Immunogenic display of diverse peptides, including a broadly cross-type neutralizing human papillomavirus L2 epitope, on virus-like particles of the RNA bacteriophage PP7

Abstract

The immunogenicity of an antigen can be dramatically increased by displaying it in a dense, multivalent context, such as on the surface of a virus or virus-like particle (VLP). Here we describe a highly versatile VLP platform for peptide display based on VLPs of the RNA bacteriophage PP7. We show that this platform can be used for the engineered display of specific peptide sequences as well as for the construction of random peptide libraries. Peptides representing the FLAG epitope, the V3 loop of HIV gp120, and a broadly cross-type neutralizing epitope from L2, the minor capsid protein of Human Papillomavirus type 16 (HPV16), were inserted into an exposed surface loop of a form of PP7 coat protein in which the two identical polypeptides of coat were fused together to form a single-chain dimer. The recombinant proteins assembled into VLPs, displayed these peptides on their surfaces, and induced high-titer antibody responses. The single-chain dimer was also highly tolerant of random 6-, 8-, and 10-amino acid insertions. PP7 VLPs displaying the HPV16 L2 epitope generated robust anti-HPV16 L2 serum antibodies after intramuscular injection that protected mice from genital infection with HPV16 pseudovirus as well as a heterologous HPV pseudovirus type, HPV45. Thus, PP7 VLPs are well-suited for the display of a wide diversity of peptides in a highly immunogenic format.

Fig. 1

The pP7K and p2P7K32 plasmids.

Fig. 2

Nucleotide and amino acid sequences of wild-type and recombinant PP7 coat proteins near the AB-loop. (A) and (B) show the N-terminal sequence of wild-type PP7 coat protein and the mutant coat protein encoded by pP7K and p2P7K32. (C) and (D) show the specific primers used to insert the FLAG, V3, HPV16 L2, and random sequences into the downstream AB-loop of the PP7 single-chain dimer. All sequences are shown 5’ to 3’; the KpnI restriction site is shown in italics and the peptide insertion is shown in bold text. All of these primers yield an insertion which is flanked by a thr residue on the N-terminal side and a Glu (the wild-type position 11 amino acid) on the C-terminal side.

Fig. 3

Agarose gel electrophoresis of whole cell lysates of 24 clones from each of the various libraries described in the text. Each of these yielded a white colony on X-Gal plates. The top half of each set is the ethidium bromide stained gel, and the bottom half is a western blot of a duplicate gel. The left-most lane in each set is the p2P7K32 control. Variations in electrophoretic mobility reflect charge differences conferred by the inserted peptides.

Fig. 4

Electrophoresis on formaldehyde/agarose gel of RNAs extracted from VLPs. The left panel shows the ethidium bromide-stained gel and a Northern blot probed with a labeled PP7 coat-specific oligonucleotide is shown in the right panel. MS2 VLP and PP7 VLP refer to the RNAs extracted from conventional VLPs from MS2 and PP7 coat protein, respectively. Sc-dimer refers to RNAs extracted from single-chain dimer VLPs. Some lanes contain RNAs produced by in vitro transcription (txpn) of MS2 and PP7 coat protein expression plasmids, as indicated.

Fig. 5

Binding of monoclonal antibodies to recombinant PP7 VLPs. 500 ng of wild-type PP7 VLPs, L2-VLPs, or FLAG-VLPs were bound to wells of an ELISA plate and then were reacted with dilutions of an anti-FLAG mAb (M2, panel A) or an anti-L2 mAb (RG-1, panel B). Binding was detected using a horseradish peroxidase-labeled goat anti-mouse IgG secondary followed by development with ABTS. Reactivity was determined by measurement of the absorbance at 405 nm (OD 405).

Fig. 6

IgG antibody responses in groups of mice immunized with wild-type PP7 VLPs, V3-VLPs, or 16L2-VLPs. End-point dilution ELISA titers against (A) a peptide representing a portion of the V3 loop from HIV_LAI conjugated to KLH or (B) a peptide representing amino acids 14–40 from HPV16 L2 conjugated to streptavidin. Results are from sera obtained three to four weeks after the second vaccination. Each datum point represents the antibody titer from an individual mouse. Lines represent the geometric mean titer for each group.

Fig. 7

Mice immunized with PP7 16L2-VLPs are protected from vaginal challenge with HPV16 or HPV45 pseudovirions. Groups of five mice were immunized two times with 10 µg 16L2-VLPs, wild-type PP7 VLPs, or HPV16 L1-VLPs formulated in incomplete Freund’s adjuvant (IFA). As an additional control, mice were immunized with IFA alone. Three weeks after the second immunization mice were intravaginally challenged with 10⁸ IU of HPV16 pseudovirus (left panel) or HPV45 pseudovirus (right panel) containing a luciferase reporter. As a control, a group of five mice were not infected. Luciferase activity was quanititated 48 hours after infection by taking images 3 min post-installation of luciferin at medium binning with a 30-s exposure. Images were then analyzed by drawing an equally sized region of interest for each mouse and measuring average radiance (photons/second/cm²/sr) within this region. Results shown are the mean average radiance for each group of five mice. Error bars represent the standard error of the mean. Lines above pairs of data indicate the percent reduction of signal in mice immunized with 16L2-VLPs relative to wild-type PP7 VLPs or the IFA control. All comparisons shown here are statistically significant (p<0.01) as calculated by T-test.

Fig. 8

Structures of the MS2 and PP7 coat protein dimers seen edge-on, with their polypeptide chains in blue and green, and the AB-loops shown in red. These structures were created using the program Polyview 3D (http://polyview.cchmc.org/polyview3d.html).