📖 Core Concepts
Core Concepts: Biochemistry I: Macromolecular Structure and Function
Personal study notes for Skipper. These notes are meant to be read slowly. Each concept builds on the one before it, and most of them reach back to something you have already learned in organic chemistry, physical chemistry, or cell biology.
0. A Note on Sources Before We Begin
Three of the course's key papers are listed in the syllabus with titles that differ from the published titles. You should cite them correctly:
- DOI 10.1126/science.181.4096.223 is Christian B. Anfinsen, "Principles that Govern the Folding of Protein Chains," Science 181:223–230 (1973). This is his Nobel lecture. It is not titled "The Structure of Proteins."
- DOI 10.1016/S0065-3233(08)60520-3 is Jane S. Richardson, "The Anatomy and Taxonomy of Protein Structure," Advances in Protein Chemistry 34:167–339 (1981). The syllabus title is correct; the author was missing.
- DOI 10.1016/S0022-2836(63)80023-6 is G. N. Ramachandran, C. Ramakrishnan, and V. Sasisekharan, "Stereochemistry of Polypeptide Chain Configurations," Journal of Molecular Biology 7:95–99 (1963). "Sterically allowed conformations" describes the content of the paper, but it is not the published title.
Getting citations exactly right is a habit worth building now. A wrong title can make a source impossible to find.
1. Introduction: What Is This Subject Really About?
You have spent several courses learning the rules that govern molecules. Organic chemistry taught you how electrons move. Physical chemistry taught you how free energy decides what happens and how quantum mechanics decides what molecules look like. Cell biology and molecular biology taught you that a cell is a crowded, organized machine run by proteins and nucleic acids. Biochemistry I is where these threads come together into one question:
How does a linear string of chemical building blocks turn into a three-dimensional machine that does precise chemical work?
The central idea of the course is the sequence → structure → function relationship. Anfinsen showed that ribonuclease A, once fully unfolded with urea and reduced with β-mercaptoethanol, could refold spontaneously into its active form. He concluded that the information needed for the three-dimensional structure is contained in the amino acid sequence itself:
"…the native conformation is determined by the totality of interatomic interactions and hence by the amino acid sequence, in a given environment." [Anfinsen, 1973]
Notice the closing phrase, "in a given environment." Proteins are not crystals of fixed shape. They are thermodynamic objects. Their folded state is a free-energy minimum, and that minimum depends on solvent, pH, temperature, ligands, and binding partners. The rest of the course grows out of this framing:
- Structure is the chain finding a low free-energy arrangement, limited by bond geometry (Ramachandran) and driven by solvent (the hydrophobic effect).
- Function comes from that structure being almost stable. Proteins are marginally stable, and the flexibility that comes with this lets them bind, change shape, and catalyze.
- Regulation comes from shifting the balance between conformations. This includes allostery, phosphorylation, and second messengers.
- Catalysis comes from an active site that binds the transition state more tightly than it binds the substrate.
A useful analogy is origami done by thermodynamics. The sequence is the crease pattern. Physics folds the paper. But unlike real origami, the finished piece keeps shivering. It flexes and partly unfolds, and the biology depends on that motion.
2. Concept One: Amino Acids as Tunable Acid–Base Chemistry
Twenty standard α-amino acids share a common scaffold: a central α-carbon bonded to an amino group, a carboxyl group, a hydrogen, and a variable side chain (R). All of them except glycine are chiral. Proteins use L-amino acids, which corresponds to the (S) configuration for every residue except cysteine. Cysteine is (R) because the sulfur raises the CIP priority of its side chain. This is a nice reminder from CHEM201 that D/L and R/S are different systems.
2.1 Ionization and the Henderson–Hasselbalch equation
You know this equation from CHEM102 and CHEM302. In biochemistry it becomes the main tool for predicting charge:
pH = pKa + log₁₀ ( [A⁻] / [HA] ) Fraction deprotonated: f(A⁻) = 1 / (1 + 10^(pKa − pH))
Approximate free amino acid values are α-COOH ≈ 2.0–2.4 and α-NH₃⁺ ≈ 9.0–9.8. At physiological pH, a free amino acid is therefore a zwitterion. The ionizable side chains matter most:
- Asp, Glu (carboxylates): pKa ≈ 3.9 and 4.1
- His (imidazole): pKa ≈ 6.0. This is the only side chain that titrates near physiological pH, which is why it appears so often in catalytic sites as a proton shuttle.
- Cys (thiol): pKa ≈ 8.3
- Tyr (phenol): pKa ≈ 10.1
- Lys (ε-amine): pKa ≈ 10.5
- Arg (guanidinium): pKa ≈ 12.5 (recent measurements suggest closer to 13.8). It is essentially always protonated.
Exact values differ slightly between Lehninger and Berg et al. Learn the trends and approximate magnitudes rather than memorizing decimals.
2.2 Isoelectric point
The isoelectric point (pI) is the pH at which the average net charge is zero. For a simple amino acid with two ionizable groups:
pI = (pKa₁ + pKa₂) / 2
For amino acids with an ionizable side chain, average the two pKa values that bracket the neutral species. For aspartate, this means averaging α-COOH and the side-chain COOH (pI ≈ 2.8). For lysine, it means averaging α-NH₃⁺ and the ε-NH₃⁺ (pI ≈ 9.7). For a whole protein, pI is found numerically, but the logic is the same. It is the basis of isoelectric focusing and of the first dimension of 2D gel electrophoresis, which connects back to CHEM302.
2.3 The deeper point: pKa values in proteins are not fixed
Tabulated pKa values describe free amino acids in water. Inside a protein, the local environment can shift them by several units:
- A lysine buried in a hydrophobic pocket has its pKa lowered. A low-dielectric environment destabilizes the charged form, so the group gives up its proton more easily.
- A carboxylate next to another negative charge has its pKa raised.
- The active-site lysine in acetoacetate decarboxylase has a pKa near 6. This makes it a good nucleophile for forming a Schiff base.
This is the first example of a theme that runs through the course: the protein fold is a way of tuning chemistry. The same functional group can be made more acidic, more basic, more nucleophilic, or more electrophilic depending on where the fold places it.
2.4 Beyond the twenty: post-translational modifications
The genetic code specifies about 20 amino acids, plus selenocysteine and pyrrolysine, which are encoded by recoded stop codons (recall BIOL201). The proteome is much more diverse than this because of post-translational modifications (PTMs):
- Phosphorylation (Ser, Thr, Tyr; His in bacterial two-component systems). This adds about 2 negative charges at physiological pH. It is reversible, fast, and the main currency of signaling.
- Glycosylation. N-linked glycans attach to Asn in the sequon Asn-X-Ser/Thr (X ≠ Pro). O-linked glycans attach to Ser/Thr. Glycosylation affects folding, trafficking, and recognition.
- Acetylation (Lys ε-amine; protein N-termini). This neutralizes lysine's positive charge, as in histone acetylation and chromatin opening.
- Methylation (Lys, Arg). It does not remove the charge. It changes shape and hydrogen-bonding capacity and creates binding sites for "reader" domains.
- Ubiquitination. The C-terminal Gly of ubiquitin forms an isopeptide bond with a Lys ε-amine through the E1→E2→E3 cascade. Lys48-linked chains generally target proteins to the proteasome. Lys63-linked and linear chains act as signaling scaffolds.
- SUMOylation. This is a ubiquitin-like modifier with similar enzymology. It usually regulates localization and protein–protein interactions rather than degradation.
- Others worth knowing: hydroxylation of Pro/Lys in collagen (requires ascorbate, hence scurvy), γ-carboxylation of Glu in clotting factors (requires vitamin K), and disulfide formation.
3. Concept Two: The Peptide Bond, a Flat Hinge with a Memory of Resonance
Amino acids are joined by condensation to form amides. You learned in CHEM202 that amides are the least reactive carboxylic acid derivatives because nitrogen's lone pair is delocalized into the carbonyl. In proteins, that same resonance becomes the main constraint on structure.
O O⁻
‖ |
Cα — C — N — Cα ⟷ Cα — C = N⁺ — Cα
| |
H H
Partial double-bond character ≈ 40%
C–N bond length ≈ 1.32 Å (typical C–N single ≈ 1.49 Å; C=N double ≈ 1.27 Å)
Rotation barrier ≈ 80–90 kJ/mol
3.1 Consequences of resonance
- Planarity. Six atoms lie in one plane: Cα(i), C, O, N, H, and Cα(i+1). The peptide unit behaves like a rigid card.
- Only two rotatable bonds per residue. These are φ (phi, rotation about N–Cα) and ψ (psi, rotation about Cα–C). The backbone is a chain of rigid plates hinged at each α-carbon. Each residue has two degrees of freedom, not three.
- A permanent dipole. The C=O is partially negative and the N–H partially positive. These are the hydrogen-bond donors and acceptors that build secondary structure.
- Cis/trans isomerism. The ω (omega) angle is near 180° (trans) for about 99.9% of non-proline peptide bonds. In the cis form, adjacent α-carbons clash. Proline is the exception. Its side chain is cyclized onto its own nitrogen, so both isomers of an X–Pro bond have comparable steric strain. About 5% of X–Pro bonds in folded proteins are cis, and the proportion is higher in unfolded peptides.
Why this matters kinetically: because the rotation barrier is about 80 kJ/mol, cis/trans prolyl isomerization takes seconds to minutes. In refolding experiments, it is often the slowest step. Cells have enzymes for it, the peptidyl-prolyl cis-trans isomerases (cyclophilins and FKBPs). These are the targets of the immunosuppressants cyclosporin A and FK506.
3.2 Nomenclature and directionality
Peptides are written from the N-terminus (free α-amine) to the C-terminus (free α-carboxylate). This matches the direction of ribosomal synthesis (recall BIOL201). Gly-Ala-Ser and Ser-Ala-Gly are different molecules. The backbone has polarity, just as DNA has 5′→3′ polarity.
4. Concept Three: The Ramachandran Plot, a Map of What Is Physically Possible
If each residue has two rotatable angles, φ and ψ, any backbone conformation can be described as a list of (φ, ψ) pairs. Ramachandran, Ramakrishnan, and Sasisekharan asked which pairs are allowed when atoms are modeled as hard spheres with standard van der Waals radii [Ramachandran, Ramakrishnan & Sasisekharan, 1963].
The answer is that most of the 360° × 360° space is forbidden by steric clashes between backbone atoms and the Cβ. The allowed regions form a few clusters:
Region φ (approx.) ψ (approx.) -------------------------------------------------------------- Right-handed α-helix (αR) −57° −47° Antiparallel β-sheet −139° +135° Parallel β-sheet −119° +113° Polyproline II (PPII) −75° +145° Left-handed helix (αL) +57° +47° (rare; mostly Gly)
4.1 Why this is important
The Ramachandran plot connects bond geometry to protein architecture. Helices and sheets are not arbitrary. They are the regular structures that fit inside the allowed regions while also satisfying backbone hydrogen bonding. The plot is also used every day as a validation tool. When a crystallographer or cryo-EM model builder deposits a structure in the PDB, residues in disallowed regions are flagged as probable errors.
4.2 The two special residues
- Glycine has no Cβ, so it can reach regions forbidden to other residues, including the left-handed αL region. It acts as the chain's flexible joint and is common in tight turns.
- Proline has a ring that fixes φ near −63° and has no amide N–H to donate a hydrogen bond. It breaks helices in their interior but is common in the first turn of a helix and in turns.
Analogy: think of the Ramachandran plot as a city map where most blocks are fenced off. You can wander a little within the open parks, but the roads between them are steep. Secondary structures are the neighborhoods people actually live in.
5. Concept Four: Secondary Structure, the Backbone Satisfying Itself
Pauling, Corey, and Branson predicted the α-helix and the pleated sheet in 1951. They built precise models using the planar peptide geometry they had measured from small-molecule crystal structures [Pauling, Corey & Branson, 1951]. Their key requirement was that every backbone N–H and C=O should be hydrogen-bonded wherever possible. A buried, unsatisfied polar group costs a lot of energy because it gave up hydrogen bonds to water when it was buried.
5.1 The α-helix
Hydrogen bond: C=O of residue i ··· H–N of residue i+4 Residues per turn: 3.6 Rise per residue: 1.5 Å Pitch (per turn): 3.6 × 1.5 Å = 5.4 Å Handedness: right-handed (for L-amino acids) Side chains: point outward and slightly toward the N-terminus
Some features that are less obvious:
- The helix dipole. All peptide dipoles point in roughly the same direction along the axis. Their sum gives a macroscopic dipole with partial positive charge at the N-terminus and partial negative charge at the C-terminus. Negatively charged residues (Asp, Glu) are often found near the N-terminus as "N-caps," and phosphate-binding sites often sit at helix N-termini.
- The ends are unsatisfied. The first four N–H groups and the last four C=O groups have no intrahelical partner. Helix capping motifs solve this.
- Amphipathic helices. With 3.6 residues per turn, residues i, i+3, i+4, and i+7 lie on the same face. A sequence that places hydrophobic residues at those positions produces a helix with one oily face and one polar face. On a helical wheel projection (100° per residue), this pattern is easy to see. Amphipathic helices form coiled coils (the heptad repeat abcdefg, with hydrophobic residues at a and d), insert into membrane surfaces, and pack against hydrophobic cores.
- Helix propensity. Ala, Leu, Glu, Met, and Lys favor helix formation. Pro and Gly disfavor it. β-branched residues (Val, Ile, Thr) prefer extended strands.
5.2 β-sheets
In β-strands, the chain is nearly fully extended at about 3.3–3.5 Å per residue. Side chains alternate above and below the sheet. Hydrogen bonds form between strands, not within them.
- Antiparallel: adjacent strands run in opposite directions. Hydrogen bonds are nearly linear and alternate between closely spaced pairs and widely spaced pairs.
- Parallel: strands run in the same direction. Hydrogen bonds are angled, so this arrangement is somewhat less stable. Connecting two parallel strands requires a long crossover connection, nearly always right-handed, often an α-helix (the β-α-β motif).
- Sheets are twisted (a left-handed twist when viewed along the strands) rather than flat. This twist is why β-barrels and the curved sheets of many enzymes exist [Richardson, 1981].
- Because side chains alternate faces, a strand with the pattern hydrophobic–polar–hydrophobic–polar produces one oily face and one polar face. This is the β-sheet equivalent of an amphipathic helix, and it is common in outer membrane β-barrels.
