This article is mentioned in:

Introduction

Cervical cancer (CC) ranks as the second leading cause of cancer-associated mortalities in developing nations and the 10th in more affluent countries, highlighting a global disparity in incidence rate (1,2). Human papilloma virus (HPV) strains 16 and 18 are responsible for ~70% of CC cases worldwide (3,4). The genomes of HPV16 and HPV18 contain six early and two late open reading frames and a substantial non-coding regulatory region (5). The early genes of HPV18 are key for viral replication and pathogenesis. In infected cells, the viral oncoproteins E6 and E7 drive malignant cell proliferation and other cancer-associated processes by inhibiting apoptosis (6). E6 binds p53 and Bcl-2, inhibiting apoptosis and enabling cells to bypass cell cycle protective checkpoints, thus promoting uncontrolled cell division (7). E7 targets retinoblastoma protein for ubiquitination, causing premature entry into the S phase and bypassing the G1-S checkpoint, which leads to unchecked cellular proliferation (8).

In CC, CD8+ T lymphocytes serve a key role in identifying and eliminating cancerous cells. These cytotoxic T cells predominantly activate the major histocompatibility complex (MHC) class I pathway, revealing a key role in the cell-mediated immune response against HPV-infected cells (9). CD8+ T lymphocytes recognise peptide-MHCs (p-MHC) on infected cells, triggering the release of cytotoxic chemicals that kill the HPV-infected cells (10). However, when patients test positive for high-risk HPV or exhibit abnormal cytological results, identifying those at a higher risk of developing high-grade cervical lesions or cancer without treatment remains a challenge (11). Insufficient screening among marginalised or hard-to-reach populations, such as immigrants or rural dwellers, increases the risk of undetected precancerous lesions and cancer, while excessive screening in wealthier communities leads to unnecessary treatment and associated risks and complications (12). Enhancing the efficacy of less frequent screening protocols is a practical strategy to expand CC screening availability for under-screened populations, potentially reducing both incidence and mortality rates.

To determine the specific viral proteins presented by naturally occurring human MHC class I molecules, identification of the unique epitopes that different alleles of human MHC class I molecules present to CD8+ T cells from the E6 and E7 oncoproteins is necessary. Studies have demonstrated the widespread presence of human leukocyte antigens (HLAs) A1, A2, A11, A24, B7 and B44, which belong to the HLA class I family, across the population (13–16). HPV-infected cervical cells, especially those undergoing precancerous or malignant transformation, exhibit upregulation of the E6 and E7 oncoproteins. This increase in expression facilitates the detection of these viral oncoproteins, which are used for diagnosing precancerous and cancerous lesions (17).

Phage display technology has advanced identification of optimal peptide or protein sequences, surpassing traditional methods such as ribosome and yeast display technologies), and providing deeper insight into molecular evolution (18). The effective presentation of antibody fragments on the primary coat protein of filamentous bacteriophages has enabled the generation of diverse antibody libraries, advancing therapeutic interventions for various diseases such as pemphigus foliaceus and paraneoplastic pemphigus) diseases (19). To produce effective T cell receptor (TCR)-like antibodies, T cells must recognise a pure, recombinant p-MHC in its native conformation (20). Obtaining a purified recombinant MHC/peptide complex folded in its native conformation, as recognised by T cells (21), is key for the successful production of TCR-like antibodies using either conventional hybridoma or phage display technology. These recombinant MHC/peptide complexes can be efficiently produced in relatively high quantities, facilitating antibody isolation. The generation process involves expressing the extracellular domains of the human leukocyte antigen (HLA) heavy chain and β2-microglobulin as inclusion bodies in E. coli, followed by in vitro refolding in the presence of the desired HLA-restricted peptide (22). The resulting peptide/MHC complexes exhibit high purity, exist in a monomeric form, and adopt the correct conformation, as confirmed by structural (23), and functional studies (24). Furthermore, these complexes can be biotinylated in a site-specific manner, a feature that enhances their utility in the in vitro selection of TCR-like antibodies and subsequent specificity characterisation (20). This approach holds promise as both a precise diagnostic tool and powerful immunotherapeutic strategy for combating HPV 18-induced cervical cancer.

Materials and methods

This present study followed a structured sequence of protocols that combined bioinformatics analyses with experimental laboratory procedures, as illustrated in the flowchart (Fig. S1).

Generation of HPV18 (E6&E7) p-MHC-A24 complex

As described by Yazdani et al (25), FASTA sequences from the National Centre for Biotechnology Information (NCBI) database (ncbi.com) were used to find peptides derived from the oncoproteins of HPV18, E6 and E7 (25). Amino acid sequences associated with HPV18 oncoproteins E6 and E7 were obtained using the accession nos. AGM34425.1 and AGM34461.1, respectively. The SYFPEITHI algorithm (Ver. 1.0) (syfpeithi.de) was used to determine how well the HPV18 peptides bound HLA-A24, as previously described (26). The selection of 9-mer peptides was predicated upon their binding scores derived from these SYFPEITHI algorithms. The peptides were synthesized by Elabscience Bionovation, Inc.

Transformation and evaluation of HLA-A24 and β-2-microglobulin (β2m) vectors

Vector transformation using the heat-shock approach was performed as described by Chang et al (27), with some adjustments. Independent transfection of the HLA-A24 and β2m chains was performed on competent cells of the bacterial strain BL21 (DE3) pLysS. A total of 200 µl of β2m-transformant culture was plated on a pre-prepared 2-YT agar plate supplemented with 100 µg/ml carbenicillin and 34 µg/ml chloramphenicol (Nacalai Tesque, Japan), 2% glucose, and 200 µl of HLA-A-24 transformant culture was plated on the pre-prepared 2-YT agar plate supplemented with 100 µg/ml carbenicillin and 2% glucose, and they were also incubated at 37°C overnight the next day. Plates were screened, and single colonies selected from each culture were inoculated into 10 ml of 2-YT broth (Sigma Aldrich, USA) containing the same antibiotics, followed by overnight incubation at 37°C. Subsequently, plasmid extraction was performed (27). HLA-A24 and β2m chains sequences were validated using Sanger sequencing method) conducted by 1st BASE, using T7 vector primers (T7 promoter and T7 terminator primers; Table SI). This involved comparing the sequences of the HLA-A24 and β2m regions with their original maps using SnapGene software V 1.1.3 (snapgene.com). Furthermore, the molecular sizes of the HLA-A24 and β2m chains were verified using PCR combined with gel electrophoresis (1% agarose gel containing 0.5 µg/ml ethidium bromide and Generuler™ 1 kb Plus ladder; Thermo Scientific, Inc.), employing DreamTaq Green PCR Master Mix (2X) with T7 vector promoter and terminator primers, following the manufacturer's instructions (Tables SII and SIII) (28). The results were captured using the Syngene Gene Flash Gel Imaging System, followed by quantification using ImageJ software V 1.54 g. Extraction of plasmids containing the HLA-A24 and β2m constructs was performed as previously described using Qiaprep Spin Plasmid Miniprep Kit; Qiagen, Germany) (29).

