**1. Introduction**

In the last decades, because of the huge advances of recombinant DNA technology, recombinant antibodies have found increasing applications in the therapy of many diseases, whether of genetic, infectious, or tumour origin. Several antibodies and antibody-based products are either approved or under investigation in clinical trials and, particularly for tumours, many of them have revolutionized classical chemotherapy based on drugs [1]. Through recombinant antibodies, it is possible to interfere with specific protein functions at DNA, RNA, or protein level. Direct targeting of pathogenic proteins can even be advantageous over the targeting of genomic sequences with an on/off mode, because it allows modulating and tailoring protein activity without affecting genomic sequences.

Currently, thanks to the ability of the mammalian immune system to produce antibodies against virtually any antigen, and to over 30 years of molecular technology studies on antibody manipulation, well-established methods allow the selection of ligands for specific protein epitopes in either intra- or extra-cellular environment. Antibody selection can be performed from recombinant antibody libraries of different kinds, even originating from animals immunized with antigens of interest. Specific antibodies can be delivered directly to the cells as purified proteins or expressed as intracellular antibodies (intrabodies) by recombinant DNA technology. Different antibody formats representing more or less extended regions of an immunoglobulin (Ig) are presently available. The small size formats, i.e., antibodies in single-chain format (scFvs) and single domain antibody or nanobodies

**Citation:** Donà, M.G.; Di Bonito, P.; Chiantore, M.V.; Amici, C.; Accardi, L. Targeting Human Papillomavirus-Associated Cancer by Oncoprotein-Specific Recombinant Antibodies. *Int. J. Mol. Sci.* **2021**, *22*,

9143. https://doi.org/10.3390/ ijms22179143

Academic Editor: Yong-Seok Heo

Received: 25 May 2021 Accepted: 20 August 2021 Published: 24 August 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

(sdAb or Nbs) [2], are the most suitable for expression as intrabodies because they are easily engineerable.

Several monoclonal antibodies (mAbs) in different formats reached the clinical stage or are in different clinical trial stages for the treatment of numerous pathologies including tumours [1,3]. We are principally interested in tumours associated to Human Papillomaviruses (HPVs), which represent a global health problem in terms of morbidity and mortality and for which many therapeutic strategies are under study. Among these, the approach based on recombinant antibodies deserves particular attention because of its potentialities related to safety, precision, and feasibility [3–5].

Here we describe, to the best of our knowledge, the different formats of recombinant antibodies against the HPV oncoproteins of Human Papillomaviruses characterized to date or currently under study and discuss whether and why they show promise for the treatment of pre-neoplastic and neoplastic lesions caused by these viruses.

#### **2. Different Antibody Formats: mAbs, scFvs and Nanobodies**

Recombinant antibody technology has undergone tremendous development in recent decades, so to prompt much progress in disease diagnosis and therapy. The use of display technologies allows in vitro selection from non-animal-derived recombinant (naïve or synthetic) repertoires (libraries) of peptides and antibody fragments in different formats such as Fab fragments (Fabs), scFvs, and Nbs. Different platforms are available such as phage display, yeast display, ribosome display, bacterial display, mammalian cell surface display, mRNA display, and DNA display. All of them mimic what occurs in vivo during antibody generation by the immune system as they rely on (1) genotypic diversity, which can be obtained by immune stimulation of a competent organism or by cloning; (2) the link existing between the genotype and phenotype; (3) selective pressure for increasing antibody specificity; and (4) amplification of specific clones originated by selective pressure. The coding sequences of binders specific for a given antigen, identified by the display technology of choice, can be expressed in prokaryotic or eukaryotic systems and tested both in vitro and in vivo for their ability to counteract the target antigen activity.

The possibility to engineer the originally identified antibody sequence represents an added value, since affinity, stability, and expression level can be improved while maintaining the desired antigen-binding properties. Furthermore, it is possible to modify the format so that the antibody could acquire new kinetic properties. Importantly, it is feasible to bypass the risk of immune reactions during clinical use by constructing antibodies from human scaffolds.

A whole IgG molecule (150 kDa) comprises heavy (H) and light (L) chains each consisting of a variable (VH and VL) and a constant (CH and CL) region covalently linked to each other and to oligosaccharides necessary for antibody effector functions and for long serum half-life. The antigen-binding regions responsible for diversity among antibodies are the complementary determining regions (CDRs), three for each VH and VL. The VH and VL joined by a disulphide bond and covalently linked to the first CH domain are obtainable by IgG proteolysis, resulting in a Fab monovalent antibody fragment (55 kDa) (Figure 1).

