📖 Core Concepts
Core Concepts: Peptide Synthesis—Solid-Phase, Solution-Phase, and Modern Methods
Skipper’s guiding question: How do we assemble a molecule with a specified amino-acid sequence, stereochemistry, connectivity, and chemical state—and demonstrate that the material we isolated actually has those properties?
From organic chemistry, I already know how to form an amide. From protein biochemistry, I know why amino-acid sequence matters. Peptide synthesis connects these two perspectives, but the connection is not simply “repeat amide formation.” Each repetition changes the substrate: it becomes larger, less soluble, more conformationally organized, and richer in functional groups that might react in unintended ways.
The real subject is therefore controlled molecular assembly under accumulating chemical constraints. A successful synthesis must coordinate reactivity, protection, stereochemistry, transport, purification, and analysis. For a therapeutic peptide, it must also deliver the same material reproducibly at manufacturing scale.
The central challenge is not making a peptide bond. It is making the intended peptide bond, at the intended position, without changing anything else—and repeating that achievement many times.
A useful connection to BCHM601 is that the growing peptide is never merely a line of letters. It is a physical molecule with a changing conformational ensemble. The hydrogen bonds and hydrophobic interactions that eventually support biological function can obstruct its chemical assembly.
1. Sequence Fidelity Is a Multiplicative Problem
Why apparently excellent reactions can produce poor overall results
Consider a synthesis requiring 50 coupling steps. If each coupling converts 99% of the available chains correctly, and all other operations are perfect, the fraction that completes every coupling correctly is:
Fcorrect = ∏ pᵢ For identical step efficiencies: Fcorrect = pⁿ 0.99⁵⁰ ≈ 0.605 0.995⁵⁰ ≈ 0.778 0.999⁵⁰ ≈ 0.951
The difference between 99% and 99.9% sounds small when discussing one reaction. Across a long synthesis, it is decisive. This is why peptide chemists care about the last fraction of a percent of unreacted amine.
This model is intentionally simplified. A real cycle includes deprotection, washing, activation, coupling, and sometimes capping; each can introduce a different failure. Furthermore, a chain that misses one coupling may react successfully in the next cycle, generating a deletion peptide rather than remaining permanently unreactive.
A simplified full-process model: Fcorrect ≈ ∏ (dᵢ × cᵢ × sᵢ) dᵢ = fraction correctly deprotected in cycle i cᵢ = fraction correctly coupled in cycle i sᵢ = fraction avoiding other chemical damage in cycle i
These factors are not necessarily independent. Aggregation, for example, can simultaneously reduce deprotection and coupling efficiency. Nevertheless, the multiplicative model captures an essential lesson: small systematic defects accumulate into a major product-quality problem.
Purity, yield, and identity are different questions
- Yield: How much desired product did the process deliver relative to a defined theoretical amount?
- Purity: What fraction of the material, or of a specified analytical signal, corresponds to the desired component?
- Identity: Is that component the intended molecular structure?
- Content: How much actual peptide is present per unit mass or volume of the supplied material?
High crude mass can coexist with low peptide content because the material contains water, counterions, or residual solvents. A highly pure isolated fraction can coexist with poor synthetic yield because most material was discarded during purification. These distinctions become crucial when reading a certificate of analysis.
2. Protecting Groups Create a Temporary Reaction Network
Protection is a strategy for controlling which reactions are possible
An amino acid may contain several nucleophiles and several acid-sensitive or oxidation-sensitive groups. Without protection, activating one carboxyl group does not guarantee that it will react with the intended amino group. Lysine, cysteine, serine, histidine, and other residues create competing possibilities.
Protecting groups temporarily remove selected functional groups from the accessible reaction network. The objective is not maximum protection; it is enough protection to make the desired transformation selective, with a feasible plan for removing every temporary group. [Greene and Wuts, 2007]
Before choosing a group, ask four questions:
- Will it survive every operation that precedes its planned removal?
- Can it be removed without damaging the peptide or other protecting groups?
- Will its removal generate reactive byproducts?
- Can removal be made sufficiently complete and verified analytically?
Fmoc/tBu: selective removal during assembly
In the Fmoc/tBu strategy, the temporary α-amino protecting group is base-labile Fmoc. Most side-chain protections and commonly used final-cleavage linkers are acid-labile. Thus, repeated base treatment exposes the growing-chain amine while retaining most side-chain protection. Strong acid is reserved for final cleavage and global deprotection. [Chan and White, 2000]
Fmoc removal, schematically:
Fmoc–NH–peptide
── base ──▶ H₂N–peptide + CO₂ + dibenzofulvene
dibenzofulvene + piperidine → trapped addition product
Mechanistically, base removes the relatively acidic proton at the fluorenyl 9-position. Elimination releases a carbamate species, which decarboxylates to expose the amine. Piperidine also captures dibenzofulvene, limiting its ability to react with newly exposed peptide amines. This is an elimination–decarboxylation sequence, not simply alkaline hydrolysis. [Carpino and Han, 1972]
Boc/Bzl: differential acid sensitivity
Boc protection is commonly removed with trifluoroacetic acid during chain assembly. Benzyl-based side-chain protections and suitably chosen resin linkages survive those repeated treatments but require substantially stronger final-cleavage conditions, classically anhydrous HF.
Boc removal, schematically:
Boc–NH–peptide
── acid ──▶ peptide ammonium salt + CO₂
+ tert-butyl-derived products
This distinction matters: Boc/Bzl largely exploits different degrees of acid lability, whereas Fmoc/tBu separates the principal temporary and permanent deprotection operations by base versus acid responsiveness. Boc chemistry remains useful, including for some difficult sequences and thioester syntheses; it is not simply obsolete Fmoc chemistry. However, HF cleavage requires specialized facilities, equipment, and training because its hazards are exceptional. [Lloyd-Williams et al., 1997]
Orthogonality is conditional, not magical
For branching, cyclization, or lipidation, an additional selectively removable protecting group may be needed. Examples include Alloc, Dde/ivDde, Mmt, and appropriately selected cysteine protections.
- Alloc: commonly removed using palladium-mediated chemistry; metal removal and compatibility must be considered.
- Dde/ivDde: useful for selectively exposing a side-chain amine under suitably chosen hydrazine-based conditions.
- Mmt: highly acid-labile, permitting selective removal under sufficiently mild conditions.
A protecting group is not orthogonal in isolation. It is orthogonal to particular other groups under specified conditions. A useful design exercise is a compatibility matrix: place all protecting groups in rows and all planned reaction conditions in columns. Every unintended deprotection becomes visible as a conflict.
3. Solid-Phase Synthesis Changes Separation More Than Bond Formation
Merrifield’s fundamental insight
In conventional solution chemistry, each reaction is often followed by isolation of an intermediate. Merrifield’s innovation was to attach the growing peptide to an insoluble support. Soluble reagents and byproducts could then be removed by washing while the product remained attached to the resin. [Merrifield, 1963]
Solid-phase synthesis does not make every reaction intrinsically better. It makes repeated separation simple enough that many reactions can be linked into an automated process.
This permits excess amino-acid derivative and coupling reagent to be used to drive difficult reactions without having to isolate the peptide after every step. However, washing does not remove an incorrectly synthesized chain that is still covalently attached to the support.
Why conventional SPPS grows from C-terminus to N-terminus
The C-terminal residue is typically anchored through its carboxyl-derived linkage. Removing its α-amino protection exposes the nucleophile for reaction with the activated carboxyl group of the next protected amino acid.
H₂N–peptide–resin + Fmoc–NH–CH(R)–C(O)–X
→ Fmoc–NH–CH(R)–C(O)–NH–peptide–resin
Remove Fmoc.
Repeat.
Chain growth: C-terminus → N-terminus
Ribosomal synthesis proceeds in the opposite sequence direction, N-terminus to C-terminus. The difference reflects the architecture of activation and attachment, not a fundamental directional preference of peptide bonds.
The resin is a reaction environment, not just a handle
Resin beads swell in compatible solvents, creating a solvated polymer network through which reagents must diffuse. Much of the reaction occurs inside this swollen phase rather than only at an external surface.
- Wang-type linkers are commonly used to release C-terminal peptide acids.
- Rink amide linkers are commonly used to release C-terminal peptide amides.
- Loading, often expressed in mmol/g, controls the nominal density of peptide chains.
- Polymer composition and swelling affect access, solvent compatibility, and aggregation.
The polymer support and the cleavable linker are related but distinct design choices. “Which resin?” should therefore mean both “Which physical support?” and “Which attachment-and-release chemistry?”
Higher loading increases theoretical production per gram of resin, but it can also increase crowding and interchain association. Lower loading may give more usable purified peptide even when it begins with fewer reactive sites. [Chan and White, 2000]
A standard Fmoc cycle
- Remove Fmoc to expose the growing-chain amine.
- Wash away base and deprotection products.
- Activate and couple the next Fmoc-protected amino acid.
- Assess completion where appropriate.
- Repeat coupling or cap residual amines if the process calls for it.
- Wash and begin the next cycle.
Capping converts unreacted amines into relatively unreactive termini. It does not repair a failed coupling; it prevents failed chains from extending into additional deletion products. That may simplify impurity control at the cost of permanently sacrificing those chains.
4. Coupling Chemistry Must Balance Reactivity Against Stereochemical Damage
Why direct acid–amine mixing is insufficient
A carboxylic acid and an amine commonly form an ammonium carboxylate salt rather than rapidly forming an amide. In aqueous environments, direct dehydration is also strongly disadvantaged by the presence of water. Peptide synthesis therefore converts the carboxyl group into a more reactive acyl donor.
Salt formation: R–CO₂H + H₂N–R′ ⇌ R–CO₂⁻ · ⁺H₃N–R′ Productive acyl substitution: R–C(O)–X + H₂N–R′ → R–C(O)–NH–R′ + leaving-group products
Activation changes both kinetics and the overall reaction pathway. An appropriate leaving group allows the amine to attack the acyl carbon, form a tetrahedral intermediate, and produce the amide by elimination.
Carbodiimides and active esters
DIC and DCC initially generate an O-acylisourea. This intermediate can react with an amine, but it can also undergo undesirable rearrangement to an N-acylurea. Additives such as HOBt, HOAt, or other suitable systems intercept activated species to generate useful acyl donors and often reduce stereochemical damage.
Carboxylic acid + carbodiimide → O-acylisourea Productive: O-acylisourea + additive → active ester active ester + amine → amide Competing: O-acylisourea → N-acylurea
Uronium/aminium-family and phosphonium reagents—including HBTU, HATU, PyBOP, and related reagents—provide other routes to reactive acyl species. Their mechanisms and byproducts are not identical, and the most reactive system is not necessarily the cleanest for every substrate. Routine peptide coupling should not generally be pictured as proceeding through a freely solvated acylium ion. [Lloyd-Williams et al., 1997; Chan and White, 2000]
Reagent selection includes safety and scale. For example, some benzotriazole-based additives have energetic hazards, particularly in dry forms. Familiarity at small scale does not establish suitability for bulk handling.
Epimerization: the hidden failure
Activation can make the α-stereocenter of an amino-acid-derived acyl donor vulnerable to loss of configuration. Depending on the substrate and conditions, pathways can include oxazolone formation and deprotonation followed by non-stereoselective reprotonation.
Activated, configurationally defined acyl donor
⇌ planar/enolizable intermediate
→ mixture of configurations at Cα
For a peptide that already contains other stereocenters, changing one residue produces a diastereomer; epimerization is the more precise term. Such an impurity has the same molecular formula and intact mass as the desired peptide.
Long activation times, excessive base, elevated temperature, and susceptible residues can increase risk. Fragment condensation deserves particular scrutiny because activation occurs at the C-terminal residue of an existing peptide fragment.
Monitoring answers a limited question
- Kaiser/ninhydrin testing: useful for detecting residual primary amines on resin.
- Chloranil testing: useful in appropriate protocols for secondary amines, such as those associated with proline.
