Limits...
Reverse Engineering of Vaccine Antigens Using High Throughput Sequencing-enhanced mRNA Display.

Guo N, Duan H, Kachko A, Krause BW, Major ME, Krause PR - EBioMedicine (2015)

Bottom Line: Thus, using mRNA display to interrogate mAbs permits high resolution identification of functional peptide antigens that direct targeted immune responses, supporting its use in vaccine reverse engineering for pathogens against which potent neutralizing mAbs are available.After the identified peptides were injected into mice, the mice produced their own antibodies with characteristics similar to the original antibody.This approach can provide previously unavailable information about antibody binding and could also be useful in developing new vaccines.

View Article: PubMed Central - PubMed

Affiliation: Division of Viral Products, Office of Vaccines Research and Review, Center for Biologics Evaluation and Research, Food and Drug Administration, 10903 New Hampshire Ave, Silver Spring, MD 20993, United States.

ABSTRACT

Unlabelled: Vaccine reverse engineering is emerging as an important approach to vaccine antigen identification, recently focusing mainly on structural characterization of interactions between neutralizing monoclonal antibodies (mAbs) and antigens. Using mAbs that bind unknown antigen structures, we sought to probe the intrinsic features of antibody antigen-binding sites with a high complexity peptide library, aiming to identify conformationally optimized mimotope antigens that capture mAb-specific epitopes. Using a high throughput sequencing-enhanced messenger ribonucleic acid (mRNA) display approach, we identified high affinity binding peptides for a hepatitis C virus neutralizing mAb. Immunization with the selected peptides induced neutralizing activity similar to that of the original mAb. Antibodies elicited by the most commonly selected peptides were predominantly against specific epitopes. Thus, using mRNA display to interrogate mAbs permits high resolution identification of functional peptide antigens that direct targeted immune responses, supporting its use in vaccine reverse engineering for pathogens against which potent neutralizing mAbs are available.

Research in context: We used a large number of randomly produced small proteins ("peptides") to identify peptides containing specific protein sequences that bind efficiently to an antibody that can prevent hepatitis C virus infection in cell culture. After the identified peptides were injected into mice, the mice produced their own antibodies with characteristics similar to the original antibody. This approach can provide previously unavailable information about antibody binding and could also be useful in developing new vaccines.

No MeSH data available.


Related in: MedlinePlus

Peptides identified by mRNA display-HTS using HCV mAb41. (A) Clone frequency distribution of the unique peptides with different copy numbers for the input 27-mer library, and 1st to 4th rounds of selection. For each peptide pool obtained after one round of selection, sequences were ranked and divided into different groups based on their copy numbers (1–10, 11–100, 101–1000, 1001–10,000, and > 10,000). The clone frequencies of the unique peptide sequences in each group were calculated and graphed. The total reads were as follows: library (2.3 × 106), 1st round (1.7 × 106), 2nd round (1.5 × 106), 3rd round (1.3 × 106), and 4th round (3.1 × 106). (B) Sequences of the wild type peptide A (pA) and peptide B (pB) from the E2 protein (aa 412–447) of HCV GT1a, and the most abundant peptides selected by mAb41 and their copy numbers are listed. QLRNSCA is the constant region present in all peptides. Dashes are introduced to display sequence alignments. The W(L/I)XX(L/I) motifs are in bold. Acidic residues are italicized. (C) Frequencies of peptides containing 1 or 2 copies of the W(L/I)XX(L/I) motif after each round of selection. (D) Peptides that show a perfect match to the wild type motif (GWLAGLF) were present among peptides selected by mAb41 but with far lower copy number. (B) and (D) are among ~ 3.1 million total sequences obtained after 4th round of selection.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4563141&req=5

f0010: Peptides identified by mRNA display-HTS using HCV mAb41. (A) Clone frequency distribution of the unique peptides with different copy numbers for the input 27-mer library, and 1st to 4th rounds of selection. For each peptide pool obtained after one round of selection, sequences were ranked and divided into different groups based on their copy numbers (1–10, 11–100, 101–1000, 1001–10,000, and > 10,000). The clone frequencies of the unique peptide sequences in each group were calculated and graphed. The total reads were as follows: library (2.3 × 106), 1st round (1.7 × 106), 2nd round (1.5 × 106), 3rd round (1.3 × 106), and 4th round (3.1 × 106). (B) Sequences of the wild type peptide A (pA) and peptide B (pB) from the E2 protein (aa 412–447) of HCV GT1a, and the most abundant peptides selected by mAb41 and their copy numbers are listed. QLRNSCA is the constant region present in all peptides. Dashes are introduced to display sequence alignments. The W(L/I)XX(L/I) motifs are in bold. Acidic residues are italicized. (C) Frequencies of peptides containing 1 or 2 copies of the W(L/I)XX(L/I) motif after each round of selection. (D) Peptides that show a perfect match to the wild type motif (GWLAGLF) were present among peptides selected by mAb41 but with far lower copy number. (B) and (D) are among ~ 3.1 million total sequences obtained after 4th round of selection.

