- Email: [email protected]
Nanopore detection of antibody prion interactions
Analytical Biochemistry 396 (2010) 36–41
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Nanopore detection of antibody prion interactions Claudia Avis Madampage a, Olga Andrievskaia b, Jeremy S. Lee a,* a b
Department of Biochemistry, University of Saskatchewan, Saskatoon, Sask., Canada S7N 5E5 Canadian Food Inspection Agency, Nepean, Ont., Canada K1A 0Y9
a r t i c l e
i n f o
Article history: Received 26 June 2009 Available online 21 August 2009 Keywords: Prions Transmissible spongiform encephalopathy Nanopore a-Hemolysin Electrochemical detection Monoclonal antibodies
a b s t r a c t In nanopore analysis, peptides and proteins can be detected by the change in current when single molecules interact with an a-hemolysin pore embedded in a lipid membrane. A prion peptide, PrP(143–169), can readily translocate through the pore, but on the addition of monoclonal antibody M2188, which binds the peptide, the number of translocations is reduced because the complex is too large to translocate. At a peptide-to-immunoglobulin G (IgG) ratio of 2:1, only bumping events were observed. The event profile of a control peptide that does not bind the antibody was unchanged. Similarly, the presence of the antibody prevents translocation of the full-length prion protein. Because a nanopore can detect a single molecule, these experiments represent an important first step towards the development of a sensitive prion detector. Ó 2009 Elsevier Inc. All rights reserved.
The prion protein has been implicated in a number of neurodegenerative diseases called transmissible spongiform encephalopathies (TSEs)1 [1,2]. These include scrapie in sheep, mad cow disease in cattle, chronic wasting disease in elk and deer, and Creutzfeldt–Jacob disease in humans [3,4]. In normal tissue, the prion protein is soluble and protease sensitive and the structure is mostly a-helical (PrPC). In the disease state, the protein aggregates and is insoluble, becomes protease resistant, and adopts a mostly b-sheet structure (PrPSC) [5,6]. The aggregated PrPSC eventually leads to neuronal apoptosis similar to other protein misfolding diseases such as Alzheimer’s and Parkinson’s diseases [7–10]. In contrast to the latter, prion diseases are transmissible because PrPSC can cause the conversion of PrPC to more PrPSC in an autocatalytic reaction [2,3,11,12]. Although the incidence of the disease in humans is low, there is evidence that it can be transmitted by consuming infected meat from livestock [1]. In addition, there is no reliable premortem test, and so the discovery of a single sick animal has led to the mandatory slaughter of the whole herd. Thus, the development of a simple diagnostic test for TSE before the occurrence of neurological symptoms is imperative. Although the concentration of PrPSC is highest in the brain, it is also found in lymphoid tissue as well as in blood, urine, saliva, and milk [13–16]. Thus, a noninvasive diagnostic test is possible, but it would need to be extremely sensitive because it is estimated that
* Corresponding author. Fax: +1 306 966 4390. E-mail address: [email protected] (J.S. Lee). 1 Abbreviations used: TSE, transmissible spongiform encephalopathy; PrPC, normal cellular prion protein; PrPSC, scrapie isoform of prion protein; bPrP, recombinant bovine prion protein; IgG, immunoglobulin G. 0003-2697/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2009.08.028
the detection limit would need to be on the order of 104 molecules per milliliter of blood during the presymptomatic phase [17]. The test would also need to be extremely specific because PrPC is present on the surface of many cell types [18]. For this reason, several groups have explored the use of antibodies that can distinguish between PrPC and PrPSC based on the exposure of different epitopes in the two forms [19–28]. One promising epitope is VYYRP, residues 171–175 in the bovine sequence, which is buried in PrPC but is available as part of an exposed b-sheet in PrPSC [19]. There is also evidence that antibodies can be used therapeutically by blocking conversion or helping to clear PrPSC from infected cells [29,30]. Here we describe the use of nanopore analysis to investigate the interaction between antibodies and prion peptides and proteins. Briefly, a single pore, such as the toxin a-hemolysin, is inserted into a lipid membrane separating two chambers filled with buffer (Fig. 1). When a voltage is applied across the pore, a constant current will result due to the passage of ions through the pore. When a large molecule diffuses close to the pore (called a bumping event) or translocates through the pore (called a translocation event), a reduction in the current, I, will be recorded for a time, T. Previous work with peptides has shown that the parameters I and T are dependent on many factors such as length, charge, hydrophobicity, and even the dipole moment of the peptide [31–38]. In addition, conformational changes can be detected [39,40]. For example, in the absence of metal ions Zn finger peptides can readily translocate, but on binding Zn(II) the peptide folds into a rigid conformation that is now too large to pass through the pore [40]. Similarly, larger molecular complexes using solid-state pores have been studied. For example, when
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41
37
Fig. 1. Antibody binding to a peptide will prevent translocation. (The antibody is not drawn to scale).