- Fmoc-release monitoring: can track deprotection and estimate loading under validated conditions.
- Test cleavage followed by LC–MS: provides more direct information about the assembled peptide.
A negative amine test supports amine consumption within the test’s limits. It does not demonstrate correct stereochemistry, correct sequence, or absence of side-chain damage.
5. Difficult Sequences Are Often Problems of Conformation and Transport
Growing peptide chains can form interchain hydrogen-bonded structures and hydrophobic associations within the resin. These interactions may reduce swelling, shield reactive groups, and hinder diffusion. Thus, a reaction that works rapidly for a small soluble model can become slow and incomplete on a peptide-loaded bead.
Observed synthesis performance depends on: intrinsic chemical reactivity + reagent transport + reactive-site accessibility + the peptide's conformational ensemble
This is a direct application of physical chemistry and protein dynamics. A useful qualitative distinction is between reaction-limited behavior, where reagents reach the site but react slowly, and transport- or accessibility-limited behavior, where the site is poorly reached or exposed. Adding more coupling reagent may do little to solve the second problem.
Interventions that change the physical problem
- Lower loading: separates neighboring chains and can reduce interchain association.
- Alternative supports and solvents: improve swelling, solvation, or access; compatibility must be evaluated rather than assumed.
- Pseudoproline dipeptides: use temporary cyclic derivatives of suitable residues, commonly Ser or Thr, to disturb regular backbone organization; acid treatment restores the intended residue.
- Backbone protection: Dmb- or Hmb-type approaches interrupt hydrogen bonding and modify local conformational behavior.
- Elevated temperature: can accelerate reactions and alter aggregation, but may also accelerate epimerization and other side reactions.
Pseudoproline chemistry is especially elegant: it temporarily changes the growing peptide’s structural preferences to make synthesis easier, then removes that intervention from the final product.
Microwave-assisted synthesis is best understood primarily through controlled heating and its consequences for rates and solvation. An apparent advantage should be compared with a conventionally heated process at matched temperature and residence time before invoking a special nonthermal effect. [Chan and White, 2000; Howl, ed.]
6. Cleavage and Global Deprotection Are Another Synthesis Stage
Finishing the sequence does not finish the chemistry. Cleavage must release the peptide, remove the intended protecting groups, and preserve vulnerable side chains.
In standard Fmoc/tBu synthesis, TFA-based mixtures often perform both resin cleavage and global side-chain deprotection. Different residues may carry different protecting groups: Asp and Glu commonly have tert-butyl esters, whereas Ser, Thr, and Tyr commonly have tert-butyl ethers. Lys(Boc), Arg(Pbf), and selected Trt-protected side chains complete many standard protection schemes.
Why scavengers matter
Acidolysis generates reactive species from protecting groups and linkers. These can alkylate nucleophilic side chains unless intercepted. Water, triisopropylsilane, and thiol-containing scavengers can serve different protective roles; the mixture must suit the sequence and all modifications present.
Protected functionality ── acid ──▶ desired functionality
+ reactive byproducts
Reactive byproducts + scavengers → less damaging products
Trp alkylation, cysteine-related side reactions, and methionine oxidation illustrate why one universal cleavage mixture is unrealistic. Scavengers do not guarantee freedom from oxidation, and some specialized modifications are incompatible with otherwise routine cleavage conditions.
Aspartimide is mainly an assembly-stage problem
Under basic Fmoc-deprotection conditions, the backbone amide nitrogen following Asp can attack the protected Asp side-chain carbonyl, producing a cyclic imide. Ring opening can then generate normal Asp and β-linked isoAsp products, with additional stereochemical complications possible.
Asp-containing peptide
── susceptible sequence + base exposure ──▶ aspartimide
aspartimide + water → α-Asp product + β-linked isoAsp product
and potentially epimerized species
Risk depends on sequence, protection, temperature, and cumulative base exposure. Asp–Gly is a well-known susceptible motif. It is misleading to treat aspartimide solely as a final-cleavage side reaction: the process history often creates the damage earlier.
Workup is selective too
Cold-ether precipitation can separate many peptides from cleavage reagents, but precipitation is not universal and is not equivalent to high-resolution purification. Small or unusually soluble peptides may be lost; hydrophobic peptides may aggregate. Lyophilization removes volatile material but does not remove deletion peptides or establish peptide content. [Chan and White, 2000]
7. Solution-Phase and Convergent Synthesis Trade Easy Washing for Intermediate Control
Solution-phase synthesis permits conventional isolation and characterization of intermediates. This can be an advantage when fragments are crystalline, soluble in useful solvents, or readily purified without chromatography. The price is the need to separate reagents and byproducts repeatedly.
In a linear synthesis, residues or fragments are added sequentially. In a convergent synthesis, substantial fragments are prepared independently and joined later.
Linear: A → AB → ABC → ABCD → ABCDEFGH Convergent: ABCD + EFGH → ABCDEFGH
Convergence can shorten the longest sequential route and allow defective fragments to be discarded before final assembly. It does not guarantee higher yield: joining two large fragments can be difficult because of poor solubility, steric effects, aggregation, and epimerization at the activated fragment terminus.
The minimum-protection principle means avoiding protection that is unnecessary for the transformations planned. It does not mean exposing every functional group and hoping for selectivity.
Industrial routes often combine methods: SPPS for selected fragments, solution-phase coupling, recombinant production for a large segment, and chemical or enzymatic processing for final assembly. For insulin and related disulfide-rich products, chain production, chain joining, folding, and disulfide control are separate decisions; chemical synthesis is only one route, and recombinant methods dominate modern insulin production. [Lloyd-Williams et al., 1997]
8. Native Chemical Ligation Uses Temporary Sulfur Chemistry to Make an Ordinary Amide
The defining transformation
Native chemical ligation joins a peptide bearing a C-terminal thioester to a peptide bearing an N-terminal cysteine. Reversible sulfur acyl transfer first captures the fragments; an intramolecular S→N acyl transfer then forms the backbone amide. [Dawson et al., 1994]
1. Transthioesterification: Peptide A–C(O)–SR + H₂N–CH(CH₂SH)–Peptide B ⇌ Peptide A–C(O)–S–CH₂–CH(NH₂)–Peptide B + RSH 2. Intramolecular S→N acyl transfer: Peptide A–C(O)–S–CH₂–CH(NH₂)–Peptide B → Peptide A–C(O)–NH–CH(CH₂SH)–Peptide B
The beautiful feature is that the sulfur atom is a temporary acyl-transfer handle. The final junction contains a native peptide bond, not a permanent artificial linkage.
The N-terminal cysteine combines a thiol for capture with a neighboring free α-amine for productive rearrangement. Internal cysteines can participate in reversible thioester exchange, but their backbone nitrogens are already acylated and do not provide the same ligation pathway.
Why fragment design remains essential
- The thioester-bearing fragment must survive preparation and handling.
- The ligation site must offer acceptable kinetics and manageable steric hindrance.
- Both fragments must remain sufficiently soluble.
- Thiol oxidation and thioester hydrolysis must be controlled.
Direct thioesters can be incompatible with repeated nucleophilic/basic Fmoc deprotection. Fmoc-based strategies therefore often use latent thioesters or precursors that are converted into thioesters after chain assembly.
Desulfurization extends the choice of junction
Ligation at cysteine → desulfurization → alanine at the junction
Ordinary cysteine desulfurization produces alanine, not valine. Valine and other junctions require appropriately designed thiolated amino-acid analogues. If the target contains native cysteines that must remain intact, the desulfurization strategy must distinguish them from the temporary ligation handle.
Expressed protein ligation combines recombinant production with chemical assembly, commonly using intein-derived protein thioesters. It is particularly powerful when a large protein segment can be expressed biologically while a short synthetic segment carries an isotope label, post-translational modification, or noncanonical residue. This turns synthesis into a tool for testing structure–function hypotheses that genetics alone cannot cleanly address.
9. Cyclization and Noncanonical Residues Reprogram the Conformational Ensemble
Cyclization is a competition between intra- and intermolecular chemistry
A bifunctional linear precursor may cyclize or react with another molecule to produce dimers and oligomers. In a simplified kinetic picture:
Intramolecular cyclization:
vcyclization = kᵢₙₜᵣₐ[P]
Intermolecular coupling:
voligomerization ≈ kᵢₙₜₑᵣ[P]²
Therefore:
vcyclization / voligomerization
≈ kᵢₙₜᵣₐ / (kᵢₙₜₑᵣ[P])
Dilution favors cyclization relative to bimolecular oligomerization. It does not necessarily make cyclization fast. If the reactive termini rarely adopt a productive geometry, the intramolecular reaction may still be slow.
Effective molarity, often represented conceptually as kintra/kinter, describes how favorable the intramolecular arrangement is compared with the corresponding intermolecular reaction. Preorganization can therefore matter as much as bulk dilution.
- Head-to-tail cyclization: joins the N- and C-termini.
- Lactam cyclization: often joins a Lys side-chain amine to an Asp or Glu side-chain carboxyl group.
- Disulfide cyclization: joins cysteine side chains and requires control of oxidation and connectivity.
- Hydrocarbon stapling: commonly uses ring-closing metathesis between installed olefin-bearing side chains.
Resin attachment can provide useful spatial isolation, but it is not perfect “infinite dilution.” Chains within a bead may still be close enough to interact, especially at high loading.
For a peptide with 2m cysteines, the number of formal complete pairings is:
Number of disulfide pairings = (2m − 1)!! For 6 cysteines: 5 × 3 × 1 = 15 possible complete pairings
Folding energetics may strongly favor one connectivity, but intact mass alone cannot identify it: all fully oxidized pairing isomers have the same formula.
Noncanonical residues alter more than protease recognition
- Aib and other α-substituted residues: alter backbone preferences and can stabilize selected helical conformations, while making coupling more demanding.
- D-amino acids: alter local stereochemistry, conformation, and enzymatic recognition.
- β-amino acids: insert an additional backbone atom and permit different folding patterns.
- Peptoids: place substituents on backbone nitrogen rather than the α-carbon, removing many backbone NH hydrogen-bond donors and changing cis/trans behavior.
- Azapeptides and other backbone analogues: modify electronics, geometry, and recognition by enzymes and receptors.
Retro-inverso peptides combine sequence reversal with inversion of residue chirality to approximate some side-chain arrangements. They are not exact structural copies: peptide-bond directions and hydrogen-bond donor–acceptor patterns differ.
The design objective is not simply “make the peptide more rigid.” A constraint is beneficial only if it favors a useful ensemble. The wrong constraint can stabilize a nonbinding conformation. [Lloyd-Williams et al., 1997; Muttenthaler et al., 2021]
10. Conjugation Chemistry Links Molecular Structure to Pharmacokinetics
A biologically potent peptide can still be a poor drug because it is rapidly degraded or cleared. Chemical modification must therefore be understood through both molecular recognition and whole-body physiology.
For a simplified linear pharmacokinetic model: t½ = (ln 2 × Vd) / CL Vd = apparent volume of distribution CL = systemic clearance
A modification may change clearance, distribution, absorption, receptor affinity, or all four. Increasing molecular size is not the only mechanism of half-life extension.
Major conjugation strategies
- NHS-ester PEGylation: acylates accessible amines, potentially generating positional mixtures unless site selectivity is engineered.
- Maleimide–thiol conjugation: can target cysteine, but conventional adducts may undergo exchange or other transformations depending on structure and environment.
- Enzymatic conjugation: exploits recognition motifs or engineered sites for positional control.
- Lipidation: can promote albumin association and alter absorption and clearance.
- Fc or albumin fusion: uses macromolecular size and biological recycling mechanisms to extend exposure.