Generation of HPV16 P-MHC-A24 complexes

HLA-A24 and β2m protein extraction was performed as described by Rodenko et al (30) and stored at 4°C for future use. The proteins) underwent examination using SDS-PAGE assay to ascertain the molecular weight of the separated proteins as described by Manns (31). SDS-resolving gel with a concentration of 10% was used to analyse HLA-A24, whereas a gel with a concentration of 15% was used to examine β2m. The molecular weights of the proteins were established by comparison with a conventional protein ladder (31), briefly the concentrations of the extracted proteins were measured using the Bradford microplate assay. Subsequently, 15 µg/lane HLA-A24 and β2m proteins, along with the appropriate volume (5 µl) of Protein Marker (PM2610 Excel Band ladder; SMOBIO Technology, Inc.), was loaded onto the corresponding SDS gel and electrophoresed at 110 V for 70 min. The results were then captured using the FluorChem Q Gel Imaging System, followed by quantification analysis of the SDS gel electrophoresis bands using ImageJ software (version 1.54 g; National Institutes of Health). The refolding of HPV18 E6 and E7 p-MHC complexes containing HLA-A24 was performed as described by Luimstra et al (32), with some adjustments. Specifically, the proteins (HLA-A24, B2M and the relevant HPV18 peptide) were added to 10 ml cold dilution buffer for each complex (Tables SIV and SV). The complexes were incubated at 4°C for 10 days, covered with aluminium foil and agitated for 30 sec daily. Highly concentrated refolded MHC complexes were recovered using ultrafiltration Amicon 4 ml spin columns with a molecular weight cut-off of 30 kDa. These complexes were stored in microcentrifuge tubes, which were covered with aluminium foil at a temperature of 4°C until they were used again.

Evaluation of generated p-MHC-A24 complex stabilization by HLA ELISA

The method of Rodenko et al (30), was modified to include six wells containing 1X PBS and refolded β2M as controls. Each well was coated with 2 µg/ml streptavidin in 100 µl 1X PBS and incubated at 4°C overnight. The wells were rinsed three times with 1X PBS-T (300 µl/well; 0.1% Tween-20). Blocking was performed using 300 µl 2% BSA (Nacalai Tesque) at room temperature for 1 h. Subsequently, the wells were washed three times with 1X PBS-T before adding 50 µg/ml each p-MHC complex in 100 µl 1X PBS. Positive control wells were supplemented with 50 µg/ml refolded β2m in 100 µl 1X PBS, while negative control wells received 100 µl 1X PBS. The plate was incubated at room temperature for 1 h at 15 × g shaking. Wells were rinsed three times with 1X PBS-T and 100 µl anti-β2m HRP (1:3,000; Cat. No. GTX40576; GeneTex, Inc.) was added to each well. Following 1 h incubation at room temperature with stirring at 15 × g shaking, wells were rinsed three times with 1X PBS-T. Finally, 100 µl 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid (ABTS) (5 mg tablet; Thermo Scientific, USA) solution was added to each well at room temperature for 10–15 min, during which the colour development was monitored. An optical plate reader was used to measure the absorbance at 405 nm.

Human DAB library screening for TCR-like antibody discovery

Preparation and purification of the KM13 helper phage were performed as described by Lee et al (33). Precipitation of the helper phage was performed using a 20% polyethylene glycol-NaCl solution (Table SVI). Colony counts and bacterial titers were calculated as follows: Colony-forming units (c.f.u.)=colony count × dilution factor ×100. The DAB library was generated and purified using the same procedure as that for the helper phage. The resulting DAB library was stored at 4°C for future use (33).

Biopanning of HPV18 P-MHC-A24 complexes against human DAB library

A traditional biopanning assay for HPV18 E6 and E7 P-MHC-A24 complexes was performed using the DAB phagemid library, following the procedure outlined by Lee et al (33), with modification as described by Dass et al (34). All ELISA experiments used PBS-T and 2% BSA as the washing and blocking solutions, respectively. The secondary antibody, anti-M13 HRP (cat. no GE27-9421-01; Sigma Aldrich, USA), was used at a 1:5,000 dilution in 2% BSA buffer for 1 h at room temperature). ABTS (5 mg/tablet; Cat. No 11204521001; Sigma Aldrich, USA) was used as the detection reagent (34). Plasmid purification of specific positive monoclonal clones identified by comparative ELISA was performed according to the manufacturer's instructions (Qiagen, Inc.; Table SVII). Sanger sequencing analysis of all positive clone plasmids was conducted by 1st BASE using M13 vector primers (Table SI). Final analysis of sequencing results was performed using ImMunoGeneTics information system®/vquest databases (imgt.org) and VBASE2 (vbase2.org), as described by Dass et al (35).

Purification of monoclonal soluble DAB protein

Successful integration of validated, sequenced plasmids into competent Escherichia coli BL21(DE3) pLysS cells (Table SVIII) was performed as described by Chang et al (27). The expression of single-domain TCR-like antibodies was performed as described by Dass et al (34) with modifications (250 ml 2-YT broth medium). Intracellular extraction of soluble, domain-specific antibody fragments was performed as described by Falgenhauer et al (36), with adjustments. Specifically, a lysozyme buffer solution was prepared by dissolving lysozyme (Sigma Aldrich; Merck KGaA) in 1X PBS to a final concentration of 300 µg/ml, total volume 15 ml. The resulting pellets were homogenised in 15 ml lysozyme buffer, cooled on ice for 1 h and subjected to ultrasonication for 5 min (10/10 sec on/off) at 3.5V. The mixture underwent two rounds of centrifugation at 10,000 × g at 4°C for 30 min. The supernatant was collected and stored at 4°C for further analysis. Samples were analysed by SDS-PAGE followed by western blotting to confirm the presence of TCR-specific antibody proteins. Briefly, the concentrations of the extracted TCR-like antibody proteins were measured using the Bradford microplate assay. Subsequently, 15 µg/lane crude TCR-like antibody protein and 5 µl/lane Protein Marker (cat. no. PM2610 ExcelBand ladder; SMOBIO Technology, Inc.) were loaded onto 15% SDS gel and subjected to electrophoresis under the conditions of 110 V for 60 min. The TCR-like antibody proteins (primary antibodies) were then transferred onto PVDF membranes. Following a 1-h blocking step at room temperature with 2% BSA (Nacalai Tesque, Japan) dissolved in PBS, the membranes were washed three times with PBS-T buffer and then incubated with an anti-c-Myc HRP (9E10) secondary antibody (1:5,000; Cat. No. NC9638329; Santa Cruz Biotechnology, USA) for 1 h at room temperature. The secondary antibody used in this study specifically detects c-Myc-tagged TCR-like antibody proteins. After incubation, the membranes were washed 3 times again with PBS-T buffer. Protein bands were visualised using the Peroxidase Stain DAB Kit (Brown Stain Kit, Cat. No. 25985-50; Nacalai Tesque, Japan), 10 min incubation at room temperature. The molecular weights of the detected proteins were determined by comparing them with the Protein Marker (PM2610 ExcelBand ladder) using ImageJ software (version 1.54 g; National Institutes of Health). Purification of TCR-like antibody proteins was performed as described by Rouet et al (37), using Protein A high-performance (HP) column according to the manufacturer's instructions (Cytiva), with adjustments. To increase the amount of purified antibody, the pH of the binding/wash buffer (20 mM sodium phosphate) was optimized to 6.8 and the elution buffer (100 mM citric acid) was set at pH 2.8. A binding buffer pH of 6.8 allowed the protein to retain activity and improved its interaction with Protein A. Similarly, a pH of 2.8 in the elution buffer facilitated complete dissociation from Protein A, improving both protein release and concentration (38). Since buffers stored at 4°C exhibited a pH change of 0.4, they were freshly prepared at room temperature. Before sample introduction (15 ml), the column was washed with 5 ml binding solution. Next the sample was passed through the column followed by two washes (5 and 2 ml, respectively), waste fractions were collected in tubes labelled W1 and W2. The elution solution was applied in runs of 5 and 2 ml, with eluted fractions collected in tubes E1 and E2. The soluble TCR-like antibodies were subsequently concentrated using a HiTrap Protein A HP column (10 kDa a molecular weight cut-off) and analysed by SDS-PAGE as previously described.

