Peptide Bonds & Primary Structure

Thermodynamics of Peptide Bond Formation

Peptide bond formation is thermodynamically unfavorable under standard conditions. The standard free energy change for hydrolysis of a peptide bond is:

This means that the equilibrium favors hydrolysis, not synthesis. In biological systems, peptide bond formation is driven by the ribosome, which couples the reaction to the hydrolysis of GTP and the energy stored in aminoacyl-tRNA bonds. The overall energetic cost of adding one amino acid residue during translation is approximately:

(accounting for aminoacyl-tRNA synthetase activation, GTP hydrolysis during elongation factor binding, and translocation steps).

Bond Characteristics

The peptide bond has several important physical properties:

• Bond length: The C-N bond in a peptide is ~1.33 Å, intermediate between a typical C-N single bond (~1.49 Å) and a C=N double bond (~1.27 Å)
• Bond dissociation energy: ~330 kJ/mol, indicating significant stability
• Planarity: The six atoms $C_\alpha$-C-O-N-H-$C_\alpha$ lie in a single plane (the peptide plane)
• Trans configuration: The $C_\alpha$ atoms on either side of the peptide bond are trans to each other in >99.9% of peptide bonds (exception: X-Pro bonds, ~10% cis)
• No free rotation: The partial double-bond character prevents rotation around the C-N bond (rotation barrier ~80 kJ/mol)

Resonance Structures

In Structure I, the C=O is a full double bond and C-N is a single bond. In Structure II, the electrons are delocalized, giving the C-N bond partial double-bond character while the C=O bond becomes a single bond with a negative charge on oxygen and a positive charge on nitrogen.

The true electronic structure is a resonance hybrid of these two forms. The fraction of double-bond character in the C-N bond can be estimated from bond length data:

This simple calculation overestimates the double-bond character. More sophisticated analyses (including electron density maps from X-ray crystallography) place it at approximately 40%.

The Peptide Plane

As a consequence of the partial double-bond character, the six atoms involved in the peptide bond unit are constrained to lie in a single plane:

• $C_\alpha^{(i)}$ -- the alpha carbon of residue $i$
• $\text{C}^{(i)}$ -- the carbonyl carbon
• $\text{O}^{(i)}$ -- the carbonyl oxygen
• $\text{N}^{(i+1)}$ -- the amide nitrogen
• $\text{H}^{(i+1)}$ -- the amide hydrogen
• $C_\alpha^{(i+1)}$ -- the alpha carbon of residue $i+1$

The dihedral angle $\omega$ (omega) describes rotation around the C-N peptide bond. For the trans configuration, $\omega = 180°$; for the cisconfiguration, $\omega = 0°$. The energy barrier to rotation about $\omega$ is approximately:

This barrier is high enough that interconversion between cis and trans is slow and usually requires enzymatic catalysis (e.g., peptidyl-prolyl isomerases).

Backbone Dihedral Angles

• $\phi$ (phi): rotation about the $\text{N} - C_\alpha$ bond, defined by the atoms $\text{C}_{i-1} - \text{N}_i - C_{\alpha,i} - \text{C}_i$
• $\psi$ (psi): rotation about the $C_\alpha - \text{C}$ bond, defined by the atoms $\text{N}_i - C_{\alpha,i} - \text{C}_i - \text{N}_{i+1}$

Both angles range from $-180°$ to $+180°$. A plot of $\psi$versus $\phi$ for all residues in a protein is called a Ramachandran plot(or Ramachandran diagram), first introduced by G.N. Ramachandran in 1963.

Steric Constraints and Allowed Regions

Not all ($\phi, \psi$) combinations are sterically allowed. Van der Waals repulsions between backbone atoms and the $C_\beta$ atom of the side chain restrict the accessible conformational space. The major allowed regions correspond to well-known secondary structures:

Special Cases: Glycine and Proline

Glycine ($\text{R} = \text{H}$) has the greatest conformational flexibility because it lacks a $C_\beta$ side chain. Its Ramachandran plot shows a much larger allowed region compared to other residues, including the upper-right quadrant (left-handed helix region). Glycine residues are therefore commonly found in tight turns and loops where other amino acids would experience steric clashes.

Proline is the most conformationally restricted amino acid. Its cyclic pyrrolidine ring constrains $\phi$ to approximately $-65° \pm 15°$. This severely limits the allowed ($\phi, \psi$) space and explains why proline is rarely found within regular $\alpha$-helices (it disrupts the hydrogen bonding pattern, since it lacks an amide hydrogen).

Backbone and Side Chains

The repeating unit of the polypeptide backbone consists of the atoms $\text{N} - C_\alpha - \text{C}$(three atoms per residue). The backbone is the structural scaffold, while the side chains (R-groups) project outward and determine the chemical properties of each position. The backbone atoms form a regular repeating pattern with a rise per residue of 3.6 Å in an $\alpha$-helix and 3.5 Å per residue in a fully extended $\beta$-strand.