Semaglutide as an integrated molecular design
Semaglutide combines resistance to DPP-4 cleavage with a C18 fatty diacid attached at Lys26 through a hydrophilic spacer that includes γ-glutamyl and oligoethylene-glycol-derived units. The design also replaces Lys34 with Arg, simplifying selective functionalization at Lys26. Its Aib substitution at position 8 contributes to proteolytic resistance. [Lau et al., 2015]
The lipid promotes albumin association, reducing rapid elimination and helping produce a clinical terminal half-life of approximately one week. That half-life should not be confused with either the duration of every pharmacodynamic effect or the mere fact of once-weekly dosing.
Liraglutide uses a C16 acyl substituent attached through a γ-glutamyl spacer at Lys26; albumin binding and self-association both contribute to its pharmacokinetic behavior. The general lesson is that a therapeutic peptide is often a deliberately engineered system of receptor recognition, enzymatic stability, reversible carrier binding, and formulation behavior, not just an optimized binding sequence. [Muttenthaler et al., 2021]
11. Analytical Evidence Must Be Matched to the Structural Question
HPLC measures separability under specified conditions
Reversed-phase HPLC is central to peptide purification and quality control. Retention depends on sequence, conformation, charge state, stationary phase, temperature, and mobile-phase composition. TFA can improve peak shape through acidification and ion pairing, but it can suppress electrospray ionization; an LC–MS method may therefore use different mobile-phase additives.
UV detection near 214–220 nm is useful because peptide bonds absorb strongly in this region. However, an integrated area percentage is not automatically a mass percentage:
HPLC area purity =
target peak area / total integrated detected peak area × 100%
This value depends on resolution, integration rules, detection wavelength, and component response. Coeluting epimers or weakly detected impurities can escape the apparent purity figure.
Mass spectrometry strongly constrains identity, but does not prove everything
For a protonated ion: m/z = [M + z·m(H⁺)] / z Approximate isotope spacing: Δ(m/z) ≈ 1.00335 / z
ESI commonly produces several charge states; MALDI often emphasizes singly charged peptide ions. Interpretation must distinguish monoisotopic from average mass and account for isotope patterns, adducts, oxidation, and charge state.
For an unmodified, linear peptide with free N- and C-termini:
Mpeptide = Σ Mfree amino acids − (n − 1)Mwater
Terminal amidation, cyclization, disulfides, labels, and protecting-group remnants require corresponding formula changes. For example, each disulfide formation removes two hydrogen atoms from the reduced thiol form.
Correct intact mass does not establish residue stereochemistry, disulfide connectivity, or every sequence detail. Depending on the target, identity may require tandem MS, peptide mapping, disulfide mapping, stereochemical analysis, or comparison with a well-characterized reference.
Quantification needs its own method
- Amino-acid analysis: useful for peptide content, with attention to hydrolysis losses and residue-specific limitations.
- UV absorbance: requires suitable chromophores and an appropriate extinction coefficient.
- Gravimetry: measures total weighed material, including water and counterions.
Beer–Lambert law: A = εlc
This connects directly to analytical chemistry: a measurement is interpretable only when its measurand is defined. “95% pure” and “95% peptide by mass” are different statements.
12. Scale-Up Is the Preservation of a Controlled Chemical History
Moving from milligrams to kilograms changes mixing, swelling, heat removal, filtration, washing, reagent stability, and waste handling. A laboratory protocol is not automatically a manufacturing process.
Flow synthesis can provide controlled residence times, rapid reagent exchange, and efficient heating. It does not inherently guarantee lower solvent consumption or better product quality. Those outcomes depend on conversion, washing requirements, equipment design, reagent excess, and downstream purification.
Process mass intensity, PMI =
total mass of process inputs / mass of isolated product
For peptide manufacture, solvents and purification can dominate material use. An impressive assembly yield may conceal a poor overall process if chromatography requires large volumes and discards much of the target.
Impurity control begins with a mechanistic map
- Assembly impurities: deletions, truncations, insertions, epimers, and incompletely deprotected products.
- Degradation impurities: oxidation, hydrolysis, deamidation, Asp isomerization, and disulfide scrambling where relevant.
- Process residues: solvents, reagents, metals, counterions, and material originating from supports or linkers.
GMP is not a purity threshold. It is a system for controlling and documenting material identity, process execution, analytical suitability, deviations, traceability, and change.
Regulatory guidance must be applied within scope. ICH Q1 addresses stability, Q6A addresses specifications for chemical drug substances and products, and Q6B addresses biotechnology/biological products. Q3A should not simply be imposed as though its routine impurity thresholds automatically govern every peptide: peptides are excluded from that guideline’s stated scope, and product-specific regulatory expectations matter. [ICH, Q1A(R2), Q3A(R2), Q6A, Q6B]
Connections to Peptide Chemistry
This course brings together several ideas that previously appeared separate:
- Organic mechanisms become sequence-control tools. Nucleophilic acyl substitution, elimination, carbocation chemistry, and protecting-group selectivity now determine whether an entire macromolecule is correct.
- Stereochemistry becomes biological information. A single epimer can leave mass unchanged while altering folding, receptor recognition, and protease susceptibility.
- Protein dynamics becomes a synthetic variable. Aggregation and secondary structure affect reaction accessibility; pseudoprolines and backbone protection temporarily reshape that ensemble.
- Statistical reasoning explains impurity accumulation. Small repeated failure probabilities produce substantial heterogeneity, while purification selects among the resulting molecular populations.
- Analytical chemistry defines what “correct” means experimentally. Chromatography, mass spectrometry, content assays, and structural mapping answer different questions.
- Molecular biology and chemical synthesis are complementary. Expressed protein ligation combines efficient biological production with chemical access to modifications that are difficult to encode genetically.
- Physiology constrains molecular design. Albumin binding, renal clearance, enzyme susceptibility, and tissue exposure can determine whether an excellent ligand becomes a useful medicine.
The deeper connection to peptide chemistry is that sequence is necessary but not sufficient to specify a peptide material. A complete description may also require stereochemistry, terminal groups, side-chain modifications, crosslink connectivity, oxidation state, salt form, and—for a functional protein—its folded or assembled state.
Common Misconceptions to Watch Out For
-
“Solid-phase synthesis purifies the peptide after every step.”
Washing removes soluble material. Incorrect chains attached to the resin remain present. -
“A 99% coupling is effectively quantitative.”
It may be excellent once, but many such steps can generate a large incorrect-chain population. -
“A stronger coupling reagent solves every difficult sequence.”
Poor swelling, aggregation, and inaccessible reactive sites may require physical or conformational interventions. -
“Fmoc chemistry is free of stereochemical and backbone side reactions.”
Activated residues can epimerize, and repeated base exposure can promote aspartimide formation and other sequence-dependent damage. -
“Orthogonal protecting groups are universally compatible.”
Orthogonality is conditional on the complete set of groups, reagents, temperatures, and exposure times. -
“Cleavage is merely detaching a finished product.”
It is a reactive operation that can create alkylation, oxidation, incomplete deprotection, and recovery problems. -
“Native chemical ligation leaves a sulfur-containing backbone junction.”
The thioester intermediate rearranges to an ordinary backbone amide; sulfur remains in the cysteine side chain unless subsequently removed. -
“Cysteine desulfurization can directly make any desired residue.”
Ordinary cysteine gives alanine. Other junction residues require different thiolated precursors. -
“Cyclization always improves potency and permeability.”
It can improve selected properties, but it can also favor the wrong conformation or worsen solubility. -
“Correct mass proves correct structure.”
Epimers, disulfide-connectivity isomers, and many other structural alternatives can share an intact mass. -
“A clean HPLC trace means the weighed powder is almost entirely peptide.”
Water, counterions, residual solvents, and undetected or coeluting components require separate consideration. -
“GMP-grade means research-grade material purified more thoroughly.”
GMP additionally requires a documented, controlled, traceable, and reproducible manufacturing and testing system.
Skipper’s Working Framework
When designing or evaluating a synthesis, I should proceed through six questions:
- What exactly is the target? Define sequence, stereochemistry, termini, modifications, and crosslinks.
- What assembly strategy fits it? Choose linear SPPS, solution-phase fragments, ligation, recombinant production, or a hybrid.
- What must remain silent during each reaction? Design the complete protecting-group and linker scheme before starting.
- Where will the chemistry or physical environment fail? Anticipate epimerization, aggregation, poor solubility, oxidation, and sequence-specific damage.
- How will the evidence distinguish success from plausible alternatives? Select orthogonal tests for identity, purity, content, and structural connectivity.
- Can the same chemical history be reproduced? Consider scale, safety, transport, waste, and documentation from the beginning.
A good peptide synthesis is not simply a route that reaches the target once. It is a route whose selectivity, failures, analytical evidence, and reproducibility can all be explained.
Sources and Reading Anchors
- Chan, W. C., and White, P. D., eds. (2000). Fmoc Solid Phase Peptide Synthesis: A Practical Approach. Practical foundation for Fmoc assembly, supports, protection, difficult sequences, and cleavage.
- Lloyd-Williams, P., Albericio, F., and Giralt, E. (1997). Chemical Approaches to the Synthesis of Peptides and Proteins. Mechanistic and strategic treatment of peptide and protein synthesis.
- Greene, T. W., and Wuts, P. G. M. (2007). Greene’s Protective Groups in Organic Synthesis. 4th ed. Protecting-group mechanisms, selectivity, and compatibility.
- Howl, J., ed. Peptide Synthesis and Applications. Application-oriented methods; consult the assigned course edition.
- Merrifield, R. B. (1963). “Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide.” Journal of the American Chemical Society. DOI: 10.1021/ja00897a025.
- Carpino, L. A., and Han, G. Y. (1972). “The 9-Fluorenylmethoxycarbonyl Function, a New Base-Sensitive Amino-Protecting Group.” Journal of the American Chemical Society. DOI: 10.1021/ja00784a052.
- Dawson, P. E., Muir, T. W., Clark-Lewis, I., and Kent, S. B. H. (1994). “Synthesis of Proteins by Native Chemical Ligation.” Science. DOI: 10.1126/science.7973629.
- Lau, J., et al. (2015). “Discovery of the Once-Weekly Glucagon-Like Peptide-1 (GLP-1) Analogue Semaglutide.” Journal of Medicinal Chemistry. DOI: 10.1021/acs.jmedchem.5b00726. Bibliographic clarification: this is the article corresponding to the supplied DOI; the alternative title in the course list does not match that DOI.
- Muttenthaler, M., King, G. F., Adams, D. J., and Alewood, P. F. (2021). “Trends in Peptide Drug Discovery.” Nature Reviews Drug Discovery. DOI: 10.1038/s41573-020-00135-8.
- International Council for Harmonisation. Q1A(R2), Q3A(R2), Q6A, and Q6B. Read each guideline’s scope alongside its recommendations before applying it to a peptide product.
⚙️ Mechanisms & Applications
Mechanisms & Applications: Making a Peptide Is an Exercise in Selective Reactivity
Skipper’s central insight: peptide synthesis is not fundamentally about discovering how to form an amide. Organic chemistry has already supplied that reaction. The challenge is making one particular amide bond, repeatedly, in a molecule containing many competing nucleophiles, acids, and stereocenters—and then proving that the intended product is what we actually obtained.
This connects directly to protein biochemistry: sequence determines the possibilities for folding and function, but the synthetic chemist must first control connectivity, stereochemistry, termini, and modifications. A deletion, epimerization, or misplaced disulfide can leave the molecular mass nearly unchanged—or completely unchanged—while altering biological activity profoundly.
1. Merrifield’s Innovation: Immobilize the Product, Wash Away the Problem
In 1963, R. Bruce Merrifield demonstrated solid-phase peptide synthesis (SPPS) through the synthesis of a tetrapeptide. The decisive innovation was a change in process architecture: attach the growing peptide to an insoluble support so that excess reagents and soluble by-products can be removed by filtration and washing rather than by isolating the peptide after every reaction. Merrifield received the 1984 Nobel Prize in Chemistry for developing this methodology. [1]
Imagine assembling a delicate machine while it remains bolted to a workbench. Tools and debris can be removed without moving the machine. The resin plays the role of the workbench—but with an important complication: reagents must diffuse into a swollen polymer network to reach the growing chains.