Exploring the ability of TCR-like antibodies to bind corresponding p-MHC-A24 complexes

A systematic evaluation of the ability of soluble, purified TCR-like antibodies to selectively bind their target was conducted using TCR-ELISA and western blotting, with adjustments based on the approach developed by Dass et al (34). To each well of a 96-well microtiter plate, 100 µl purified TCR-like antibody (targeted and non-targeted) diluted in 1X PBS at a concentration of 10 µg/ml was added. The background control was 1X PBS, while the negative control used the non-target 16 kDa p-MHC-A24 complex. The plates were incubated overnight at 4°C. The wells were rinsed three times with PBS-T, followed by blocking with 2% BSA for 1 h, 80 × g at room temperature. Following another washing step (3 times with PBS-T), 100 µl HPV18 p-MHC-A24 complexes in 1X PBS at a concentration of 10 µg/ml was added to the wells containing the antibodies and incubated for 1 h, 80 × g at room temperature. Following three washes with PBS-T), 100 µl streptavidin-HRP, (1:5,000 in 2% BSA; cat. No. 21130; Thermo Fisher Scientific, USA) was added to each well and incubated 1 h, 80 × g at room temperature. A final 3 times wash with PBS-T was performed and 100 µl ABTS (5 mg/tablet; Cat. No 11204521001; Sigma Aldrich, USA) substrate was added to each well. The plates were placed in the dark, the colour development was monitored, and absorbance was measured at 405 nm after 15 min at room temperature. Western blot analysis of antibody-antigen interactions was performed as described by Dass et al (34), with adjustments. The concentrations of the purified TCR-like antibody proteins were measured using the Sigma Bradford microplate assay. Subsequently, 15 µg/lane each purified TCR-like antibody protein and 5 µl of Protein Marker (PM2610 ExcelBand ladder; SMOBIO Technology, Inc.) were loaded onto a 15% SDS gel (two gels). The first gel was stained overnight with Coomassie Brilliant Blue, while the second gel included purified TCR-like antibody proteins (primary antibodies) was transferred directly onto a PVDF membrane using a wet transfer system (100 V for 60 min in 1X cold Towbin buffer) for western blot analysis. Following 1 h blocking step at room temperature with 2% BSA (Nacalai Tesque, Japan) dissolved in PBS, then incubated 1 h at room temperature with 10 µg/ml of the HPV18 p-MHC-A24 complexes in 1X PBS (5 ml total) under gentle agitation. The membranes were washed three times with PBS-T buffer and then incubated with streptavidin-HRP diluted in 2% BSA (1:5,000; Cat. No. 21130; Thermo Scientific, USA) for 1 h at room temperature. The streptavidin-HRP reagent specifically detects biotin-tagged HLA-A24 chain protein. After incubation, the membranes were washed three more times with PBS-T buffer. Protein bands were visualised using the Peroxidase Stain DAB Kit (Brown Stain Kit, Cat. No. 25985-50; Nacalai Tesque, Japan), 10 min incubation at room temperature. The molecular weights of the detected protein bands were determined by comparing them with the Protein Marker (PM2610 ExcelBand ladder) using ImageJ software (version 1.54 g; National Institutes of Health).

Statistical analysis

A parametric one-way ANOVA was conducted using GraphPad Prism 10 software (Dotmatics), followed by Tukey's post hoc test to analyse the results. P<0.05 was considered to indicate a statistically significant difference. Results are presented as the mean ± standard deviation from three independent experiments. ELISA was performed using sextuplicate wells for each sample and control.

Results

Bioinformatics analysis for generation of HPV18 P-MHC-A24 complexes

Using the NCBI FASTA database, peptides derived from the E6 and E7 oncoproteins of human papillomavirus type 18 (HPV18) were identified. The affinity of these peptides for binding to MHC-A24 was assessed using the SYFPEITHI algorithm. The binding scores for each peptide sequence were notably high. The synthesis of the selected peptides achieved a purity of 98% (Table I).

Table I. Binding scores of HPV18 E6 and E7 oncoproteins to HLA-A24 using the National Centre for Biotechnology Information database and SYFPEITHI method.

Table I.

Binding scores of HPV18 E6 and E7 oncoproteins to HLA-A24 using the National Centre for Biotechnology Information database and SYFPEITHI method.

HPV oncoprotein	Peptide sequence	SYFPEITHI score	Purity, %
HPV18 E6	V Y C K T V L E L	24	98
HPV18 E7	V D L L C H E Q L	16	98

[i] HPV, human papilloma virus; HLA, human leukocyte antigen.

Transformation and evaluation of HLA-A24 and β2m vectors

The sequence integrity of the HLA-A24 and β2m chains was verified through analysis of sequencing data (Fig. S2, Fig. S3, Fig. S4, Fig. S5, Fig. S6, Fig. S7). These data were compared with the original vector maps using SnapGene software (Table SIX). The results showed complete match with the original sequence (Figs. 1 and 2). The sizes of the synthesised HLA-A24 and β2m chains were largely in accordance with their corresponding sizes as indicated in the vector system map. Specifically, the HLA-A24 chain was 1,144 bp, while the β2m chain was 495 bp (Fig. 3A).

