Identification of high affinity peptides for capturing norovirus capsid proteins

Abstract

Here, we present a platform where novel short and linear peptide motifs that enable binding to norovirus capsid proteins are selected by M13 phage display. The best peptide which recognizes recombinant norovirus 6H-P2 proteins has a sequence of ‘QHIMHLPHINTL’ and the apparent dissociation binding constants (K_d,app) of the selected peptides was found to be 185 nM of affinity.

---

Noroviruses (Norwalk-like viruses) are one of the major causative factors of viral gastroenteritis in humans.^1,2 It contains single-stranded RNA consisting of three open reading frames (ORFs). In fact, ORF1 encodes non-structural proteins, while ORF2 encodes the major (VP1) and minor (VP2) capsid proteins. Among ORF2, VP1 has two major domains containing the shell (S) and the protruding (P) domain. The P domain is structurally categorized into the P1 and P2 sub-domains. In some of the structure-to-function studies with crystallographic and mutational approaches,^2–4 the P1 sub-domain was connected with the S domain inside, while the P2 sub-domain was positioned on the outermost with an exposed shape and was involved in the norovirus binding to the host cells with immunogenic activity. Even though several studies have been reported for efficient identification and biosensing of norovirus,^3,5,6 it still has some bottlenecks including no reliable cell culture with laborious preparation steps for ready-to-use in a real test.

In an effort to solve these limitations, an efficient diagnostic method has been developed for the early detection of norovirus from various resources including polluted water, food and other types of samples.^6–8 For example, antibody-based immunoassay, reverse transcription (RT)-PCR and other methods are currently being studied.^6,7,9 However, only one immunoassay has been clinically approved and extensively used for norovirus detection.⁷ It relies mainly on antibodies as molecular binders and requires multi-step processing for sample preparation and labeling of the samples in some cases.¹⁰ Polyclonal antibodies are relatively cheap to produce by immunizing animals, but are heterogeneous to binding properties, while monoclonal anti-bodies are expensive to produce, but homogeneous. Both antibodies easily lose their binding properties under unfavorable environmental conditions such as high temperature, pH and other harsh circumstances.^10,11 Several efforts have been made to address these bottlenecks on the molecular sensing including the use of aptamers¹² and peptides.^7,11 Compared to antibody-based immunoassay, unique and short linear peptides have proven to be effective method for biosensing platform technology.^7,10,11 (a) One of major benefits is the small size of the binding peptides which can be synthesized easily and cost-effectively; (b) peptides are more stable and amenable than antibodies to manipulate at the molecular scale and this can facilitate the creation of a novel diagnostic biosensor, (c) peptides have lower immunogenicity and higher surface density and it can thus be employed for novel functional biomaterials. Therefore, improving current diagnostic technology is needed for simple, sensitive and accurate detection with high sensitivity and specificity.

Biopanning of M13 phage displayed peptide libraries is a powerful technology that enables the identification of novel peptide motifs that bind to almost limitless targets of interest.^10,13 There have been some reports where M13 phages are employed as the sensing probes on clinical application,⁷ imaging probes on cancer therapy,¹⁴ novel scaffold materials for tissue engineering¹⁵ and energy generation.¹⁶ Thus, small and recognizing peptides can be used in a wider variety of sensing material for high-throughput capturing and detection of biomarkers.

In this study, we utilized M13 phage display library to identify novel peptide motifs that bind to recombinant norovirus capsid VP1 proteins. We selected the P1 and P2 subdomain from P domain as target antigens for the use of norovirus detection. Therefore, we constructed the plasmid, pET22b-6H-P1 (amino acids 406–520) and pET22b-6H-P2 (amino acids 279–405) fusion protein expression vectors. The recombinant 6H-P1 fusion proteins were expressed as insoluble fraction in E. coli BL21(DE3) and therefore it was denatured with 8 M urea and refolded by dialysis. Unlikely, the recombinant 6H-P2 proteins were successfully purified as soluble form with Ni-NTA column chromatography (Fig. S1, S2 and S3†). Finally purified two fusion proteins were used as target proteins for further studies. At this point, we first verified the reactivity of two recombinant proteins, 6H-P1 and 6H-P2 with interacting anti-norovirus capsid VP1 antibody (Fig. S4†). It was shown that recombinant 6H-P2 proteins play an important role in antibody binding compared to recombinant 6H-P1 proteins. Although we have no obvious clue in these results, it is probably due to the most exposed region of the P2 domain and thus the epitope of this polyclonal antibody may be close to P2 domain.