5.3 β-turns and loops
A β-turn reverses chain direction over four residues (i to i+3), with a hydrogen bond from C=O(i) to N–H(i+3). Gly and Pro are enriched in turns because of their unusual Ramachandran access. Turns and loops are usually on the surface, and they are where many binding sites, antibody epitopes, and protease cleavage sites are found.
5.4 Fibrous proteins as secondary structure scaled up
- α-Keratin: coiled coils of amphipathic helices, cross-linked by disulfides. A "permanent wave" in hair works by reducing these disulfides, reshaping the hair, and reoxidizing them.
- Collagen: not an α-helix. Each chain forms a left-handed polyproline-II-like helix with about 3 residues per turn, and three chains wind into a right-handed triple helix. Every third residue must be glycine (Gly-X-Y repeat, often Gly-Pro-Hyp), because only Gly is small enough to fit in the crowded center of the triple helix. 4-Hydroxyproline stabilizes the helix mainly through stereoelectronic (gauche) effects that pre-organize the pyrrolidine ring pucker. It was originally thought to work through water bridges. Its formation by prolyl 4-hydroxylase requires ascorbate. In osteogenesis imperfecta, a single Gly substitution can produce brittle bones.
- Elastin: extensively cross-linked (desmosine) and largely disordered. Its elasticity is mostly entropic, like rubber. Stretching it orders the chains and exposes hydrophobic residues. The chains snap back because disorder is favored. This connects directly to the polymer entropy you studied in CHEM301.
6. Concept Five: Tertiary Structure and Protein Stability, a Delicate Balance
6.1 Proteins are only marginally stable
This is one of the most counterintuitive facts in biochemistry. A folded protein contains hundreds of hydrogen bonds, many salt bridges, and thousands of van der Waals contacts. Yet the net free energy of unfolding for a typical small globular protein is only about:
ΔG°(unfold) = G(U) − G(N) ≈ +20 to +60 kJ/mol Two-state equilibrium: N ⇌ U K_unf = [U]/[N] = exp(−ΔG°_unf / RT) At 298 K with ΔG°_unf = 40 kJ/mol: K_unf ≈ exp(−16.1) ≈ 1 × 10⁻⁷
That is about the energy of two or three hydrogen bonds. Folding is a small difference between very large opposing terms. Favorable enthalpy from interactions and the large favorable solvent entropy gain of the hydrophobic effect are opposed by the large conformational entropy lost when a floppy chain is locked into one structure.
Why would evolution choose marginal stability? Partly because function needs flexibility. Allostery, induced fit, and catalysis all depend on motion. Partly because proteins need to be degradable so the cell can turn them over. And partly because selection only acts until a protein is "stable enough," so mutations drift stability down toward that threshold.
6.2 The hydrophobic effect: the dominant driving force
Kauzmann argued that the burial of nonpolar side chains away from water is the main force driving folding [Kauzmann, 1959]. The key point is that the hydrophobic effect is mainly about water, not about attractions between oil molecules.
- Water next to a nonpolar surface cannot form its usual hydrogen-bond network in all directions. It adopts more ordered arrangements around the solute, which are often loosely described as "cage-like" or clathrate-like.
- When nonpolar groups cluster together, the total nonpolar surface area shrinks and the ordered water is released. Near room temperature, ΔS of the solvent is large and positive, and this dominates.
- Transfer of nonpolar groups into water has a large positive heat capacity change (ΔCp > 0). As a result, the hydrophobic effect depends strongly on temperature. Its enthalpic and entropic parts change with temperature, and it weakens at low temperature.
A surprising consequence: cold denaturation. Because ΔCp(unfolding) is positive, a plot of ΔGunf against temperature is a downward-opening curve with a maximum. Proteins therefore unfold at high temperature and also, in principle, at low temperature. Cold denaturation is real and has been observed experimentally. This follows directly from the Gibbs–Helmholtz relationships you derived in CHEM301:
ΔG(T) = ΔH_m (1 − T/T_m) + ΔCp [ (T − T_m) − T ln(T/T_m) ]
6.3 The other contributors
- Hydrogen bonds. Each is roughly 4–20 kJ/mol in isolation. In folding, though, a protein–protein hydrogen bond mostly replaces hydrogen bonds to water, so the net gain is small. Hydrogen bonds contribute more to specificity than to raw stability. An unsatisfied buried donor or acceptor is strongly destabilizing.
- Electrostatics / salt bridges. Coulomb's law applies (recall PHYS202), but the effective dielectric constant varies from about 80 in water to about 4 in a protein core. Surface salt bridges contribute little because water screens them and desolvation costs offset them. Buried salt bridges can contribute more, but desolvating the charges is expensive.
- Van der Waals interactions. Each one is weak (about 0.4–4 kJ/mol), but protein cores are packed about as densely as organic crystals, so the total is large. Good packing is a large part of why native structures are unique.
- Disulfide bonds. These are covalent Cys–Cys links formed by oxidation. They stabilize mainly by reducing the entropy of the unfolded state. They are found mostly in extracellular proteins, because the cytosol is reducing (high glutathione).
6.4 Cooperativity
Small globular proteins usually unfold in an all-or-none, two-state manner. Plot fraction unfolded against temperature or denaturant and you see a sharp sigmoidal transition, not a gradual one. The interactions are interdependent: losing a few contacts destabilizes the neighbors, and the whole structure gives way together. It resembles a phase transition in a small system, and it ties to the partition-function thinking from CHEM301.
6.5 Domains, folds, and databases
Larger proteins are built from domains, which are compact units, often 50–250 residues, that can often fold independently. Richardson's taxonomy organized the observed structures into a limited set of architectures: all-α, all-β, α/β (alternating, like the TIM barrel and the Rossmann fold), and α+β [Richardson, 1981]. Her hand-drawn ribbon diagrams set the visual conventions still used today. The Protein Data Bank (PDB) holds experimentally determined structures. Classification databases such as SCOP and CATH group them by fold. A striking result is that the number of distinct folds seems to be limited, perhaps a few thousand, even though the number of possible sequences is enormous. Nature reuses its scaffolds.
7. Concept Six: Protein Folding: Paradox, Funnel, and Chaperones
7.1 Levinthal's paradox
Levinthal pointed out that a random search of conformational space would never finish. Here is a version of the calculation:
Assume 100 residues, 3 backbone states per residue: Conformations ≈ 3¹⁰⁰ ≈ 5 × 10⁴⁷ Sample at ~10¹³ conformations per second (bond vibration timescale): Time ≈ 5 × 10³⁴ s ≈ 10²⁷ years Real proteins fold in microseconds to seconds.
The way out is that folding is not a random search. Partial structures that are native-like are slightly more stable than non-native ones, so the search is biased. Even a small bias at each step makes the problem manageable.
7.2 The energy landscape and the folding funnel
The modern view [Dill & Chan, 1997] replaces a single "pathway" with an energy landscape shaped like a funnel. The width of the funnel represents conformational entropy, and its depth represents free energy. Unfolded chains start at the wide rim and have many possible routes downhill. As they descend, entropy decreases and favorable contacts accumulate. The surface is rough, with local minima where chains can become trapped as misfolded intermediates or molten globules.
Analogy: imagine marbles released at the rim of a bumpy, roughly conical bowl. You cannot predict any single marble's path, but nearly all of them reach the bottom. Some get stuck in small dips along the way.
7.3 Chaperones: not instructions, but protection
Anfinsen's principle says that the sequence determines the structure. Chaperones do not violate this. They contain no folding information. In the crowded cytosol, where macromolecule concentrations reach about 300 g/L, newly made chains expose hydrophobic surfaces that can aggregate. Chaperones prevent this, and they rescue chains that have become kinetically trapped.
- Hsp70 (bacterial DnaK): binds short extended hydrophobic segments. ATP hydrolysis, regulated by Hsp40 (DnaJ) and nucleotide exchange factors, cycles it between low-affinity (ATP-bound) and high-affinity (ADP-bound) states. It holds and releases chains repeatedly.
- Hsp90: acts late in folding. Its clients are often signaling proteins, including kinases and steroid hormone receptors. It is an anticancer drug target because tumor cells depend on it to stabilize mutant oncoproteins.
- GroEL/GroES (chaperonin): GroEL is two stacked heptameric rings, and each ring encloses a cavity. A substrate binds to the hydrophobic rim. ATP and GroES binding then cause a large conformational change that buries the hydrophobic residues, makes the cavity lining hydrophilic, and releases the substrate into a sealed chamber. There it can fold in isolation for about 10 seconds, like folding in a private room.
7.4 Misfolding diseases: amyloids and prions
Amyloid fibrils are a generic, highly stable structure that many unrelated polypeptides can adopt. It is a cross-β structure in which β-strands run perpendicular to the fibril axis and side chains interdigitate in a dry "steric zipper." Examples include Aβ in Alzheimer's disease, α-synuclein in Parkinson's disease, and islet amyloid polypeptide in type 2 diabetes. For many sequences, the amyloid state may be the global free-energy minimum at high concentration. The native fold may be kinetically protected rather than thermodynamically supreme. This refines Anfinsen's principle in an important way.
Prions (PrPSc) go further. The misfolded conformer acts as a template that converts native PrPC into more PrPSc. This is infectious conformational information without a nucleic acid. It is consistent with the central dogma, because no sequence information is transferred, but it reveals a form of inheritance carried by protein structure.
8. Concept Seven: Quaternary Structure and Allostery: Hemoglobin as the Model
8.1 Myoglobin versus hemoglobin
Myoglobin is a single chain with one heme. Its O₂ binding follows a simple hyperbola:
Y = [O₂] / (K_d + [O₂]) = pO₂ / (P₅₀ + pO₂) Myoglobin P₅₀ ≈ 2.8 torr Hemoglobin P₅₀ ≈ 26 torr
Hemoglobin (α₂β₂) binds O₂ cooperatively, so its binding curve is sigmoidal. Binding one O₂ increases the affinity of the remaining sites. This allows hemoglobin to load nearly fully in the lungs (about 100 torr) and unload a large fraction in tissues (about 20–40 torr). A hyperbolic binder with the same P₅₀ could not deliver as much. Recall from BIOL302 how this curve shifts during exercise.
8.2 The Hill equation
Y = pO₂ⁿ / (P₅₀ⁿ + pO₂ⁿ) Linearized: log [ Y / (1 − Y) ] = n · log(pO₂) − n · log(P₅₀) n_H = 1 no cooperativity (myoglobin) n_H > 1 positive cooperativity (hemoglobin, n_H ≈ 2.8–3.0) n_H < 1 negative cooperativity or heterogeneous sites
The Hill coefficient is an empirical measure. It is always ≤ the number of sites (4 for hemoglobin). It equals the number of sites only for perfect, infinitely cooperative binding. Hemoglobin's value of about 3 means it is highly but not perfectly cooperative.
8.3 The structural mechanism
In deoxy-Hb, the Fe²⁺ is high-spin and slightly too large to sit in the porphyrin plane, so it sits about 0.4 Å out of plane toward the proximal His (F8). When O₂ binds, the iron becomes low-spin and moves into the plane. It pulls His F8 and the F helix with it, and this movement is transmitted to the α₁β₂ interface. The tetramer then switches from the T (tense) state, which has low affinity and extra salt bridges, to the R (relaxed) state, which has high affinity [Perutz, 1970]. This is the spin-state chemistry from your CHEM303 work on ligand fields, operating as a molecular switch.
Heterotropic effectors stabilize the T state and shift the curve to the right:
- H⁺ (the Bohr effect): protonation of His146β and other groups stabilizes T-state salt bridges. Acidic, active tissue receives more O₂.
- CO₂: forms carbamates at N-termini, and also contributes H⁺ through carbonic anhydrase.
- 2,3-Bisphosphoglycerate (2,3-BPG): binds in the central cavity of T-state Hb, which is lined with positively charged residues. Fetal Hb (α₂γ₂) has Ser in place of His143, binds BPG weakly, and so has higher O₂ affinity. This is how O₂ moves from mother to fetus.
8.4 Two models of allostery
MWC (concerted / symmetry) model [Monod, Wyman & Changeux, 1965]:
- The protein exists in an equilibrium between T and R states even when no ligand is bound (L = [T₀]/[R₀]).
- All subunits switch together, so symmetry is preserved. Hybrid states such as one R subunit with three T subunits are not allowed.
- The ligand binds R more tightly (c = KR/KT < 1 for dissociation constants) and therefore shifts the equilibrium toward R. This is conformational selection.
α(1 + α)ⁿ⁻¹ + L·c·α(1 + c·α)ⁿ⁻¹
Y = ───────────────────────────────────── α = [S]/K_R
(1 + α)ⁿ + L(1 + c·α)ⁿ
KNF (sequential / induced-fit) model [Koshland, Némethy & Filmer, 1966]:
- Ligand binding induces a conformational change in the subunit it binds to.
- That change alters the interfaces with neighboring subunits and changes their affinity.
- Hybrid states are allowed. Because interactions can make neighbors more or less receptive, this model can explain negative cooperativity, which MWC cannot.
Which is right? Hemoglobin fits MWC reasonably well, though careful studies show deviations. Many systems combine both. In modern terms, the two models are limiting cases of a general ensemble in which ligands reshape the distribution of pre-existing conformations. Allostery is a population shift.
9. Concept Eight: Enzyme Catalysis: Binding the Transition State
9.1 The magnitude of the problem
Enzymes accelerate reactions by factors of 10⁶ to 10¹⁷. Orotidine 5′-monophosphate decarboxylase has an uncatalyzed half-life of about 78 million years. It accelerates its reaction about 10¹⁷-fold, and the enzyme turns substrate over in milliseconds [Radzicka & Wolfenden, 1995]. Catalysts do not change ΔG° or Keq (recall CHEM102 and CHEM301). They lower ΔG‡ by providing a different pathway.
9.2 Rate enhancement and free energy
From transition-state theory (CHEM303):
k = (k_B T / h) · exp(−ΔG‡ / RT) Rate enhancement: k_cat / k_uncat = exp(ΔΔG‡ / RT) At 298 K: RT ln 10 ≈ 5.7 kJ/mol → every 5.7 kJ/mol of transition-state stabilization gives a 10-fold rate increase → a 10¹⁰ enhancement requires only ≈ 57 kJ/mol
About 57 kJ/mol is roughly the energy of a few hydrogen bonds and electrostatic interactions. Enzymes reach enormous rate enhancements by placing modest, ordinary interactions in exactly the right positions.