Mentions: We next performed mRNA display using a 27-mer library against HCV mAb41, for which a previous phage display experiment identified a WL binding motif that aligns with the W437 and L438 residues in the wild type sequence pA of the HCV GT1a E2 protein (Duan et al., 2012). We carried out four rounds of mRNA display selection and sequenced the selected peptides by HTS after each round. The clone frequency distribution (Fig. 2A) indicated an increase in the abundance of certain unique peptides after the 3rd round of selection, which followed a pre-clearing step with protein G beads in the absence of the selection antibody, indicating enrichment of these peptides. A fourth round of selection further enriched the most common sequences selected in the third round. Identified peptides were ranked by copy number (Fig. 2B), and named based on their rank (with peptide p41_1 being the most abundant peptide binder). The most abundant mRNA display-enriched peptides show a W(L/I)XX(L/I) motif, which aligns with W437, L438 and L441 residues in the wild type sequence. The W(L/I) residues within the motif are similar to the WL identified by previous phage display selection. Different from the phage display results, mRNA display often identified a second (L/I) residue within this motif that aligns with L441 (Fig. 2B–C). Although the frequency of W(L/I)XX(L/I) (2.54%, 57,925 reads containing W(L/I)XX(L/I) among 2,278,952 total reads) in the original 27mer library was a bit higher than its expected frequency (1.65%) within a random library, due to an unintentional bias towards inclusion of tryptophan-encoding codons in the commercially prepared input library, a ~ 60% of selected peptides from the 4th round of selection contained at least one copy of W(L/I)XX(L/I) (Fig. 2C), indicating preferential selection of mAb41 for this motif.


Reverse Engineering of Vaccine Antigens Using High Throughput Sequencing-enhanced mRNA Display.

Guo N, Duan H, Kachko A, Krause BW, Major ME, Krause PR - EBioMedicine (2015)

Peptides identified by mRNA display-HTS using HCV mAb41. (A) Clone frequency distribution of the unique peptides with different copy numbers for the input 27-mer library, and 1st to 4th rounds of selection. For each peptide pool obtained after one round of selection, sequences were ranked and divided into different groups based on their copy numbers (1–10, 11–100, 101–1000, 1001–10,000, and > 10,000). The clone frequencies of the unique peptide sequences in each group were calculated and graphed. The total reads were as follows: library (2.3 × 106), 1st round (1.7 × 106), 2nd round (1.5 × 106), 3rd round (1.3 × 106), and 4th round (3.1 × 106). (B) Sequences of the wild type peptide A (pA) and peptide B (pB) from the E2 protein (aa 412–447) of HCV GT1a, and the most abundant peptides selected by mAb41 and their copy numbers are listed. QLRNSCA is the constant region present in all peptides. Dashes are introduced to display sequence alignments. The W(L/I)XX(L/I) motifs are in bold. Acidic residues are italicized. (C) Frequencies of peptides containing 1 or 2 copies of the W(L/I)XX(L/I) motif after each round of selection. (D) Peptides that show a perfect match to the wild type motif (GWLAGLF) were present among peptides selected by mAb41 but with far lower copy number. (B) and (D) are among ~ 3.1 million total sequences obtained after 4th round of selection.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4563141&req=5