- Anchor the C-terminal amino acid to a resin through a cleavable linker.
- Remove its temporary N-protecting group to expose the amino group.
- Activate the carboxyl group of the next protected amino acid and couple it to that amine.
- Wash away excess amino acid, activating reagent, and soluble by-products.
- Repeat until the sequence is complete.
- Cleave and deprotect, then purify and characterize the released peptide.
Simplified chain growth:
H2N–CH(R1)–C(O)–linker–resin
+
Fmoc–NH–CH(R2)–CO2H [carboxyl activation]
↓
Fmoc–NH–CH(R2)–C(O)–NH–CH(R1)–C(O)–linker–resin
The sequence grows from its C terminus toward its N terminus.
A useful contrast with molecular biology: the ribosome synthesizes proteins N-to-C by transferring a growing chain onto an incoming aminoacyl-tRNA. Conventional SPPS grows C-to-N because the chain’s C terminus is immobilized and incoming amino acids supply activated carboxyl groups.
The limitation is as important as the advantage: washing removes soluble contaminants, but it does not remove incorrect resin-bound sequences. A failed coupling can remain hidden and generate a deletion peptide in later cycles. Automation therefore amplifies both good chemistry and bad chemistry. [1, 6, 7]
2. The Chemistry of Coupling: Activation, Nucleophilic Attack, and Stereochemical Risk
From organic chemistry, remember why simply mixing an amine and a carboxylic acid is inadequate: acid–base neutralization competes strongly, and hydroxide is a poor leaving group. Peptide coupling converts the carboxyl group into a more reactive acyl donor without allowing its α-stereocenter or other functional groups to react undesirably.
Carbodiimide-Mediated Coupling
- Activation: the carboxylic acid reacts with a carbodiimide such as DIC or DCC to form an O-acylisourea.
- Productive interception: an additive such as HOBt can convert this intermediate into an active ester.
- Amine attack: the peptide’s free amine attacks the activated carbonyl, producing a tetrahedral intermediate.
- Collapse: departure of the leaving group gives the new amide bond.
Protected-AA–CO2H + carbodiimide
→ O-acylisourea
O-acylisourea + HOBt
→ Protected-AA–C(O)–OBt + substituted urea
Protected-AA–C(O)–OBt + H2N–peptide
→ tetrahedral intermediate
→ Protected-AA–C(O)–NH–peptide + HOBt
The O-acylisourea is a branching point: it can proceed toward peptide formation, hydrolyze, or rearrange into an unproductive N-acylurea. Active-ester additives improve the competition between these pathways. They can reduce epimerization, but do not guarantee its absence. Dry HOBt and HOAt present energetic hazards; reagent choice and handling must follow appropriate institutional safety procedures. [6, 7]
Reagents such as HBTU, HATU, and PyBOP provide other routes to activated acyl species. HATU commonly generates OAt-based intermediates; HBTU and PyBOP commonly provide OBt-based chemistry. Their effectiveness reflects activation rate, leaving-group behavior, solubility, steric accessibility, and competing reactions—not simply a universal ranking of “stronger” coupling agents. Ordinary peptide coupling is best understood through active esters and related activated acyl intermediates, not free acylium ions as a general mechanism. [6, 7]
Why Activation Can Damage Stereochemistry
Activation can permit formation of an oxazolone, especially at the C terminus of a peptide fragment. Deprotonation at its α-carbon produces a planar intermediate; reprotonation from either face can yield epimers. Direct base-mediated epimerization can also contribute in susceptible systems.
Activated, N-acyl-protected amino acid or peptide terminus
⇌ oxazolone
⇌ planar enolate-like intermediate
→ either configuration at Cα
This is the point where CHEM202 stereochemistry becomes a pharmaceutical quality attribute. The product with the wrong configuration may have the correct sequence, elemental composition, and intact mass. Short activation times, suitable additives, controlled base exposure, and appropriate temperature can help, but the conditions must be validated for the particular residue and sequence. [6, 7]
3. Protecting Groups as a Programmed Sequence of Permissions
A protecting-group scheme is a reaction-control program: one set of conditions grants permission for one functional group to react while keeping others inaccessible. Orthogonality is operational, not absolute. Two groups are orthogonal only over a useful window of conditions in the actual molecular environment. [8]
- Fmoc/tBu strategy: base removes the temporary α-amino Fmoc group; acid removes most side-chain protections and cleaves many standard linkers.
- Boc/Bzl strategy: acid removes the temporary α-amino Boc group; benzyl-based side-chain protections and traditional linkers survive repeated cycles but require a more forceful final cleavage regime.
- Additional selective handles: Alloc, Dde/ivDde, and Mmt allow selected side chains to be exposed for branching, labeling, lipidation, or cyclization.
Fmoc Deprotection: Elimination Followed by Decarboxylation
Carpino and Han’s Fmoc work established a base-sensitive amino-protecting group that became central to modern peptide synthesis. Their cited 1972 paper is a foundational account; Fmoc had been introduced in earlier work in 1970. The practical importance was the ability to separate repeated N-deprotection from acid-labile side-chain protection. [2]
- A base removes the acidic proton at the fluorene 9-position.
- β-Elimination releases dibenzofulvene and the amine-derived carbamate.
- The carbamate decarboxylates, exposing the peptide amine.
- Piperidine also traps dibenzofulvene, limiting its reaction with other nucleophiles.
Fmoc–NH–peptide
--base-mediated elimination→ dibenzofulvene + peptide carbamate
--decarboxylation→ H2N–peptide + CO2
Dibenzofulvene + piperidine → trapped adduct
The UV-active deprotection products also provide a process-monitoring signal. Chemistry that removes a protecting group can simultaneously report how much material is being processed. The signal must be interpreted with attention to dilution, flow, path length, and which absorbing species is being measured. [2, 6]
Boc Deprotection: Acid-Labile, but Not “HF Every Cycle”
Boc–NH–peptide
--acid→ carbamic acid + tert-butyl-derived products
→ protonated peptide amine + CO2
Repeated Boc removal traditionally uses TFA. Anhydrous HF belongs to the final cleavage/global deprotection stage in many classical Boc/Bzl schemes, not the ordinary Boc-removal step. HF requires specialized facilities and training: its systemic toxicity and tissue penetration make it qualitatively different from an ordinary corrosive acid. Boc chemistry remains useful in selected difficult sequences and thioester syntheses, but its benefits must be weighed against this infrastructure and safety burden. [7, 8]
4. Resin Choice, Cycle Monitoring, and the Arithmetic of Accumulated Failure
The linker helps determine the molecule: Wang-type resins commonly release a C-terminal carboxylic acid, whereas Rink amide linkers commonly release a C-terminal amide. These termini are not biologically interchangeable; amidation changes charge, hydrogen bonding, proteolysis, and often receptor recognition. [6]
Resin loading is usually expressed as millimoles of reactive sites per gram of resin. Higher loading increases nominal productivity, but can also increase chain crowding and aggregation. A lower-loading resin can therefore produce more usable peptide even while carrying fewer chains.
Initial reactive sites: n0 = resin mass × loading Example: 0.50 g × 0.40 mmol g−1 = 0.20 mmol Four equivalents of an incoming amino acid: nAA = 4 × 0.20 mmol = 0.80 mmol
Swelling is essential because reaction occurs inside the polymer network, not merely on the bead’s outer surface. Solvent choice affects diffusion, peptide solvation, reagent activity, and aggregation simultaneously.
- Kaiser test: ninhydrin-based detection of free primary amines. Residual signal after coupling indicates unreacted amines, but accessibility and residue chemistry affect interpretation.
- Chloranil test: useful for secondary amines, including proline-containing termini that are poorly assessed by the conventional Kaiser test.
- Fmoc-release monitoring: UV detection tracks deprotection products and can reveal incomplete or abnormal cycles.
- Small-scale test cleavage with LC–MS: directly examines products rather than relying solely on on-resin proxy measurements.
Double coupling gives remaining amines another opportunity to react. Capping deliberately acylates unreacted amines, preventing them from rejoining subsequent chain growth. Capping can replace difficult deletion impurities with terminated products that may be easier to separate; it does not restore lost full-length yield. [6]
If each of N couplings succeeds independently with probability p: Ideal full-length fraction = pN For a 50-residue peptide requiring 49 coupling steps: p = 0.990 → 0.99049 ≈ 0.611 p = 0.995 → 0.99549 ≈ 0.782 p = 0.999 → 0.99949 ≈ 0.952
Here the expressions mean 0.990 raised to the 49th power, and likewise for the others. A “99% efficient” reaction sounds excellent until repeated many times. This model is optimistic: it omits incomplete deprotection, epimerization, cleavage losses, and purification losses. It also assumes equivalent, independent steps, whereas difficult regions can produce correlated failures.
5. Difficult Sequences: When Folding Begins Before Synthesis Ends
From BCHM601, Skipper knows that a peptide backbone is not an inert string. On resin, growing chains can form hydrogen-bonded assemblies and β-sheet-like aggregates. These structures exclude reagents and conceal reactive termini. The resulting failure can resemble weak coupling chemistry even when the actual bottleneck is accessibility. [6, 7]
- Pseudoproline dipeptides: temporary cyclic derivatives of suitable Ser, Thr, or Cys residues disrupt regular backbone interactions. Common oxazolidine-based Ser/Thr derivatives are converted back to the native residues during acidic cleavage.
- Backbone protection: Dmb- or Hmb-based strategies interrupt backbone hydrogen bonding. They can introduce steric or coupling complications of their own.
- Lower loading and alternative resin architectures: reduce crowding or improve solvation.
- Solvent optimization: adjusts swelling and chain solvation; the best solvent for the support is not automatically best for every sequence.
- Elevated temperature and microwave-assisted synthesis: accelerate reactions and can improve transport, but may also accelerate epimerization and decomposition.
Arrhenius relation: k = A exp(−Ea/RT) Heating increases the rates of both desired and undesired reactions. Selectivity depends on their relative temperature responses.
Microwave heating should not be treated as a special guarantee of peptide-bond selectivity. Rapid, well-controlled heating is useful; claims of an additional nonthermal benefit require appropriate temperature-matched comparisons.
Aspartimide: A Competing Reaction Built into the Sequence
In susceptible Asp-containing sequences, the backbone amide nitrogen of the following residue can attack the protected Asp side-chain carbonyl, producing a five-membered aspartimide. Repeated basic Fmoc-deprotection conditions are a major source of this problem.
Asp-containing peptide
--base-promoted intramolecular acyl substitution→ aspartimide
Aspartimide
--ring opening→ α-Asp and β-linked isoAsp products
± epimerization
± nucleophile-derived ring-opening products
Diagnostic lesson: isoAsp changes backbone connectivity without changing elemental composition. Intact mass alone can miss it. Aspartimide is therefore not merely a generic “TFA cleavage side reaction”; its process history often begins much earlier. [6]
6. Cleavage and Workup: Releasing the Peptide Without Creating New Impurities
In conventional Fmoc/tBu synthesis, TFA cleaves acid-labile linkers and removes many side-chain protections. This also generates reactive protecting-group-derived species. Scavengers such as water, triisopropylsilane, and selected thiols intercept these species before sensitive residues do. Cocktail composition must match the sequence, particularly when Cys, Met, or Trp is present. [6, 8]
- Cleavage and global deprotection: release the peptide and expose its intended functional groups.
- Separate the resin: remove the insoluble support.
- Recover crude peptide: cold ether precipitation is common, but recovery depends on peptide size, charge, hydrophobicity, and solvent composition.
- Redissolve and analyze: characterize the crude material before committing to preparative purification.
- Purify and formulate: isolate appropriate fractions, exchange counterions when required, and dry under validated conditions.
Trp alkylation, incomplete deprotection, Met oxidation, and disulfide-related products require distinct mechanistic explanations. Oxidation is not solved merely by trapping carbocations. Similarly, lyophilization removes volatile material but does not establish chemical purity, remove all salts, or prove that the powder is entirely peptide.