Figure 1. Sequence analysis of the HLA-A24 chain. HLA-A24-T7 promoter and terminator sequencing results (600 bp on each side) were compared with the original HLA-A24 map. Sanger sequencing was conducted using T7 vector primers. HLA, human leukocyte antigen.

Figure 2. Sequence analysis of the β2m chain. β2m-T7 promoter and terminator sequencing results (300 bp on each side) were compared with the original β2m chain map. Sanger sequencing was conducted using T7 vector primers. β2m, β-2-microglobulin.

Figure 3. Evaluation of P-MHC-A2402 complex components. (A) Agarose gel electrophoresis visualizing the PCR amplification of HLA-A24 and β2m chains. L, 1 KB plus DNA ladder; 1, HLA-A24 chain at ~1,144 bp; 2, β2m chain at ~495 bp; 3, negative control (ddH2O). (B) SDS-PAGE analysis of inclusion body fractions of heavy and light chains. L, pre-stained protein ladder (9–245 kDa); 1, HLA-A24 fraction band at ~35 kDa or β2m fraction band at ~12 kDa. HLA-ELISA analysis of the stability of the generated p-MHC complex. The OD 405 nm results indicate successful formation of the HPV18 (C) E6 and (D) E7 p-MHC-A24 complexes. ***P<0.001. HLA, human leukocyte antigen; β2m, β-2-microglobulin; OD, optical density.

Generation of HPV18 p-MHC-A24 complexes

Quantification of proteins from both chain fractions was performed using the Bradford assay for each chain. SDS-PAGE was employed to examine the HLA-A24 and β2M protein fractions for the determination of target protein sizes. HLA-A24 chains had the expected protein size of ~35 kDa (Fig. 3B), whereas the β2M chain was ~12 kDa (Fig. 3B).

Evaluation of generated p-MHC stabilisation by HLA ELISA

ELISA indicated increased levels of p-MHC complexes compared with both the negative and positive controls, validating the development of stable p-MHC complexes. ELISA yielded values of 0.0586 for the negative control and 0.5291 for the positive control. The corresponding values for the E6 p-MHC and the E7 p-MHC complexes were 0.7481and 0.7059, respectively (Fig. 3C and D). This indicated successful formation of stable p-MHC complexes through refolding of the HPV18 peptides with the β2M light chain and the HLA-A24 heavy chains.

Biopanning of HPV18 p-MHC-A24 complexes against human DAB library

After preparing and purifying the biopanning phages, the titer of the KM13 helper phage was 5×1012 c.f.u. The size of the main antibody library was confirmed by titration to be 4×1012 c.f.u. Biopanning of HPV18 p-MHC-A24 complexes was carried out using a library of human DABs. E6 and E7 p-MHC-A24 exhibited enrichment ratios of 1.6×104 and 4.5×105, respectively (data not shown). These values suggest an enrichment of >50%.

Polyclonal antibody phage ELISA

Polyclonal ELISA revealed HPV18 E6 p-MHC-A24 values of 0.349 and 0.984 in the first and second rounds, respectively, and HPV18 E7 p-MHC-A24 values of 0.462 and 1.445 in the first and second rounds, respectively (Fig. 4). Based on these results, the phages from the second round were used for monoclonal analysis of each target.

Figure 4. Polyclonal domain antibody phage ELISA. Following background normalization, there was greater phage enrichment against the HPV18-E6 and E7 p-MHC complexes. Round 2 yielded a higher number of recovered phages compared with round 1, as indicated by OD 405 nm measurements. ***P<0.001. OD, optical density; HPV, human papilloma virus; p-MHC, peptide-major histocompatibility complex.

Monoclonal antibody phage ELISA

A total of 94 clones were analysed in the monoclonal antibody assay. Of these, 28 clones were verified as positive for the HPV18 complex E6 p-MHC-A24 (Fig. 5), while 14 clones tested positive for the HPV18 complex E7 p-MHC-A24 (Fig. 6). To select clones that demonstrated high binding affinity, which will be selected for further analysis in future steps, the results with values <0.4 nm was excluded after subtracting background values, which were uniformly low for all clones.

Figure 5. Monoclonal domain antibody phage ELISA against HPV18 E6 p-MHC-A24 complex. A total of 28 positive clones were selected (*) based on an optical density 405 nm cut-off value >0.4, following background subtraction during comparative ELISA. HPV, human papilloma virus; p-MHC, peptide-major histocompatibility complex.

Figure 6. Monoclonal domain antibody phage ELISA against HPV18 E7 p-MHC-A24 complex. A total of 14 positive clones were selected (*) using the optical density 405 nm cut-off value >0.4, following background subtraction in the comparative ELISA step. HPV, human papilloma virus; p-MHC, peptide-major histocompatibility complex.

Comparative ELISA

Prior to sequencing analysis, comparative ELISA was conducted to evaluate positive clones against target and non-target HLA-A24 peptide complexes and decrease false positives. A total of nine clones (A11, B12, C11, D10, D11, E1, E9, G1 and H11) tested positive for the HPV18 E6 p-MHC-A24 complex (Fig. 7A) and eight (A9, E5, E7, F9, F10, G1, G8 and G12) tested positive for the HPV18 E7 P-MHC-A24 complex (Fig. 7B). Results with values <0.4 nm were excluded after adjusting for background value, as the background values for all clones were uniformly low.

Figure 7. Comparative ELISA of selected monoclonal clones against the targets and non-target (negative control). Comparative ELISA was performed to select (*) specific binders of human papilloma virus 18 (A) E6 and (B) E7 peptide-major histocompatibility complex-A24 complex.

Antibody sequencing

Sequence analysis identified three positive clones against HPV18 p-MHC complexes, each presenting a complete sequence without frameshift mutations or stop codons, as confirmed by VBASE2 and IMGT/VQUEST testing. H11 clone, corresponding to the HPV18 E6 p-MHC-A24 complex, was designated as H11/Ab, while E5 and G8 clones, corresponding to the HPV18 E7 P-MHC-A24 complex, were designated E5/Ab and G8/Ab, respectively. Analysis confirmed the absence of stop codons and in-frame junctions in these positive clones. The TCR structure, including the variable, diversity (D), and joining (J) gene segments, was characterized. These diversification mechanisms contribute to the vast TCR repertoire and are responsible for encoding the complementarity-determining regions (CDRs), with each TCR containing three CDRs (CDR1, CDR2 and CDR3). Among these, CDR3 exhibits the highest variability (Table II). According to the IMGT, the H11/Ab clone comprised 195 amino acids, while the E5/Ab and G8/Ab clones contained 183 and 160 amino acids, respectively, encompassing all complementarity determining regions (CDRs) and frame sections. Sequence analysis using the IMGT numbering tool identified the positions and lengths of CDR segments in the clones (Table III). The amino acid frequencies within the CDRs of these clones were calculated based on IMGT numbering (Fig. 8). The genetic maps of TCR-like antibodies were generated using sequencing data (Fig. S8, Fig. S9, Fig. S10). The length of the CDRs varied between antibodies. CDR1 regions in the H11/Ab, E5/Ab and G8/Ab TCR-like antibodies were each 24 bp in length. The CDR2 regions measured 21 in H11/Ab and 24 bp in E5/Ab and G8/Ab. Meanwhile, the lengths of the CDR3 regions were 57 in H11/Ab, 54 in E5/Ab and 39 bp in G8/Ab (Fig. 9). These sequencing results (which confirmed the full lengths of the CDR segments in the selected clones) support further investigation into the soluble form of these clones, to exploration the binding ability of TCR-like antibodies to their corresponding P-MHC-A24 complexes using TCR-like antibody ELISA and Western blot techniques.

