Synthesis and chemoselective intramolecular crosslinking of a HER2-binding affibody

Structural guidelines for stabilization of α-helical coiled coils via PEG stapling

Macrocyclization or stapling is one of the most well-known and generally applicable strategies for enhancing peptide/protein conformational stability and target binding affinity. However, there are limited structure- or sequence-based guidelines for the incorporation of optimal interhelical staples within coiled coils: the location and length of an interhelical staple is either arbitrarily chosen or requires significant optimization. Here we explore the impact of interhelical PEG stapling on the conformational stability and proteolytic resistance of a model disulfide-bound heterodimeric coiled coil. We demonstrate that (1) interhelical PEG staples are more stabilizing when placed farther from an existing disulfide crosslink; (2) e/g′ staples are more stabilizing than f/b′ or b/c′ staples; (3) PEG staples between different positions have different optimal staple lengths; (4) PEG stapling tolerates variation in the structure of the PEG linker and in the mode of conjugation; and (5) the guidelines developed here enable the rational design of a stabilized PEG-stapled HER-2 affibody with enhanced conformational stability and proteolytic resistance.

Here we identify key criteria for designing PEG-stapled coiled coils with increased conformational and proteolytic stability.

Stability Enhancement of a Dimeric HER2-Specific Affibody Molecule through Sortase A-Catalyzed Head-to-Tail Cyclization

Natural backbone-cyclized proteins have an increased thermostability and resistance towards proteases, characteristics that have sparked interest in head-to-tail cyclization as a method to stability-enhance proteins used in diagnostics and therapeutic applications, for example. In this proof-of principle study, we have produced and investigated a head-to-tail cyclized and HER2-specific ZHER2:342 Affibody dimer. The sortase A-mediated cyclization reaction is highly efficient (>95%) under optimized conditions, and renders a cyclic ZHER3:342-dimer with an apparent melting temperature, Tm, of 68 °C, which is 3 °C higher than that of its linear counterpart. Circular dichroism spectra of the linear and cyclic dimers looked very similar in the far-UV range, both before and after thermal unfolding to 90 °C, which suggests that cyclization does not negatively impact the helicity or folding of the cyclic protein. The cyclic dimer had an apparent sub-nanomolar affinity (Kd ~750 pM) to the HER2-receptor, which is a ~150-fold reduction in affinity relative to the linear dimer (Kd ~5 pM), but the anti-HER2 Affibody dimer remained a high-affinity binder even after cyclization. No apparent difference in proteolytic stability was detected in an endopeptidase degradation assay for the cyclic and linear dimers. In contrast, in an exopeptidase degradation assay, the linear dimer was shown to be completely degraded after 5 min, while the cyclic dimer showed no detectable degradation even after 60 min. We further demonstrate that a site-specifically DyLight 594-labeled cyclic dimer shows specific binding to HER2-overexpressing cells. Taken together, the results presented here demonstrate that head-to-tail cyclization can be an effective strategy to increase the stability of an Affibody dimer.

Strategic Trastuzumab Mediated Crosslinking Driving Concomitant HER2 and HER3 Endocytosis and Degradation in Breast Cancer

Efficacious anticancer therapies for targeting plasma membrane receptors with antibody based therapeutics are often contingent on sufficient endocytic delivery of receptor and conjugate to lysosomes. This results in downregulation of receptor activity and, in the case of antibody-drug conjugates (ADCs), intracellular release of a drug payload. The oncogenic receptor HER2 is a priority therapeutic target in breast cancer. Known as an "endocytosis resistant" receptor, HER2 thwarts the receptor downregulating efficiency of the frontline treatment trastuzumab and reduces the potential of trastuzumab-based therapies such as trastuzumab-emtansine. We previously demonstrated that strategically inducing trastuzumab and HER2 crosslinking in breast cancer cells promoted endocytosis and lysosomal delivery of the HER2-trastuzumab complex, stimulating downregulation of the receptor. Here we reveal that HER3, but not EGFR, is also concomitantly downregulated with HER2 after crosslinking. This is accompanied by strong activation of MEK/ERK pathway that we show does not directly contribute to HER2/trastuzumab endocytosis. We show that crosslinking induced trastuzumab endocytosis occurs via clathrin-dependent and independent pathways and is an actin-dependent process. Detailed ultrastructural studies of the plasma membrane highlight crosslinking-specific remodelling of microvilli and induction of extensive ruffling. Investigations in a cell model of acquired trastuzumab resistance demonstrate, for the first time, that they are refractory to crosslinking induced HER2 endocytosis and downregulation. This implicates further arrest of HER2 internalisation in developing trastuzumab resistance. Overall our findings highlight the potential of receptor crosslinking as a therapeutic strategy for cancer while exposing the ability of cancer cells to develop resistance via endocytic mechanisms.

