Part 3: Amino Acids & Protein Structure

Amino Acid Chemistry

Amino acids are the monomeric building blocks of proteins — the workhorses of biology. There are 20 standard amino acids encoded by the genetic code, each sharing a common backbone but differing in their side chain (R group), which determines their chemical personality. Understanding amino acid chemistry is essential because every aspect of protein behavior — folding, catalysis, binding, regulation — ultimately traces back to the properties of these 20 molecules.

Stereochemistry: L vs D Configuration

Except for glycine (R = H), all amino acids have a chiral α-carbon and exist as two enantiomers: L (levo) and D (dextro). Biological proteins use exclusively L-amino acids — one of the deepest mysteries in biochemistry. Why did life choose L over D? The answer remains unknown, though some theories invoke circularly polarized light from neutron stars or parity violation in the weak nuclear force.

Analogy: Think of L and D amino acids as left and right shoes. Although chemically identical in most reactions (like left and right shoes weigh the same), they behave differently in a chiral environment (your feet). Enzymes are "chiral environments" — built from L-amino acids, they only accommodate L-substrates, just as your left foot only fits a left shoe.

Classification

The 20 amino acids are classified by the chemical properties of their R groups, which determine how they interact with water, other amino acids, and substrates:

Ionization & pI

Amino acids are polyprotic acids — they have multiple ionizable groups. At different pH values, these groups gain or lose protons, changing the overall charge of the molecule:

Titration of a Simple Amino Acid (Alanine)

Alanine has two ionizable groups: α-COOH (pK_a1 = 2.34) and α-NH₃⁺ (pK_a2 = 9.69). As we titrate from low to high pH:

1. pH < 2: Both groups protonated → H₃N⁺−CHR−COOH → net charge = +1
2. pH = 2.34 (pK_a1): Half of −COOH deprotonated → midpoint of first titration plateau
3. pH = 6.01 (pI): Zwitterion → H₃N⁺−CHR−COO⁻ → net charge = 0
4. pH = 9.69 (pK_a2): Half of −NH₃⁺ deprotonated → midpoint of second titration plateau
5. pH > 10: Both groups deprotonated → H₂N−CHR−COO⁻ → net charge = −1

import numpy as np

# Amino acid ionization calculator
# Calculate charge at any pH for amino acids with known pKa values

def amino_acid_charge(pH, pKa_values):
    """Calculate net charge at given pH
    For each group: charge contribution = -1 (for COOH) or +1 (for NH3+)
    Fraction protonated = 1 / (1 + 10^(pH - pKa))
    """
    charge = 0
    for pKa, group_type in pKa_values:
        frac_protonated = 1 / (1 + 10**(pH - pKa))
        if group_type == 'acid':   # -COOH: protonated = 0 charge, deprotonated = -1
            charge += -1 * (1 - frac_protonated)
        elif group_type == 'base':  # -NH3+: protonated = +1, deprotonated = 0
            charge += +1 * frac_protonated
    return charge

# Define amino acids with pKa values
amino_acids = {
    'Alanine':    [(2.34, 'acid'), (9.69, 'base')],
    'Glutamate':  [(2.19, 'acid'), (4.25, 'acid'), (9.67, 'base')],
    'Lysine':     [(2.18, 'acid'), (8.95, 'base'), (10.53, 'base')],
    'Histidine':  [(1.82, 'acid'), (6.00, 'base'), (9.17, 'base')],
}

print("Amino Acid Charges at Different pH Values")
print("-" * 65)
print(f"{'Amino Acid':12s} | {'pH 1':>6s} | {'pH 4':>6s} | {'pH 7':>6s} | {'pH 10':>6s} | {'pI':>5s}")
print("-" * 65)

for name, pKas in amino_acids.items():
    charges = [amino_acid_charge(ph, pKas) for ph in [1, 4, 7, 10]]
    # Find pI (pH where charge is closest to 0)
    ph_range = np.arange(0, 14, 0.01)
    charges_all = [amino_acid_charge(ph, pKas) for ph in ph_range]
    pI = ph_range[np.argmin(np.abs(charges_all))]
    print(f"{name:12s} | {charges[0]:+6.2f} | {charges[1]:+6.2f} | {charges[2]:+6.2f} | {charges[3]:+6.2f} | {pI:5.2f}")