Figure 8. CDR amino acid frequencies. Amino acid frequencies of the H11, E5 and G8 T cell receptor-like antibody CDRs, categorized according to ImMunoGeneTics numbering. CDR, complementarity determining region; aa, amino acid.

Figure 9. Sequencing map of the H11/Ab, E5/Ab and G8/Ab T cell receptor-like antibodies according to ImMunoGeneTics numbering. Sanger sequencing was performed using M13 vector primers.

Table II. VBASE2 and IMGT-vquest analysis of TCRs.

Table II.

VBASE2 and IMGT-vquest analysis of TCRs.

Antibody clone	V-gene family	D segment	J segment
H11\Ab	IGHV3	IGHD3-22*01	IGHJ1*01
E5\Ab	IGHV3	IGHD2-15*01	IGHJ5*02
G8\Ab	IGHV3	IGHD3-9*01	IGHJ4*02

[i] IMGT, ImMunoGeneTics information system^®; TCR, T cell receptor.

Table III. Region sequence for positive TCRs.

Table III.

Region sequence for positive TCRs.

A, H11/Ab
Region	Residue	Length, amino acids
Leader	1-34	34
HFR1	3-60	26
CDR-1	61-68	8
HFR2	6-85	17
CDR-2	86-92	7
HFR3	93-130	38
CDR-3	131-149	19
HFR4	150-160	11
Tail	160-195	35
B, E5/Ab
Region	Residue	Length, amino acids
Leader	1-34	34
HFR1	35-59	25
CDR-1	60-67	8
HFR2	68-84	17
CDR-2	85-92	8
HFR3	93-130	38
CDR-3	131-148	18
HFR4	149-159	11
Tail	159-183	24
C, G8/Ab
Region	Residue	Length, amino acids
Leader	1-8	8
HFR1	9-33	25
CDR-1	34-41	8
HFR2	42-58	17
CDR-2	59-66	8
HFR3	67-104	38
CDR-3	105-117	13
HFR4	118-128	11
Tail	128-160	32

[i] HFR, heavy chain framework regions; CDR, complementarity determining region.

Purification of soluble TCR-like antibodies

Prior to purifying TCR-like antibodies, the expression of target proteins was assessed using western blot analysis. The expected size of the DAB, ~15 kDa, was confirmed by western blot analysis with anti-c-Myc HRP (Fig. 10B). DABs were isolated using protein A affinity chromatography and SDS-PAGE was employed to confirm the successful isolation of each specific target. SDS-PAGE analysis identified the expected single DAB band (~15 kDa) in all the specific antibodies (Fig. 10A).

Figure 10. Investigation of TCR-like domain antibodies in soluble form. (A) SDS-PAGE analysis of purified soluble fractions of H11/Ab, E5/Ab, and A12/Ab. L, pre-stained protein ladder; C, crude antibody sample; W1, concentrated flow from the first column wash; W2, concentrated flow from the second column wash; E1, concentrated flow from the first elution (purified soluble antibody fractions); E2, concentrated flow from the second elution. (B) Western blot detection of soluble TCR-like domain antibody proteins in crude samples. (C) Western blot analysis to assess the binding capacities of H11/Ab, E5/Ab and G8/Ab TCR-like domain antibodies with the respective HPV18 P-MHC complexes. ELISA was conducted to evaluate the binding capacities of (D) H11/Ab, (E) E5/Ab and (F) A12/Ab TCR-like antibodies against their respective p-MHC-A24 complex. ***P<0.001. HPV, human papilloma virus; p-MHC, peptide-major histocompatibility complex; TCR, T cell receptor.

Ability of TCR-like antibodies to bind corresponding p-MHC-A24 complexes

Western blot analysis revealed a 15 kDa band associated with all TCR-like antibodies, including H11/Ab, E5/Ab and G8/Ab (Fig. 10C). These results confirmed the strong binding affinity of the antibodies towards their specific targets, thereby demonstrating the functional capability of the TCR in the soluble antibodies. Additionally, ELISA of the TCR DABs demonstrated interactions with the target p-MHC complexes. In the presence of a negative control (non-target p-MHC complex), the optical densities for H11/Ab, E5/Ab and G8/Ab were 1.024, 1.440 and 1.17 nm, respectively, at 405 nm (Fig. 10D).

Discussion

The present study aimed to design TCR-like antibodies that selectively target the HPV18 E6 and E7 oncoprotein peptides using DAB) library, driven by the demonstrated efficacy of TCR-like antibodies in treating various conditions, including malignancy, viral infections and autoimmune disorders (39). In targeting cervical cancer, peptides derived from the HPV 18 oncoprotein E6 and E7 serve as antigenic targets for TCR-like antibody generation against cervical carcinogenesis (Fig. S11) (40). The development of these antibodies focused on the primary p-MHC complex, which presents the HPV18 E6 and E7 oncoprotein peptides for immune recognition. Peptide selection was based on their binding affinity to HLA-A24. As a result, the TCR-like antibodies in the present study may serve as promising candidates for the early detection of latent CC (41).

HPV18 E6 and E7 oncoprotein peptides were selected as specific targets for TCR-like antibodies based on their predicted affinity for HLA-A24 using SYFPEITHI computational methods. SYFPEITHI predictions are derived from known sequences and natural ligands, with an emphasis on common amino acids located at anchor sites (42). The scoring system assigns values to amino acids based on their roles as anchors or preferred residues. Ideal anchors receive a score of 10, atypical anchors score 6–8 points, auxiliary anchors 4–6 points and preferred residues 1–4 points. Amino acids that reduce binding affinity are assigned scores ranging from −1 to −3 (43).

HPV18 p-MHC-A24 complexes were generated by refolding α chains and incorporating β2m light chain residues and HPV18 peptides in a cold buffer solution. Efficient refolding of HLA-A24 was observed with the selected peptides when using the optimal ligand (30). ELISA confirmed the successful formation of each p-MHC complex. Anti-β2m HRP was used instead of streptavidin HRP due to the high selectivity of anti-β2m HRP for the β2m light chain, which prevents detection of unbound heavy chains and ensures maximal accuracy (35).

