An anti-HER2 nanobody binds to its antigen HER2 via two independent paratopes

https://doi.org/10.1016/j.ijbiomac.2021.04.032Get rights and content

Highlights

  • An experiment-driven model was built that predicted nanobody paratopes.

  • A10 is the first nanobody to have two distinct paratopes.

  • The model was validated experimentally with A10 negative mutants.

Abstract

High-resolution structural data of complexes between antibodies and membrane receptors still represent a demanding task. In this study, we used complementary sets of experimental data to obtain a structural model of the complex formed by the human epidermal growth factor receptor 2 (HER2) and its specific nanobody A10. First we identified by NMR the residues that bind or rearrange as a consequence of the complex formation. In parallel, the complex was cross-linked, digested and the resulting peptides were characterized by mass-spectrometry to define maximal distance restraints between HER2 and A10 amino acids in their complex. These independent datasets guided a docking process, refined by molecular dynamics simulations, to develop a model of the complex and estimate per-residue free-energy contributions. Such a model explains the experimental data and identifies a second, non-canonical paratope, located in the region opposite to the conventional nanobody paratope, formed by the hypervariable loop regions LH1 and LH3. Both paratopes contributed substantially to the overall affinity by binding to independent HER2 epitopes. Nanobody mutants with substitution of key interaction residues, as indicated by the model, possess significantly lower affinity for HER2. This is the first described case of a “natural” biparatopic nanobody, directly selected by in-vitro panning.

Introduction

With more than 2 million diagnosed cases and roughly 600,000 deaths per year worldwide, breast cancer is the most frequent type of malignant tumor and the main oncologic cause of mortality in women [1]. The human epidermal growth factor receptor 2 (HER2) is a membrane receptor [2] over-expressed in around 15–20% of patients diagnosed with breast cancer and correlates with aggressive forms of the disease [3]. As an accessible biomarker present on the cell surface, HER2 represents a preferential target for antibody-based diagnostics and therapy [[4], [5], [6]] and the availability of the monoclonal antibodies trastuzumab and pertuzumab has significantly improved the curative options for this breast cancer sub-class [7]. Specifically, trastuzumab binds to a HER2 epitope close to the cell membrane and blocks the downstream signalling cascade whereas pertuzumab impairs the receptor dimerization by recognizing an apical region of the extra-cellular domain [8,9]. However, acquired drug resistance, and consequent disease recurrence, is frequent in patients treated with such antibodies [10,11]. Consequently, the identification of new binders, to be used alone or in combination, would provide the necessary alternative options. In particular, several anti-HER2 nanobodies have been isolated and characterized [[12], [13], [14], [15]].

Nanobodies are single-domain antibody fragments (13–15 kDa), structurally stable, simple to engineer and considerably smaller than conventional IgGs, a feature that make them ideal reagents for in-vivo imaging and super-resolution fluorescence microscopy, [16,17] as well as to be arranged into multimeric structures [18]. A still under-appreciated, additional advantage of nanobodies is that, due to their short sequence (120–130 residues), they enable a relatively fast in-silico optimization of their biophysical properties [[19], [20], [21]]. Considering the rapid improvement of computational speed, this approach has good chances to outscore in a few years the conventional methods based on random mutagenesis and secondary panning of the hits obtained from the primary biopanning [22]. However, in-silico modelling requires structural data to guide the process of model generation. In the case of the high affinity anti-HER2 nanobody A10 [14], no structural information was available but preliminary competition results obtained by Surface Plasmon Resonance (data not shown) indicated that its epitope was partially overlapping with that of trastuzumab.

While high-resolution structures of soluble proteins, alone or in complex with antibody fragments, are relatively straightforward to accomplish, the determination of the detailed structure of membrane proteins in complex with their binding antibodies is still a time-consuming and technically demanding work. The 3D structure of the HER2 ectodomain is available [23,24]. Nonetheless, obtaining native-like antigen in sufficient amounts for crystallography is quite cumbersome and expensive. We therefore looked for an alternative method to allow reliable modelling of the HER2-receptor/anti-HER2 A10 nanobody complex.

Specifically, we combined the complementary information gained from nuclear magnetic resonance (NMR) spectroscopy [25] and mass spectrometry on covalently cross-linked HER2/A10 complex (XL-MS) [26], to guide an antigen-antibody docking step. The docked model was then refined by molecular-dynamics (MD) simulations and the individual contribution of each involved amino acid was predicted by molecular mechanics energies combined with the generalized Born and surface area continuum solvation (MM-GBSA) analysis. This integrated approach allowed obtaining an accurate model of the HER2/A10 complex that was validated by testing rationally designed A10 mutants. This structural characterization, in turn, permitted to predict contact residues between HER2 and A10, a necessary step to support rational design of mutants with desired binding capacity.