Peptide Bonds & Primary Structure

Amino acids polymerize into proteins through peptide bonds — covalent amide linkages formed by a condensation reaction between the α-carboxyl group of one amino acid and the α-amino group of the next, releasing one molecule of water.

Peptide Bond Characteristics

The peptide bond is not a simple single bond. Due to resonance between the C=O and C−N bonds, the peptide bond has approximately 40% double-bond character, which has profound structural consequences:

Primary Structure

Primary structure is simply the linear sequence of amino acids in a polypeptide chain, read from the N-terminus (free amino group) to the C-terminus (free carboxyl group). It is fully determined by the gene encoding the protein.

The primary structure dictates all higher levels of structure. A single amino acid change can have devastating consequences — as in sickle cell disease, where Glu6Val in hemoglobin's β-chain causes polymerization of deoxy-HbS, distorting red blood cells into a sickle shape.

Secondary Structure

Secondary structure refers to regular, repeating local folding patterns stabilized by hydrogen bonds between backbone N−H and C=O groups. These are not side-chain interactions — they involve only the peptide backbone, making them universal structural elements. The two major types — the α-helix and the β-sheet — were predicted by Linus Pauling and Robert Corey in 1951, before any protein crystal structure had been solved.

Alpha-Helix (α-Helix)

The α-helix is a right-handed spiral in which each backbone C=O forms a hydrogen bond with the N−H of the residue four positions ahead (i → i+4). This creates a tightly wound helix with 3.6 residues per turn and a rise of 5.4 Å per turn (1.5 Å per residue).

Beta-Sheet (β-Sheet)

β-sheets form when two or more polypeptide segments (β-strands) align side-by-side with hydrogen bonds running perpendicular to the strand direction. Each strand is nearly fully extended (3.5 Å rise per residue), and side chains alternate above and below the sheet plane.

Other Secondary Structure Elements

Beyond helices and sheets, proteins contain β-turns (tight 180° turns involving 4 residues, often with Pro and Gly), loops (irregular connectors that often form functional sites — enzyme active sites, antibody CDR loops), and 310 helices (tighter helices with i → i+3 H-bonds, less stable, found at helix termini).

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Generate a Ramachandran plot showing allowed secondary structure regions
phi_helix = np.random.normal(-57, 8, 200)
psi_helix = np.random.normal(-47, 8, 200)

phi_beta = np.random.normal(-135, 12, 150)
psi_beta = np.random.normal(135, 12, 150)

phi_left = np.random.normal(57, 10, 30)
psi_left = np.random.normal(47, 10, 30)

fig, ax = plt.subplots(1, 1, figsize=(8, 7))
ax.scatter(phi_helix, psi_helix, alpha=0.5, s=15, c='#BF092F', label='α-helix')
ax.scatter(phi_beta, psi_beta, alpha=0.5, s=15, c='#16476A', label='β-sheet')
ax.scatter(phi_left, psi_left, alpha=0.5, s=15, c='#3B9797', label='Left-handed helix')

ax.set_xlim(-180, 180)
ax.set_ylim(-180, 180)
ax.set_xlabel('φ (phi) [degrees]', fontsize=12)
ax.set_ylabel('ψ (psi) [degrees]', fontsize=12)
ax.set_title('Ramachandran Plot — Allowed Backbone Conformations', fontsize=14)
ax.axhline(y=0, color='gray', linewidth=0.5)
ax.axvline(x=0, color='gray', linewidth=0.5)
ax.legend(fontsize=10)
ax.grid(True, alpha=0.3)

plt.tight_layout()
plt.savefig('ramachandran_plot.png', dpi=150)
plt.show()
print("α-helix center: φ = -57°, ψ = -47°")
print("β-sheet center: φ = -135°, ψ = +135°")
print("Glycine: allowed in ALL regions (smallest R-group)")
print("Proline: restricted to small region (rigid ring)")