f0010: Peptides identified by mRNA display-HTS using HCV mAb41. (A) Clone frequency distribution of the unique peptides with different copy numbers for the input 27-mer library, and 1st to 4th rounds of selection. For each peptide pool obtained after one round of selection, sequences were ranked and divided into different groups based on their copy numbers (1–10, 11–100, 101–1000, 1001–10,000, and > 10,000). The clone frequencies of the unique peptide sequences in each group were calculated and graphed. The total reads were as follows: library (2.3 × 106), 1st round (1.7 × 106), 2nd round (1.5 × 106), 3rd round (1.3 × 106), and 4th round (3.1 × 106). (B) Sequences of the wild type peptide A (pA) and peptide B (pB) from the E2 protein (aa 412–447) of HCV GT1a, and the most abundant peptides selected by mAb41 and their copy numbers are listed. QLRNSCA is the constant region present in all peptides. Dashes are introduced to display sequence alignments. The W(L/I)XX(L/I) motifs are in bold. Acidic residues are italicized. (C) Frequencies of peptides containing 1 or 2 copies of the W(L/I)XX(L/I) motif after each round of selection. (D) Peptides that show a perfect match to the wild type motif (GWLAGLF) were present among peptides selected by mAb41 but with far lower copy number. (B) and (D) are among ~ 3.1 million total sequences obtained after 4th round of selection.
Mentions: We next performed mRNA display using a 27-mer library against HCV mAb41, for which a previous phage display experiment identified a WL binding motif that aligns with the W437 and L438 residues in the wild type sequence pA of the HCV GT1a E2 protein (Duan et al., 2012). We carried out four rounds of mRNA display selection and sequenced the selected peptides by HTS after each round. The clone frequency distribution (Fig. 2A) indicated an increase in the abundance of certain unique peptides after the 3rd round of selection, which followed a pre-clearing step with protein G beads in the absence of the selection antibody, indicating enrichment of these peptides. A fourth round of selection further enriched the most common sequences selected in the third round. Identified peptides were ranked by copy number (Fig. 2B), and named based on their rank (with peptide p41_1 being the most abundant peptide binder). The most abundant mRNA display-enriched peptides show a W(L/I)XX(L/I) motif, which aligns with W437, L438 and L441 residues in the wild type sequence. The W(L/I) residues within the motif are similar to the WL identified by previous phage display selection. Different from the phage display results, mRNA display often identified a second (L/I) residue within this motif that aligns with L441 (Fig. 2B–C). Although the frequency of W(L/I)XX(L/I) (2.54%, 57,925 reads containing W(L/I)XX(L/I) among 2,278,952 total reads) in the original 27mer library was a bit higher than its expected frequency (1.65%) within a random library, due to an unintentional bias towards inclusion of tryptophan-encoding codons in the commercially prepared input library, a ~ 60% of selected peptides from the 4th round of selection contained at least one copy of W(L/I)XX(L/I) (Fig. 2C), indicating preferential selection of mAb41 for this motif.

Bottom Line: Thus, using mRNA display to interrogate mAbs permits high resolution identification of functional peptide antigens that direct targeted immune responses, supporting its use in vaccine reverse engineering for pathogens against which potent neutralizing mAbs are available.After the identified peptides were injected into mice, the mice produced their own antibodies with characteristics similar to the original antibody.This approach can provide previously unavailable information about antibody binding and could also be useful in developing new vaccines.

View Article: PubMed Central - PubMed

Affiliation: Division of Viral Products, Office of Vaccines Research and Review, Center for Biologics Evaluation and Research, Food and Drug Administration, 10903 New Hampshire Ave, Silver Spring, MD 20993, United States.

ABSTRACT

Unlabelled: Vaccine reverse engineering is emerging as an important approach to vaccine antigen identification, recently focusing mainly on structural characterization of interactions between neutralizing monoclonal antibodies (mAbs) and antigens. Using mAbs that bind unknown antigen structures, we sought to probe the intrinsic features of antibody antigen-binding sites with a high complexity peptide library, aiming to identify conformationally optimized mimotope antigens that capture mAb-specific epitopes. Using a high throughput sequencing-enhanced messenger ribonucleic acid (mRNA) display approach, we identified high affinity binding peptides for a hepatitis C virus neutralizing mAb. Immunization with the selected peptides induced neutralizing activity similar to that of the original mAb. Antibodies elicited by the most commonly selected peptides were predominantly against specific epitopes. Thus, using mRNA display to interrogate mAbs permits high resolution identification of functional peptide antigens that direct targeted immune responses, supporting its use in vaccine reverse engineering for pathogens against which potent neutralizing mAbs are available.

Research in context: We used a large number of randomly produced small proteins ("peptides") to identify peptides containing specific protein sequences that bind efficiently to an antibody that can prevent hepatitis C virus infection in cell culture. After the identified peptides were injected into mice, the mice produced their own antibodies with characteristics similar to the original antibody. This approach can provide previously unavailable information about antibody binding and could also be useful in developing new vaccines.

No MeSH data available.


Related in: MedlinePlus