*4.9. Tumour-Promoting Inflammation*

Through various mechanisms, E5, E6, and E7 are all involved in the development of chronic inflammation, a major cofactor of malignant transformation. In fact, the expression of genes involved in the inflammatory response increases due to E6/E7 activity. Persistent expression of oncoproteins leads to changes in the release of several pro-inflammatory cytokines (e.g., IL-6 and IL-18) and chemokines with a well-characterized role in inflammation and carcinogenesis. By acting in an autocrine manner, cytokines affect keratinocyte differentiation, proliferation, and secretion of other soluble mediators while, in a paracrine manner, they lead to the increase of infiltrating inflammatory cells. This tumour microenvironment contributes to tumour growth, angiogenesis, resistance to apoptosis, and local immune surveillance [61]. In addition, HPV oncoproteins stimulate cyclo-oxygenase-2 (COX-2) production with the consequent activation of the COX-prostaglandin (PG) pathway, which is considered the main cause of HPV-induced inflammation [62].

#### **5. Therapeutic Recombinant Antibodies Targeting HPV Oncoproteins**

Safe and non-invasive interventions without side effects would be desirable for the treatment of HPV-related lesions. In order to be effective also in immunodeficient patients, therapeutic strategies possibly not involving the individual immune response would be recommended.

An effective and timely treatment of pre-neoplastic lesions could avoid their progression toward invasive cancer. At the same time, a well-timed and effective therapy for already established tumours could ameliorate patient prognosis. Another important therapeutic area of intervention could be the prevention of metastases deriving from surgically removed HPV tumours.

Much progress has been made in the last years to develop therapeutic vaccines against HPV-associated tumours. The main platforms include peptide- and protein-based vaccines; DNA virus- and RNA virus-based vectors; bacterial vectors; cell-based, DNA-based, and RNA-based vaccines; and vaccines combining two of the mentioned platforms. Other lines of research involve the combined use of therapeutic vaccines with other treatment modalities such as PD-1/PD-L1 axis inhibitors or other checkpoint inhibitors, HDAC inhibitors or other treatments. Many clinical trials are ongoing, with some having even reached Phase II. They are reviewed in detail elsewhere [25]. Nevertheless, critical issues

remain to be solved such as anti-vector immunity, HLA-restriction of peptides, or the difficulty of identifying valid outcomes to compare the efficacy of these strategies.

In view of their crucial role in the onset and progression of HPV-driven tumours [63], E5, E6, and E7 proteins represent ideal targets for alternative anti-tumour therapeutic approaches based on protein knock out or knock down methods. In this context, recombinant antibodies seem to represent a valid therapeutic opportunity. Currently, a number of recombinant antibodies against the HPV oncoproteins are available in different formats. In recent years, such antibodies have been tested for their therapeutic potential against HPV-associated disease. They are summarized here referring to their targets.

#### *5.1. E6-Specific Recombinant Antibodies*

Many studies demonstrated that E6 protein, with its multiple direct and indirect interactions, is an undiscussed druggable target [64]. Since its X-ray structure in complex with E6-associated protein (E6AP), the ubiquitin ligase involved in p53 polyubiquitination, and p53 has been resolved, the possibility of inhibiting this complex has been increasingly investigated through several methods among which those based on specific recombinant antibodies appear promising.

#### 5.1.1. ScFvs and mAbs

The 1F1 and 6F4 (F4) scFvs, derived from mAbs obtained by mice immunization with the GST-HPV16E6 fusion protein and targeting the E6 N-terminus, were able to hamper p53 degradation in vitro by inhibiting the formation of the E6/p53 complex [65]. Lagrange et al. characterized three further anti-16E6 mAbs (1F5, 3B8, 3F8) targeting the 16ZD2 zinc-binding domain, which were able to bind E6 through a shared 16 amino acid sequence. By comparing the activity of these mAbs to that of scFv F4, they found opposite effects, with the mAbs being unable to affect the E6AP-dependent and able to affect the E6AP-independent binding of p53, possibly as a consequence of an antibody-induced conformational change at the E6AP-binding site of E6 [66]. The capacity of intracellular folding and cytosolic stability/solubility of scFvF4 was improved by mutagenesis, obtaining the IF4-P41L scFv [66]. Such scFv expressed by adenoviral system was able to cause specific apoptosis of HPV16-positive cells in a way proportional to the scFv solubility and not related to p53 rescue, showing not to depend on the block of p53 degradation [67] (Figure 3).