Tertiary & Quaternary Structure

Tertiary structure is the overall three-dimensional shape of a single polypeptide chain — the complete spatial arrangement of all its atoms. While secondary structure involves only backbone interactions, tertiary structure is driven primarily by interactions among side chains (R-groups), especially the burial of hydrophobic residues in the protein's interior.

Forces Stabilizing Tertiary Structure

Protein Domains & Motifs

Large proteins are often organized into independently folding units called domains (typically 50–350 residues). Domains are evolutionary units — they can be shuffled between proteins through gene recombination. Common structural motifs include:

• β-α-β motif: Two parallel β-strands connected by an α-helix (found in Rossmann fold — NAD⁺-binding)
• Greek key motif: Four antiparallel β-strands arranged like the Greek key pattern on pottery
• β-barrel: Closed cylinder of antiparallel β-strands (porins, GFP)
• Helix-turn-helix: DNA-binding motif in transcription factors
• Coiled-coil: Two α-helices wound around each other (leucine zippers, keratin)
• TIM barrel (β/α barrel): Eight parallel β-strands surrounded by eight α-helices — the most common enzyme fold

Quaternary Structure

Quaternary structure describes the arrangement of multiple polypeptide subunits (protomers) into a functional multi-subunit complex. Not all proteins have quaternary structure — only those composed of two or more chains.

Protein Folding & Chaperones

How does a polypeptide chain — a flexible polymer with an astronomically large number of possible conformations — fold into a single, precise three-dimensional structure in milliseconds to seconds? This is the protein folding problem, one of the greatest challenges in molecular biology.

Levinthal's Paradox & the Folding Funnel

Cyrus Levinthal (1969) calculated that if a 100-residue protein sampled all possible conformations randomly (3¹⁰⁰ possibilities, each taking 10⁻¹³ s), it would take longer than the age of the universe to find the native state. Yet proteins fold in microseconds to seconds. The resolution is that folding does NOT involve random searching.

Protein Misfolding & Disease

When proteins fail to fold correctly, the consequences can be devastating. Misfolded proteins can aggregate into toxic structures:

Molecular Chaperones

In the crowded cellular environment (300–400 mg/mL protein), newly synthesized polypeptides face a high risk of misfolding and aggregation. Molecular chaperones are proteins that assist folding without becoming part of the final structure. They don't provide folding information — they prevent aggregation and give polypeptides a chance to find their native state.

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Simulate a protein folding energy funnel
theta = np.linspace(0, 8 * np.pi, 500)
r = np.linspace(5, 0.3, 500)
z = np.linspace(0, -10, 500)

# Add noise to simulate roughness of energy landscape
noise = np.random.normal(0, 0.15, 500)
z_rough = z + noise * (r / 5)

fig = plt.figure(figsize=(10, 7))
ax = fig.add_subplot(111, projection='3d')
x = r * np.cos(theta)
y = r * np.sin(theta)

ax.plot(x, y, z_rough, color='#BF092F', linewidth=1.5, alpha=0.8)
ax.scatter([0], [0], [z[-1]], color='#3B9797', s=200, zorder=5, label='Native state')
ax.scatter(x[0:3], y[0:3], z_rough[0:3], color='#132440', s=100, zorder=5, label='Unfolded ensemble')

ax.set_xlabel('Conformational coordinate 1')
ax.set_ylabel('Conformational coordinate 2')
ax.set_zlabel('Free Energy (G)')
ax.set_title('Protein Folding Energy Funnel', fontsize=14)
ax.legend(fontsize=10)

plt.tight_layout()
plt.savefig('folding_funnel.png', dpi=150)
plt.show()
print("Folding funnel: broad top (many conformations) → narrow bottom (native state)")
print("Molten globule intermediate: compact but without fixed side-chain packing")
print("Folding time: microseconds (small proteins) to seconds (large multidomain)")