A few decades ago, it was observed that the N-terminal IgG fragment including the VH and VL retains the same antigen-binding capacity as the whole IgG molecule. The so-called scFvs (27 kDa) lack the constant regions and include only the VH and VL linked by a short peptide consisting of a sequence of glycine and serine residues such as (Gly4Ser)3. This arrangement provides flexibility, hydrophilicity, and resistance to proteases digestion. The linker length can be modified to favour or not the formation of multimers. In fact, shortening the linker to 3–12 amino acids prevents the formation of monomeric forms supporting inter-molecular VH-VL combinations also in different orientations, with spontaneous formation of a scFv dimer called "diabody" (60 kDa), where each of the two antigen-binding sites are formed by the VH of one scFv and the VL of the other one (Figure 1). The linkage of two scFvs in a unique molecule forms a tandem scFv. Both diabodies and tandem scFvs can have two different binding specificities, and in this case, they are called bispecific (bs). Interestingly, even the VH and VL arrangement in the scFv fusion protein can influence the binding activity, and it is currently possible to predict the best functional structure so as to design scFvs that meet all requirements by molecular modelling using a computer-aided homology method [6,7].

**Figure 1.** Schematic representation of the structure of conventional and camelidae monoclonal antibodies and of different antibody fragments. On the left, the whole monoclonal antibody (**top**) and camelid antibody (**bottom**) structures are represented. The variable light (VL) and heavy (VH) chains, as well as the constant light (CL) and heavy (CH1, CH2, CH3) chains are indicated. The complementarity determining regions responsible for antigen binding, three for each VL and each VH, are represented by stripes highlighted in different colours according to the different antigen specificity. Arrows indicate different monospecific and bispecific antibody fragments derived from the original antibody molecules with their nomenclature and molecular weight. The VH and VL in the different arrangements are connected by peptide linkers of 3–12 amino acids represented by black curved lines.

The CDRs of a scFv are embedded in an amino acid scaffold of either human or animal origin, according to the library utilized for selection. Of note, CDRs with specific binding activities can be isolated and grafted onto different scaffolds suitable for the purposes of interest. Both scFv and Fab fragments can be engineered into stable oligomers to increase binding avidity and widen antigen specificity. Specific applications of these formats are the recruitment of T-cells to tumours in immunotherapy, viral retargeting in gene therapy, and targeting of multiple antigens for a synergic/additive effect. All the mentioned antibody formats, having a size of 15–80 kDa (Figure 1), show an easy tumour penetration and are cleared from the bloodstream flow more quickly with respect to the full size IgGs (150 kDa, Figure 1). Furthermore, their genes can be easily manipulated to modify their stability, specificity, and affinity for the antigens.

Antibody engineering also allows cloning of scFv sequences into eukaryotic vectors equipped with intracellular localization signals, for the scFv expression in specific cell compartments as intrabodies. These can reach and recognize target antigens in the cellular compartments where they are located, with outcomes ranging from direct antigen blockade to indirect impairment of its activity through delocalization, to its targeted degradation [8,9].

The discovery of the smallest format of antibody fragments, the Nbs, expanded the possibilities of targeting intracellular antigens through biotechnology. Nbs derive from Camelidae species (e.g., llamas, dromedaries and camels), which in their antibody repertoire have IgG lacking both light chains and CH1 domains (heavy-chain-only antibodies: HCAbs) [10]. The variable domains of these HCAbs are named VHHs and can be isolated as single-domain antibodies (sdAbs), which are small-sized (~15 kDa) but retain the antigen-binding capability of the full-size antibody. Thanks to the small size, VHHs can easily penetrate tissues and access cryptic epitopes [11,12]. They are more soluble and capable of efficient folding with respect to conventional mAbs, which renders them suitable for high-yield production in *E. coli* and even for delivery to or expression in infected cells as intrabodies. Nbs can resist a wide pH range and high temperatures, and some of them tolerate the presence of organic compounds. Despite the non-human origin, VHHs are rarely immunogenic due to the small size and high sequence homology to the human VH3 gene family, which avoids the necessity of humanization for translation into clinic [13]. The small size also favours rapid renal clearance and facilitates the in vivo application in diagnosis rather than therapy, as the latter use requires prolongation of their half-life, which is approximately 2 h. Nevertheless, VHHs targeting haematological, oncological, infectious, inflammatory/auto-immune, bone and neurological diseases are already being evaluated in clinical trials, while the humanized VHH Caplacizumab (CabliviTM) was recently approved in Europe and USA for the treatment of acquired thrombotic thrombocytopenic purpura [14–16].