Section snippets

Production and purification of A10 and HER2-Fc

A10, fused to a 6xHis sequence and a C-tag at the C-terminus [27], was expressed and purified as described previously [28], with minor modifications. E. coli BL21 (DE3) cells, pre-transformed with a chloramphenicol-resistance plasmid carrying the sequences coding for DsbC and sulfhydryl oxidase, were further transformed with the A10 C-tag expression vector. Three growth media supplemented with ampicillin (100 μg/mL) and chloramphenicol (34 μg/mL) were used, depending on the final applications:

Experimental restraints used for docking and refinement of HER2 extracellular domain complex/A10 nanobody

The anti-HER2 nanobody A10 was produced and purified to homogeneity by standard protocols (see Materials and Methods section and Fig. S1) and quantities could be scaled up easily. HER2 ectodomain was fused to Fc for simplifying its purification by Protein A affinity chromatography, but the yields of the recombinant construct were low, despite several attempts of optimization, and prevented the possibility to crystallize the HER2/A10 complex. Fusion of HER2 with human growth hormone (hGH)

Conclusions

In-vitro panning is a potent methodology for rapid isolation of recombinant nanobodies from libraries differing for design and display format, but rarely results in the recovery of binders with excellent biophysical and functional features. Such primary panning binders can be considered as hits that require further maturation into lead reagents with improved quality in terms of affinity, yields, stability or sequence humanization [51]. The conventional wet-lab approach, based on hypermutation

Authors' contribution

Conceptualization: AS, RG, AdM; Investigation, Methodology & Data curation: DU, MO, MK, CI, MS, GB, GI, BM, SO; Funding acquisition: AS, RG, AdM; Supervision AS, RG, AdM, SF, PS; Writing - draft: DU, MO, MK, GB, MS, AS, RG, AdM, SF, PS; Writing - review & editing: DU, MO, AS, RG, AdM.

Declaration of competing interest

the authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

A.S. acknowledges financial support by the DFG (RTG 2467, project number 391498659 “Intrinsically Disordered Proteins – Molecular Principles, Cellular Functions, and Diseases”). A.d.M. acknowledges financial support by the grants ARRS/N4-0046 and ARRS/J4-9322 provided by the Javna Agencija za Raziskovalno dejavnost Republike Slovenije. Christian Ihling is acknowledged for assistance with mass spectrometry measurements.

References (55)

  • M. Hu et al.

    In vitro affinity maturation to improve the efficacy of a hypoxia-inducible factor 1α single-domain intrabody

    Biochem. Biophys. Res. Commun.

    (2020)
  • A. Ahmad

    Breast cancer statistics: recent trends

    Adv. Exp. Med.

    (2019)
  • P. van der Geer et al.

    Receptor protein-tyrosine kinases and their signal transduction pathways

    Ann. Rev. Cell Biol.

    (1994)
  • N. Iqbal et al.

    Human epidermal growth factor receptor 2 (HER2) in cancers: overexpression and therapeutic implications

    Mol. Biol. Int.

    (2014)
  • E.R. Sauter

    Reliable biomarkers to identify new and recurrent cancer

    Eur. J. Breast. Health.

    (2017)
  • S. Aebi et al.

    Primary breast cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up

    Ann. Oncol.

    (2015)
  • R.V. Kholodenko et al.

    Antibody fragments as potential biopharmaceuticals for cancer therapy: success and limitations

    Curr. Med. Chem.

    (2019)
  • S.K. Pal et al.

    Targeting HER2 epitopes

    Semin. Oncol.

    (2006)
  • J.T. Garrett et al.

    Novel engineered trastuzumab conformational epitopes demonstrate in vitro and in vivo antitumor properties against HER-2/neu

    J. Immunol.

    (2007)
  • X. Deng et al.

    Comparative analysis of evolutionarily conserved motifs of epidermal growth factor receptor 2 (HER2) predicts novel potential therapeutic epitopes

    PLoS One

    (2014)
  • D. de Melo Gagliato et al.

    Mechanisms of resistance and sensitivity to anti-HER2 therapies in HER2+ breast cancer

    Oncotarget

    (2016)
  • I. Vaneycken et al.

    Preclinical screening of anti-HER2 nanobodies for molecular imaging of breast cancer

    FASEB J.

    (2011)
  • S. Djender et al.

    Bacterial cytoplasm as an effective cell compartment for producing functional VHH-based affinity reagents and Camelidae IgG-like recombinant antibodies

    Microb. Cell Factories

    (2014)
  • S. Moutel et al.

    NaLi-H1: a universal synthetic library of humanized nanobodies providing highly functional antibodies and intrabodies

    Elife

    (2016)
  • T. Pleiner et al.

    Nanobodies: site-specific labeling for super-resolution imaging, rapid epitope-mapping and native protein complex isolation

    Elife

    (2015)
  • A. de Marco

    Recombinant expression of nanobodies and nanobody-derived immunoreagents, protein Expr

    Purif

    (2020)
  • M.A. Soler et al.

    Molecular dynamics simulations and docking enable to explore the biophysical factors controlling the yields of engineered nanobodies

    Sci. Rep.

    (2016)
  • Cited by (5)

    View full text