9.3 Pauling's insight: complementarity to the transition state
Pauling proposed that enzymes are complementary not to the substrate but to the transition state [Pauling, 1946]. If an enzyme bound the substrate tightly, it would trap it in a low-energy well and raise the barrier for the step that follows. Binding energy should be spent where it lowers the activation barrier, which means at the transition state. Jencks developed this into the idea that intrinsic binding energy is "used" to pay for strain, desolvation, and loss of entropy, and is only fully released at the transition state [Jencks, 1975].
Practical consequences:
- Transition-state analogs are extremely potent inhibitors. Examples include peptide boronic acids for serine proteases and statine-containing inhibitors for aspartyl proteases, as in HIV protease drugs.
- Catalytic antibodies raised against transition-state analogs show measurable catalysis.
9.4 The catalytic toolkit
- General acid–base catalysis: proton transfer within the transition state, often by His, Asp, Glu, or Lys. It avoids building up unstable charged intermediates. You saw specific acid/base catalysis in CHEM202. General catalysis uses groups other than H₃O⁺ and OH⁻ and so works at neutral pH.
- Covalent catalysis: a transient covalent bond forms between enzyme and substrate, such as an acyl-enzyme (serine proteases), a Schiff base (PLP enzymes, aldolase), or a thioester (cysteine proteases).
- Metal ion catalysis: Lewis-acid activation, generation of a bound hydroxide nucleophile (Zn²⁺ in carbonic anhydrase lowers the pKa of bound water from about 15.7 to about 7), charge shielding, and redox.
- Proximity and orientation: holding reactants at the correct distance and angle. This converts a bimolecular encounter into an "intramolecular" reaction. Effective molarities can reach far above anything achievable in solution.
- Electrostatic preorganization: the active site is a pre-arranged dipolar environment that already "expects" the charge distribution of the transition state. Warshel has argued that this is the largest single contributor in many enzymes.
9.5 The paradigm mechanism: chymotrypsin and the catalytic triad
Chymotrypsin hydrolyzes peptide bonds on the C-terminal side of large hydrophobic residues (Phe, Tyr, Trp). These residues fit its deep hydrophobic S1 pocket. The mechanism combines nearly every principle above.
Catalytic triad: Asp102 ··· His57 ··· Ser195
ACYLATION
1. Substrate binds; aromatic side chain enters the S1 pocket.
2. His57 (general base) removes the proton from Ser195-OH;
Ser195-O attacks the scissile carbonyl carbon.
Asp102 orients His57 and stabilizes its imidazolium form.
3. Tetrahedral intermediate (oxyanion) forms. The O⁻ is stabilized
by backbone N–H groups of Gly193 and Ser195 → the OXYANION HOLE.
4. His57-H⁺ (now a general acid) protonates the leaving amine;
C–N bond breaks → ACYL-ENZYME + released amine (C-terminal fragment).
DEACYLATION
5. Water enters; His57 deprotonates it; OH⁻ attacks the acyl carbon.
6. Second tetrahedral intermediate, again stabilized by oxyanion hole.
7. Collapse releases the carboxylic acid (N-terminal fragment);
Ser195 regenerated.
Experimental evidence:
- Burst kinetics with p-nitrophenyl acetate: a rapid initial "burst" of p-nitrophenolate, equal to the amount of enzyme, is followed by a slower steady state. This shows that a fast acylation is followed by rate-limiting deacylation and is direct kinetic evidence for the acyl-enzyme [Hartley & Kilby, 1954].
- DIPF (diisopropyl fluorophosphate) labels only Ser195. This shows that the serine is unusually reactive, made so by its environment rather than by any intrinsic property.
Specificity by pocket design: trypsin has Asp189 at the base of its S1 pocket and cleaves after Lys and Arg. Elastase has Val216 and Thr226, which narrow the pocket, so it cleaves after small residues such as Ala and Gly. The catalytic machinery is the same, and the S1 pocket determines specificity. Subtilisin, a bacterial enzyme with an unrelated fold, uses the same triad. This is convergent evolution arriving at one chemical solution twice.
The other protease families use the same problem-solving logic with different nucleophiles:
- Cysteine proteases (papain, caspases, cathepsins): a Cys–His dyad. A thiolate is the nucleophile and forms a thioester acyl-enzyme.
- Aspartyl proteases (pepsin, renin, HIV protease): two Asp residues activate a water molecule directly. There is no covalent intermediate.
- Metalloproteases (thermolysin, carboxypeptidase A, MMPs): a Zn²⁺-bound water acts as the nucleophile, and the metal stabilizes the oxyanion.
10. Concept Nine: Michaelis–Menten Kinetics: The Mathematics of Saturation
10.1 Derivation (steady-state, Briggs–Haldane)
Michaelis and Menten [Michaelis & Menten, 1913] assumed rapid equilibrium binding. Briggs and Haldane [Briggs & Haldane, 1925] generalized this with the steady-state assumption, which is the version you should know:
k₁ k_cat
E + S ⇌ ES ─────→ E + P
k₋₁
Assumptions: (i) [S] ≫ [E]_T so [S] ≈ [S]₀
(ii) initial rates, so reverse reaction from P is negligible
(iii) steady state: d[ES]/dt = 0
d[ES]/dt = k₁[E][S] − (k₋₁ + k_cat)[ES] = 0
Conservation: [E]_T = [E] + [ES]
Solve: [ES] = [E]_T [S] / (K_M + [S]), where K_M ≡ (k₋₁ + k_cat) / k₁
v₀ = k_cat[ES] = V_max [S]
───────────── with V_max = k_cat [E]_T
K_M + [S]
This is the same steady-state approximation you used in CHEM303 for Lindemann unimolecular kinetics. The mathematics is shared, and only the chemical context differs.
10.2 What the parameters actually mean
- Vmax: the rate when all enzyme is saturated. It depends on how much enzyme is present, so it is not an intrinsic property of the enzyme.
- kcat (turnover number): Vmax/[E]T, the number of substrate molecules converted per active site per second at saturation. It ranges from about 0.5 s⁻¹ (lysozyme) to about 10⁶ s⁻¹ (carbonic anhydrase).
- KM: the [S] at which v₀ = Vmax/2. It is equal to the dissociation constant KS = k₋₁/k₁ only when kcat ≪ k₋₁. In general it is not a pure measure of affinity. A useful interpretation is that it is roughly the substrate concentration around which the enzyme responds most sensitively. Cellular [S] is often near KM.
- kcat/KM (specificity constant): the apparent second-order rate constant for E + S → E + P at low [S]. When several substrates compete for the same enzyme, the ratio of their rates equals the ratio of their kcat/KM values. This is what "specificity" means quantitatively.
At [S] ≪ K_M: v₀ ≈ (k_cat / K_M) [E]_T [S] (second-order) Upper limit: k_cat/K_M ≤ k₁ ≈ 10⁸ – 10⁹ M⁻¹ s⁻¹ (diffusion-controlled encounter)
Catalytic perfection: enzymes such as triose phosphate isomerase, fumarase, and superoxide dismutase have kcat/KM near the diffusion limit. Every encounter with substrate leads to product. Evolution can no longer improve them chemically. Only the rate of diffusion, which you can estimate from the Smoluchowski equation in physical chemistry, limits them.
10.3 Measuring the parameters
Measure initial rates at a range of [S], ideally from about 0.2 KM to about 5 KM, then fit the hyperbola by nonlinear regression. Your MATH201 training applies here: weighting and error structure matter. The Lineweaver–Burk double-reciprocal plot is still taught because it makes inhibition patterns easy to see:
1/v₀ = (K_M / V_max)(1/[S]) + 1/V_max slope = K_M/V_max y-intercept = 1/V_max x-intercept = −1/K_M
It distorts errors badly, however. Taking reciprocals magnifies the noise in low-[S] points, which are the least precise. Use it for visualization, not for fitting.
10.4 Enzyme inhibition: a unified framework
Reversible inhibitors can bind free enzyme (E), the ES complex, or both. Define:
α = 1 + [I]/K_I (inhibitor binds free E; K_I = [E][I]/[EI])
α′ = 1 + [I]/K_I′ (inhibitor binds ES; K_I′ = [ES][I]/[ESI])
General (mixed) rate law:
V_max [S]
v₀ = ──────────────────
α K_M + α′ [S]
Apparent parameters: V_max,app = V_max / α′ K_M,app = α K_M / α′
| Type | Binds | Apparent Vmax | Apparent KM | Lineweaver–Burk pattern |
|---|---|---|---|---|
| Competitive | E only (α′ = 1) | unchanged | increases (αKM) | lines intersect on y-axis |
| Uncompetitive | ES only (α = 1) | decreases | decreases | parallel lines |
| Mixed | both (α ≠ α′) | decreases | increases or decreases | intersect left of y-axis |
| Noncompetitive (pure) | both, equally (α = α′) | decreases | unchanged | intersect on x-axis |
IC₅₀ vs. KI: IC₅₀ is the inhibitor concentration that halves activity under specific assay conditions. KI is a true equilibrium constant. For a competitive inhibitor, the Cheng–Prusoff relationship links them:
IC₅₀ = K_I (1 + [S]/K_M) (competitive)
An IC₅₀ reported without the substrate concentration cannot be compared across assays. Dixon plots (1/v vs. [I] at several fixed [S]) give −KI from the intersection point for competitive inhibitors.
Physical intuition: a competitive inhibitor is overcome by flooding the system with substrate, so Vmax is unchanged. An uncompetitive inhibitor locks ES. Removing ES pulls the binding equilibrium forward, which lowers KM,app, but the trapped ES cannot turn over, which lowers Vmax.
11. Concept Ten: Coenzymes, Nucleic Acids, and Signals, Extending the Chemistry
11.1 Coenzymes: borrowing chemistry that amino acids lack
The 20 side chains do acid–base chemistry and nucleophilic chemistry well. They cannot easily carry electrons in pairs, stabilize carbanions, or move one-carbon units. Coenzymes, often derived from vitamins, fill these gaps:
- NAD⁺/NADH, NADP⁺/NADPH (niacin): hydride (H⁻) transfer at C4 of the nicotinamide ring. Two-electron redox. NADH absorbs at 340 nm, which is the basis of countless spectrophotometric assays (CHEM302).
- FAD/FADH₂, FMN (riboflavin): one- or two-electron redox. These link two-electron metabolism to one-electron carriers such as iron–sulfur clusters and quinones.
- Coenzyme A (pantothenate): forms acyl thioesters. Thioesters are activated toward both nucleophilic acyl substitution and α-carbon enolization, because sulfur's poor 3p–2p overlap gives weaker resonance than in oxygen esters. This is CHEM202 chemistry applied in metabolism.
- Pyridoxal phosphate (PLP) (B₆): forms a Schiff base with amino acids and acts as an "electron sink" that stabilizes carbanions at Cα. This allows transamination, decarboxylation, and racemization. Which bond breaks is controlled by orienting it perpendicular to the π-system (Dunathan's stereoelectronic hypothesis).
- Thiamine pyrophosphate (TPP) (B₁): its thiazolium C2 is acidic and forms a carbanion (an ylide) that adds to carbonyls and acts as an acyl anion equivalent. This is an "umpolung" catalyst used in pyruvate dehydrogenase and transketolase.
- Biotin: carries activated CO₂ (as carboxybiotin) for carboxylations.
- Tetrahydrofolate (THF): carries one-carbon units at several oxidation levels (methyl, methylene, formyl). In thymidylate synthase, N⁵,N¹⁰-methylene-THF donates a CH₂ and a hydride to convert dUMP into dTMP, and in the process it becomes dihydrofolate. 5-Fluorouracil (as FdUMP) traps the enzyme covalently, and methotrexate blocks the dihydrofolate reductase that regenerates THF. Both are cancer chemotherapies that target this cycle.
11.2 Nucleic acid structure
B-DNA [Watson & Crick, 1953] is a right-handed double helix with about 10.5 base pairs per turn, a rise of about 3.4 Å per base pair, and a diameter of about 20 Å. The glycosidic bonds of a base pair are not diametrically opposite, so the backbones are closer on one side. This creates a wide, shallow major groove and a narrow minor groove. The major groove displays a distinctive pattern of hydrogen-bond donors, acceptors, and methyl groups for each base pair (A·T, T·A, G·C, C·G). Proteins can "read" the sequence there without unwinding the helix, using α-helices inserted into the groove, as in helix-turn-helix, zinc finger, and leucine zipper proteins. The stability of the duplex comes largely from base stacking (van der Waals and hydrophobic effects), not only from Watson–Crick hydrogen bonds. RNA, with its 2′-OH, prefers the A-form and folds into complex tertiary structures, as in tRNA, ribozymes, and the ribosome. Recall from BIOL201 that the ribosome's peptidyl transferase center is RNA.
11.3 Signal transduction: allostery and covalent modification at scale
Signaling uses the concepts above to carry information. A ligand binds a receptor (conformational selection). The receptor activates an effector, for example a G protein exchanging GDP for GTP, which acts as a timer. Effectors produce second messengers (cAMP, cGMP, IP₃, DAG, Ca²⁺) that allosterically activate protein kinases (PKA, PKC, CaMKII). Kinases phosphorylate targets, and phosphatases reverse the modification. Each enzymatic step multiplies the signal, so one hormone molecule can lead to millions of product molecules. This amplification is how a nanomolar hormone signal produces a large physiological response, as you saw in BIOL302. Kinases themselves are regulated by an activation loop whose phosphorylation shifts their conformational equilibrium, which is allostery again.
12. Connections to Peptide Chemistry
Everything in this course has a mirror image in the laboratory. That is where chemists make, modify, and study peptides directly. These connections point toward advanced courses in chemical biology, peptide synthesis, and drug discovery.
12.1 Building peptides: the chemistry behind the amide bond
Forming an amide from an acid and an amine is thermodynamically uphill in water and slow at ambient temperature. The ribosome solves this by using aminoacyl-tRNA esters, which are activated acyl donors. Chemists solve it with coupling reagents such as carbodiimides (DCC, DIC), uronium/aminium salts (HBTU, HATU), and phosphonium salts (PyBOP). These convert the carboxylic acid into an active ester. The logic is CHEM202 acyl substitution, made efficient.
Solid-phase peptide synthesis (SPPS) [Merrifield, 1963] anchors the C-terminal residue to an insoluble resin and adds amino acids one at a time. Note that synthesis runs C → N, the opposite of ribosomal synthesis (N → C). Excess reagents are washed away after each step. This requires orthogonal protecting groups:
- Fmoc on the α-amine (removed by base, piperidine), with acid-labile side-chain groups (tBu, Trt, Boc, Pbf), all removed at the end with TFA.