Proteomimetics as protein-inspired scaffolds with defined tertiary folding patterns

Proteins have evolved as a variable platform that provides access to molecules with diverse shapes, sizes and functions. These features have inspired chemists for decades to seek artificial mimetics of proteins with improved or novel properties. Such work has focused primarily on small protein fragments, often isolated secondary structures; however, there has lately been a growing interest in the design of artificial molecules that mimic larger, more complex tertiary folds. In this Perspective, we define these agents as 'proteomimetics' and discuss the recent advances in the field. Proteomimetics can be divided into three categories: protein domains with side-chain functionality that alters the native linear-chain topology; protein domains in which the chemical composition of the polypeptide backbone has been partially altered; and protein-like folded architectures that are composed entirely of non-natural monomer units. We give an overview of these proteomimetic approaches and outline remaining challenges facing the field.

Exploring alternative antibody scaffolds: Antibody fragments and antibody mimics for targeted drug delivery

The field of targeted therapeutics has benefitted immeasurably from the development of high-affinity antibodies. These important ligands have facilitated the development of effective therapies, particularly when conjugated to potent cytotoxic payloads i.e. in antibody-drug conjugates (ADCs). The success of ADCs is evidenced by rapid adoption within the pharmaceuticals community; many major companies have dedicated ADC research programmes. However, despite the advantages, the field of ADCs has failed to live up to its full potential. Studies have emerged suggesting that traditional IgG scaffolds may not be the optimal format for targeted payload delivery. In response, the protein engineering community has begun to explore alternative high-binding protein scaffolds as antibody mimics. In this short review I will summarise the generation, modification, and application of emerging antibody fragments and synthetic antibody mimics, with a focus on their use as drug carriers. The review aims to highlight the advantages of antibody mimics, and how they could be employed to overcome the issues and limitations of traditional ADCs.

Intramolecular Thioether Crosslinking to Increase the Proteolytic Stability of Affibody Molecules

Protein therapeutics suffer from low oral bioavailability, mainly due to poor membrane permeability and digestion by gastrointestinal proteases. To improve proteolytic stability, intramolecular thioether crosslinks were introduced into a three-helix affibody molecule binding the human epidermal growth factor receptor (EGFR). Solid-phase peptide synthesis was used to produce an unmodified control protein domain and three different crosslinked protein domain variants: one with a thioether crosslink between the N-terminal lysine residue and a cysteine residue in the second loop region (denoted K4), a second with a crosslink between the C-terminal lysine residue and a cysteine residue in the first loop region (denoted K58), and a third with crosslinks in both positions (denoted K4K58). Circular dichroism (CD) and surface-plasmon-resonance-based (SPR-based) biosensor studies of the protein domains showed that the three-helix structure and high-affinity binding to EGFR were preserved in the crosslinked protein domains. In vitro digestion by gastrointestinal proteases demonstrated that the crosslinked protein domains showed increased stability towards pepsin and towards a combination of trypsin and chymotrypsin.