Post-Translational Modifications

Most proteins are not functional the instant they leave the ribosome. Post-translational modifications (PTMs) are chemical changes made to proteins after synthesis that regulate their activity, localization, interactions, and lifespan. Over 400 types of PTMs are known, and most proteins carry multiple modifications simultaneously.

import numpy as np

# Protein PTM catalog: modifications and their effects
ptm_catalog = {
    'Phosphorylation': {
        'residues': ['Ser', 'Thr', 'Tyr'],
        'mass_change_da': 79.966,
        'charge_change': -2,
        'reversible': True,
        'prevalence': '~30% of all human proteins are phosphorylated'
    },
    'N-Glycosylation': {
        'residues': ['Asn (in Asn-X-Ser/Thr motif)'],
        'mass_change_da': 'Variable (1000-3000 Da typical)',
        'charge_change': 0,
        'reversible': False,
        'prevalence': '~50% of proteins are glycosylated'
    },
    'Ubiquitination': {
        'residues': ['Lys'],
        'mass_change_da': 8565,
        'charge_change': 0,
        'reversible': True,
        'prevalence': 'Targets ~80% of short-lived proteins'
    },
    'Acetylation': {
        'residues': ['Lys', 'N-terminus'],
        'mass_change_da': 42.011,
        'charge_change': 0,
        'reversible': True,
        'prevalence': '~85% of human proteins are N-terminally acetylated'
    }
}

for ptm, info in ptm_catalog.items():
    print(f"\n{'='*50}")
    print(f"  {ptm}")
    print(f"{'='*50}")
    print(f"  Target residues  : {', '.join(info['residues'])}")
    print(f"  Mass change      : {info['mass_change_da']} Da")
    print(f"  Charge change    : {info['charge_change']:+d}" if isinstance(info['charge_change'], int) else f"  Charge change    : {info['charge_change']}")
    print(f"  Reversible       : {'Yes' if info['reversible'] else 'No'}")
    print(f"  Prevalence       : {info['prevalence']}")

# Calculate how phosphorylation changes a peptide's mass
peptide_mass = 1500.0  # Da (example small peptide)
n_phospho = 3
modified_mass = peptide_mass + (n_phospho * 79.966)
print(f"\nOriginal peptide mass: {peptide_mass:.1f} Da")
print(f"After {n_phospho}x phosphorylation: {modified_mass:.1f} Da")
print(f"Mass shift detectable by mass spectrometry: +{n_phospho * 79.966:.1f} Da")

Practice Exercises

Amino Acids & Protein Structure Worksheet

Conclusion & Next Steps

Amino acids are the molecular alphabet from which all proteins are written. From the 20 standard amino acids — with their unique side chains governing charge, polarity, size, and reactivity — polypeptide chains are assembled through rigid, planar peptide bonds. These chains then fold through a hierarchy of increasingly complex structures:

• Primary: The linear sequence — the blueprint dictated by DNA
• Secondary: Local backbone patterns (α-helices and β-sheets) stabilized by backbone H-bonds
• Tertiary: The complete 3D fold driven by hydrophobic effect, H-bonds, ionic interactions, disulfide bonds, and van der Waals forces
• Quaternary: Multi-subunit assemblies enabling cooperativity and allosteric regulation

The protein folding problem — how sequence encodes structure — has been spectacularly addressed by computational approaches like AlphaFold, but the biological machinery of chaperones and quality control remains essential for in vivo folding. Post-translational modifications add another layer of regulation, expanding the functional repertoire of the proteome far beyond what the ~20,000 human genes alone could achieve.