- The older Boc/Bzl strategy (Boc removed with TFA, final cleavage with HF).
The yields add up quickly. At 99% efficiency per step, a 50-mer is made in only about 0.99100 ≈ 37% overall yield, counting coupling plus deprotection as two steps per residue. This compounding loss is why long proteins are made by joining fragments.
12.2 Racemization: the danger of activating the carboxyl
Activating the C-terminal carbonyl of an N-acyl amino acid can cause cyclization to an oxazolone (azlactone). Its Cα–H is acidic, which allows epimerization. This is why urethane protecting groups such as Fmoc and Boc are used. They suppress oxazolone formation, and additives such as HOBt and Oxyma suppress it further. Cys and His are especially prone to racemization. Compare this to how carefully the ribosome and the aminoacyl-tRNA synthetases preserve L-chirality.
12.3 Native chemical ligation: joining unprotected fragments
Native chemical ligation [Dawson, Muir, Clark-Lewis & Kent, 1994] joins a peptide C-terminal thioester to a peptide with an N-terminal cysteine. Transthioesterification is followed by a spontaneous S→N acyl shift through a five-membered transition state, giving a native peptide bond. This is proximity catalysis (Concept Eight) built into a synthetic method. It has the same thioester logic as CoA and as the acyl-enzymes of cysteine proteases. Nature's own version, intein splicing, uses a related N→S (or N→O) acyl shift.
12.4 Peptide conformation by design
- Short linear peptides are usually disordered in water, because one helical turn cannot pay its entropic cost. Stapled peptides (hydrocarbon cross-links at i and i+4 or i and i+7, which sit on the same helix face) and lactam bridges lock helices in place, increasing protease resistance and target affinity.
- Cyclization restricts conformational entropy, as disulfides do in natural proteins. Cyclosporin A and many natural product peptides are cyclic, often N-methylated. N-methylation removes N–H donors and can greatly improve membrane permeability.
- D-amino acids and Aib (α-aminoisobutyric acid, which strongly favors helices) reach Ramachandran regions that natural L-residues rarely use. They also resist proteolysis.
- Peptidomimetics and transition-state isosteres underpin HIV protease inhibitors, the boronic acid bortezomib (a proteasome inhibitor), and the nitrile warhead of nirmatrelvir (a covalent SARS-CoV-2 Mpro inhibitor). Each is a direct application of Pauling's transition-state complementarity.
12.5 Reading peptides: sequencing
Edman degradation [Edman, 1950] uses phenyl isothiocyanate (PITC) to label the N-terminal amine. Mild acid then cleaves off the first residue as a thiazolinone, which rearranges to a phenylthiohydantoin (PTH) derivative identified by HPLC. Repetitive yield limits the method to about 30–50 residues, and a blocked N-terminus (for example, acetylated) stops it entirely. Proteins are therefore first fragmented by specific cleavage: trypsin after Lys/Arg (not before Pro), chymotrypsin after aromatic residues, cyanogen bromide (CNBr) after Met. The sequence is then assembled from overlapping fragments.
Today, tandem mass spectrometry (MS/MS) dominates (connect to CHEM302). Collision-induced dissociation breaks peptide bonds to give b-ions (N-terminal fragments) and y-ions (C-terminal fragments). Mass differences between consecutive ions identify residues. Leu and Ile, both 113.08 Da, cannot be distinguished by low-energy CID. MS also detects PTMs as characteristic mass shifts: +80 Da for phosphorylation, +42 Da for acetylation, +14 Da for methylation, and +114 Da for the Gly-Gly remnant left on lysine after trypsin digestion of ubiquitinated proteins.
13. Common Misconceptions to Watch Out For
- "Hydrogen bonds are the main force holding proteins together." The hydrophobic effect dominates. Protein hydrogen bonds mostly replace hydrogen bonds to water, so their net contribution to stability is modest. They matter enormously for specificity, and a buried, unsatisfied hydrogen bond is highly destabilizing.
- "The hydrophobic effect is an attraction between nonpolar groups." It is mostly a property of water. Near room temperature, it is driven by the entropy gained when ordered water around nonpolar surfaces is released. Direct van der Waals contacts between nonpolar groups contribute, but they are not the main source.
- "Proteins are rigid structures." Crystal structures are time- and ensemble-averaged snapshots. Proteins are dynamic ensembles with motions from femtoseconds (bond vibrations) to seconds (domain movements and prolyl isomerization). Intrinsically disordered proteins, about 30% of eukaryotic proteins having substantial disordered regions, have no single fixed structure at all.
- "KM is the dissociation constant of the ES complex." It equals KS only when kcat ≪ k₋₁. In general KM = (k₋₁ + kcat)/k₁. A low KM does not prove tight binding.
- "Lineweaver–Burk plots are a good way to determine KM and Vmax." They are useful for visualization, but they distort errors severely. Fit the untransformed data by nonlinear regression.
- "An enzyme with a higher kcat is always the better catalyst." At physiological (often sub-saturating) substrate concentrations, kcat/KM governs the rate. Specificity between competing substrates depends on kcat/KM, not on kcat alone.
- "Enzymes bind their substrates as tightly as possible." Tight substrate binding would be counterproductive, because it creates a thermodynamic well that raises the barrier to the transition state. Enzymes are optimized to bind the transition state tightly [Pauling, 1946; Jencks, 1975].
- "Catalysts change the equilibrium." They do not. Enzymes accelerate the forward and reverse reactions by the same factor. ΔG° and Keq are unchanged.
- "Noncompetitive inhibition means the inhibitor binds somewhere other than the active site." The classification is kinetic (how KM and Vmax change), not structural. Pure noncompetitive inhibition (α = α′) is actually rare. Most such inhibitors are mixed. An allosteric inhibitor can show competitive kinetics if its binding prevents substrate binding.
- "Chaperones tell proteins how to fold." Chaperones contain no structural information. They prevent aggregation and rescue kinetically trapped chains. The native structure is still encoded in the sequence [Anfinsen, 1973].
- "The native state is always the global free-energy minimum." This is usually true for a single chain at low concentration under physiological conditions. At high concentration, amyloid aggregates may be more stable. Some proteins, such as serpins, are trapped in metastable folds that are functionally important. Anfinsen's "thermodynamic hypothesis" is an excellent first approximation, not an absolute law.
- "Collagen is an α-helix" or "collagen chains are α-helical." Each collagen chain is a left-handed polyproline-II-type helix (about 3 residues per turn), and three chains wind into a right-handed superhelix. The term "α-chain" in collagen nomenclature refers to the chain type, not to the α-helix.
- "The peptide bond can rotate freely like any single bond." Resonance gives it about 40% double-bond character and a rotation barrier of about 80 kJ/mol. It is planar and almost always trans, except for some X–Pro bonds.
- "A Hill coefficient of 2.8 means 2.8 sites are occupied cooperatively." The Hill coefficient is an empirical measure of cooperativity. Hemoglobin has 4 sites, and nH ≈ 2.8–3.0 means strong but imperfect cooperativity. It is not a count of interacting sites.
- "The pKa of a side chain is a fixed property." Protein environments shift pKa values by several units. This tuning is one of the main tools enzymes use, as in the low-pKa lysine of acetoacetate decarboxylase and the Zn²⁺-activated water of carbonic anhydrase.
14. Closing Reflection: The Unifying Idea
Looking back over these concepts, one theme connects them: proteins are thermodynamic and kinetic systems whose behavior emerges from weak interactions arranged in precise positions. No single bond in a folded protein is special. The covalent chemistry is the ordinary amide chemistry you learned in organic chemistry. What is remarkable is the arrangement: a polar group placed exactly here, an oily pocket exactly there, a histidine at exactly the right distance from a serine. The fold turns ordinary chemistry into extraordinary chemistry.
The protein is a device for placing functional groups in space with sub-ångström precision, and then letting them move just enough.
As you move into Biochemistry II (metabolism), notice that every pathway is a sequence of the catalytic strategies here: covalent catalysis, coenzyme chemistry, and allosteric regulation. They are combined into networks and regulated through the same population-shift logic that governs hemoglobin.
References Cited
- Anfinsen, C. B. (1973). Principles that govern the folding of protein chains. Science, 181, 223–230. DOI: 10.1126/science.181.4096.223
- Berg, J. M., Tymoczko, J. L., Gatto, G. J., & Stryer, L. (2019). Biochemistry (9th ed.). W. H. Freeman.
- Briggs, G. E., & Haldane, J. B. S. (1925). A note on the kinetics of enzyme action. Biochemical Journal, 19, 338–339.
- Dawson, P. E., Muir, T. W., Clark-Lewis, I., & Kent, S. B. H. (1994). Synthesis of proteins by native chemical ligation. Science, 266, 776–779.
- Dill, K. A., & Chan, H. S. (1997). From Levinthal to pathways to funnels. Nature Structural Biology, 4, 10–19.
- Edman, P. (1950). Method for determination of the amino acid sequence in peptides. Acta Chemica Scandinavica, 4, 283–293.
- Hartley, B. S., & Kilby, B. A. (1954). The reaction of p-nitrophenyl esters with chymotrypsin and insulin. Biochemical Journal, 56, 288–297.
- Jencks, W. P. (1975). Binding energy, specificity, and enzymic catalysis: the Circe effect. Advances in Enzymology and Related Areas of Molecular Biology, 43, 219–410.
- Kauzmann, W. (1959). Some factors in the interpretation of protein denaturation. Advances in Protein Chemistry, 14, 1–63.
- Koshland, D. E., Némethy, G., & Filmer, D. (1966). Comparison of experimental binding data and theoretical models in proteins containing subunits. Biochemistry, 5, 365–385.
- Merrifield, R. B. (1963). Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. Journal of the American Chemical Society, 85, 2149–2154.
- Michaelis, L., & Menten, M. L. (1913). Die Kinetik der Invertinwirkung. Biochemische Zeitschrift, 49, 333–369.
- Monod, J., Wyman, J., & Changeux, J.-P. (1965). On the nature of allosteric transitions: a plausible model. Journal of Molecular Biology, 12, 88–118.
- Nelson, D. L., & Cox, M. M. (2021). Lehninger Principles of Biochemistry (8th ed.). W. H. Freeman.
- Pauling, L. (1946). Molecular architecture and biological reactions. Chemical and Engineering News, 24, 1375–1377.
- Pauling, L., Corey, R. B., & Branson, H. R. (1951). The structure of proteins: two hydrogen-bonded helical configurations of the polypeptide chain. Proceedings of the National Academy of Sciences USA, 37, 205–211.
- Perutz, M. F. (1970). Stereochemistry of cooperative effects in haemoglobin. Nature, 228, 726–739.
- Radzicka, A., & Wolfenden, R. (1995). A proficient enzyme. Science, 267, 90–93.
- Ramachandran, G. N., Ramakrishnan, C., & Sasisekharan, V. (1963). Stereochemistry of polypeptide chain configurations. Journal of Molecular Biology, 7, 95–99. DOI: 10.1016/S0022-2836(63)80023-6
- Richardson, J. S. (1981). The anatomy and taxonomy of protein structure. Advances in Protein Chemistry, 34, 167–339. DOI: 10.1016/S0065-3233(08)60520-3
- Watson, J. D., & Crick, F. H. C. (1953). Molecular structure of nucleic acids: a structure for deoxyribose nucleic acid. Nature, 171, 737–738.
⚙️ Mechanisms & Applications
Mechanisms & Applications: From Molecular Structure to Biological Action
Skipper, the organizing question for this section is not simply “What does this molecule look like?” It is “How does its structure reshape the probabilities of binding, reaction, folding, and regulation?” Your organic chemistry explains electron movement; physical chemistry explains energy landscapes and rates; molecular biology explains biological context. Biochemistry brings these descriptions together. A protein is simultaneously a chemical substance, a fluctuating physical object, and a participant in a cellular system.
A useful habit throughout: separate thermodynamics—which states are favored—from kinetics—how rapidly states interconvert. An enzyme, a chaperone, and an allosteric ligand influence different parts of this picture. Confusing their roles is the source of many plausible but incorrect explanations.
1. Ionization: The First Mechanistic Decision
Before drawing a catalytic mechanism or predicting peptide solubility, ask which groups are protonated. Protonation determines charge, nucleophilicity, hydrogen-bonding behavior, and often biological activity. The familiar amino-acid structures in a table are not permanent molecular identities; they represent members of a pH-dependent ensemble. [1,2]
HA ⇌ H⁺ + A⁻ pH = pKa + log₁₀([A⁻]/[HA]) Fraction deprotonated = 1 / (1 + 10^(pKa − pH)) Fraction protonated = 1 / (1 + 10^(pH − pKa))
These concentration expressions assume approximately ideal behavior or appropriately defined conditional constants. For a histidine side chain with pKa = 6.0, about 9% is protonated at pH 7.0. But 6.0 is not a universal property of every histidine: nearby charges, solvent exposure, and hydrogen bonds can shift its pKa substantially. A catalytic histidine works partly because its protein environment tunes the accessibility of its protonated and unprotonated states.
- Identify all ionizable groups. Include free termini, side chains, and relevant modifications.
- Assign plausible pKa values. Free-amino-acid values are starting estimates, not exact protein values.
- Calculate fractional charges. An acidic group contributes approximately −fdeprotonated; a basic group contributes +fprotonated.
- Sum the contributions. For independent sites, the resulting charge function Q(pH) estimates the mean net charge.
- Find the isoelectric point. Solve Q(pH) = 0. Averaging two pKa values works only when those values appropriately bracket the neutral species; averaging every pKa does not.
Application: pH selection is central to peptide purification and formulation. Near a protein’s pI, reduced electrostatic repulsion often increases aggregation risk, although solubility is not determined by net charge alone. Ion-exchange chromatography exploits charge differences; isoelectric focusing separates proteins according to pI. A formulation scientist must also consider chemical degradation: changing pH can improve solubility while accelerating deamidation or hydrolysis.
Connection to CHEM301: a pKa shift is an energetic statement. At 298 K, shifting an acid dissociation pKa by one unit changes its standard dissociation free energy by approximately 2.303RT ≈ 5.7 kJ mol−1. A modest local interaction can therefore change protonation populations tenfold.