Monoclonal antibodies were selected using the antibody phage display technique, based on biopanning analysis of HPV18 p-MHC-A24 complexes against a library of antibodies targeting the human domain. Effective phage enrichment was achieved after two rounds of selection, eliminating the need for further rounds (44). Significant ELISA results confirmed successful phage retrieval and packaging, facilitated by a precise trypsin-mediated elution. Trypsin was used to remove background contamination from the trypsin-sensitive helper phage protein III due to its ability to cleave the c-Myc tag located between the antibody fragment and the phage coat protein (protein III) (33). The screening for monoclonal antibodies involved the selection of 94 clones with low background values. To ensure the production of highly specific antibodies targeting HPV18 p-MHC-A24 complexes, comparative ELISA was conducted on all clones. Sequencing data were analyzed using the IMGT tool to align the core set with the CDR and framework region placements, confirming that the length of each region fell within the expected range.

The binding affinity of TCR-like antibodies to target complexes was evaluated using western blotting and ELISA. In the western blot analysis, a notable interaction between the TCR-like antibodies and targets was demonstrated on the membrane surface, following the detection of biotinylated HLA-A24 heavy chain within the p-MHC complex using HRP-conjugated streptavidin. These findings confirmed the high affinity of the TCR-like antibodies for their target complexes. To assess their selectivity, two controls were employed in the ELISA screenings. The first control, targeting non-p-MHC-A24 complexes, ensured the TCR-like antibodies did not bind unrelated p-MHC segments. Additionally, background control was used to verify the absence of non-specific binding to the p-MHC complex. TCR-like DABs exhibited no non-specific binding to unrelated p-MHC complexes or unintended interaction, demonstrating a high level of selectivity for their target p-MHC complexes.

TCR-like antibodies serve a role in identifying cancer-associated proteins, as demonstrated in previous clinical studies (45–48). The development of TCR-like antibodies has gained substantial attention (49–52). Increasingly, studies have explored their potential medical applications, aiming to harness these antibodies for immunotherapy and diagnosis of various types of disease as melanoma and prostate and Breast cancer (48,50,53–56). This may advance TCR-cancer therapy research, particularly if these antibodies exhibit enhanced efficacy against antigenic targets (57,58). Recent advancements in new-generation immunotherapy have potential application in cancer treatment (59–66).

The ability of TCR-like antibodies to recognize tumor-specific or -associated antigens, several of which are intracellular and presented on MHC molecules (40,46), provides a foundation for effective immunotherapy of CC (67). These antigens derive from tumor-specific and oncogenic viral proteins or neoantigens, which are processed by the cellular proteasome into shorter antigenic peptide sequences. These peptides are transported into the endoplasmic reticulum, where they bind MHC class I molecules. The resulting p-MHC complex is transferred via the Golgi apparatus and displayed on the cell surface (21,68,69). TCRs recognize intracellular CC antigenic peptides presented on MHC class I molecules of cancer cells, thereby eliciting immune responses, including phagocytosis, antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity and other immune components. When combined with immunotoxins, TCR-like antibody fusion molecules or chemotherapy, these antibodies contribute to a therapeutic strategy for CC (67). Unlike T cell-based immunotherapy, TCR-like antibodies are unaffected by T cell depletion, overcoming a key limitation and enhancing therapeutic efficacy (34).

TCR-like antibodies have traditionally been challenging to generate. However, new advances, including improved antigen production methods, have facilitated the development of TCR-like antibodies targeting an expanding repertoire of tumor and viral antigens (39,70–72). These antigens include Wilms' tumor gene1 (WT1), α-fetoprotein (AFP), preferentially expressed antigen in melanoma (PRAME), NY-ESO-1, MAGEA1, telomerase reverse transcriptase (hTERT), TCRγ alternate reading frame protein (TARP), tyrosinase, hCG β, p53, p68 MIF, proteinase 3, MAGE3 and Epstein Barr virus (EBV) proteins (50). The primary methods for isolating TCR-like antibodies are immunization followed by hybridoma and in vitro screening through phage display. Additionally, single B cell sorting and cloning offer an alternative approach for TCR-like antibody isolation, further broadening the potential for development (73). TCR-like antibody fusion molecules also show promise for enhancing cancer immunotherapy. These antibodies can be paired with single-chain variable fragment (scFv) dimers, known as diabodies or tandem scFvs, which increase affinity by doubling interactions between tumor and effector immune cells, thus enhancing cytotoxicity and overall therapeutic effectiveness (74). Another promising approach involves developing chimeric antigen receptors (CARs) incorporating TCR-like antibodies. CARs enable cytotoxic T lymphocytes to selectively destroy tumor cells. Designed with the intracellular domain of the CD3 protein essential for T cell activation, TCR-like antibody CARs specifically recognize p-MHC molecules on cancer cells, triggering T cell-mediated tumor destruction (75). Furthermore, priority should be given to developing faster, more cost-effective and widely accessible diagnostic methods at the population level, leveraging the high potential of TCR-like antibodies to selectively diagnose cancer antigens with precision.

Discovery of CC biomarkers has unveiled new opportunities for diagnostic and therapeutic advances. These biomarkers not only improve histopathological screening accuracy but also enhance risk assessment, enable improved monitoring of high-risk and post-treatment patients and support personalized treatment planning (76–78). Additionally, certain biomarkers serve as key antigenic targets for TCR-like antibody development (78). For example, HPV oncoproteins E6 and E7, which are key drivers of cervical carcinogenesis, are ideal candidates. The E6 oncoprotein helps cells proliferate by blocking the p53 tumor suppressor, which normally stops cells from dying and the cell cycle from stopping. Meanwhile, the E7 oncoprotein inactivates the retinoblastoma (Rb) tumour suppressor, contributing to cell immortalization (79). Another distinctive target is cancer-testis (CT) antigens, which are normally restricted to testis germ cells in healthy individuals but are upregulated in several types of malignancy, including CC and ovarian, breast, renal and colorectal cancer (80). The differential expression of CT antigens between healthy and cancerous cells makes them valuable for diagnosis and as therapeutic targets for TCR-like antibody generation. CT antigens in CC include sperm-associated antigen (SPAG), B cell receptor-associated protein 31, MAGE family proteins (A3, A4, A6 and A12), GAGE-3/6 (tumor-associated antigens), PRAME and LAGE-1 (53,81,82). TCR-like human IgG1 antibody (Pr20) targeting the PRAME peptide ALY mediates ADCC against leukemia cells in vivo, validating the potential of CT antigens for CC immunotherapy (83). Similar results have been observed with other TCR-like antibodies targeting tumor antigens expressed in CC, emphasizing the potential of using the same antigenic targets for therapeutic antibody generation (84–87). Contrary to earlier hypotheses, previous analysis of patient samples of CC suggests that HPV is not responsible for MHC class I downregulation (88). An analysis of >800 CC datasets from The Cancer Genome Atlas previously revealed that MHC class I expression in HPV-positive CC cells is comparable with that in HPV-negative tumors and non-cancerous cells (89).

