges (Why Selectivity Works)

Carbon-13 shifts span ~0–200 ppm.

The difference is large → selective excitation is possible.

In fact: It is difficult to excite both simultaneously!

📦 5. 3D Spectra and “Strips”

True spectra are cubes (3D):

• ¹H axis
• ¹⁵N axis
• ¹³C axis

But cubes are hard to interpret.

Instead: We extract strips, typically perpendicular to nitrogen dimension.

Each strip:

• Has an HSQC peak (H–N)
• Shows carbon shifts in the third dimension

This converts a 3D cube into a series of readable 2D panels.

🧠 6. Sequential Assignment Strategy

For each residue:

1. Identify its Cα
2. Find which strip has the same value as its “previous Cα”
3. Connect them
4. Walk residue by residue through the sequence

It’s like puzzle pieces:

• Each strip has a “knob” (previous shift)
• And a “hole” (own shift)

Matching them reconstructs the sequence.

⚠️ Risk: Different residues may share similar Cα shifts. Solution: Combine multiple nuclei (Cα + C' + Cβ)

Overlap becomes extremely unlikely.

🧬 7. Using Chemical Shift Statistics

Databases show:

• Average Cα, Cβ shifts
• Standard deviations

Examples:

• Threonine & Serine
  • High Cβ shift
  • Because Cβ bonded to oxygen
• Glycine
  • No Cβ
  • Very distinctive Cα shift

This allows tentative residue-type prediction before full assignment.

🌿 8. Sidechain Assignment (Carbon-based TOCSY)

Now we extend beyond backbone.

Since the protein is ¹³C-labeled, we can perform carbon TOCSY transfers.

🧪 H(CCCO)NH

Concept:

• Start at sidechain H
• Transfer to C
• Spread via carbon TOCSY through sidechain
• Transfer to CO
• Then N
• Then H

What is recorded?

• N and H shifts
• Sidechain hydrogen shifts of previous residue

Carbons in parentheses → not recorded.

🧪 (H)C(CCO)NH

Same transfer, but:

• Record carbon shifts
• Hydrogens in parentheses

Now you see:

• All sidechain carbon shifts
• Linked to backbone HN of previous residue

⚠️ Proline Problem

All HN-based experiments fail for residues preceding proline.

Why? Proline has no amide proton (HN).

Solution: Use HCCH experiments, which:

• Do not depend on HN
• Correlate hydrogen and carbon within sidechains

🌸 Aromatic Residues

Aromatic sidechains:

• Hard to assign
• Special experiments required

Because:

• Different coupling network
• Ring systems complicate transfers

🕒 9. Practical Reality

To fully assign a protein:

You typically record:

• HNCA
• HNCOCA
• HNCO
• HNCACO
• Cβ equivalents
• Hα / Hβ equivalents
• Sidechain experiments
• Aromatic-specific experiments

Time required: ➡️ Often one month or more of spectrometer time.

This is not casual work.

🎯 10. Final Assignment Logic

After backbone + sidechain assignment:

1. Identify short stretches of connected residues
2. Compare with primary sequence
3. Find unique sequence patterns
4. Map spin-system numbers to residue numbers

Same logic as homonuclear assignment — but now with much more information.

🧩 Big Picture

This process:

• Is labor-intensive
• Requires isotopic labeling
• Requires multiple 3D spectra
• Requires chemical shift statistics

But it is essential for:

• Protein structure determination
• Dynamics studies
• Binding studies
• Functional analysis

Without resonance assignment: No structural biology via NMR.