2. Building and Modifying Peptides: Chemistry Becomes a Manufacturing Process
The peptide bond is an amide. Resonance gives the C–N bond partial double-bond character, making the peptide unit approximately planar and strongly restricting rotation about that bond. Most backbone flexibility instead resides in φ, rotation about N–Cα, and ψ, rotation about Cα–C′. The peptide-bond dihedral ω is usually near 180° for trans; cis peptide bonds are uncommon, but appreciably more frequent before proline. [1,2]
A crucial distinction from introductory reaction schemes: cells do not efficiently make proteins by allowing free amino acids to condense in water. Translation uses chemically activated aminoacyl-tRNAs. The growing chain is transferred from an ester-linked donor to the amino group of the next aminoacyl-tRNA. The ribosome organizes this chemistry; ATP expenditure during tRNA charging establishes the activated substrates.
Industrial peptide synthesis solves the same activation problem differently.
One cycle of Fmoc solid-phase peptide synthesis: 1. Begin with a protected growing peptide anchored through its C terminus to an insoluble resin. 2. Remove the N-terminal Fmoc protecting group. → Expose a reactive α-amino group. 3. Activate the carboxyl group of the incoming Fmoc-protected amino acid. 4. Couple: resin–peptide–NH₂ + activated amino acid → resin–peptide–NH–CO–CH(R)–NH–Fmoc 5. Wash away soluble reagents and by-products. 6. Repeat; finally cleave from the resin and remove side-chain protections using chemistry appropriate to the resin and protecting-group scheme.
Conventional solid-phase synthesis therefore assembles the sequence from C terminus toward N terminus, whereas ribosomal synthesis proceeds N to C. Robert Bruce Merrifield’s 1963 introduction of solid-phase peptide synthesis made repeated coupling-and-washing cycles practical and ultimately automatable. [10]
Idealized fraction of correctly coupled product after n couplings: F = yⁿ For 99% yield per coupling and 50 couplings: F = 0.99⁵⁰ ≈ 0.61
This calculation assumes independent, identical couplings and ignores other losses. Its lesson is industrially important: excellent individual reactions can still generate a difficult final mixture. Incomplete coupling, epimerization, oxidation, and sequence-dependent side reactions make purification and analytical verification indispensable.
Therapeutic applications: peptide design often balances affinity against proteolysis, renal clearance, solubility, and manufacturing complexity. Cyclization, selected D-amino-acid substitutions, N-methylation, terminal modification, and lipidation can improve particular properties—but may also disrupt receptor recognition. The long action of lipidated GLP-1 analogues illustrates how modifying pharmacokinetics can be as important as improving receptor binding.
Post-translational modifications are chemically distinct control mechanisms:
- Phosphorylation: protein kinases transfer a phosphoryl group, usually from ATP, to an acceptor such as serine, threonine, or tyrosine. The resulting charge and recognition surface can alter conformation or recruit binding partners.
- Lysine acetylation: converts a usually positively charged amino group into a neutral amide. Lysine methylation, by contrast, generally preserves positive charge while changing recognition and steric properties.
- Glycosylation: adds heterogeneous carbohydrate structures that can influence folding, trafficking, circulation time, and immune recognition. Glycoform control is consequently a major issue in recombinant biologic manufacture.
- Ubiquitination: an ATP-dependent E1 activation step, E2 conjugation step, and E3-mediated substrate-selection step typically attach ubiquitin’s C terminus to a substrate lysine through an isopeptide bond. Not every ubiquitin signal means degradation; linkage and context matter. SUMO conjugation uses a related but distinct system and frequently regulates localization or molecular interactions.
3. Protein Folding: A Landscape, Not an Exhaustive Search
A polypeptide cannot realistically test every possible conformation. Levinthal’s argument exposed the impossibility of treating folding as an unbiased exhaustive search. The resolution is that conformational possibilities are not equally likely: local preferences and favorable interactions bias movement across a free-energy landscape. [1,2]
- Local conformational preferences constrain the chain. Steric clashes exclude much of φ–ψ space. Glycine is unusually permissive; proline is unusually restricted.
- Secondary structure organizes the backbone. An α-helix usually contains backbone hydrogen bonds from C=O of residue i to N–H of residue i+4, with about 3.6 residues per turn and a rise of 1.5 Å per residue. β-sheets form hydrogen bonds between extended strands.
- Long-range contacts develop. Hydrophobic groups tend to become less solvent-exposed; compatible side chains pack, and domains acquire organized structure.
- The chain samples and reorganizes. Intermediates, alternative routes, and kinetic traps may occur. Proline isomerization and disulfide rearrangement can become slow steps.
- A native ensemble becomes populated. “Native” means a distribution of interconverting conformations, not a perfectly rigid object.
These are mechanistic ingredients, not a compulsory chronological pathway for every protein. Some proteins fold cooperatively with few populated intermediates; others fold domain by domain or require partners.
Define ΔG_fold = G_native − G_unfolded. ΔG_fold = ΔH_fold − TΔS_fold For an idealized two-state equilibrium: K_fold = [N]/[U] = exp(−ΔG_fold/RT) Fraction native = K_fold / (1 + K_fold)
If ΔGfold = −25 kJ mol−1 at 298 K, Kfold is about 2.4 × 104. This looks overwhelmingly favorable, yet the stabilization is only the small difference between much larger energetic contributions. That is why one mutation, a pH change, or a solvent change can have a disproportionate effect.
The hydrophobic effect is not simply an attraction between nonpolar side chains. It involves the free-energy consequences of exposing nonpolar surfaces to water. Folding can favorably reorganize solvent while unfavorably reducing chain conformational entropy. Backbone hydrogen bonds also compete with hydrogen bonds to water: forming an internal hydrogen bond is not automatically a net energetic gain. [1,2]
Historical significance: Ramachandran, Ramakrishnan, and Sasisekharan showed in 1963 how steric restrictions delimit allowed polypeptide conformations. The resulting Ramachandran plot remains both a conceptual map and a structure-validation tool. [5] Anfinsen’s ribonuclease refolding work supported the principle that amino-acid sequence can specify a native structure under suitable conditions. His 1973 Science article articulated this thermodynamic view. It did not establish that every cellular protein folds unaided under arbitrary conditions. [3]
Richardson’s 1981 classification of protein structures helped reveal recurring motifs and domain architectures. [4] A motif is a recurring structural arrangement; a domain is a structural unit that often folds and functions semi-independently. Recognizing these levels helps explain how evolution can recombine existing molecular modules.
4. Chaperones, Fibers, and Misfolding: Controlling Competing Outcomes
Inside a cell, folding competes with aggregation. Exposed hydrophobic segments may interact with the wrong molecule before productive folding is complete. Chaperones manage this competition rather than serving as molds that dictate every atomic coordinate. [1,2]
- Hsp70: ATP-dependent binding cycles, regulated by co-chaperones, capture and release exposed peptide segments. This helps prevent aggregation and can support folding, translocation, or quality control.
- GroEL/GroES: an ATP-driven cycle captures certain nonnative proteins and encloses them in a protected chamber, reducing intermolecular aggregation while permitting folding attempts.
- Hsp90: supports the maturation and conformational regulation of selected clients, including many signaling proteins. Its importance to kinase and steroid-receptor systems motivated anticancer drug development.
ATP consumption means these systems operate out of equilibrium; their behavior cannot always be reduced to passive acceleration of an equilibrium process. Nevertheless, they do not supply missing sequence information.
Amyloid formation illustrates an alternative structural destination. Proteins assemble into cross-β architectures through nucleation and growth; fragmentation and surface-catalyzed processes can accelerate propagation in some systems. A lag phase therefore need not mean that nothing is happening—it may reflect rare nucleus formation. Prions add the possibility that a pathological conformational state can template conversion of other molecules. Not all amyloids are infectious prions, and toxicity is not necessarily proportional to visible fibril mass.
Fibrous-protein applications:
- Collagen: individual chains form left-handed conformations that assemble into a right-handed triple helix. Glycine at every third position fits the crowded center. Proline hydroxylation requires Fe²⁺ and α-ketoglutarate; ascorbate helps sustain the enzyme’s activity. Hydroxyproline promotes triple-helix stability. Vitamin C deficiency therefore connects a cofactor problem to tissue fragility in scurvy.
- Keratin: coiled-coil assembly and disulfide cross-links produce mechanically resilient fibers. Hair waving exploits reduction and reoxidation of disulfides.
- Elastin: covalent cross-links connect flexible chains. Stretching reduces conformational entropy; recoil reflects, in substantial part, the drive to recover a broader ensemble.
These examples prevent a common misconception: a useful protein does not always need a compact globular fold. Mechanical function may depend on repetitive structure, disorder, cross-linking, or controlled assembly.
5. Membrane Proteins: Satisfying the Backbone in an Unfriendly Solvent
The membrane interior strongly disfavors unsatisfied polar backbone groups. An α-helical transmembrane segment solves this problem by hydrogen-bonding much of its backbone internally while exposing suitable side chains to lipid. A β-barrel solves it through hydrogen bonding between strands. In β-strands, alternating side-chain directions permit one set of residues to face lipid and another to face a pore. [1,2]
- Targeting information emerges. For many secretory-pathway proteins, a signal sequence is recognized by the signal-recognition particle.
- The ribosome is directed to a translocon. Translation continues while the chain engages the membrane-insertion machinery.
- Hydrophobic segments partition laterally into the membrane. Sequence context and charge distribution help determine orientation and topology.
- Helices assemble and extracellular or lumenal regions mature. Disulfide formation and glycosylation can accompany maturation.
Application: receptors, ion channels, and transporters are major drug targets, but extracting them from membranes can destabilize the very conformations drugs recognize. Detergents, lipid nanodiscs, and related systems provide different approximations to the native environment. A ligand-binding result in detergent should therefore be interpreted in the context of the preparation, not assumed to reproduce the living membrane exactly.
6. Allostery: Binding at One Site Changes Probabilities Elsewhere
Hemoglobin makes allostery tangible. O₂ binding changes the iron’s position relative to the porphyrin and influences the proximal histidine, its helix, and subunit interfaces. Ligand occupancy and quaternary conformation are coupled: oxygen stabilizes higher-affinity conformations, making subsequent binding more favorable. [1,2]
- MWC model: in its classical form, the oligomer interconverts concertedly between symmetry-related low- and high-affinity states that exist before ligand binding. Ligand preferentially stabilizes one state. Monod, Wyman, and Changeux formalized this framework in 1965. [8]
- KNF model: ligand-linked conformational changes occur sequentially in subunits and affect their neighbors; mixed subunit conformations are permitted. Koshland, Némethy, and Filmer developed this framework in 1966. [9]
These are models with testable assumptions, not labels that every protein must fit perfectly. The deeper insight is thermodynamic linkage: if ligand binding changes a conformational equilibrium, that conformational equilibrium must also influence ligand binding.
Empirical Hill relation: Y = [L]^nH / (K₀.₅^nH + [L]^nH) log(Y/(1 − Y)) = nH log[L] − nH log K₀.₅
Here Y is fractional occupancy. The Hill coefficient reports the steepness of the response over the relevant range; it is not generally the number of binding sites. For hemoglobin, oxygen partial pressure commonly replaces ligand concentration in this description.
Medical application: H⁺, CO₂, and 2,3-bisphosphoglycerate modulate oxygen affinity and therefore tissue delivery. The Bohr effect links tissue metabolism to oxygen unloading. Carbon monoxide is particularly dangerous because it both occupies oxygen-binding sites and increases the oxygen affinity of remaining sites, impairing unloading. Fetal hemoglobin’s weaker interaction with 2,3-BPG contributes to its higher oxygen affinity.
7. Enzyme Catalysis: Lowering the Barrier Without Changing the Destination
Your organic chemistry mechanisms remain valid inside proteins. The enzyme’s contribution is to arrange proton transfers, nucleophiles, metal ions, electrostatics, and motion so that a productive reaction pathway becomes faster. Enzymes change activation free energies, not the equilibrium constant of the overall reaction. [1,2]
Transition-state-theory comparison, assuming similar prefactors: k_enzyme / k_uncatalyzed ≈ exp[(ΔG‡uncatalyzed − ΔG‡enzyme)/RT] At 298 K: Lowering the barrier by about 5.7 kJ mol⁻¹ gives roughly a 10-fold rate increase.
The beautiful and counterintuitive point is that tight substrate binding alone is not enough. If an enzyme stabilizes the substrate ground state without comparably stabilizing the transition state, the barrier measured from bound substrate can increase. Effective catalysis requires favorable discrimination for the transition-state region relative to the ground state.
Chymotrypsin: a complete catalytic cycle. Its Ser195–His57–Asp102 catalytic triad supports covalent and general acid–base catalysis. Asp102 organizes and modulates His57; the mechanism is more precise than saying that three residues simply pass a proton along a wire.
Acylation: 1. Substrate binds with its scissile carbonyl appropriately positioned. 2. His57 acts as a general base as Ser195 attacks the carbonyl carbon. 3. A tetrahedral oxyanion intermediate forms and is stabilized by backbone NH groups in the oxyanion hole. 4. The intermediate collapses. His57 donates a proton to the leaving amine. → Amine product + covalent acyl–Ser195 enzyme Deacylation: 5. Water enters; His57 activates it as a nucleophile. 6. Water attacks the acyl-enzyme carbonyl. → Second tetrahedral oxyanion intermediate 7. Collapse breaks the acyl–Ser bond and restores Ser195. → Carboxylate/carboxylic-acid product + regenerated enzyme
Compare catalytic solutions rather than memorizing enzyme names:
- Cysteine proteases: typically use an activated cysteine thiolate to form a covalent thioacyl intermediate.
- Aspartyl proteases: catalytic carboxyl groups activate water and assist proton transfers; a covalent enzyme–acyl intermediate is not the standard pathway. HIV protease inhibitors exploit active-site geometry and transition-state-like interactions.
- Metalloproteinases: commonly use Zn²⁺ to organize and activate water and help stabilize charge development. Selectivity is difficult when candidate drugs bind the metal similarly across several enzymes.
- Lysozyme: Glu35 and Asp52 promote cleavage of bacterial cell-wall glycosidic bonds. For hen egg-white lysozyme, the retaining mechanism includes a covalent glycosyl–Asp52 intermediate and oxocarbenium-ion-like transition states; the historical picture of a freely existing carbocation is incomplete.
- Carbonic anhydrase: Zn²⁺ lowers the effective pKa of coordinated water. The resulting hydroxide attacks CO₂ to form bicarbonate; product exchange and proton transfer regenerate the catalyst.