Courtête et al. delivered the anti-16E6 4C6 mAb to HPV16- and HPV18-positive cells and found a specific p53 accumulation in the nucleus of HPV16-positive cells, which was favoured in the presence of a network of scFv peptide dimers linked through COOHterminal Cysteine residues (Figure 3). Interestingly, cell proliferation was hampered but apoptosis was not restored, and a synergistic effect was obtained by co-delivery of silencing RNA targeting E6 [68].

GTE6-1, a 16E6 binder selected from a scFv library constructed by Griffin et al., was able to bind to the first zinc finger of E6 with high affinity. GTE6-1 was able to recognize specifically both partially denatured and native E6 and to inhibit E6-mediated degradation of p53 in vitro [69] (Figure 3). To evaluate the capability of antibodies in different formats to hamper the E6 activity, Griffin et al. expressed the GTE6-1 scFv also as a diabody and a triabody in a number of cell lines whose proliferation depends on E6 and E7, and compared such capability to that of peptides containing the E6-binding motif ELLG. Only the scFv format induced significant nuclear apoptosis and p53 rescue in HPV16-positive cells. The reason for the poor biological effect of diabody and triabody probably relies on the size-dependent inability to diffuse through nuclear pores. The ELLG-containing peptides exhibited high target avidity but were not effective as inhibitors of E6 function.

**Figure 3.** Schematic representation of the known effects of anti-E6 intrabodies expressed in HPV-positive cells. The effects of the intracellular expression of specific single-chain antibody fragments (scFvs) (GTE6-1 and IF4-P41L) and monoclonal antibodies (mAbs) (4C6 and F127-6G6) are shown. The binding of I7nuc to E6 inhibits its nuclear export and the subsequent cytoplasmic degradation of p53. Similarly, E6 binding by the GTE6-1 scFv or 4C6 and F127-6G6 mAbs inhibits p53 degradation, preventing the resistance to apoptosis and leading to the uncontrolled cell proliferation characteristic of HPV-positive cells. The rescue of nuclear p53 levels activates the transcription of genes involved in the induction of apoptosis and in the control of cell proliferation. The pro-apoptotic effect of IF4-P41L scFv, apparently p53-independent, is also shown.

By delivering the 16E6-targeting F127-6G6mAb to HPV16-positive cells by sonoporation, Togtema et al. were able to reduce the E6-mediated p53 degradation but not to induce apoptosis [70] (Figure 3). Nevertheless, the effect was transient probably due to the inability of molecules as large as mAbs to penetrate the cell nucleus, and the outcome was different in the two HPV16-positive cell lines utilized, suggesting that different treatment plans might be necessary for in vivo tumour therapy. No mAb effect was observed in HPV-negative cells, confirming the safety of a mAb-based treatment, effective only in tumour cells.

Direct selection of scFvs as intrabodies is appropriate to identify stable binders able to recognize intracellular antigens such as E6 and E7. Indeed, scFvs unstable in reducing the intracellular environment will spontaneously exclude themselves from selection. In our studies, we selected from the SPLINT library [71] the anti-16E6 scFvI7 intrabody by Intracellular Antibody Capture Technology (IACT), which allows performing selection in the intracellular environment. In the light of E6 activity in the cell nucleus, we provided scFvI7 with a signal for nuclear localization (NLS), and tested the I7nuc effect in HPV16 positive cells. When co-transfecting the same cells with I7nuc and the recombinant E6, we observed I7nuc/E6 co-localization in cell nucleus. I7nuc caused a partial rescue of p53, which accumulated in cell nucleus and was able to markedly hamper cell proliferation and induce apoptosis and necrosis of SiHa cells [71] (Figure 3).

We then investigated the I7nuc antitumour activity in mouse models for HPV tumours based on the injection of HPV16-positive tumour cells in C57BL6 mice. The scFv capability to either prevent cancer development from scFv-expressing tumour cells, or to hinder cancer progression by delivery to already established tumours, was evaluated [71,72]

(Figure 4). We observed a clear impairment of tumour growth in all mice injected with TC-1 tumour cells expressing I7nuc by retroviral transduction before inoculation into mice, with 60% of them completely protected from tumour onset for the 4 months of observation time [71]. In a different experimental setting, we delivered the scFvI7-expressing plasmids by electroporation to newly implanted tumours. Even in this case, it was possible to hamper tumour growth, providing the proof of principle for a scFv-based early treatment of cancer. The result was confirmed using two different HPV16-positive tumour cells, namely C3 and TC-1 cells, and also by employing higher amounts of TC-1 cells to mimic tumours that are more aggressive. Through histology and immunohistochemistry, we showed that the antitumour activity is based on the induction of tumour cell death by apoptosis [72].