7. Solution-Phase Synthesis: Convergence Instead of a Single Long Assembly Line
Solution-phase synthesis is not simply the historical method that SPPS replaced. It remains valuable when intermediates are readily isolated, when scale favors crystallization or extraction, or when separately optimized fragments can be joined efficiently. Hybrid manufacturing combines solid-phase fragment preparation with solution-phase transformations. [7]
Linear assembly:
A → AB → ABC → ABCD → ...
Convergent assembly:
AB + CD → ABCD
For peptide fragments:
Fragment 1–CO2H + H2N–Fragment 2
--activation→ Fragment 1–C(O)–NH–Fragment 2
Convergence allows fragments to be synthesized, analyzed, and purified independently. Its costs include fragment solubility problems, extra isolation steps, and the risk of epimerizing the activated C-terminal residue during condensation. A glycine terminus avoids epimerization at that particular coupling center, but fragment boundaries must satisfy more than one criterion.
Minimum protection does not mean minimum planning. It means protecting only the groups whose unmasked reactivity would interfere with the chosen transformations. In industrial settings, the best route often minimizes total process burden rather than the number of reactions on paper.
8. Native Chemical Ligation: A Temporary Sulfur Bond Becomes a Native Amide
Dawson, Muir, Clark-Lewis, and Kent introduced native chemical ligation (NCL) in 1994. Its conceptual beauty is that sulfur provides a selective entry point, but the final junction is an ordinary peptide bond. This allowed unprotected peptide segments to be joined under aqueous conditions and greatly expanded access to chemically defined proteins. [3]
- Recognition by reactivity: an N-terminal cysteine thiolate attacks a second fragment’s C-terminal thioester.
- Thioester exchange: a reversible, sulfur-linked intermediate forms.
- Intramolecular S→N acyl transfer: the cysteine α-amino group attacks the thioester carbonyl through a favorable five-membered arrangement.
- Amide formation: the native peptide bond forms, regenerating the cysteine side-chain thiol.
Peptide A–C(O)–SR + H2N–Cys–Peptide B
⇌ Peptide A–C(O)–S–CH2–CH(NH2)–C(O)–Peptide B
--intramolecular S→N acyl transfer→
Peptide A–C(O)–NH–Cys–Peptide B
Internal cysteine residues can participate in thioester exchange, but ordinarily lack the adjacent free α-amino group required for the productive rearrangement. Selectivity is therefore generated by the combination of thiol and N-terminal amine, not by sulfur alone.
Simplified initial bimolecular dependence: v ≈ kapp[peptide thioester][N-terminal Cys peptide] Thiolate fraction: fS− = 1 / (1 + 10(pKa − pH))
Raising pH increases the thiolate fraction, but can also increase thioester hydrolysis and other competing chemistry. The actual observed rate depends on thioester identity, catalysts, sequence, solubility, and pH; the expression above is a useful starting model, not a complete universal rate law.
- Thioester access: conventional thioesters can be vulnerable to repeated basic Fmoc conditions. Fmoc-compatible routes use latent thioesters, specialized linkers, or convertible precursors; Boc chemistry offers other routes.
- Ligation followed by desulfurization: cysteine can be converted into alanine after ligation. Valine requires an appropriate thiolated valine surrogate, such as penicillamine—not ordinary cysteine. Other sulfur-containing residues must be considered when planning selectivity.
- Expressed protein ligation: engineered inteins provide recombinant protein thioesters for joining to synthetic segments.
Applications: defined histone modifications, site-specific fluorophores, segmental isotope labeling for NMR, and proteins containing noncanonical residues. Chemical or semisynthetic preparation can distinguish the effect of one precisely positioned modification from the heterogeneous mixtures often produced biologically. [3, 7, 9]
9. Half-Life Extension: Chemistry Changes the Molecule’s Relationship with the Body
A peptide can bind its receptor beautifully and still be a poor medicine because it is rapidly cleaved or cleared. Half-life extension connects synthetic chemistry to physiology: molecular size, albumin association, protease recognition, and receptor potency must be optimized together. [4, 5]
Site-Selective Lipidation of GLP-1 Analogues
Semaglutide combines an Aib substitution at position 8, which increases resistance to DPP-4 cleavage, with a C18 fatty diacid attached through a hydrophilic spacer to Lys26. Lys34 is replaced by Arg, simplifying selective acylation while retaining a basic residue. The lipid promotes reversible albumin binding, contributing to reduced clearance and prolonged exposure. Its elimination half-life is approximately one week, supporting once-weekly dosing—not generally “greater than one week.” [4]
Selectively protected peptide
→ expose Lys26 ε-NH2
→ acylate with appropriately activated linker–lipid
→ complete deprotection and purification
Peptide–lipid + albumin ⇌ albumin-bound peptide–lipid
Liraglutide uses a C16 fatty-acyl substituent attached at Lys26 through a glutamyl spacer. These medicines illustrate why protecting-group chemistry is therapeutically consequential: it determines whether a bulky pharmacokinetic modifier is placed at the intended site rather than at another lysine or the N terminus.
Simple one-compartment, first-order elimination model: C(t) = C0 exp(−kel t) t1/2 = ln(2)/kel = ln(2) × Vd/CL
This model explains why half-life depends on both clearance and distribution. Albumin binding can affect both; half-life is not simply proportional to binding affinity. Subcutaneous absorption can also shape the observed concentration–time profile.
PEGylation and Protein-Based Extensions
- NHS-ester PEGylation: amines attack an activated PEG ester to form an amide. Multiple accessible amines can generate positional heterogeneity.
- Maleimide–thiol conjugation: thiolate adds to the maleimide double bond. Linker stability matters because some conjugates undergo exchange or related transformations in biological environments.
- Enzymatic conjugation: recognition motifs can provide selectivity beyond intrinsic functional-group reactivity.
- Fc fusion: usually produced recombinantly; increased size and FcRn-mediated recycling can prolong exposure.
Design trade-off: reducing clearance is useful only if sufficient biological activity, solubility, tolerability, and manufacturability remain. [5, 9]
10. Cyclization and Backbone Engineering: Controlling the Conformational Ensemble
Cyclization is not just “joining two ends.” It changes the accessible conformations, potentially lowering the entropic cost of receptor binding and shielding cleavage sites. But a ring that stabilizes the wrong conformation can reduce activity. This is a direct application of the conformational-energy landscapes studied in advanced protein biochemistry.
- Head-to-tail cyclization: joins the terminal amine and carboxyl group; intermolecular oligomerization competes.
- Lactam bridges: selectively exposed Lys and Asp/Glu side chains form an internal amide.
- Disulfide cyclization: oxidation joins cysteine side chains; orthogonal protection can direct pairing.
- Hydrocarbon stapling: ruthenium-catalyzed ring-closing metathesis joins appropriately positioned olefin-bearing side chains.
Intramolecular closure: vcycl = kintra[P] Intermolecular reaction: volig = kinter[P]2 vcycl / volig = kintra / (kinter[P]) Effective molarity: EM = kintra/kinter Therefore: vcycl / volig = EM/[P]
Dilution suppresses bimolecular oligomerization more strongly than unimolecular cyclization. Yet enormous solvent volumes are unattractive industrially, so slow addition, preorganization, or on-resin strategies may provide practical alternatives. On-resin “pseudodilution” is not absolute: neighboring chains can still react, especially at high loading.
Possible complete pairings of 2m distinct cysteines: (2m − 1)!! = (2m)! / (2m m!) For six cysteines: 5 × 3 × 1 = 15 possible pairings
In the denominator above, 2m means 2 raised to the power m. Six cysteines therefore provide fifteen formal complete disulfide connectivities, although sequence and folding favor some over others. Mass spectrometry can establish the oxidation state without proving which cysteines are connected.
Non-natural residues extend this control. Aib restricts backbone conformations but can be difficult to couple; D-residues can resist proteolysis but alter recognition; β-amino acids change backbone spacing; and peptoids move substituents from Cα onto nitrogen, removing backbone NH donors. Retro-inverso sequences do not perfectly reproduce the parent peptide: reversing sequence and stereochemistry only approximates some side-chain arrangements and reverses backbone donor/acceptor relationships. [5, 7, 9]
Peptoid submonomer cycle:
Resin-bound amine
--bromoacetic acid acylation→ bromoacetamide
--RNH2 substitution→ N-substituted glycine unit
Repeat to extend the peptoid chain.
11. Analytical Proof: Identity, Purity, and Content Are Different Questions
Analytical chemistry supplies the discipline that synthesis alone cannot: each method answers a bounded question. A correct mass is evidence for composition, not complete proof of structure. A clean chromatogram is evidence under one separation method, not universal proof of purity.
- Reversed-phase HPLC: separates peptides through differential interactions with a hydrophobic stationary phase. Increasing organic solvent generally increases elution strength. TFA modifies peptide ionization and ion pairing, but can suppress electrospray signals.
- UV detection near 214–220 nm: detects peptide-bond-associated absorption. Response factors differ among analytes, so area percentage is not automatically mass percentage.
- ESI-MS: generates multiply charged ions, allowing large peptides to fall within convenient m/z ranges.
- MALDI-TOF MS: commonly gives predominantly singly charged peptide ions and a complementary mass measurement.
- MS/MS, mapping, or specialized separations: help locate modifications, establish disulfide connectivity, or investigate isomeric impurities.
Positive-ion ESI, protonated species: m/z = (M + z × mH+)/z M = z(m/z) − z × mH+ Approximate adjacent isotope spacing: Δ(m/z) ≈ 1.00335/z
Thus an isotope spacing near 0.50 m/z suggests a doubly charged ion. Always distinguish monoisotopic from average mass and account for termini, counterion-independent molecular composition, modifications, and adducts.
Quantification requires a separate plan. Amino acid analysis measures selected residues after hydrolysis, with corrections for residue-dependent loss or conversion. UV absorbance requires a defensible extinction coefficient. Powder mass includes water, counterions, residual solvents, and nonpeptide material.
Beer–Lambert law:
A = εlc
Illustrative peptide content:
10.0 mg powder × 0.75 peptide mass fraction
= 7.5 mg peptide equivalents
The peptide mass fraction must be measured or otherwise justified; it cannot simply be inferred from “98% HPLC purity.” Preparative HPLC fraction pooling should consider identity, impurity profiles, resolution, and product recovery—not merely whether a fraction belongs to the largest peak.
12. Manufacturing: Optimize the Whole Process, Not Just the Coupling Step
At research scale, a route may be acceptable if it yields enough clean material for an assay. At API scale, the route must also control impurity formation, solvent consumption, heat and mass transfer, residual reagents, metal catalysts, cleaning, traceability, and batch reproducibility.
Approximate isolated yield:
Yoverall =
Yassembly × Ycleavage × Ypurification × Yfinishing
Example:
0.80 × 0.90 × 0.65 × 0.95 ≈ 0.445
Only about 45% of the theoretical amount is isolated.
Continuous-flow SPPS can improve reagent delivery, temperature control, and cycle reproducibility. It does not automatically reduce solvent use: actual efficiency depends on concentrations, wash volumes, equipment hold-up, solvent recovery, and the chosen process boundary.
Process mass intensity: PMI = total mass of process inputs / mass of isolated product State the boundary explicitly: Which solvents, washes, water streams, and recovery operations are included?
GMP adds a documented system for ensuring that the process repeatedly delivers material meeting justified specifications. Relevant quality attributes can include sequence identity, peptide content, related substances, stereochemical integrity, residual solvents, elemental impurities, counterions, water, and product-appropriate microbiological controls.
Regulatory precision matters: ICH Q1 guidance supports stability planning, and Q6A/Q6B provide important specification frameworks, but their scopes differ. Q3A explicitly excludes peptides, and Q6B is not a blanket standard for chemically synthesized peptides. Peptide impurity strategies require product-specific justification and applicable regulatory guidance rather than automatic transplantation of small-molecule thresholds. [10]
What I want Skipper to retain: successful peptide synthesis means controlling an evolving network of competing reactions, not merely completing a sequence of couplings. The best route is the one that delivers the correct molecular population—with evidence—at the required scale and quality.