To identify linear peptide motifs that specifically bind to recombinant 6H-P1 or 6H-P2 fusion proteins, we biopanned an M13 phage library against both two proteins. The enrichment after each round is listed in Tables S1 and S2.† As expected, the yields of eluted phage clones increased during biopanning experiments in both cases. For the recombinant 6H-P1 proteins, the percent yields increased from 1.48 × 10⁻⁵ after the first round to 6.86 × 10⁻¹ after the fifth round. When biopanning against recombinant 6H-P2 proteins, the percent yield increased from 3.68 × 10⁻⁶ after the first round to 2.28 × 10⁻¹ after the fifth round.

Samples of the eluted phages from the fourth and fifth rounds of biopanning were used to purify plaques of individual phage clones and to analyze DNA sequencing. In the case of the recombinant 6H-P1 proteins, 20 clones (9 from the 5th round and 11 from the 4th round) were used to prepare genomic DNA for sequencing analysis (Table S3†). The sequencing results showed that 10 of the 20 sequences obtained were found to be identical (5R5A12 with amino acid sequence IRPHRMRMLIQM). In the recombinant 6H-P2 proteins, 65 clones were prepared for DNA sequencing and from these 15 readable sequences were obtained. The sequences showed up 7 within the 15 clones (5R7A2 with amino acid sequence LSITSLRIMRLQ and 5R7A10 with amino acid sequence QHIMHLPHINTL). Since these are the dominant sequence to recombinant 6H-P2 proteins, only two clones were used for further characterization.

ELISA was performed to determine the binding affinity of the 5 phage clones for their target proteins. The results showed that two phages (5R7A2 and 5R7A10) appeared to bind with higher affinity to recombinant 6H-P2 proteins, while the affinities of phages (4R5A9, 5R5A9 and 5R5A12) against recombinant 6H-P1 proteins were less (Fig. 1a and b). Among the 5 positive phage clones, 5R7A2 and 5R7A10 clones were found to be potential peptide candidates for targets.

Fig. 1 ELISA assays of the selected phage clones. (a) and (c): 5R7A2, (b) and (d): 5R7A10.

To study the binding affinity of the selected phage clones at different protein concentrations, 5R7A2 and 5R7A10 clones were tested by ELISA. As shown in Fig. 1c and d, only the 5R7A10 clones appeared to be better than the 5R7A2 clones, indicating that the relative binding affinity of 5R7A2 clones was much less to the same target proteins. In detail, the binding affinity of 5R7A10 clones increased with the increase in recombinant 6H-P2 protein concentrations. At the recombinant protein concentration of 62.5 μg mL⁻¹, the interaction of phage clones displaying peptides with proteins was almost completed. The sequences have a several basic amino acids like His and they are rich in hydrophobic amino acids like Ile, Met, Leu and Pro, suggesting that hydrophobic interactions is exploited in the binding of the recombinant norovirus capsid proteins. And, the sequence displaying on 5R7A10 clones contains a single Pro which may stabilize peptide conformation or flexibility.

To measure the apparent dissociation binding constants (K_d,app) of the 5R7A10 clones, we performed in-direct ELISA protocol, as described previously.¹⁰ As shown in Fig. 2a, the K_d,app values of 5R7A10 clones were determined to be 185 nM of binding affinity for their targets. We also investigated the effects of serum for binding against target proteins. Interestingly, binding affinity of the best peptides, 5R7A10 were still comparable after incubation with 0.1% serum, even though binding affinity was slightly decreased in the presence of serum (Fig. 2b). Therefore, we concluded that the best peptide (5R7A10) identified by biopanning of M13 phages are found to be potential molecular binders for detection of norovirus in contaminated food or water.