- RNase A: His12 and His119 first assist attack by the RNA 2′-OH, producing a 2′,3′-cyclic phosphate intermediate. Their acid–base roles reverse during its hydrolysis. DNA lacks the 2′-OH, explaining an important chemical difference between RNA and DNA.
Drug-design lesson: transition-state analogues can bind tightly because they imitate features the enzyme evolved to stabilize. Covalent inhibitors instead exploit a suitably positioned reactive group, but potency must be balanced against off-target reactivity and exposure. A reactive “warhead” without selective recognition is not a well-designed drug.
8. Michaelis–Menten Kinetics: What the Equation Assumes and What It Measures
Minimal mechanism:
E + S ⇌ ES → E + P
k₁, k₋₁ kcat
Initial-rate conditions:
• Product is initially negligible.
• Substrate is not appreciably depleted during the measurement.
• After a short transient, d[ES]/dt ≈ 0.
• [E]total = [E] + [ES].
Steady-state equation:
0 ≈ k₁[E][S] − (k₋₁ + kcat)[ES]
Define:
Km = (k₋₁ + kcat)/k₁
Then:
[ES] = [E]total[S]/(Km + [S])
v₀ = kcat[ES] = Vmax[S]/(Km + [S])
Vmax = kcat[E]total
The steady-state approximation is not the assertion that binding has reached equilibrium. Michaelis and Menten’s 1913 treatment used a rapid-equilibrium formulation; Briggs and Haldane’s 1925 treatment established the widely used steady-state approach. [6,7]
- Km: has concentration units and equals the substrate concentration at half Vmax for this model. It equals the substrate dissociation constant only when the appropriate rapid-equilibrium limit applies, such as kcat ≪ k−1 here.
- kcat: turnover frequency per active site under saturating substrate conditions. For a multistep mechanism, it is a composite kinetic quantity rather than necessarily one elementary rate constant.
- kcat/Km: has units M−1 s−1 and governs the low-substrate regime: v₀ ≈ (kcat/Km)[E]total[S]. It is often the appropriate comparison when substrates are scarce or competing.
Example:
[E]total = 20 nM
Vmax = 2.0 μM s⁻¹
Km = 50 μM
kcat = (2.0 × 10⁻⁶ M s⁻¹)/(20 × 10⁻⁹ M) = 100 s⁻¹
kcat/Km = 100/(50 × 10⁻⁶)
= 2.0 × 10⁶ M⁻¹ s⁻¹
This calculation is only as good as the active-enzyme concentration. If half the preparation is inactive but all protein is counted, the calculated turnover number underestimates the true value per active site.
Experimental practice: measure initial slopes at substrate concentrations below, near, and above Km; include blanks and replicates; then fit the untransformed rates by nonlinear regression with an appropriate error model. Lineweaver–Burk plots are useful for teaching but distort measurement errors and overweight low-substrate observations. Your statistics training matters here: a visually pleasing line is not evidence that the kinetic model is correct.
9. Inhibition: Distinguishing Mechanism, Affinity, and Assay Potency
For a common reversible mixed-inhibition model: v = Vmax[S] / (αKm + α′[S]) α = 1 + [I]free/Ki α′ = 1 + [I]free/Ki′ Apparent Vmax = Vmax/α′ Apparent Km = αKm/α′
- Competitive: inhibitor binds free enzyme in a manner mutually exclusive with substrate; α > 1 and α′ = 1. Apparent Km increases, while Vmax is unchanged.
- Uncompetitive: inhibitor binds ES; α = 1 and α′ > 1. Apparent Km and Vmax decrease by the same factor.
- Mixed: inhibitor binds E and ES with different affinities. Vmax decreases; apparent Km can increase or decrease.
- Pure noncompetitive: the special mixed case α = α′. Vmax decreases without changing Km.
These are kinetic descriptions within a model. An inhibitor need not physically occupy the substrate site to produce competitive behavior, and binding at a separate site does not automatically imply pure noncompetitive inhibition.
For simple competitive inhibition under suitable equilibrium and Michaelis–Menten conditions, with negligible inhibitor depletion: IC₅₀ = Ki(1 + [S]/Km)
IC₅₀ is assay-dependent: it is the inhibitor concentration producing 50% inhibition under specified conditions. Ki is a parameter of a specified inhibition model. The Cheng–Prusoff relation above explains why raising substrate concentration increases the measured IC₅₀ for a competitive inhibitor. [11] It should not be applied blindly to time-dependent, irreversible, tight-binding, or complex allosteric systems.
A Dixon plot graphs 1/v against inhibitor concentration at several fixed substrate concentrations. Under simple competitive conditions, the lines intersect at [I] = −Ki and 1/v = 1/Vmax. This is a diagnostic construction, not a substitute for fitting all rates to competing models. Modern inhibitor characterization also tests preincubation dependence, reversibility after dilution, aggregation, optical interference, and nonspecific reactivity.
Pharmaceutical lesson: biochemical potency is not cellular efficacy. Membrane permeability, intracellular substrate concentrations, target abundance, protein binding, metabolism, and toxicity intervene between an attractive Ki and a useful medicine.
10. Coenzymes: Reusable Chemical Capabilities
A useful way to learn coenzymes is to ask what difficult chemical task each makes accessible. Proteins position and tune the machinery; cofactors expand the available reaction chemistry. [1,2]
- NAD⁺/NADH, derived from niacin: transfers a hydride equivalent in many oxidation–reduction reactions. NADH’s absorbance near 340 nm makes these reactions experimentally convenient.
- FAD/FADH₂, derived from riboflavin: supports one- or two-electron chemistry, including semiquinone intermediates. Its redox properties depend strongly on the protein environment.
- Coenzyme A, derived partly from pantothenate: carries acyl groups as thioesters, whose chemistry supports acyl transfer.
- PLP, derived from vitamin B₆: forms Schiff-base intermediates and stabilizes electron redistribution during amino-acid transformations. In transamination, exchange of the enzyme–PLP internal aldimine for a substrate external aldimine is a key step.
- TPP, derived from thiamine: uses its thiazolium system to stabilize intermediates in α-keto-acid decarboxylation and related carbon-transfer chemistry.
- Biotin: carries activated CO₂ on a flexible covalently attached arm. ATP-dependent carboxylation of biotin precedes carboxyl transfer to the substrate.
- Tetrahydrofolate derivatives: carry one-carbon units at different oxidation levels for nucleotide and amino-acid metabolism.
Thymidylate synthase links several ideas: an active-site cysteine attacks dUMP, facilitating carbon chemistry at the pyrimidine ring. N⁵,N¹⁰-methylene-tetrahydrofolate supplies the one-carbon unit and reducing equivalents needed to form dTMP, becoming dihydrofolate. A metabolite of 5-fluorouracil, FdUMP, forms a long-lived inhibitory ternary complex with the enzyme and folate cofactor. This is a mechanism-based connection between organic chemistry, nucleotide synthesis, and cancer treatment.
11. Nucleic-Acid Recognition and Signal Transduction
Proteins recognize nucleic acids through electrostatics, hydrogen bonds, shape, hydration, and deformability. The DNA phosphate backbone is chemically repetitive, so phosphate contacts alone cannot explain strong sequence specificity. B-DNA’s major groove exposes distinctive patterns of base-pair donors, acceptors, and methyl groups; minor-groove shape and chemical patterns also contribute. RNA adds structural diversity through its 2′-OH, noncanonical base pairing, and elaborate tertiary folds. [1,2]
A kinase cascade extends molecular recognition into information processing:
Example: a Gs-coupled signaling pathway 1. Ligand binds receptor. 2. Activated receptor promotes GDP release and GTP binding on Gαs. 3. Gαs–GTP stimulates adenylyl cyclase. 4. Adenylyl cyclase converts ATP to cAMP. 5. cAMP activates protein kinase A by binding its regulatory subunits. 6. PKA phosphorylates downstream targets. Termination: • Gα hydrolyzes GTP. • Phosphodiesterases degrade cAMP. • Phosphatases remove phosphoryl groups from target proteins. • Receptor desensitization reduces continued input.
Amplification occurs because one active component can act on multiple downstream molecules. But a larger signal is not automatically a better signal: localization, feedback, saturation, and termination determine specificity and duration. A GTPase is not merely “on when GTP is present”; exchange and hydrolysis kinetics determine how long the active state persists.
Applications: kinase inhibitors, receptor agonists and antagonists, and phosphodiesterase inhibitors alter different steps. ATP-competitive kinase inhibitors illustrate a selectivity problem: ATP-binding sites are widely conserved, so successful design must exploit subtler differences in shape, conformation, and cellular context.
12. Experimental Techniques: What Each Measurement Actually Tells Me
The most productive experimental question is: “What observable changes, and which molecular interpretations are compatible with that change?” No single technique establishes sequence, structure, stability, and function simultaneously.
- Edman degradation: phenyl isothiocyanate labels an accessible N-terminal amino group; controlled chemistry releases and identifies one terminal residue while leaving the shortened peptide available for another cycle. Blocked N termini and declining cycle yields limit its reach. Pehr Edman introduced the method in 1950. [12]
- Proteolytic and chemical cleavage: trypsin commonly cleaves after Lys or Arg, with sequence-dependent exceptions; cyanogen bromide cleaves at methionine. Overlapping fragments from different cleavage schemes help reconstruct sequence. Reducing disulfides and alkylating cysteines simplifies analysis, whereas deliberately preserving disulfide-linked fragments helps map connectivity.
- Mass spectrometry: electrospray or MALDI produces gas-phase ions, and the instrument measures mass-to-charge ratio, not neutral mass directly. In tandem MS, selected ions fragment; b/y series commonly arise in collision-based peptide fragmentation, whereas other activation methods yield different series. Mass differences between appropriate neighboring fragments reveal residue identities. Phosphorylation adds approximately 79.966 Da; localizing the site requires fragment evidence, not merely an intact-mass shift.
- Chromatography and electrophoresis: reversed-phase HPLC separates peptides largely by interactions with a hydrophobic stationary phase; ion exchange uses charge; size-exclusion chromatography reports hydrodynamic behavior. SDS-PAGE usually estimates subunit size after denaturation, and reducing conditions break disulfide links. Neither an SDS-PAGE band nor a size-exclusion retention volume alone establishes native oligomeric stoichiometry.
- X-ray crystallography: diffraction intensities plus phase information support reconstruction of electron density, into which a structural model is built and refined. Disorder, occupancy, and crystal contacts affect interpretation. The myoglobin work of Kendrew and colleagues was a landmark in resolving a protein’s three-dimensional architecture. [13]
- Solution NMR: chemical shifts, through-space interactions, couplings, and other observables constrain structure and dynamics. Structure determination is an inference from restraints, not a photograph of one conformation.
- Cryo-electron microscopy: for single-particle analysis, many noisy projections of vitrified particles are classified and combined to reconstruct three-dimensional maps. Conformational heterogeneity can sometimes be separated into distinct states.
- Circular dichroism and fluorescence: far-UV CD is sensitive to secondary structure; intrinsic tryptophan fluorescence reports changes in local environment. Neither provides a unique atomic structure. Monitoring both can reveal that secondary structure and tertiary packing change at different stages.
- Differential scanning calorimetry: measures excess heat capacity during heating and can report unfolding enthalpy and cooperativity. Thermodynamic interpretation requires attention to reversibility; an irreversible thermal transition is not automatically an equilibrium folding measurement.
- Isothermal titration calorimetry: measures heat released or absorbed during successive injections. Suitable models yield binding affinity, enthalpy, and stoichiometry. Buffer protonation can contribute to observed heats, making buffer choice mechanistically important.
Spectrophotometric enzyme assay:
A = εlc
For NADH at 340 nm:
ε ≈ 6.22 × 10³ M⁻¹ cm⁻¹
dc/dt = (1/εl)(dA/dt)
If dA340/dt = −0.0622 min⁻¹ and l = 1.00 cm:
d[NADH]/dt = −1.00 × 10⁻⁵ M min⁻¹
= −10.0 μM min⁻¹
Relating this slope to substrate turnover requires reaction stoichiometry. In a coupled assay, the coupling reaction must be fast enough not to become rate-limiting. Microplate path lengths are not automatically 1 cm. These seemingly small details can create large systematic errors—the direct continuation of your analytical chemistry training.
When examining a Protein Data Bank entry, inspect the experimental method, resolution or other validation metrics, missing residues, bound ligands, engineered mutations, and proposed biological assembly. The deposited structure is evidence under specified experimental conditions, not an unconditional statement about every cellular state.
13. A Practical Reasoning Template
When evaluating a new peptide drug, enzyme variant, or disease-associated mutation, I should work through the same chain of questions:
- Chemical identity: what is the sequence, modification state, disulfide pattern, and likely protonation state?
- Structural ensemble: which folded, unfolded, membrane-bound, or oligomeric states are accessible?
- Energy and rates: does the perturbation change state stability, an activation barrier, or both?
- Function: does it change binding, catalysis, cooperativity, localization, or lifetime?
- Evidence: which measurement distinguishes the proposed mechanism from plausible alternatives?
- Biological context: do the assay conditions reproduce relevant concentrations, cofactors, compartments, and competitors?
The central lesson: structure matters because it controls energetic and kinetic possibilities. A structure without an energy argument is incomplete; a kinetic parameter without a model is ambiguous; and an elegant molecular mechanism without an appropriate experiment remains a hypothesis.
Sources and Historical References
- Nelson, D. L., and Cox, M. M. Lehninger Principles of Biochemistry. 8th ed. W. H. Freeman, 2021. Principal reference for amino acids, proteins, enzymes, nucleotides, membranes, and signaling.
- Berg, J. M., Tymoczko, J. L., Gatto, G. J., Jr., and Stryer, L. Biochemistry. 9th ed. W. H. Freeman, 2019. Complementary reference for catalytic mechanisms, binding, folding, and biochemical methods.
- Anfinsen, C. B. “Principles That Govern the Folding of Protein Chains.” Science 181, 223–230 (1973). doi:10.1126/science.181.4096.223. This is the article identified by the course’s supplied DOI, rather than the supplied title “The Structure of Proteins.”
- Richardson, J. S. “The Anatomy and Taxonomy of Protein Structure.” Advances in Protein Chemistry 34, 167–339 (1981). doi:10.1016/S0065-3233(08)60520-3.