Despite the availability of conventional diagnostic methods, the incidence of CC continues to rise (90): The numbers of estimated cervical cancer cases and deaths are projected to rise by 56.8 and 80.7%, respectively, by 2050. Furthermore, the anticipated increase in early-onset CC is predominantly observed in countries with a low Human Development Index), whereas a decline in disease burden is expected in countries with a high Human Development Index HDI) (90). The Visual Inspection with Acetic Acid (VIA) test and Pap smear are commonly used procedures for CC screening (91). However, these methods have limited effectiveness, particularly in resource-poor regions (92). Comparative analyses of Pap smears and VIA tests reveal that VIA has sensitivity of ~80 and an accuracy of 63.4, whereas the Pap smear demonstrates a sensitivity of ~50 and an accuracy of 69.9%. While VIA offers greater sensitivity, its overall accuracy remains limited in CC screening. The Pap smear requires the visual identification of cellular changes near the optical resolution limit, making it challenging to detect small and localized precancerous lesions (93). Additionally, the number of diagnostic pre-malignant cells in a specimen may be minimal and ~1,000 fields of view must be examined under 10× magnification to analyze the entire sample. The time required for this varies depending on sample complexity but generally ranges from 5 to 10 min/sample. Fatigue-associated guidelines of American Society of Cytopathology Workload recommend that cytotechnicians work <7 h/day (94). Additionally, the molecular methods, such as PCR assays for HPV diagnosis, selectively amplify DNA targets but present additional challenges (95). HPV molecular testing identifies viral genes rather than oncoproteins, making it incapable of determining the stage of CC. For example, HPV DNA detection in CC specimens has sensitivity of 37.9% (96). In resource-limited settings, test specificity is key, as follow-up procedures for false-positive results increase costs and may lead to unnecessary treatment. Low specificity contributes to overtreatment, placing strain on healthcare systems (95). While therapeutic vaccines for CC offer a multifaceted approach, they are associated with limitations. These include potential toxicity in immunocompromised patients, the development of neutralizing antibodies that diminish efficacy with repeated doses, HLA restrictions that impede universal vaccination and the need for booster components to enhance low-grade immunoglobulin responses (40). Although HPV vaccines effectively decrease levels of HPV-associated cancer precursors and minimize the need for therapeutic interventions, they are linked to increased risks of severe nervous system disorder (97,98), and other adverse effects (99–101). The benefits of HPV vaccination remain uncertain when weighed against these risks, as several studies have prioritized efficacy over safety (102,103). Furthermore, limited access to comprehensive clinical study reports and trial data prevents thorough assessment of potential adverse effects (104). Moreover, current vaccines primarily target HPV strains 16 and 18, which account for ~70% of CC cases, providing only partial protection (105). The high cost of vaccines restricts their use in public health programs in low-income countries. This highlights the need for innovative approaches to detect precursor lesions and address CC (104). Given these challenges, there is need for novel strategies for early detection of CC. Diagnostic techniques that detect intracellular markers using TCR-like antibodies offer a promising approach for the early and accurate diagnosis of CC.

TCR-like antibodies hold promise for the future of CC immunotherapy. However, numerous challenges must be addressed to optimize their effectiveness in treatment. One key challenge is the HLA restriction of epitopes, as TCR-like antibodies are specific to particular types of HLA (106). This specificity means that the antibodies are only beneficial for individuals with certain types of HLA. Moreover, prevalence of HLA types varies based on factors such as geographical location and disease susceptibility (107). For example, the HLA-A2 gene is prevalent globally, while HLA-A11 and HLA-A24 are concentrated in Asian populations (108,109). Additionally, certain HLA genes, such as HLA-DQB10602 and HLA-DRB11501, are associated with susceptibility to HPV infection (110). These factors highlight the importance of selecting HLAs during TCR-like antibody development to maximize their applicability. Previous studies have explored multiple HLAs, including HLA-A2, HLA-A11, and HLA-A24, to ensure broader population coverage (111–113). A second challenge of TCR-like antibodies, particularly in cancer therapy, is the suppression of antigen presentation on HLA molecules by tumors (114). Various strategies have been proposed to optimize the use of TCR-like antibodies. These includes using cytokines such as IFNγ to increase the expression of proteasome activators as low molecular mass polypeptides (LMP2, LMP7), multicatalytic endopeptidase complex (MECL-1), transporter associated with antigen processing complex (TAP1/TAP2) and MHC heavy chains, with TNFα helping to stabilize and enhance MHC functionality (115). LMP2, LMP7 and MECL-1 facilitate protein degradation and peptide generation for cytotoxic T cell presentation (116), while TAP1/TAP2 transport foreign peptides to the endoplasmic reticulum for binding with MHC I. The resulting p-MHC I complexes are then presented on the cell surface, initiating an immune response (117). Another strategy involves administering low-dose chemotherapeutic agents, ionizing radiation or topotecan, all of which increase MHC class I expression in breast cancer cells (118). A third strategy involves using pathway inhibitors, such as MEK inhibitors (targeting the MAPK pathway) and erlotinib (targeting the EGFR pathway), to enhance MHC class I expression in esophageal and gastric cancer (119). In cases with low p-MHC presentation, TCR-like antibody fusion molecules offer a promising solution. For example, in a mouse xenograft model with minimal HLA-A antigen presentation activated humoral immune responses targeting the cancer-testis antigen SPAG9 (120). Similarly, a TCR-like antibody conjugated with a therapeutic agent shows cytotoxic activity against breast and colon cancer cells, even under low target MHC density, validating the efficacy of TCR-like antibodies in conditions of reduced antigen presentation (121).

A third challenge of TCR-like antibodies is unknown specificity of tumour-infiltrating lymphocytes (TILs). Adoptive T cell transfer is a therapeutic strategy using ex vivo-expanded TILs), which effectively induce tumor regression following reinfusion into patients with cancer (122). Despite response rates of 20–50% (123), widespread use is limited by lengthy production times and low commercial viability (124). Furthermore, the unknown specificity of TILs complicate outcome predictions, fostering interest in therapies such as CAR T cells (125) or T cells engineered with TCRs targeting known cancer antigens (126). TCR-based gene therapy has gained feasibility through studies demonstrating redirected T cell specificity via αβTCR gene transduction, enabling both antiviral and antitumor responses (127–129). Trials with DMF4 TCRs (130) and DMF5 TCRs (131) targeting HLA-A0201 MART-1 melanoma peptides have showed that high-affinity TCRs enhance efficacy but can cause autotoxicity (132). The aforementioned studies indicate that the success of TCR-based gene therapy relies on the transfer of T cells expressing high-affinity TCRs. This is supported by research showing an association between TCR binding affinity and functional responses, underscoring the importance of affinity optimization to enhance therapeutic efficacy (133). As a result, strategies to design high-affinity TCRs for broader application have been developed as deep mutational scan (134) and Yeast display strategy (135).