**Figure 4.** Antitumour effect of anti-E6 and anti-E7 recombinant antibodies in vivo. The effect of the anti-16E7 (43M2SD with localization in the endoplasmic reticulum, ER) and anti-16E6 scFvs (I7nuc with localization in the cell nucleus) in HPV-driven tumour mouse models is shown on the left. Mice tumours were electroporated after injection of scFv-expressing plasmids, resulting in the induction of apoptosis and large necrotic areas in the tumour mass due to caspase 3 activation. The effect of intratumour injection of anti-16E6 anti-16E7 mAbs in HPV16-positive tumour-bearing mice is shown on the right. The significant inhibition of tumour growth by C1P5 and TVG701Y might be due to the complement C3 deposition.

In view of a possible direct use of antibodies in protein format for therapeutic purposes, we expressed the anti-16E6 scFv coding sequences, provided or not with the NLS, in prokaryotes. We tested the stability, reactivity, and specificity towards 16E6, of I7 and I7nuc proteins purified from *E. coli* in soluble form. The scFvs in protein format delivered to HPV16-positive cell lines were able to recognize the endogenous monomeric E6 in the cell nucleus and hampered the proliferation of these cells. These results could have interesting implications for therapy also in consideration of the high safety of scFv proteins, even though a prolonged administration over time would be necessary as the proteins are subject to degradation [73].

A recent study by Jiang et al. utilized anti-16 E6 and -16 E7 mAbs in an experimental murine model based on HPV16-positive CaSki cells implanted in Balb/c nude mice. Two different doses of the anti-16E6 C1P5 and anti-16E7 TVG701Y mAbs were delivered via intraperitoneal or intratumour injections and both showed significant ability to specifically inhibit tumour growth at an extent comparable to the Cisplatin chemotherapeutic agent. Inhibition of tumour growth was virus-specific and suggested that the mechanism underlying the mAb activity consists of a specific effect causing complement deposition and a non-specific effect on macrophage polarization [74] (Figure 4). Accumulation of complement in tumour tissue facilitates the elimination of cancer cells due to opsonisation, and significantly activates complementary pathways, thus promoting surveillance by the immune system.

#### 5.1.2. Nanobodies

Nanobodies represent the new generation of anti-HPV recombinant antibodies (Figure 1). In view of the preferential localization of 16E6 protein in the cell nucleus [75], and of their advantageous properties of thermal and chemical stability, VHHs are increasingly attracting interest for the targeting of E6. Indeed, they can penetrate into the nucleus through nuclear pores and interact with epitopes inaccessible to conventional mAbs. Three VHHs binding to the recombinant E6 with nanomolar affinities were identified from two llama, immune VHH phage display libraries by Togtema et al. [76]. The capacity of the selected VHHs to bind the native E6 derived from HPV16-positive biological samples had not yet been determined at the time of the study, nor had the bound E6 epitopes been characterized.

More recently, a different 16E6-targeting nanobody was isolated and characterized [77]. Zhang et al. advocated the possible therapeutic use of such nanobody to counteract HPVinduced tumours given its capacity to inhibit both the proliferation of HPV16 -positive cells in vitro and the growth of xenograft tumours in nude mice.

Celegato et al. explored a different approach also based on nanobodies and equally aimed at neutralizing the E6 ability to degrade p53. VHHs against the degradationbinding domain (DBD) of p53 were developed and shown to stabilise nuclear p53 in HeLa cells, which harbour the HPV18 genome, with a specific effect for HPV-positive cells. Nevertheless, the VHHs were unable to rescue the p53 tumour-suppressive functions. The authors hypothesized that this was due to inhibition of p53 transactivation associated with an increased cell proliferation and viability, and highlighted that anti-p53 DBD VHHs were able to modulate protein properties even if not reaching the desired antiproliferative effect [78].
