The fundamental biological aspect of peptide vaccination is that T-cell responses require specific T-cell receptor recognition of the displayed oligopeptide-epitope. The antigen epitope corresponds to only a small part of the complete protein (polypeptide) antigen. Immunogenicity of synthetic peptide-based vaccines can be greatly affected by the method of delivery. For example, the efficiency of cytotoxic T-cell activation and induction of anti-tumor immune responses is enhanced if peptides are encapsulated in liposomes or covalently conjugated to adjuvants. Such vaccine enhancements ensure efficient capture of antigenic peptides by specialized antigen-presenting cells (APCs) which perform effective proteolytic processing for MHC class I and class II presentation to CD8+ cytotoxic- and CD4+ helper-T cells respectively. However, the development of optimal peptide delivery modalities is not trivial and to a large extent remains a process of trial-and-error based on time-intensive and indirect read-out systems. Most of what is known about in vivo processing routes of peptides is based on murine models and little data are available in humans. Furthermore, sequence and amino acid composition can alter the physical properties and immunological behavior of individual peptides. Mechanisms of intracellular routing and processing of administered peptides in APC require in depth examination.
Fig. 1 Preparation and mechanism of peptide cancer vaccine.1,2
Linear peptides display high susceptibility to breakdown by chemical reactions and enzymatic action. The peptide's immunogenic epitopes can degrade in aqueous buffers when hydrolysis occurs with Asn/Gln residues deamidation or Met and Cys side chains oxidation throughout storage. Peptidases in serum (aminopeptidases, chymotrypsin, carboxypeptidases) will degrade the peptide following injection (half-lives of the order of minutes), efficiently clearing the antigen before it can be taken up by APCs. During lyophilisation or reconstitution the thermal stress may drive peptide aggregation and β-sheet formation which decreases the effective dose and uncovers hidden epitopes that could generate off-target antibodies. Cyclic-peptide engineering avoids many of these problems: Through macrocyclisation free N- and C-termini that make peptides vulnerable to exoproteases get eliminated while D-amino acids or N-methylated residues protect internal amide bonds disulfide or thioether bridges create a resistant rigid structure to maintain peptide stability at high temperatures. Within accelerated stability experiments cyclic peptides preserve 95% purity through 6 months while linear peptides fall beneath 80% purity halfway through a month.
B-cell epitopes that bind with high affinity need longer sequences beyond the usual 8–20 residue range typical of linear peptides. Cross-linking of B-cell receptors is difficult, as are efficient uptake and presentation by dendritic cells. They generate weak or tolerogenic immune responses as a result. In addition, most peptides are promiscuous for only a handful of HLA alleles; for a given epitope this can mean that in an outbred human population only 10–30 % of patients can be covered. This enables immune escape and limits any universal applicability. Cyclic peptides, in contrast, adopt a more constrained conformation that can better mimic the native structure of protein epitopes, which improves B-cell recognition and affinity maturation. Multimeric cyclic constructs, which are generated by dendrimerization or self-assembly of the peptides into nanofibrils, present a high-density, multivalent array of the same epitope. Such constructs are effective at clustering B-cell receptors and so trigger antibody production. In addition, the cyclic scaffold can be used to present both B-cell epitopes and promiscuous T-helper epitopes on the same molecule, which ensures CD4⁺ help and long-lived memory without the need for large carrier proteins that can dilute specificity or cause unwanted anti-carrier immunity.
The more complex the design, the more challenging the synthesis of peptide antigens becomes. In particular, long, heavily modified and multi-epitope sequences are associated with stepwise SPPS of 30-mer "long peptides" being prone to aggregation, incomplete couplings and aspartimide formation, which reduces crude purity below 30 % and complicates downstream purification. Protecting groups for side-chains must tolerate repeated coupling cycles and then be removed cleanly under mild conditions to avoid side reactions such as oxidation or racemization. Cyclisation also presents challenges: head-to-tail lactamization of large peptides requires high-dilution or microfluidic pseudo-dilution to minimize intermolecular oligomerization, while formation of disulfide bridges requires controlled redox chemistry to avoid incorrect pairing of cysteines. The presented platform addresses many of these issues. Orthogonal protection strategies (Dde/ivDde for Lys, Acm/Trt for Cys) allow site-selective deprotection and on-resin cyclisation, with the cyclic product then being purified using a two-step RP-HPLC protocol with core–shell columns, achieving >95 % final purity. Real-time monitoring by UPLC-MS during synthesis also highlights incomplete couplings immediately, allowing re-coupling to occur before losses are cumulative.