A fourth challenge is cross-reactivity. Clinical trials have evaluated the safety and efficacy of adoptive T cell transfer using TCR-transduced T cells (136–139). While this approach holds potential for treating cancer with ‘off-the-shelf’ T cell products applicable to patients sharing the same HLA haplotype, the development of such therapy is hindered by immunotoxicity risks. Two key trials have highlighted the need for caution in TCR-based therapy. In the first, a high-avidity TCR targeting a MAGE-A3-derived peptide in HLA-A0201 transgenic mice achieved clinical responses in five of nine mice. However, three mice experienced neurological toxicity, and two fatalities occurred (140) due to TCR cross-recognition of a related brain-expressed MAGE-A12 peptide, differing by a single residue (141). In the second trial, a TCR targeting a MAGE-A3-derived peptide presented by HLA-A0101 (142), originally derived from a vaccinated patient (143), and engineered to enhance its affinity (144), was administered to two patients with myeloma and melanoma. Both patients suffered cardiac arrest and died shortly after receiving the T cell infusion; despite a 55% sequence overlap with the MAGE-A3 peptide, the TCR cross-reacted with titin, a protein expressed in contracting cardiomyocytes (145). Numerous strategies have been developed to enhance the identification of TCR cross-reactivity including combinatorial peptide libraries, yeast display and DNA barcode-labeled MHC multimers. These approaches aim to identify specific amino acids within peptide sequences that are key for TCR recognition. By integrating this information with human proteome and HLA presentation data, these strategies provide a predictive framework to identify potential cross-reactivity risk (146).

The combinatorial peptide library method is a valuable tool for determining the amino acid requirements for TCR recognition (147), offering insight into the specific interactions between TCRs and peptides (148). Although effective for detailed functional analysis, this approach requires large quantities of TCR-expressing cells and does not yield a relative ranking of interactions. To complement this method, smaller-scale assays may serve as an early-stage selection criterion to identify TCRs that may cross-react with endogenous peptides, enabling a more efficient and targeted assessment of TCR specificity (146). The yeast display system focuses on direct p-MHC-TCR interactions, providing a more accurate depiction of TCR binding degeneracy compared with combinatorial peptide libraries and DNA barcode-labeled MHC multimer strategies (149). This technique, which uses random peptide sequences and unbiased screening, identifies TCR-specific peptide targets, including those with previously unknown specificity (146). However, this method has limitations, such as the inability to equally display all peptide sequence positions and cover all peptide variants. Additionally, its applicability is restricted to a limited number of MHC molecules and specialized laboratories (150). Strategies that use cellular systems, such as T cell clones or TCR-transduced T cells, bear similarities to the yeast display system and offer notable advantages. Methods such as signaling and antigen-presenting bifunctional receptors and trogocytosis-based systems have effectiveness in antigen discovery and evaluating the breadth of TCR recognition (151). These approaches hold potential for identifying TCR-specific antigens and assessing cross-recognition risk, making them promising tools for advancing T cell therapy (146). TCR fingerprinting, which uses DNA barcode-labelled MHC multimers, is a powerful technique for determining TCR amino acid preferences for specific peptide sequences (152). This method generates libraries of peptide variants and analyzes the binding hierarchy of TCRs, enabling the creation of a positional scoring matrix to determine TCR specificity (153). While this approach offers flexibility, sensitivity and ease of implementation, it has limitations, such as costs of peptide synthesis and the necessity of a pre-established peptide target for library construction. Furthermore, the method may not identify novel p-MHC targets for TCRs with unknown specificity and single amino acid substitutions might not capture all potential TCR interactions (154). Despite these constraints, TCR fingerprinting has shown promise in identifying TCRs with lower cross-recognition risk, particularly for mutation-derived neoepitopes, establishing it as a valuable first-selection criterion for clinical applications (154). Combining experimental approaches with structural data and in silico modelling may provide more comprehensive understanding of TCR binding degeneracy and cross-recognition potential (142). Translating molecular interaction points of TCRs into cross-recognition potential is another promising strategy. Understanding the molecular interaction points of a TCR can identify cross-recognized peptides derived from endogenous proteins, offering a tool to assess the cross-recognition potential of a TCR before clinical application. This approach could decrease risk of severe side effects associated with TCR therapy. To evaluate this risk, insights from analysis of TCR interactions with p-MHC should be explored in silico using tools such as Find Individual Motif Occurrences (155) or Scan Prosite (156), which identify peptide sequences in the human proteome that align with the molecular interaction points of a TCR and are therefore at risk of cross-recognition (146). Despite these challenges, ongoing advancements and refinements offer viable solutions. He et al (157) demonstrated that a Fab-immunotoxin containing Pseudomonas exotoxin could recognize the melanoma antigenic peptide MART-1 presented by the HLA-A2*01 molecule, inducing cell death in human melanoma cells (157). Further validation of TCR-like antibody-immunotoxins came with the development of another Fab-immunotoxin targeting TCR gamma alternative reading frame protein, an antigenic protein associated with breast and prostate cancer, which effectively induced cell cytotoxicity (158). Additionally, two TCR-mimic antibody derivatives conjugated with the cytotoxic agent monomethyl auristatin E (MMAE), ESK-MMAE, and Q2L-MMAE, were designed to target WT1) oncoprotein (19). TCR-like antibodies represent a novel and promising class of therapeutic tool, holding potential for improving diagnosis and immunotherapy, particularly for CC (40).

The present study identified three TCR-like antibodies through a systematic screening process targeting the HPV18 (E6 and E7) oncoprotein antigens presented by MHC-A24. Analysis of the soluble forms of TCR-like antibodies revealed notable binding affinity for their respective HPV18 targets. The findings of the present study may contribute to the enhancement and optimization of early detection methods and immunotherapeutic strategies for CC.

Future studies should investigate all HPV variants associated with CC, specifically HPV 16 and 18. Integrating a range of HLAs, including A2, A11 and A24, which are part of the MHC class I family may facilitate applicability of this diagnostic tool to the global population, regardless of individual HLA variations. Antibodies should also be tested on cancer cell line models, such as CaSki, HeLa, SiHa, ME180 and C-33A cells, before proceeding to the clinical evaluation phase.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgements

Not applicable.

Funding

The present study was supported by Ministry of Higher Education, Malaysia, Higher Education Centre of Excellence (grant no. A305-KR-AKH002-0004401005-0000).

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

BAS performed experiments, analyzed data and wrote the manuscript. SAD edited the manuscript, conceived the study and designed the experiments. RSR analyzed data and edited the manuscript. GJT and VB conceived the study and designed the experiments. All authors have read and approved the final manuscript. BS and VB confirm the authenticity of all the raw data.

Ethics approval and consent to participate

Not applicable.

Patient consent for publications

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References