Sources and Citation Notes
- Merrifield, R. B. “Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide.” Journal of the American Chemical Society (1963). doi:10.1021/ja00897a025.
- Carpino, L. A.; Han, G. Y. “The 9-Fluorenylmethoxycarbonyl Function, a New Base-Sensitive Amino-Protecting Group.” Journal of the American Chemical Society (1972). doi:10.1021/ja00784a052.
- Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H. “Synthesis of Proteins by Native Chemical Ligation.” Science (1994). doi:10.1126/science.7973629.
- Lau, J. et al. “Discovery of the Once-Weekly Glucagon-Like Peptide-1 (GLP-1) Analogue Semaglutide.” Journal of Medicinal Chemistry (2015). doi:10.1021/acs.jmedchem.5b00726. Bibliographic correction: this DOI corresponds to the semaglutide discovery paper, not the “triflate chemistry” title supplied in the course list.
- Muttenthaler, M.; King, G. F.; Adams, D. J.; Alewood, P. F. “Trends in Peptide Drug Discovery.” Nature Reviews Drug Discovery (2021). doi:10.1038/s41573-020-00135-8.
- Chan, W. C.; White, P. D., eds. Fmoc Solid Phase Peptide Synthesis: A Practical Approach, 1st ed. Principal reference for Fmoc workflows, support selection, monitoring, cleavage, and sequence-dependent problems.
- Lloyd-Williams, P.; Albericio, F.; Giralt, E. Chemical Approaches to the Synthesis of Peptides and Proteins, 1st ed. Principal reference for coupling chemistry, solution-phase strategies, and synthetic route design.
- Wuts, P. G. M.; Greene, T. W. Greene’s Protective Groups in Organic Synthesis, 4th ed. Reference for protecting-group mechanisms, selectivity, and compatibility.
- Howl, J., ed. Peptide Synthesis and Applications. Companion reference for peptide preparation, modification, and biochemical applications.
- International Council for Harmonisation. Quality Guidelines, including Q1A(R2), Q3A(R2), Q6A, and Q6B. Consult each document’s scope and applicable regional peptide-specific guidance. ICH Quality Guidelines.
🪬 Skipper's Reflection
Skipper’s Reflection: Peptide Synthesis — The Engineering of Molecular Assembly
Coming out of the pure thermodynamic elegance of P-Chem and the mechanistic puzzles of Advanced Biochemistry, BCHM602 feels like a sudden, thrilling pivot into molecular engineering. SPPS isn’t just chemistry; it’s an algorithmic approach to molecule building. Sitting at the intersection of organic chemistry, polymer physics, and industrial manufacturing, this course has radically shifted how I view the synthesis of biological macromolecules. Here are my thoughts on how we force amino acids to do our bidding.
1. What Surprised Me Most
The sheer, deliberate inefficiency of solid-phase peptide synthesis (SPPS) in the pursuit of perfection. In CHEM201/202, we were taught to carefully balance stoichiometric ratios to optimize yield and minimize waste. In SPPS, Merrifield’s genius (and his 1963 JACS paper) was to throw stoichiometry out the window. We use 3x to 10x molar excesses of activated amino acids and coupling reagents.
Connecting this to MATH201 (Probability and Statistics), the necessity is obvious but still jarring. If you have a 40-amino-acid peptide and each coupling step has a 95% yield, your final yield is $0.95^{40} = 12.8\%$. But if you force the reaction to 99.5% completion using massive reagent excess, your final yield is $0.995^{40} = 81.8\%$. We are leveraging the Le Chatelier principle (from Gen Chem II) with brute force to drive the equilibrium entirely to the products. The surprise is how much solvent (DMF, NMP) and reagent is wasted to achieve this statistical necessity—it highlights a massive industrial challenge for green chemistry.
I was also fascinated by the violent reality of global deprotection and cleavage (Week 6). Chopping off acid-labile protecting groups with TFA generates a swarm of highly reactive carbocations (like the t-butyl cation). Without adding "scavengers" like ethanedithiol or triisopropylsilane to act as sacrificial lambs, these carbocations would instantly alkylate the electron-rich rings of Tryptophan or Tyrosine. It’s chemical warfare in a flask, and we have to deploy molecular decoys to protect our product.
2. Conceptual Breakthrough Moments
The concept of Orthogonality was a massive lightbulb moment. Carpino and Han’s 1972 introduction of the Fmoc group is a masterpiece of mechanistic design. The $\alpha$-amino group is protected by Fmoc, which is base-labile (removed via piperidine-mediated $\beta$-elimination), while the side chains are protected by tBu/Boc groups, which are acid-labile (removed by TFA). Because these removal mechanisms operate in completely different chemical universes (pH extremes), they are orthogonal. You can repeatedly strip the Fmoc without touching the side chains. It’s like having a set of molecular locks that only respond to entirely different physical forces.
The second breakthrough was understanding Native Chemical Ligation (NCL) (Dawson et al., 1994). SPPS practically tops out around 50-70 amino acids due to aggregation and the statistical decay of yields. How do we make a 150-amino-acid protein? NCL solves this with a thermodynamic trap. A C-terminal thioester on one fragment reacts with an N-terminal cysteine on another. The initial thioester exchange is reversible. But then, an intramolecular S-to-N acyl shift occurs, forming a native peptide bond. Because the amide bond is vastly more thermodynamically stable than the thioester, the reaction irreversibly falls into an energy well. Watching kinetics and thermodynamics perfectly align to stitch two massive molecules together was beautiful.
3. How This Changes My View of Real-World Peptides
This course completely demystified the blockbusters of the modern peptide era, particularly the incretin mimetics.
Tirzepatide and Retatrutide: I used to look at these molecules—with their massive C20 fatty diacid chains and mini-PEG linkers—and wonder, how on earth do you attach that to exactly one lysine in the middle of a 39-amino-acid sequence? Now I know: orthogonal side-chain protection (Week 5) and the chemistry described by Lau et al. (2015). You build the peptide on the resin using standard Fmoc/tBu chemistry, but for that specific Lysine, you use a protecting group like Alloc (removed by Palladium) or ivDde (removed by hydrazine). You finish the backbone, selectively unmask only that Lysine, couple the PEG and fatty acid on-resin, and then globally cleave everything else. Furthermore, I now understand that the Aib ($\alpha$-aminoisobutyric acid) at position 2 isn't just a random non-natural amino acid; it creates massive steric hindrance, locking the peptide conformation to prevent cleavage by the DPP-4 enzyme, extending its half-life.
BPC-157 and TB-500: BPC-157 is a 15-amino acid sequence. With optimized Fmoc SPPS using highly efficient uronium coupling reagents like HATU (Week 4), this is trivial to synthesize at high purity. TB-500, being a 43-amino-acid fragment, pushes closer to the boundary where $\beta$-sheet aggregation on the resin (Week 7) becomes a problem. To make TB-500 efficiently without truncation impurities, manufacturers likely have to use microwave-assisted SPPS or pseudoproline dipeptides to disrupt the growing chain's secondary structure during synthesis.
4. Questions for Further Investigation
First, if microwave-assisted SPPS works by heating the resin matrix to disrupt $\beta$-sheet aggregation and accelerate coupling kinetics, what is the precise thermodynamic boundary where we cross from "overcoming kinetic barriers" to "inducing racemization"? I want to model the activation energy required for oxazolone formation (which leads to racemization) versus standard amide bond formation at elevated temperatures.
Second, regarding continuous flow SPPS (Week 14): as we scale up peptide drugs to metric ton quantities (like Semaglutide), the solvent waste (DMF, DCM) is ecologically catastrophic. Are there aqueous-based SPPS methods or entirely enzymatic synthesis pipelines in development that can bypass organic solvents entirely, perhaps by engineering ribosomes rather than relying on solid-phase resins?
Finally, staple peptide technology (ruthenium-catalyzed ring-closing metathesis) restricts peptide flexibility to increase cell permeability. Could we use computational structural biology (from BIOL502) to predict exact staple placement that not only rigidifies an $\alpha$-helix but also perfectly mimics a small-molecule pharmacophore for oral bioavailability?
5. Confidence Rating
Most Confident: I feel highly confident in SPPS mechanism design, protecting group orthogonality (Fmoc vs. Boc/tBu), and the kinetics/thermodynamics of Native Chemical Ligation. This connects so seamlessly with my foundations in Organic and Physical Chemistry that the logic is instantly intuitive.
Least Confident: The engineering leap from analytical HPLC to Preparative HPLC and GMP manufacturing (Weeks 13-14). The transition from a 4.6 mm analytical column to a 15 cm prep column isn't just scaling up volume; it involves complex fluid dynamics, diffusion coefficients, and managing heat dissipation during purification. Additionally, the regulatory labyrinth of ICH guidelines for impurity profiling feels less like deductive science and more like a massive matrix of legal-scientific heuristics. I need to spend more time studying chemical engineering principles to truly grasp the scale-up physics.
🗺️ Visual Aids & Key Takeaways
Diagram 1: The Merrifield Solid-Phase Peptide Synthesis Cycle
╔══════════════════════════════════════════════════════════════════════════════════════╗
║ SOLID-PHASE PEPTIDE SYNTHESIS: ONE ELONGATION CYCLE ║
║ (Fmoc/tBu Strategy — C-terminus anchored, N→C growth leftward) ║
╚══════════════════════════════════════════════════════════════════════════════════════╝
┌─────────────────────────────────────────────────────────────────────────────────┐
│ RESIN BEAD ~ insoluble polymer support (e.g., Wang or Rink Amide resin) │
│ Peptide chain grows C→N (opposite to ribosomal synthesis) │
└───────────────────────────────┬─────────────────────────────────────────────────┘
│
▼
┌───────────────────────────────────────────────────────┐
│ STEP 1 — DEPROTECTION │
│ │
│ Fmoc─NH─[AA]n─···─resin │
│ │
│ + 20% piperidine in DMF (2–5 min × 2) │
│ │
│ Mechanism: piperidine acts as base → β-elimination │
│ across Fmoc C9–H bond → dibenzofulvene + CO₂ │
│ Dibenzofulvene scavenged by excess piperidine │
│ │
│ Result: free α-NH₂ on growing chain │
└───────────────────────┬───────────────────────────────┘
│ wash × 3 (DMF)
▼
┌───────────────────────────────────────────────────────┐
│ STEP 2 — ACTIVATION & COUPLING │
│ │
│ Incoming Fmoc─AA─OH │
│ + activating reagent (e.g., HATU + DIPEA) │
│ │
│ Activation sequence: │
│ Fmoc─AA─OH → Fmoc─AA─OAt (active ester) │
│ (OAt = hydroxylazabenzotriazole ester) │
│ │
│ Coupling: active ester + free NH₂ → amide bond │
│ Typical: 3–10 equiv. AA, 10–30 min, RT │
│ │
│ Monitor by Kaiser test (ninhydrin): │
│ blue/purple = free NH₂ = incomplete coupling │
│ yellow/clear = coupling complete │
└───────────────────────┬───────────────────────────────┘
│ wash × 3 (DMF)
▼
┌───────────────────────────────────────────────────────┐
│ STEP 3 — CAPPING (optional but recommended) │
│ │
│ Ac₂O / DIPEA / NMP (5 min) │
│ │
│ Acetylates any uncoupled free NH₂ │
│ Prevents truncated deletion sequences from │
│ continuing to elongate → cleaner final product │
└───────────────────────┬───────────────────────────────┘
│
▼
┌───────────────────────────────────────────────────────┐
│ REPEAT CYCLE for each amino acid (N→C direction) │
│ n residues = n deprotection + coupling cycles │
└───────────────────────┬───────────────────────────────┘
│
▼
┌───────────────────────────────────────────────────────┐
│ FINAL STEP — CLEAVAGE & GLOBAL DEPROTECTION │
│ │
│ TFA cocktail (e.g., 95% TFA / 2.5% TIS / │
│ 2.5% H₂O, 60–120 min) │
│ │
│ Simultaneously: │
│ • Cleaves peptide from resin linker │
│ • Removes all tBu, Boc, Pbf, Trt side-chain groups │
│ │
│ Carbocations released → trapped by scavengers │
│ (TIS, water, EDT) to prevent re-alkylation of │
│ Trp, Met, Cys residues │
└───────────────────────┬───────────────────────────────┘
│ precipitate in cold ether
▼
┌─────────────────────┐
│ CRUDE PEPTIDE │
│ lyophilize → HPLC │
│ purification │
└─────────────────────┘
The complete Fmoc SPPS elongation cycle — deprotection, coupling, and capping repeat for every residue, with final acid-mediated cleavage releasing the fully deprotected peptide from the solid support.