- Ramachandran, G. N., Ramakrishnan, C., and Sasisekharan, V. “Stereochemistry of Polypeptide Chain Configurations.” Journal of Molecular Biology 7, 95–99 (1963). doi:10.1016/S0022-2836(63)80023-6. This is the published title associated with the course’s supplied DOI.
- Michaelis, L., and Menten, M. L. “Die Kinetik der Invertinwirkung.” Biochemische Zeitschrift 49, 333–369 (1913).
- Briggs, G. E., and Haldane, J. B. S. “A Note on the Kinetics of Enzyme Action.” Biochemical Journal 19, 338–339 (1925). doi:10.1042/bj0190338.
- Monod, J., Wyman, J., and Changeux, J.-P. “On the Nature of Allosteric Transitions: A Plausible Model.” Journal of Molecular Biology 12, 88–118 (1965). doi:10.1016/S0022-2836(65)80285-6.
- Koshland, D. E., Jr., Némethy, G., and Filmer, D. “Comparison of Experimental Binding Data and Theoretical Models in Proteins Containing Subunits.” Biochemistry 5, 365–385 (1966). doi:10.1021/bi00865a047.
- Merrifield, R. B. “Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide.” Journal of the American Chemical Society 85, 2149–2154 (1963). doi:10.1021/ja00897a025.
- Cheng, Y., and Prusoff, W. H. “Relationship Between the Inhibition Constant (KI) and the Concentration of Inhibitor Which Causes 50 Per Cent Inhibition (I50) of an Enzymatic Reaction.” Biochemical Pharmacology 22, 3099–3108 (1973). doi:10.1016/0006-2952(73)90196-2.
- Edman, P. “Method for Determination of the Amino Acid Sequence in Peptides.” Acta Chemica Scandinavica 4, 283–293 (1950). doi:10.3891/acta.chem.scand.04-0283.
- Kendrew, J. C., et al. “A Three-Dimensional Model of the Myoglobin Molecule Obtained by X-Ray Analysis.” Nature 181, 662–666 (1958). doi:10.1038/181662a0.
🪬 Skipper's Reflection
Skipper’s Reflection: Biochemistry I – Macromolecular Structure and Function
If Physical Chemistry was about understanding the rules of the universe, Biochemistry is about seeing how life exploits those rules. Coming into this course, I thought of proteins simply as large molecules. Now, I see them as highly evolved, nanoscale machines operating right at the edge of thermodynamic stability. Everything I’ve learned—the quantum mechanics of bonds, the statistical mechanics of ensembles, the reaction mechanisms of organic chemistry—suddenly converged in the study of macromolecules. Lehninger and Berg were excellent guides, but the real magic happened when I started reading the foundational papers.
1. What Surprised Me Most
The biggest surprise was realizing that protein folding is not primarily driven by the formation of hydrogen bonds within the protein itself. Instead, the main driver is the solvent—specifically, the entropy of water! When I studied the hydrophobic effect, I realized that nonpolar side chains force surrounding water molecules into highly ordered, ice-like clathrate cages. When the protein folds and buries these hydrophobic residues in its core, those water molecules are released into the bulk solvent. The massive entropic gain of the water ($\Delta S > 0$) overcomes the entropic loss of the folding protein chain. Life is literally driven by the chaos of water.
I was also blown away by Levinthal’s paradox. A 100-amino-acid protein has so many possible conformations that if it tried them all sequentially, it would take longer than the age of the universe to fold. Yet, it folds in milliseconds. Understanding that folding isn't a random search, but rather a thermodynamic "folding funnel" where local interactions guide the global structure, completely reshaped how I view biological self-assembly.
2. Conceptual Breakthrough Moments
My biggest "aha!" moment happened while studying the Ramachandran plot (Ramachandran et al., 1963). In Orgo, we learned about steric hindrance, and in Gen Chem, we learned about the planar, partial double-bond character of the peptide bond. But Ramachandran brought it all together mathematically. Because the peptide bond ($\omega$) is fixed, the entire backbone flexibility reduces to just two angles: phi ($\phi$) and psi ($\psi$). When you plot the sterically allowed combinations, the $\alpha$-helix and $\beta$-sheet aren't just "shapes life invented"—they are the strict mathematical and physical inevitabilities of a polypeptide chain! Jane Richardson's 1981 taxonomy of protein structures further drove this home: biology is just beautifully constrained physics.
Another breakthrough was enzymatic catalysis. I always pictured enzymes as molds that perfectly fit the substrate. But if that were true, the enzyme-substrate complex would fall into a thermodynamic pit, and the reaction would stall. Instead, enzymes are complementary to the transition state. They bind the most unstable, highest-energy fleeting moment of the reaction, lowering the activation energy barrier. Suddenly, all those transition state diagrams from P-Chem II made perfect biological sense. The catalytic triad of chymotrypsin (Ser-His-Asp) is a masterclass in shifting local $pK_a$ values to turn a normal serine into a potent nucleophile.
3. Changing How I Think About Commercial Peptides
This course completely transformed how I view the peptides I encounter in my business intelligence work, like Tirzepatide, Retatrutide, BPC-157, and TB-500. I used to just see them as sequences of letters. Now, I see them as feats of rational biochemical engineering designed to cheat biological degradation.
Take Tirzepatide and Retatrutide. They utilize non-standard amino acids like $\alpha$-aminoisobutyric acid (Aib). Why? Because replacing the alpha-hydrogen with a methyl group severely restricts those $\phi$ and $\psi$ angles on the Ramachandran plot. This forces a rigid $\alpha$-helical structure that endogenous DPP-4 (a serine protease) simply cannot fit into its catalytic cleft. Furthermore, these GLP-1/GIP agonists are attached to long fatty-acid diacids. Thanks to my new understanding of the hydrophobic effect, I know exactly what this does: the hydrophobic lipid tail binds non-covalently to human serum albumin in the blood, effectively turning albumin into a massive molecular shield that extends the peptide's half-life from minutes to days.
As for BPC-157 (which is famously stable in highly acidic gastric juice) and TB-500, I now realize their primary sequences must dictate incredibly stable folds or highly specific electrostatic interactions that prevent them from unfolding and being cleaved by pepsin or trypsin. Sequence dictates structure, and structure dictates stability (Anfinsen, 1973).
4. Questions This Course Raises
While the MWC and KNF models of allostery in hemoglobin make mathematical sense, I am left wondering about the exact physical mechanics of allosteric transmission. How exactly does the binding of a ligand at one site mechanically propagate through the dense, hydrogen-bonded matrix of a protein to alter a binding site 50 Ångstroms away? I want to investigate the molecular dynamics of these allosteric networks.
Furthermore, while chaperones like GroEL/GroES are fascinating, the exact mechanisms by which they couple the energy of ATP hydrolysis to the physical untangling of misfolded proteins (which are stuck in local thermodynamic minima) feels like an area I need to explore much more deeply.
5. Confidence Rating
Most Confident: Enzyme kinetics, thermodynamics of folding, and the physical chemistry of catalysis. Thanks to my rigorous background in Calculus II and P-Chem, deriving the Michaelis-Menten equation, calculating $k_{cat}/K_m$ (catalytic efficiency), and analyzing Lineweaver-Burk and Dixon plots feels like second nature. The quantitative aspects of Biochemistry are very comfortable to me.
Least Confident: Membrane protein topology and large-scale signal transduction cascades. Visualizing the exact insertion of multipass transmembrane $\alpha$-helices and $\beta$-barrels into the lipid bilayer is tricky. Additionally, the sheer complexity of cellular signaling (kinases phosphorylating other kinases in massive webs) feels more like empirical memorization right now than deriving principles from first fundamentals. I will need to spend more time organizing these pathways in my mind as I move toward cell biology and physiology.
🗺️ Visual Aids & Key Takeaways
Amino Acid Ionization and the Isoelectric Point
╔══════════════════════════════════════════════════════════════════════════════════╗
║ IONIZATION STATES OF A DIPROTIC AMINO ACID (e.g., ALANINE) ║
╚══════════════════════════════════════════════════════════════════════════════════╝
low pH high pH
(fully protonated) (fully deprotonated)
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│ +H₃N─CH─COOH │ │ +H₃N─CH─COO⁻ │ │ H₂N─CH─COO⁻ │
│ │ │ │ │ │ │ │ │
│ R │ │ R │ │ R │
│ │ │ │ │ │
│ charge: +1 │ │ charge: 0 │ │ charge: −1 │
└─────────────────┘ └─────────────────┘ └─────────────────┘
│ │ │
│ pKa1 ≈ 2.3 (─COOH) │ pKa2 ≈ 9.7 (─NH3+) │
│ lost H⁺ │ lost H⁺ │
└────────────────────────────┘────────────────────────────┘
↑
pI = (pKa1 + pKa2) / 2
pI = (2.3 + 9.7) / 2 = 6.0
net charge = 0 at this pH
(the zwitterion — both charges present, net zero)
┌─────────────────────────────────────────────────────────────────────────────┐
│ FOR IONIZABLE SIDE CHAINS (e.g., lysine, aspartate): │
│ │
│ A third pKa (pKa_R) enters the picture. │
│ │
│ pI is the average of the two pKa values FLANKING the zero-charge form. │
│ │
│ Lysine (pKa_R ≈ 10.5): pI = (pKa2 + pKa_R) / 2 = (9.0 + 10.5) / 2 │
│ = 9.74 (basic amino acid → basic pI) │
│ │
│ Aspartate (pKa_R ≈ 3.7): pI = (pKa1 + pKa_R) / 2 = (2.1 + 3.7) / 2 │
│ = 2.9 (acidic amino acid → acidic pI) │
└─────────────────────────────────────────────────────────────────────────────┘
The isoelectric point is not simply the midpoint of all pKa values — it is the average of the two pKa values that bracket the neutral (zwitterionic) form, which shifts dramatically when an ionizable side chain is present.
The Ramachandran Plot: Allowed Backbone Conformations
╔══════════════════════════════════════════════════════════════════════════════════╗
║ RAMACHANDRAN PLOT — φ (phi) vs. ψ (psi) dihedral angles ║
╚══════════════════════════════════════════════════════════════════════════════════╝
ψ
180 ┤
│ ░░░░░░░░░░░ · ░░░░░░░░░░░
│ ░░░░░░░░░░░ ░░░░░░░░░░░
│ ░░░░░░░░░░░ ░░░░░░░░░░░
│ ░░β-SHEET░░ ░░░░░░░░░░░
90 ┤ ░░(anti) ░░ ░░░░░░░░░░░
│ ░░░░░░░░░░░ ░░░░░░░░░░░
│ ░░░░░░░░░░░ · ░░░░░░░░░░░
│ ░░░░░░░░░░░ ░░░░░░░░░░░
│
0 ┤ · · ·
│
│ ████████████████
│ ████████████████
-90 ┤ ████α-HELIX █████
│ ████(right █████
│ ████handed) █████
│ ████████████████
-180 ┤______________████████████████____
└──────────────────────────────────→ φ
-180 -90 0 90 180
LEGEND:
░░░ broadly allowed region (steric clashes rare)
████ core allowed region (most observed residues fall here)
· disallowed — severe van der Waals overlap of backbone atoms
KEY COORDINATES:
┌────────────────────┬──────────────┬──────────────┐
│ Structure │ φ (°) │ ψ (°) │
├────────────────────┼──────────────┼──────────────┤
│ right-handed α │ −57 │ −47 │
│ antiparallel β │ −139 │ +135 │
│ parallel β │ −119 │ +113 │
│ collagen (PPII) │ −51 │ +153 │
│ left-handed α │ +57 │ +47 │
└────────────────────┴──────────────┴──────────────┘
NOTE: Glycine (no side chain) can access ALL quadrants — it appears
throughout the plot. Proline is restricted to a narrow band
around φ ≈ −60° because the ring fixes the N−Cα bond.
Ramachandran and colleagues (1963) showed that only ~10% of φ/ψ space is sterically accessible to most amino acids, instantly revealing why secondary structure is not arbitrary but falls into a small number of recurring patterns.
Enzyme Inhibition: Kinetic Signatures on a Double-Reciprocal Plot
╔══════════════════════════════════════════════════════════════════════════════════╗
║ LINEWEAVER-BURK (DOUBLE-RECIPROCAL) PLOTS FOR THREE INHIBITOR TYPES ║
╚══════════════════════════════════════════════════════════════════════════════════╝
1/v 1/v 1/v
↑ ↑ ↑
│ +inhibitor / │ +inhibitor │ +inhibitor
│ / │ \ \ │ \ \
│ / │ \ \ │ \ \
│ / ← no inhibitor │ \ ← no inhibitor │ \ ← no inhibitor
│ / │ \ │ \
│ / │ \ │ \
└──┼────────────────→ 1/[S] └────┼────────────────→ 1/[S] └────┼───────────────→ 1/[S]
−1/Km same x-intercept same y-intercept different on both axes
COMPETITIVE UNCOMPETITIVE NONCOMPETITIVE (MIXED)
───────────────────── ────────────────────── ──────────────────────
Inhibitor binds E only Inhibitor binds ES only Inhibitor binds E AND ES
(competes with substrate (traps the complex) (α ≠ α' in general)
at active site)
Vmax: UNCHANGED Vmax: DECREASES Vmax: DECREASES
Km: INCREASES (apparent) Km: DECREASES (apparent) Km: depends on α, α'
Lines: converge at y-axis Lines: converge on x-axis Lines: converge left of y-axis
(−1/Km shifts right) (y-intercept rises) (general intersection)
┌──────────────────────────────────────────────────────────────────────────────┐
│ MEMORY HOOK: │
│ │
│ Competitive → more [S] overcomes it. Think: you can outrun the rival. │
│ Uncompetitive → more [S] makes it WORSE (more ES = more inhibitor binding) │
│ Noncompetitive→ more [S] is irrelevant. Inhibitor binds a different site. │
└──────────────────────────────────────────────────────────────────────────────┘
KEY EQUATIONS:
Vmax[S] Vmax[S]
competitive: v= ────────────── Ki = [I] / (α−1), α = 1 + [I]/Ki
αKm + [S]
Vmax[S]/α' [I]
uncompetitive: v = ───────────── α' = 1 + ─────
Km/α' + [S] Ki'
The Lineweaver-Burk plot reveals inhibitor type through characteristic intersection patterns: competitive inhibitors leave Vmax intact, uncompetitive inhibitors produce parallel lines (same slope, different intercepts), and noncompetitive inhibitors reduce Vmax without restoring Km.