Use of the Nanofitin Alternative Scaffold as a GFP-Ready Fusion Tag

With the continuous diversification of recombinant DNA technologies, the possibilities for new tailor-made protein engineering have extended on an on-going basis. Among these strategies, the use of the green fluorescent protein (GFP) as a fusion domain has been widely adopted for cellular imaging and protein localization. Following the lead of the direct head-to-tail fusion of GFP, we proposed to provide additional features to recombinant proteins by genetic fusion of artificially derived binders. Thus, we reported a GFP-ready fusion tag consisting of a small and robust fusion-friendly anti-GFP Nanofitin binding domain as a proof-of-concept. While limiting steric effects on the carrier, the GFP-ready tag allows the capture of GFP or its blue (BFP), cyan (CFP) and yellow (YFP) alternatives. Here, we described the generation of the GFP-ready tag from the selection of a Nanofitin variant binding to the GFP and its spectral variants with a nanomolar affinity, while displaying a remarkable folding stability, as demonstrated by its full resistance upon thermal sterilization process or the full chemical synthesis of Nanofitins. To illustrate the potential of the Nanofitin-based tag as a fusion partner, we compared the expression level in Escherichia coli and activity profile of recombinant human tumor necrosis factor alpha (TNFα) constructs, fused to a SUMO or GFP-ready tag. Very similar expression levels were found with the two fusion technologies. Both domains of the GFP-ready tagged TNFα were proved fully active in ELISA and interferometry binding assays, allowing the simultaneous capture by an anti-TNFα antibody and binding to the GFP, and its spectral mutants. The GFP-ready tag was also shown inert in a L929 cell based assay, demonstrating the potent TNFα mediated apoptosis induction by the GFP-ready tagged TNFα. Eventually, we proposed the GFP-ready tag as a versatile capture and labeling system in addition to expected applications of anti-GFP Nanofitins (as illustrated with previously described state-of-the-art anti-GFP binders applied to living cells and in vitro applications). Through a single fusion domain, the GFP-ready tagged proteins benefit from subsequent customization within a wide range of fluorescence spectra upon indirect binding of a chosen GFP variant.

Design, Preparation, and Characterization of PNA-Based Hybridization Probes for Affibody-Molecule-Mediated Pretargeting

In radioimmunotherapy, the contrast between tumor and normal tissue can be improved by using a pretargeting strategy with a primary targeting agent, which is conjugated to a recognition tag, and a secondary radiolabeled molecule binding specifically to the recognition tag. The secondary molecule is injected after the targeting agent has accumulated in the tumor and is designed to have a favorable biodistribution profile, with fast clearance from blood and low uptake in normal tissues. In this study, we have designed and evaluated two complementary peptide nucleic acid (PNA)-based probes for specific and high-affinity association in vivo. An anti-HER2 Affibody-PNA chimera, Z(HER2:342)-SR-HP1, was produced by a semisynthetic approach using sortase A catalyzed ligation of a recombinantly produced Affibody molecule to a PNA-based HP1-probe assembled using solid-phase chemistry. A complementary HP2 probe carrying a DOTA chelator and a tyrosine for dual radiolabeling was prepared by solid-phase synthesis. Circular dichroism (CD) spectroscopy and UV thermal melts showed that the probes can hybridize to form a structured duplex with a very high melting temperature (T(m)), both in HP1:HP2 and in Z(HER2:342)-SR-HP1:HP2 (T(m) = 86-88 °C), and the high binding affinity between Z(HER2:342)-SR-HP1 and HP2 was confirmed in a surface plasmon resonance (SPR)-based binding study. Following a moderately fast association (1.7 × 10(5) M(-1) s(-1)), the dissociation of the probes was extremely slow and <5% dissociation was observed after 17 h. The equilibrium dissociation constant (K(D)) for Z(HER2:342)-SR-HP1:HP2 binding to HER2 was estimated by SPR to be 212 pM, suggesting that the conjugation to PNA does not impair Affibody binding to HER2. The biodistribution profiles of (111)In- and (125)I-labeled HP2 were measured in NMRI mice, showing very fast blood clearance rates and low accumulation of radioactivity in kidneys and other organs. The measured radioactivity in blood was 0.63 ± 0.15 and 0.41 ± 0.15%ID/g for (125)I- and (111)In-HP2, respectively, at 1 h p.i., and at 4 h p.i., the kidney accumulation of radioactivity was 0.17 ± 0.04%ID/g for (125)I-HP2 and 3.83 ± 0.39%ID/g for (111)In-HP2. Taken together, the results suggest that a PNA-based system has suitable biophysical and in vivo properties and is a promising approach for pretargeting of Affibody molecules.