Fig. 2 (a) Apparent binding constants (K_d,app) of the best peptides following biopanning, (b) Effects of serum for binding to target proteins.

Conclusions

In this study, novel and linear peptide motifs which bind to norovirus capsid VP1 proteins were identified and characterized. The best peptide recognizing recombinant 6H-P2 proteins had a sequence of QHIMHLPHINTL. Based on the kinetic ELISA assay, the apparent binding constants of the best peptides was found to be nanomolar affinity for their target. These results suggest that evolved or manipulated peptide derivatives from this study may be useful for the detection of norovirus in a miniaturized peptide biosensor with high sensitivity and selectivity. In this regard, the characterization of synthetic peptides away from the phage particles both in solution and upon immobilization by using QCM as well as SERS is now under consideration.

Acknowledgements

This study was supported by the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2011-0010312) and by the Korea government (MSIP) (no. NRF-2014R1A2A2A01005612).

Notes and references

1. M. Tan, P. Fang, T. Chachiyo, M. Xia, P. Huang, Z. Fang, W. Jiang and X. Jiang, Virology, 2008, 382, 115–123.
2. W. Bu, A. Mamedova, M. Tan, M. Xia, X. Jiang and R. S. Hegde, J. Virol., 2008, 82, 5340–5347.
3. M. Tan and X. Jiang, J. Virol., 2005, 79, 14017–14030.
4. T. Yoda, Y. Terano, A. Shimada, Y. Suzuki, K. Yamazaki, N. Sakon, I. Oishi, E. T. Utagawa, Y. Okuno and T. Shibata, J. Med. Virol., 2000, 60, 475–481.
5. G. S. Hansman, D. W. Taylor, J. S. McLellan, T. J. Smith, I. Georgiev, J. R. H. Tame, S.-Y. Park, M. Yamazaki, F. Gondaira, M. Miki, K. Katayama, K. Murata and P. D. Kwong, J. Virol., 2012, 86, 3635–3646.
6. S. Ishida, S. Yoshizumi, T. Ikeda, M. Miyoshi, M. Okano and T. Okui, J. Med. Virol., 2008, 80, 913–920.
7. J. D. Rogers, N. J. Ajami, B. G. Fryszczyn, M. K. Estes, R. L. Atmar and T. Palzkill, J. Clin. Microbiol., 2013, 51, 1803–1808.
8. D. Sano, T. Matsuo and T. Omura, Appl. Environ. Microbiol., 2004, 70, 3434–3442.
9. A. Giamberardino, M. Labib, E. M. Hassan, J. A. Tetro, S. Springthorpe, S. A. Sattar, M. V. Berezovski and M. C. DeRosa, PLoS One, 2013, 8, e79087.
10. J. P. Park, D. M. Cropek and S. Banta, Biotechnol. Bioeng., 2010, 105, 678–686.
11. J. Wu, J. P. Park, K. Dooley, D. M. Cropek, A. C. West and S. Banta, PLoS One, 2011, 6, e24948.
12. R. Beier, C. Pahlke, P. Quenzel, A. Henseleit, E. Boschke, G. Cuniberti and D. Labudde, FEMS Microbiol. Lett., 2014, 351, 162–169.
13. S. Swaminathan and Y. Cui, RSC Adv., 2012, 2, 12724–12727.
14. J. W. Kehoe and B. K. Kay, Chem. Rev., 2005, 105, 4056–4072.
15. S. Y. Yoo, J. W. Oh and S. W. Lee, Langmuir, 2011, 28, 2166–2172.
16. B. Y. Lee, J. Zhang, C. Zueger, W. J. Chung, S. Y. Yoo, E. Wang, J. Meyer, R. Ramesh and S. W. Lee, Nat. Nanotechnol., 2012, 7, 351–356.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details including construction of plasmid, expression, purification and biopanning of M13 phages. See DOI: 10.1039/c5ra09655c

---