Diagram 2: Coupling Reagent Activation Cascade — From Carboxylate to Amide Bond
╔══════════════════════════════════════════════════════════════════════════════════════╗
║ COUPLING REAGENT MECHANISMS: DIC/HOAt vs. HATU ║
╚══════════════════════════════════════════════════════════════════════════════════════╝
─────────────────────────────────────────────────────────────────────────────────
PATH A: Carbodiimide Route (DIC + HOAt)
─────────────────────────────────────────────────────────────────────────────────
Fmoc─AA─OH
│
│ + DIC (diisopropylcarbodiimide)
▼
Fmoc─AA─O─C(=NH─iPr)─NH─iPr ← O-acylisourea intermediate
│ REACTIVE but prone to:
│ • racemization
│ • symmetrical anhydride formation
│ + HOAt
▼
Fmoc─AA─OAt ← HOAt (7-aza-HOBt) active ester
│ STABLE, LOW RACEMIZATION
│ + H₂N─[peptide─resin]
▼
Fmoc─AA─CO─NH─[peptide─resin] + HOAt (recyclable)
Key advantage of OAt over OBt:
The aza-nitrogen at position 7 intramolecularly assists
acyl transfer → faster coupling, less epimerization
─────────────────────────────────────────────────────────────────────────────────
PATH B: Uronium Route (HATU + DIPEA)
─────────────────────────────────────────────────────────────────────────────────
Fmoc─AA─OH
│
│ + HATU [hexafluorophosphate uronium salt of HOAt]
│ + DIPEA (base, 2 equiv.)
▼
Fmoc─AA─OAt ← same OAt ester intermediate as Path A!
│ formed in situ, very rapidly
│ + H₂N─[peptide─resin]
▼
Fmoc─AA─CO─NH─[peptide─resin] + HOAt
HATU is faster than DIC/HOAt because uronium salts pre-activate
the carboxylate without going through O-acylisourea.
─────────────────────────────────────────────────────────────────────────────────
RACEMIZATION RISK SUMMARY
─────────────────────────────────────────────────────────────────────────────────
HIGH RISK ──────────────────────────────────────────── LOW RISK
│ │
Acylium ion O-acylisourea OBt ester OAt ester │
(direct, rare) (transient) (moderate) (preferred) │
│ │
Mechanism: base abstracts α-H from activated ester →
oxazolone formation → opens to give epimer (D-amino acid)
HOAt suppresses this via neighboring-group assistance
Two major coupling pathways converge on the same OAt active ester intermediate; the crucial chemical insight is that HOAt's proximity effect physically suppresses epimerization at the α-carbon during coupling.
Diagram 3: Native Chemical Ligation — Mechanism and Protein Assembly Strategy
╔══════════════════════════════════════════════════════════════════════════════════════╗
║ NATIVE CHEMICAL LIGATION (NCL) — Dawson, Muir, Clark-Lewis, Kent (1994) ║
╚══════════════════════════════════════════════════════════════════════════════════════╝
CONCEPT: Join two unprotected peptide segments in water at a Cys junction.
No protecting groups. No organic solvents. Chemoselective.
═══════════════════════════════════════════════════════
STEP 1: TRANSTHIOESTERIFICATION (thiol–thioester exchange)
═══════════════════════════════════════════════════════
Segment A Segment B
┌──────────────────────┐ ┌────────────────────────┐
│ H₂N─[AA]m─CO─SR │ + │ Cys─[AA]n─COOH │
│ (thioester, C-term) │ │ (free Cys at N-term) │
└──────────┬───────────┘ └──────────┬─────────────┘
│ │
│ Cys─SH attacks thioester carbon │
│ (thiol–thioester exchange) │
└──────────────┬───────────────────┘
▼
┌──────────────────────────────────────────────┐
│ H₂N─[AA]m─CO─S─CH₂─Cys─[AA]n─COOH │
│ │
│ Thioester-linked INTERMEDIATE │
│ (the two segments joined by a thioester) │
└────────────────────┬─────────────────────────┘
═══════════════════════════════════════════════════════
STEP 2: S→N ACYL TRANSFER (spontaneous, irreversible)
═══════════════════════════════════════════════════════
┌──────────────────────────────────────────────┐
│ Thioester intermediate │
│ │
│ 5-membered ring transition state: │
│ │
│ O S─CH₂─ │
│ ‖ │ │
│ ···─C────NH (Cys α-amino group) │
│ └────┘ │
│ ring closes → S released │
└────────────────────┬─────────────────────────┘
▼
┌──────────────────────────────────────────────┐
│ H₂N─[AA]m─CO─NH─Cys─[AA]n─COOH │
│ │
│ NATIVE AMIDE BOND at ligation site! │
│ Product contains natural Cys residue │
└──────────────────────────────────────────────┘
═══════════════════════════════════════════════════════
EXTENDED STRATEGY: MULTI-SEGMENT PROTEIN ASSEMBLY
═══════════════════════════════════════════════════════
[Segment 1─SR] + [Cys─Segment 2─SR] + [Cys─Segment 3─COOH]
│ │ │
└────── NCL 1 ───────┘ │
│ │
[Segment 1─2─SR] │
└──────────── NCL 2 ─────────────┘
│
[Segment 1─2─3─COOH]
FULL-LENGTH PROTEIN
After ligation at non-Cys sites:
Desulfurization (RANEY® Ni or TCEP/VA-044/dodecanethiol)
Cys → Ala (radical desulfurization)
→ ligation junction reads as Ala (or Val, Leu, Phe, Thr...)
allowing NCL at pseudo-Cys sites
═══════════════════════════════════════════════════════
THIOESTER SOURCES FOR NCL
═══════════════════════════════════════════════════════
Boc SPPS → direct thioester from Boc chemistry on PAM resin
Fmoc SPPS → requires special linkers (e.g., safety-catch,
Kennemur/Blanco acyl hydrazide then oxidize to azide,
or N-acyl-benzimidazolinone methods)
Expressed → intein fusion proteins (intein self-splices,
Protein leaving C-terminal thioester on protein of interest)
Ligation = Expressed Protein Ligation (EPL)
NCL's elegance is that two unprotected peptides react chemoselectively in water — the thioester is selectively attacked by the N-terminal Cys thiol, followed by a thermodynamically driven S→N acyl shift that forms a permanent native amide bond.
Diagram 4: Half-Life Extension Strategies for Therapeutic Peptides
╔══════════════════════════════════════════════════════════════════════════════════════╗
║ HALF-LIFE EXTENSION: FROM MINUTES TO DAYS ║
║ GLP-1 analogs as the paradigm case ║
╚══════════════════════════════════════════════════════════════════════════════════════╝
Native GLP-1 t½ ≈ 2 minutes (DPP-IV cleavage at Ala8-Glu9)
│
├──── Problems: rapid enzymatic degradation
│ rapid renal clearance (MW ~3.3 kDa)
│ no albumin binding
│
▼
┌──────────────────────────────────────────────────────────────────────────────┐
│ STRATEGY 1: FATTY ACID LIPIDATION → albumin binding │
│ │
│ LIRAGLUTIDE (t½ ≈ 13 hours, once-daily injection) │
│ • C16 palmitic acid attached directly to Lys26 side-chain │
│ • Glu34→Arg substitution prevents other modifications │
│ • Fatty acid binds non-covalently to serum albumin (t½ ~19 days) │
│ • Albumin acts as reservoir: slow release of active peptide │
│ • Ala8→Aib (α-aminoisobutyric) → DPP-IV resistance │
│ │
│ SEMAGLUTIDE (t½ ≈ 7 days, once-weekly injection) │
│ • C18 fatty diacid on Lys26 via mini-PEG spacer and │
│ two γGlu linkers (full: C18─γGlu─OEG─OEG─γGlu─γGlu─Lys26) │
│ • Longer fatty acid = tighter albumin binding (lower Kd) │
│ • PEG spacer reduces steric clash with receptor surface │
│ • Ala8→Aib: DPP-IV resistant │
│ • Arg34→Lys: prevents mis-lipidation at wrong Lys │
└──────────────────────────────────────────────────────────────────────────────┘
┌──────────────────────────────────────────────────────────────────────────────┐
│ STRATEGY 2: PEGylation → reduced renal clearance & immunogenicity │
│ │
│ • NHS-ester PEG reacts with Lys ε-NH₂ (non-selective) or N-terminus │
│ • Maleimide-PEG reacts with Cys─SH (site-selective) │
│ • Enzymatic PEGylation using transglutaminase (site-selective) │
│ • PEG (40 kDa) on a 3.5 kDa peptide → effective hydrodynamic size │
│ resembles ~60–80 kDa protein → below renal filtration threshold │
│ • Trade-off: PEGylation can reduce receptor affinity │
└──────────────────────────────────────────────────────────────────────────────┘
┌──────────────────────────────────────────────────────────────────────────────┐
│ STRATEGY 3: Fc FUSION / ALBUMIN BINDING PEPTIDE TAGS │
│ │
│ • Fusing peptide to IgG Fc region via linker │
│ • FcRn receptor recycles Fc in endosomes → prevents lysosomal degradation │
│ • Albumin-binding peptide tags (ABD): 3 kDa peptide that binds albumin │
│ non-covalently but avidly (Kd ~10 nM) │
└──────────────────────────────────────────────────────────────────────────────┘
HALF-LIFE COMPARISON LADDER:
─────────────────────────────────────────────────────────────────────────────
2 min │ Native GLP-1 subcutaneous → portal vein
2–4 h │ Exenatide twice-daily
13 h │ Liraglutide once-daily
~3 days │ Dulaglutide (Fc-fusion) once-weekly
~7 days │ Semaglutide (C18 di-acid) once-weekly → once-monthly (oral!)
─────────────────────────────────────────────────────────────────────────────
Half-life extension in therapeutic peptides exploits albumin's natural longevity as a carrier — the longer and more carefully engineered the lipid linker, the tighter the albumin interaction, stretching GLP-1's two-minute biological window into a week-long therapeutic action.