While peptide vaccines are typically viewed as platform technologies for simple and globally scalable products, the costs of synthesis scale dramatically with both peptide length and chemical complexity. A 25-mer with 3 disulfide bridges or multiple ncAAs can cost >$200/g at the pilot scale due to the high consumption of coupling reagents, resin volumes, and solvent waste. At the same time, personalized neoantigen vaccines have to be produced on an individual patient basis with 20–30 unique peptides manufactured under GMP within a 4-6-week timeframe, straining capacity and quality assurance systems to their limits. To meet these challenges, flow-chemistry-enabled SPPS slashes solvent use by 80 % and speeds up cycle times to<4 min per residue. Convergent fragment condensation enables the parallel synthesis of shorter fragments, which are cyclised and ligated to each other in solution, reducing the overall resin volumes and reagent consumption by ~50 %. Disposable GMP suites with single-use reactors and validated cleaning procedures in between production campaigns allow for scaling of production runs from milligrams for first-in-human applications to multi-kilogram batches for late-stage supply, ensuring a cost-of-goods of < $50 even for structurally complex cyclic vaccines.
Benefits of cyclic peptide vaccines include well-defined antigens, non-toxicity and non-allergenicity, and induction of humoral and cellular immunity. The basic principle for their immunogenicity is major histocompatibility complex (MHC) presentation of antigenic peptides to T-cells, which confers an adaptive immune response to bacteria.
Peptide cyclization locks linear antigens into conformationally constrained macrocycles that are more stable to a range of environmental factors during vaccine processing, storage, and delivery. The peptide backbone closure eliminates free N- and C-termini which serve as natural targets for aminopeptidases and carboxypeptidases thus extends serum half-life from minutes to hours. The presence of intramolecular hydrogen bonds and rigid β-turn/β-hairpin structures protect labile amide bonds from proteases and cyclic peptides with these features preserved >95 % purity after half-year stability trials at 40 °C/75 % RH while linear compounds fell under 80 % purity after four weeks. The reduced conformational entropy in cyclic peptides minimizes their inclination towards thermal aggregation or β-sheet formation which disrupts antigen presentation. Cyclic scaffolds are also inherently more robust to the lyophilisation and reconstitution process without degradation, thus being compatible with both liquid and dry-powder vaccine formulations. Use of non-natural amino acids (e.g. D-amino acids, N-methylated residues) can add additional chemical stability against racemization and oxidative damage during long-term storage. These factors together work towards significant reductions in cold-chain requirements and excipient load while also maintaining batch consistency which is important for worldwide vaccine distribution.
The natural structural organization of cyclic antigens enables them to accurately mimic native pathogen-derived epitopes better than flexible linear peptides. This leads to greater affinity for B-cell receptors and MHC molecules, and a concomitant reduction in the entropy loss on binding which is less severe for the cyclic compared to the linear construct. Dissociation constants of cyclic peptides are often 10- to 100-fold lower than their linear counterparts. The cyclic geometry also further restricts the antigen and rigidly presents its key hydrophobic and electrostatic patches to ensure optimal recognition by MHC-II and cross-presentation pathways. Cyclic peptides can also be multimerized on dendrimeric scaffolds or self-assembled into nanofibrillar structures to display high-density antigen arrays with avidly cross-linking capacity for the B-cell receptor. These design features translate into vigorous germinal-center reactions and long-lived plasma cells. Recent work has also shown that cyclic lipopeptide conjugates engage not only the antigen but also the TLR2/4 on dendritic cells in their built-in adjuvant capacity to boost co-stimulatory signals and cytokine secretion in the absence of external and structurally heterogeneous adjuvants such as alum. The improved avidity and intrinsic adjuvanticity of cyclic constructs have led to improved neutralizing antibody titers in mouse and non-human primate models of Group A Streptococcus and SARS-CoV-2 and protective efficacy against homologous challenge in the latter, often after a single immunization. Cyclic peptides thus bridge the specificity spectrum between small-molecule haptens and full-size protein antigens while also providing a better safety profile than live-attenuated or viral-vector vaccines.