Intramolecular thioether crosslinking of therapeutic proteins to increase proteolytic stability

Protein-based pharmaceuticals typically display high selectivity and low toxicity, but are also characterized by low oral availability, mainly because of enzymatic degradation in the gastrointestinal tract and poor permeability across the intestinal wall. One way to increase the proteolytic stability of peptides and proteins is by intramolecular crosslinking, such as the introduction of disulfide bridges. However, disulfide bridges are at risk of thiol-disulfide exchange or reduction during production, purification, and/or therapeutic use, whereas thioether bridges are expected to be stable under the same conditions. In this study, thioether crosslinking was investigated for a 46 aa albumin-binding domain (ABD) derived from streptococcal protein G. ABD binds with high affinity to human serum albumin (HSA) and has been proposed as a fusion partner to increase the in vivo half-lives of therapeutic proteins. In the study, five ABD variants with single or double intramolecular thioether bridges were designed and synthesized. The binding affinity, secondary structure, and thermal stability of each protein was investigated by SPR-based biosensor analysis and CD spectroscopy. The proteolytic stability in the presence of the major intestinal proteases pepsin (found in the stomach) and trypsin in combination with chymotrypsin (found in pancreatin secreted to the duodenum by the pancreas) was also investigated. The most promising crosslinked variant, ABD_CL1, showed high thermal stability, retained high affinity in binding to HSA, and showed dramatically increased stability in the presence of pepsin and trypsin/chymotrypsin, compared to the ABD reference protein. This suggests that the intramolecular thioether crosslinking strategy can be used to increase the stability towards gastrointestinal enzymes.

N-terminal engineering of amyloid-β-binding Affibody molecules yields improved chemical synthesis and higher binding affinity

The aggregation of amyloid-β (Aβ) peptides is believed to be a major factor in the onset and progression of Alzheimer's disease. Molecules binding with high affinity and selectivity to Aβ-peptides are important tools for investigating the aggregation process. An Aβ-binding Affibody molecule, ZAβ3 , has earlier been selected by phage display and shown to bind Aβ(1-40) with nanomolar affinity and to inhibit Aβ-peptide aggregation. In this study, we create truncated functional versions of the ZAβ3 Affibody molecule better suited for chemical synthesis production. Engineered Affibody molecules of different length were produced by solid phase peptide synthesis and allowed to form covalently linked homodimers by S-S-bridges. The N-terminally truncated Affibody molecules ZAβ3 (12-58), ZAβ3 (15-58), and ZAβ3 (18-58) were produced in considerably higher synthetic yield than the corresponding full-length molecule ZAβ3 (1-58). Circular dichroism spectroscopy and surface plasmon resonance-based biosensor analysis showed that the shortest Affibody molecule, ZAβ3 (18-58), exhibited complete loss of binding to the Aβ(1-40)-peptide, while the ZAβ3 (12-58) and ZAβ3 (15-58) Affibody molecules both displayed approximately one order of magnitude higher binding affinity to the Aβ(1-40)-peptide compared to the full-length Affibody molecule. Nuclear magnetic resonance spectroscopy showed that the structure of Aβ(1-40) in complex with the truncated Affibody dimers is very similar to the previously published solution structure of the Aβ(1-40)-peptide in complex with the full-length ZAβ3 Affibody molecule. This indicates that the N-terminally truncated Affibody molecules ZAβ3 (12-58) and ZAβ3 (15-58) are highly promising for further engineering and future use as binding agents to monomeric Aβ(1-40).