Molecular Weight Calculations

The average molecular weight of an amino acid residue (after peptide bond formation, i.e., minus water) is approximately 110 Da (daltons). For a polypeptide of $n$ residues, each peptide bond releases one water molecule ($M_r = 18$ Da), so the total molecular weight is:

where $M_i$ is the molecular weight of the $i$-th free amino acid. Using the average residue weight approximation:

Here 128 Da is the average molecular weight of a free amino acid (weighted by typical amino acid composition), and each peptide bond removes 18 Da of water. The residue weight of ~110 Da is the standard approximation used in biochemistry.

Edman Degradation

Developed by Pehr Edman in 1950, this method sequentially removes and identifies the N-terminal residue. The reagent phenylisothiocyanate (PITC) reacts with the free $\alpha$-amino group under mildly alkaline conditions:

Treatment with anhydrous acid cleaves the PTC-amino acid derivative as a thiazolinone, which is then converted to the more stable phenylthiohydantoin (PTH)-amino acid and identified by HPLC. The cycle is repeated on the shortened peptide. Key limitations:

• Practical limit of ~50-60 residues per run (cumulative yield losses, ~2-5% per cycle)
• If the overall yield per cycle is $Y$, after $n$ cycles the remaining signal is $Y^n$
• Blocked N-termini (e.g., acetylated) cannot be sequenced by Edman degradation
• Requires ~1-5 pmol of pure protein

For a 95% yield per cycle after 60 cycles: $0.95^{60} = 0.046$ (only 4.6% of original signal remains).

Mass Spectrometry

Modern protein sequencing relies heavily on tandem mass spectrometry (MS/MS). The protein is digested into peptides (typically with trypsin, which cleaves after Lys and Arg), and the peptides are analyzed by:

• MALDI-TOF: Matrix-assisted laser desorption/ionization with time-of-flight detection. Measures $m/z$ ratios to determine peptide masses
• ESI-MS/MS: Electrospray ionization with tandem mass analysis. Selected peptide ions are fragmented (CID) and the fragment masses reveal the sequence
• Peptide mass fingerprinting: The pattern of peptide masses is compared to theoretical digests of known proteins in databases

The mass accuracy of modern instruments allows resolution of individual amino acids. For example, distinguishing leucine (131.17 Da) from isoleucine (131.17 Da) requires additional fragmentation methods (e.g., ETD or high-energy CID to generate $d$-ions and $w$-ions).

Sanger's Method (Historical)

Frederick Sanger developed the first complete protein sequencing method in the 1950s, determining the sequence of bovine insulin (51 residues, 2 chains). His approach used:

• 2,4-dinitrofluorobenzene (DNFB) to label the N-terminal residue
• Partial acid hydrolysis to generate overlapping fragments
• Paper chromatography to separate and identify DNP-labeled amino acids
• Assembly of overlapping fragment sequences to reconstruct the complete sequence

Sanger received the Nobel Prize in Chemistry in 1958 for this work. His method demonstrated that each protein has a unique, genetically determined amino acid sequence.

Amino Acid Composition vs. Sequence

It is important to distinguish between amino acid composition (the total number of each amino acid present, without regard to order) and amino acid sequence (the precise order from N- to C-terminus). For a peptide of $n$ residues, the number of possible unique sequences is:

For a modest 100-residue protein, $20^{100} \approx 1.27 \times 10^{130}$ possible sequences exist -- a number far exceeding the total number of atoms in the observable universe ($\sim 10^{80}$). This astronomical diversity explains how natural selection can produce proteins with extraordinarily specific functions.

Phosphorylation

The most common regulatory PTM. A phosphate group ($\text{PO}_4^{3-}$) is transferred from ATP to the hydroxyl group of Ser, Thr, or Tyr by protein kinases:

Phosphorylation introduces two negative charges at physiological pH, which can cause conformational changes, alter binding interfaces, or create docking sites for phospho-binding domains (e.g., SH2, 14-3-3). The reaction is reversed by phosphatases. Approximately 30% of all eukaryotic proteins are phosphorylated at any given time.

Glycosylation

The attachment of sugar moieties to proteins. Two major types exist:

• N-linked glycosylation: Oligosaccharide attached to the amide nitrogen of Asn in the sequence $\text{Asn-X-Ser/Thr}$ (where X is any amino acid except Pro). Occurs in the ER.
• O-linked glycosylation: Sugars attached to the hydroxyl group of Ser or Thr. Occurs primarily in the Golgi apparatus.

Glycosylation affects protein folding, stability, cell-cell recognition, and immune evasion. More than 50% of all human proteins are glycosylated.

Other Major PTMs