The most common disadvantage of linear peptide antigens is proteolytic instability: serum proteases reduce their half-lives in vivo to mere minutes, curtailing the antigen's exposure time to APCs. Cyclic peptides overcome this bottleneck via several, complementary mechanisms. The lack of terminal charges removes recognition sites for exopeptidases, while backbone rigidity sterically prohibits binding of endoproteases such as trypsin, chymotrypsin or elastase. Peptides gain protection from water-mediated degradation when intramolecular disulfide bonds as well as lactam or hydrocarbon staples enforce a compact hydrophobic structure upon them. Head-to-tail cyclised 15-mer peptides maintain over 90 % integrity after 24-hour exposure to 90 % human serum whereas their linear versions are almost entirely degraded in 30 minutes. The use of non-natural building blocks – D-amino acids, α,α-dialkyl glycines or N-methylated residues – can also be used to disrupt protease cleavage sites, providing a further degree of protection. This longer circulation time is important for lymphatic drainage and dendritic cell antigen presentation, and improves both humoral and cellular immunity. Importantly, this proteolytic resilience has not translated into off-target toxicity: cyclic peptides have been well tolerated in a number of Phase I trials. In brief, the greater enzymatic stability of cyclic peptides results in lower antigen doses, reduced dosing frequency, and more efficacious vaccine profiles against a wide range of targets.
We use advanced solid-phase peptide synthesis (SPPS) to produce vaccine-quality cyclic peptides quickly and consistently. Syntheses are typically initialized on high-loading Rink amide MBHA or 2-chlorotrityl resins (0.65–1.14 mmol/g) to ensure adequate loading, even for long sequences (≥30 residues). Double-coupling of each amino acid using HATU/HOAt/DIPEA activation under microwave irradiation (50–90 °C, 4 min/cycle) to drive stepwise yields >99 % and minimize racemization (<0.2 %). Lipopeptide antigens are obtained by using lipidated building blocks (e.g. C16 lipoamino acids) introduced as pre-synthesised Fmoc-DDe-C16 derivatives to enable direct on-resin conjugation without requiring post-synthetic solution-phase conjugation. Use of orthogonal protecting groups (Allyl/Alloc, Mtt, ivDde) allows site-selective modifications (e.g. azide or alkyne handles for further click conjugation) with the peptide still on-resin, without the need for additional purification steps.
As the immunogenic strength of a peptide epitope is highly conformation-dependent, the cyclisation chemistry is designed to capture the antigen in a native-like, pathogen-mimetic conformation. Head-to-tail lactamisation is used for short, helical epitopes; the reactions are pushed to completion at 2–5 mM under high-dilution flow conditions to minimize oligomerization and routinely reach 70–80 % isolated yield. For larger or loop-like epitopes, side-chain-to-side-chain disulfide or thioether bridges are introduced through on-resin strategies that preserve stereochemistry while avoiding post-synthetic manipulations. CuAAC click cyclisation is used when additional chemical handles (fluorophores, lipids, or adjuvants) need to be included.
Each vaccine candidate cyclic peptide is analyzed by a three-tier QC cascade that is compliant with ICH Q7 and FDA vaccine regulations. The first-tier analytical technique is UPLC-MS that determines molecular identity and quantifies individual impurities to ≤0.5 %. The second-tier analysis adds high-resolution MS/MS to assign fragmentation patterns and confirm the integrity of the cyclisation site; residual linear precursor, if >0.1 %, results in re-purification. The third tier involves 2D-NMR (1H–13C HSQC, ROESY) and circular dichroism to confirm that the constrained conformation faithfully reproduces the native conformation of the viral epitope, which is a necessary (but not sufficient) condition for functional antibody induction. Aggregates or host-cell contaminants are then removed by size-exclusion chromatography and subsequent endotoxin and bioburden testing. All data are recorded in an electronic batch record with electronic signature and can be directly transferred to regulatory submissions.
Peptide Synthesis Services at Creative Peptides
References