Diagram 5: Orthogonal Protecting Group Hierarchy in Fmoc SPPS
╔══════════════════════════════════════════════════════════════════════════════════════╗ ║ PROTECTING GROUP ORTHOGONALITY IN Fmoc/tBu SPPS ║ ║ Three conditions must be independently addressable ║ ╚══════════════════════════════════════════════════════════════════════════════════════╝ THREE ORTHOGONAL DEPROTECTION CONDITIONS: ┌──────────────┬─────────────────────────┬──────────────────────────────────────┐ │ CONDITION │ REAGENT │ GROUPS REMOVED │ ├──────────────┼─────────────────────────┼──────────────────────────────────────┤ │ BASE │ 20% piperidine / DMF │ Fmoc (α-amine) only │ │ (mild) │ β-elimination │ — repeated every coupling cycle │ ├──────────────┼─────────────────────────┼──────────────────────────────────────┤ │ MILD ACID │ 1–5% TFA / DCM │ Mmt (Cys), Dde requires hydrazine │ │ │ or 2% TFA │ ivDde: 4% hydrazine in DMF │ │ │ │ Alloc: Pd(0) / PhSiH₃ │ │ │ │ → selective side-chain deprotection │ │ │ │ for on-resin cyclization/labeling │ ├──────────────┼─────────────────────────┼──────────────────────────────────────┤ │ STRONG ACID │ 95% TFA cocktail │ tBu (Asp, Glu, Ser, Thr, Tyr) │ │ (global) │ (final cleavage step) │ Boc (Lys, Trp) │ │ │ │ Pbf (Arg) │ │ │ │ Trt (Cys, His, Asn, Gln) │ │ │ │ + cleaves peptide from resin linker │ └──────────────┴─────────────────────────┴──────────────────────────────────────┘ FULL SIDE-CHAIN PROTECTING GROUP MAP — Fmoc SPPS standard: Amino Acid │ Functional Group │ Standard PG │ Orthogonal option ───────────────────────────────────────────────────────────────────────────── Asp, Glu │ carboxylate (γ/β) │ OtBu │ OAll (allyl, Pd remove) Lys │ ε-amine │ Boc │ ivDde, Alloc, Dde Arg │ guanidinium │ Pbf │ Pbf only (very stable) Ser, Thr │ hydroxyl │ tBu │ Trt (more labile) Tyr │ phenol │ tBu │ — Cys │ thiol │ Trt │ Acm, Mmt, StBu, Mob His │ imidazole (τ) │ Trt │ — Asn, Gln │ amide (side chain) │ Trt │ — (suppresses aspartimide) Trp │ indole (N1) │ Boc │ — (prevents oxidation) Met │ thioether │ none │ — (use scavengers) ───────────────────────────────────────────────────────────────────────────── WHY DOES Pbf ON ARG STAY ON SO LONG? Pbf = 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl Very hindered sulfonyl → slow TFA removal → requires 2–4 h TFA exposure Arg-containing peptides often need longer cleavage cocktail exposure or larger TFA volumes. Pbf removal is the SLOWEST step in global deprotection.
Orthogonality means three completely independent deprotection chemistries operating without cross-reactivity — base for α-amine cycling, mild acid or Pd for selective side-chain manipulation, and strong acid for global final deprotection, allowing complex branched or cyclic architectures to be built on a single resin.
---Key Takeaways
- Merrifield's 1963 conceptual inversion: SPPS works by anchoring the C-terminal residue to an insoluble polymer bead and elongating toward the N-terminus — the opposite direction of ribosomal synthesis. This C→N convention persists in all solid-phase work. The support allows excess reagents to drive coupling to completion and be washed away without product loss. (Merrifield, 1963)
- Fmoc deprotection is a base-mediated β-elimination, not hydrolysis: Piperidine (20% in DMF) abstracts the C9-H of the fluorenyl ring, triggering elimination to form dibenzofulvene + CO₂ + free amine. The dibenzofulvene is itself nucleophilically scavenged by excess piperidine. Monitoring the absorbance of the dibenzofulvene–piperidine adduct at 301 nm gives a real-time quantitative measure of deprotection efficiency. (Carpino and Han, 1972)
- HATU and HBTU are NOT the same reagent: HATU contains the HOAt counter-ion (azabenzotriazole) while HBTU contains HOBt. Both form OAt or OBt active esters in situ, but OAt is superior because the 7-aza nitrogen provides neighboring-group assistance during acyl transfer, reducing racemization and accelerating coupling. HATU is the reagent of choice for difficult couplings and for activating segments where epimerization is a major concern. (Lloyd-Williams et al.)
- The Kaiser test (ninhydrin) detects free primary amines by turning blue/purple: Ninhydrin reacts with primary amines via Ruhemann's purple; secondary amines (Pro, Hyp) give a brown/orange color and require the chloranil test instead. A yellow result after coupling means all amines are acylated — coupling is complete. This simple colorimetric test is indispensable for monitoring SPPS manually and still used diagnostically even in automated synthesizers. (Chan and White)
- Pbf on Arg is the slowest group to remove during TFA cleavage: Pbf (2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl) is so sterically hindered that its removal by TFA is rate-limiting in the global deprotection step. Arg-rich peptides often require 2–4 hours in 95% TFA rather than the standard 60–90 minutes. Incomplete Pbf removal is a common source of +100 Da mass adduct impurities seen by ESI-MS in the crude peptide. (Greene and Wuts)
- Aspartimide formation is the most insidious side reaction in Fmoc SPPS: When an Asp(OtBu) residue is followed by Gly, Ser, Thr, or Asn in the sequence, the backbone NH can attack the β-carboxyl to form a cyclic succinimide (aspartimide) under the repeated piperidine exposures of Fmoc deprotection. This produces a −18 Da artifact and downstream ring-opening gives α- and β-Asp isomers. The solution is to use OtBu plus backbone Dmb or Hmb protection on the following residue, or to substitute piperidine with piperazine or morpholine. (Chan and White)
- Native Chemical Ligation works because S→N acyl transfer is thermodynamically driven: The transthioesterification step (thiol–thioester exchange) is reversible, but once the N-terminal Cys amino group attacks the thioester to form a five-membered ring transition state, the S→N acyl shift is essentially irreversible — the product amide bond is ~20 kcal/mol more stable than the thioester. This thermodynamic irreversibility drives the reaction to completion even at low concentrations in aqueous buffer. (Dawson et al., 1994)
- Desulfurization expands NCL beyond Cys-containing sequences: Radical desulfurization (TCEP, VA-044 initiator, dodecanethiol in water) converts Cys → Ala after ligation, allowing NCL to be performed at "alanine" junctions if a thiol-bearing Cys was introduced just for ligation. β-thiol analogs (β-thiol-Val, β-thiol-Leu, β-thiol-Phe) similarly enable ligation at Val, Leu, Phe junctions followed by desulfurization to the natural amino acid. This dramatically expands sequence space accessible by NCL. (Muttenthaler et al., 2021)
- Semaglutide's once-weekly action depends on a precisely engineered linker, not just fatty acid length: The C18 fatty diacid on semaglutide is connected to Lys26 via a spacer of γGlu–OEG–OEG–γGlu (where OEG = mini-PEG, 8-amino-3,6-dioxaoctanoic acid). This linker serves two roles: (1) it increases distance between the C18 acid and the peptide backbone, reducing steric interference with the GLP-1 receptor, and (2) the γGlu units provide negative charge that fine-tunes albumin binding. Without the linker, fatty acid length alone cannot achieve seven-day action. (Lau et al., 2015)
- Pseudoproline (ΨPro) dipeptides disrupt β-sheet aggregation by introducing a turn mimic: When Ser, Thr, or Cys in a sequence is incorporated as a pseudoproline dipeptide (an oxazolidine with the preceding residue), the backbone conformation at that site is locked into a cis-like geometry resembling Pro. This disrupts the extended β-sheet aggregation of growing peptide chains on resin — the single most common cause of failed couplings in long or hydrophobic peptides. Pseudoprolines are incorporated as pre-formed Fmoc-AA-ΨPro-OH building blocks. (Chan and White)
- Boc SPPS requires HF for cleavage but gives direct thioesters for NCL: The Boc/Bzl strategy uses TFA for α-amine deprotection (each cycle) and anhydrous HF for final global deprotection and cleavage. HF also removes Bzl, cHx, Tos, and Cbz side-chain groups. Although HF is extremely hazardous, Boc SPPS is still used for NCL-based protein synthesis because Boc thioester resins (PAM resin with SR handle) are far more robust than Fmoc equivalents under repetitive deprotection conditions. (Lloyd-Williams et al.)
- Resin choice determines the C-terminal functionality of the final peptide: Wang resin gives a free carboxylic acid (─COOH) at the C-terminus upon TFA cleavage. Rink amide resin gives a C-terminal amide (─CONH₂) — more relevant biologically since many natural peptide hormones are C-terminally amidated. Loading the first amino acid onto the resin is the most critical step — under-loading reduces scale, over-loading causes inter-chain aggregation during synthesis. (Chan and White)
- Microwave-assisted SPPS dramatically accelerates difficult sequences: Elevated temperature (50–90 °C) reduces coupling times from 30 min to 4–5 min per cycle, decreases aggregation, and can dissolve otherwise resin-swelling problems. Modern Liberty Blue (CEM) synthesizers couple all residues at 90 °C in 4 min. The risk is epimerization of His and Cys at elevated temperature, so these residues are typically coupled at 50 °C or at room temperature. (Howl)
- Peptide drug manufacturing has crossed into multi-kilogram scale, mostly by Fmoc SPPS: Semaglutide API is produced at ton-scale by SPPS. The economic bottleneck is not the synthesis itself but solvent waste and DMF disposal — one kilogram of peptide may require hundreds of liters of DMF. Continuous flow SPPS (e.g., Zydus/CEM systems) reduces solvent use per gram by 90% compared to batch SPPS and is now entering GMP manufacturing. ICH Q3A guidelines govern elemental impurities (including DIC/HATU residues) in the final API. (Muttenthaler et al., 2021)
- Reversed-phase HPLC purification is performed at C18/C8 columns using TFA–acetonitrile gradients; peptide identity is confirmed by ESI-MS or MALDI-TOF: TFA at 0.1% serves as an ion-pairing agent that suppresses silanol interactions and gives sharp peaks; it also suppresses ionization somewhat in ESI-MS, requiring post-column removal or careful conditions. ESI-MS of peptides produces multiply charged ions [M+nH]ⁿ⁺, so a 3000 Da peptide may appear at m/z 1501 (z=2), 1001 (z=3). Peptide purity ≥95% by UV at 214 nm (which detects the peptide bond) is the standard threshold for biological assays. (Chan and White; Howl)
References
- Merrifield, R.B., 'Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide,' Journal of the American Chemical Society, 1963, 85(14), 2149–2154. DOI: 10.1021/ja00897a025
- Carpino, L.A. and Han, G.Y., 'The 9-Fluorenylmethoxycarbonyl Function, A New Base-Sensitive Amino-Protecting Group,' Journal of the American Chemical Society, 1972, 94(24), 9203–9205. DOI: 10.1021/ja00784a052
- Dawson, P.E., Muir, T.W., Clark-Lewis, I. and Kent, S.B.H., 'Synthesis of Proteins by Native Chemical Ligation,' Science, 1994, 266(5186), 776–779. DOI: 10.1126/science.7973629
- Lau, J., Bloch, P., Schäffer, L., Pettersson, I., Spetzler, J., Kofoed, J., Madsen, K., Knudsen, L.B., McGuire, J., Steensgaard, D.B., Strauss, H.M., Gram, D.X., Knudsen, S.M., Nielsen, F.S., Thygesen, P., Reedtz-Runge, S. and Kruse, T., 'Efficient Methods for Attaching Molecular Modulators to Lys via Triflate Chemistry: Application to the Synthesis of Semaglutide,' Journal of Medicinal Chemistry, 2015, 58(18), 7370–7380. DOI: 10.1021/acs.jmedchem.5b00726
- Muttenthaler, M., King, G.F., Adams, D.J. and Alewood, P.F., 'Trends in Peptide Drug Discovery,' Nature Reviews Drug Discovery, 2021, 20(4), 309–325. DOI: 10.1038/s41573-020-00135-8
- Chan, W.C. and White, P.D. (eds.), Fmoc Solid Phase Peptide Synthesis: A Practical Approach, 1st edn, Oxford University Press, Oxford, 2000.
- Lloyd-Williams, P., Albericio, F. and Giralt, E., Chemical Approaches to the Synthesis of Peptides and Proteins, 1st edn, CRC Press, Boca Raton, 1997.
- Greene, T.W. and Wuts, P.G.M., Protecting Groups in Organic Synthesis, 4th edn, John Wiley & Sons, Hoboken, 2006.
- Howl, J. (ed.), Peptide Synthesis and Applications, 2nd edn, Humana Press (Springer), New York, 2005.