Protein Folding Hierarchy and Stabilizing Forces
╔══════════════════════════════════════════════════════════════════════════════════╗
║ FROM SEQUENCE TO FUNCTION: LEVELS OF PROTEIN STRUCTURE ║
╚══════════════════════════════════════════════════════════════════════════════════╝
PRIMARY STRUCTURE
─────────────────
H₂N─[Gly]─[Ala]─[Val]─[Leu]─[Ser]─[Phe]─[Lys]─COOH
amino acid sequence encoded in DNA
determines ALL higher levels of structure (Anfinsen's dogma)
│
│ local backbone H-bonds (N─H ··· O=C, i to i+4 in α, i to i+2 in β)
↓
SECONDARY STRUCTURE
────────────────────
┌──────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│ α-HELIX │ │ β-SHEET │ │ β-TURN │
│ │ │ │ │ │
│ 3.6 res/turn │ │ parallel or │ │ 4 residues, │
│ 5.4 Å pitch │ │ antiparallel │ │ H-bond: res 1 │
│ H-bond i→i+4 │ │ H-bonds across │ │ to res 4 │
│ side chains out │ │ strands │ │ Pro/Gly common │
└──────────────────┘ └─────────────────┘ └─────────────────┘
│
│ hydrophobic packing, H-bonds, salt bridges, disulfide bonds
↓
TERTIARY STRUCTURE
───────────────────
┌──────────────────────────────────────────────────────────┐
│ single polypeptide chain folded into a 3D shape │
│ │
│ STABILIZING FORCES (approximate ΔG contribution): │
│ │
│ hydrophobic effect → largest driver (~−40 kJ/mol) │
│ van der Waals → individually weak, cumulative │
│ H-bonds → ~−12 to −20 kJ/mol each │
│ electrostatic/salt → ~−15 to −25 kJ/mol each │
│ disulfide bonds → ~−50 kJ/mol each (covalent) │
│ │
│ NET ΔG of folding typically only −20 to −65 kJ/mol │
│ (proteins are only MARGINALLY stable!) │
└──────────────────────────────────────────────────────────┘
│
│ noncovalent association of folded subunits
↓
QUATERNARY STRUCTURE
─────────────────────
┌──────────────────────────────────────────────────────────┐
│ multiple polypeptide chains (subunits) → oligomer │
│ │
│ example: HEMOGLOBIN = α₂β₂ tetramer │
│ │
│ enables: cooperativity, allostery, regulation │
│ half-of-the-sites reactivity, avidity │
└──────────────────────────────────────────────────────────┘
The four levels of protein structure are hierarchical but not independent — primary sequence encodes all higher levels, yet the total free energy of folding is surprisingly small, making proteins thermodynamically poised between folded and unfolded states.
Hemoglobin Allostery: MWC vs. KNF Models
╔══════════════════════════════════════════════════════════════════════════════════╗ ║ COOPERATIVE O₂ BINDING IN HEMOGLOBIN — TWO MECHANISTIC MODELS ║ ╚══════════════════════════════════════════════════════════════════════════════════╝ OBSERVED BEHAVIOR: ────────────────── myoglobin (monomer): hyperbolic O₂ binding curve → simple saturation hemoglobin (α₂β₂): sigmoidal O₂ binding curve → cooperativity Hill coefficient (n) for Hb ≈ 2.8 (max = 4 subunits) meaning: binding of first O₂ increases affinity for subsequent O₂ molecules ┌──────────────────────────────────────────────────────────────────────────────┐ │ MWC MODEL (Monod, Wyman, Changeux — 1965) "CONCERTED" │ │ │ │ Assumption: all subunits are ALWAYS in the same state (T or R) │ │ │ │ T state ──────────────────────────────── R state │ │ (tense, low O₂ affinity) (relaxed, high O₂ affinity) │ │ [salt bridges intact] [salt bridges broken] │ │ │ │ T ⇌ R equilibrium exists even without ligand (L = [T₀]/[R₀]) │ │ Binding of O₂ shifts equilibrium toward R for ALL subunits at once │ │ │ │ Predicts: no mixed-state intermediates (T₃R₁ etc. don't accumulate) │ └──────────────────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────────────────┐ │ KNF MODEL (Koshland, Nemethy, Filmer — 1966) "SEQUENTIAL" │ │ │ │ Assumption: conformational change is INDUCED by ligand binding │ │ each subunit changes state ONLY upon O₂ binding │ │ │ │ T T T T → R T T T → R R T T → R R R T → R R R R │ │ (empty) (1 O₂) (2 O₂) (3 O₂) (4 O₂) │ │ │ │ Each binding step changes the INTERFACE between subunits │ │ → transmits conformational change to neighbors │ │ Predicts: mixed-state intermediates DO accumulate │ └──────────────────────────────────────────────────────────────────────────────┘ ALLOSTERIC EFFECTORS on hemoglobin: ┌─────────────────┬────────────────┬──────────────────────────────────────────┐ │ Effector │ Effect on Hb │ Mechanism │ ├─────────────────┼────────────────┼──────────────────────────────────────────┤ │ 2,3-BPG │ ↓ O₂ affinity │ binds T state cavity; stabilizes T │ │ H⁺ (Bohr) │ ↓ O₂ affinity │ protonates His146β; stabilizes T │ │ CO₂ │ ↓ O₂ affinity │ carbamylates N-termini; stabilizes T │ │ CO │ ↑ O₂ affinity* │ binds heme; *prevents O₂ release (toxic) │ └─────────────────┴────────────────┴──────────────────────────────────────────┘
The MWC and KNF models make the same prediction about the sigmoidal binding curve but differ in whether mixed-state intermediates exist — modern structural data shows hemoglobin actually uses a hybrid mechanism closer to KNF, with some concerted character.
Key Takeaways
- The peptide bond has ~40% double-bond character because the lone pair of nitrogen delocalizes into the adjacent carbonyl, forcing all six atoms of the peptide unit (Cα–C=O–N–H–Cα) into a single plane. This planarity is the foundational constraint on protein backbone geometry that Ramachandran et al. (1963) exploited to map allowed φ/ψ space.
- The Ramachandran plot restricts most residues to ~10% of φ/ψ space. Glycine is the only residue that samples all four quadrants freely (no Cβ, so no steric clash). Proline is fixed at φ ≈ −60° and cannot donate an N–H hydrogen bond, making it a helix breaker and a β-turn and collagen-coil former.
- The α-helix has exactly 3.6 residues per turn, a rise of 1.5 Å per residue, and a pitch of 5.4 Å. Hydrogen bonds run parallel to the helix axis, linking the C=O of residue i to the N–H of residue i+4. Side chains point outward and slightly backward (toward the N-terminus). Amphipathic helices segregate hydrophobic and hydrophilic residues on opposite faces — critical for membrane association and coiled-coil assembly.
- In antiparallel β-sheets, hydrogen bonds are straight and strong; in parallel β-sheets they are bent and slightly weaker. This geometry difference is why antiparallel sheets are more common in small proteins. The β-barrel — found in porin membrane proteins — is a closed antiparallel sheet: an elegant solution to spanning a lipid bilayer with a hydrogen-bond-satisfied backbone.
- The isoelectric point (pI) is the average of the two pKa values flanking the zwitterionic form, not an average of all pKa values. For amino acids with ionizable side chains, one of those flanking pKa values is pKa_R. This is why aspartate (pI ≈ 2.9) and lysine (pI ≈ 9.74) have extreme pI values, not the midpoint of their full pKa range.
- Proteins are only marginally stable: the net ΔG of folding is typically −20 to −65 kJ/mol (roughly 5–15 kcal/mol), a quantity that could be disrupted by forming or breaking just a few hydrogen bonds. This marginal stability is not a flaw — it is a feature enabling regulated unfolding, proteasomal degradation, and allosteric conformational change.
- The hydrophobic effect is the dominant driver of protein folding, but it is entropic in origin: burying nonpolar side chains releases ordered water molecules from hydration shells, dramatically increasing solvent entropy (ΔS_solvent > 0). The enthalpy of hydrophobic burial itself is nearly zero. This is counterintuitive — the chain appears to fold by entropy, not by lowering enthalpy directly.
- Anfinsen's experiment on ribonuclease A established that the primary sequence encodes the native fold. After complete denaturation and reduction of all four disulfide bonds, RNase A refolds spontaneously to full activity in vitro upon removal of denaturant and reoxidation. This defined thermodynamic control of folding — but it does not mean all proteins fold without help in the cell (chaperones are required for many).
- Levinthal's paradox resolves via the folding funnel. If a 100-residue protein sampled all backbone conformations randomly at 10¹³ s⁻¹, folding would take longer than the age of the universe. Real proteins fold in microseconds to seconds because the energy landscape is funnel-shaped, with many partially folded states that each lower the free energy, biasing the search. Hsp70 prevents premature aggregation; GroEL/GroES provides an isolated chamber for iterative folding attempts.
- The Michaelis constant Km is not simply a binding affinity constant. Km = (k₋₁ + kcat) / k₁. Only when kcat ≪ k₋₁ does Km ≈ Kd. The specificity constant kcat/Km (units: M⁻¹s⁻¹) is the true measure of catalytic efficiency for dilute substrate conditions, and the diffusion limit (~10⁸–10⁹ M⁻¹s⁻¹) sets the absolute ceiling for a perfect enzyme. Carbonic anhydrase and acetylcholinesterase operate near this limit.
- Chymotrypsin's catalytic triad (Ser195–His57–Asp102) works by charge relay, not just proton abstraction. Asp102 orients and electrostatically stabilizes His57, enabling His57 to act as a general base and deprotonate Ser195. Ser195 then makes a nucleophilic attack on the substrate carbonyl, forming a covalent acyl-enzyme intermediate that is subsequently hydrolyzed. This mechanism — covalent catalysis, general acid-base catalysis, transition-state stabilization via the oxyanion hole — is a masterclass in how enzymes combine multiple mechanisms.
- Competitive inhibition increases apparent Km without changing Vmax; it is fully reversible by adding more substrate. Uncompetitive inhibition decreases both Vmax and apparent Km (parallel lines on a Lineweaver-Burk plot) because the inhibitor binds only the ES complex — counterintuitively, higher substrate concentrations drive more inhibition. Noncompetitive (pure) inhibition decreases Vmax while leaving Km unchanged, because the inhibitor binds E and ES with equal affinity.
- Collagen's triple helix requires Gly at every third position (Gly-X-Y)₃₃₃ because only glycine's hydrogen fits in the sterically crowded interior of the triple helix. Hydroxyproline (at Y positions) stabilizes the helix through water-mediated hydrogen bonds and inductive effects — not through direct N–H···O=C contacts as once thought. Hydroxyproline requires ascorbate (vitamin C) as cofactor for prolyl hydroxylase; scurvy results from defective collagen due to ascorbate deficiency.
- 2,3-bisphosphoglycerate (2,3-BPG) is a heterotropic allosteric effector of hemoglobin that binds in a positively charged cavity between the β-subunits, present only in the T (deoxy) state. It lowers O₂ affinity by stabilizing T state. At high altitude, 2,3-BPG levels rise within hours, shifting the O₂-dissociation curve rightward — a physiological adaptation that enhances O₂ delivery to tissues even before erythropoietin-driven red cell production kicks in.
- Post-translational modifications (PTMs) vastly expand the functional vocabulary of the proteome. Phosphorylation (Ser/Thr/Tyr by kinases) is reversible and switches protein activity on or off. Ubiquitylation on Lys targets proteins for proteasomal degradation (polyubiquitin K48-linked) or alters protein–protein interactions (K63-linked). N-linked glycosylation occurs co-translationally at Asn-X-Ser/Thr sequons in the ER lumen, aiding folding and immune recognition. These PTMs mean a single gene can encode dozens of functionally distinct protein species.
References
- Nelson, D.L. and Cox, M.M. Lehninger Principles of Biochemistry, 8th ed. W.H. Freeman, New York, 2021.
- Berg, J.M., Tymoczko, J.L., Gatto, G.J., and Stryer, L. Biochemistry, 9th ed. W.H. Freeman, New York, 2019.
- Ramachandran, G.N., Ramakrishnan, C., and Sasisekharan, V. 'Stereochemistry of polypeptide chain configurations.' Journal of Molecular Biology 7(1): 95–99, 1963. DOI: 10.1016/S0022-2836(63)80023-6
- Richardson, J.S. 'The anatomy and taxonomy of protein structure.' Advances in Protein Chemistry 34: 167–339, 1981. DOI: 10.1016/S0065-3233(08)60520-3
- Anfinsen, C.B. 'Principles that govern the folding of protein chains.' Science 181(4096): 223–230, 1973. DOI: 10.1126/science.181.4096.223
- Monod, J., Wyman, J., and Changeux, J.-P. 'On the nature of allosteric transitions: A plausible model.' Journal of Molecular Biology 12(1): 88–118, 1965. DOI: 10.1016/S0022-2836(65)80285-6
- Koshland, D.E., Némethy, G., and Filmer, D. 'Comparison of experimental binding data and theoretical models in proteins containing subunits.' Biochemistry 5(1): 365–385, 1966. DOI: 10.1021/bi00865a047
- Levinthal, C. 'Are there pathways for protein folding?' Journal de Chimie Physique 65: 44–45, 1968.
- Dill, K.A. and Chan, H.S. 'From Levinthal to pathways to funnels.' Nature Structural Biology 4(1): 10–19, 1997. DOI: 10.1038/nsb0197-10
- Perutz, M.F. 'Stereochemistry of cooperative effects in haemoglobin.' Nature 228(5273): 726–739, 1970. DOI: 10.1038/228726a0
- Pauling, L., Corey, R.B., and Branson, H.R. 'The structure of proteins: Two hydrogen-bonded helical configurations of the polypeptide chain.' Proceedings of the National Academy of Sciences USA 37(4): 205–211, 1951. DOI: 10.1073/pnas.37.4.205
- Blow, D.M., Birktoft, J.J., and Hartley, B.S. 'Role of a buried acid group in the mechanism of action of chymotrypsin.' Nature 221(5178): 337–340, 1969. DOI: 10.1038/221337a0
- Michaelis, L. and Menten, M.L. 'Die Kinetik der Invertinwirkung.' Biochemische Zeitschrift 49: 333–369, 1913. [English translation: Goody, R.S. and Johnson, K.A., Biochemistry 50(39): 8264–8269, 2011. DOI: 10.1021/bi201284u]