an antibody Fab fragment binds to its target, the current blockade increases and so the free and bound forms can be distinguished [41–43]. In addition, an assay for the nanopore detection of hormone/antibody complexes that compares well with other immunoassays has been described [44]. Recently, it has been demonstrated that proteins can translocate the a-hemolysin pore, especially in the presence of denaturing agents that aid in protein unfolding [45,46]. Here we provide evidence that PrPC can also translocate even though it contains a disulfide bond. It has been shown that both b-hairpins and disulfide-linked a-helical hairpins can translocate, and so a single disulfide bond is not sufficient to prevent translocation (Ref. [39] and unpublished results). However, as shown in Fig. 1, the translocation events are eliminated on binding antibody; thus, antibody specificity can still be investigated.
Materials and methods The following peptides were purchased from Sigma (St. Louis, MO, USA) with a purity greater than 98%: Prp(143–169): SAMSRPLIHFGNDYEDRYYRENMYRYP (the epitope for antibody M2188 is underlined) (net charge +1). PrP(168–178): YPNQVYYRPMD (net charge 0). The recombinant bovine prion protein (bPrP, net charge +8), residues 25–242 with a C-terminal His5 tag, and monoclonal antibody M2188 were expressed and purified as described previously [27]. The peptides were dissolved in 1 M KCl and 10 mM KPi buffer (pH 7.8) at 5 mg/ml, and bPrP was dissolved in 10 mM Tris–HCl buffer (pH 7.8) at 1 mg/ml.
Fig. 2. Current traces for PrP(143–169) with antibody (A) and without antibody (B). Typical bumping and translocation events are indicated. In the presence of antibody, the event frequency is decreased and those events that do occur are mostly bumping.
38
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41
Fig. 3. Current blockade histograms for the peptides: (A) PrP(143–169); (B) PrP(143–169) with M2188, 2:1 ratio; (C) subtraction of M2188 from panel B; (D) PrP(168–178); (E) PrP(168–178) with M2188, 4:1 ratio; (F) subtraction of M2188 from panel E.
Nanopore analysis was performed as described in detail previously with an a-hemolysin pore that self-assembles into a lipid membrane [31,38,40]. The electrolyte solution was 1.0 M KCl and 10 mM KPi buffer (pH 7.8) for the peptides or 10 mM Tris–HCl buffer (pH 7.8) for bPrP. Peptide (5 ll) or bPrP (20 ll) was incubated with varying concentrations of antibody for 1 h. Once stable pore insertion was detected (fewer than three pores), these solutions were added to a 1.5-ml cis chamber proximal to the aperture. The experiments were carried out at 22 ± 1 °C with an applied potential of 100 mV at a bandwidth of 10 KHz. The blockade current populations were obtained by fitting the blockade current distribution with the Gaussian function. The lifetime data were obtained by fitting each blockade duration distribution with a single exponential function [31,38,40].
Results and discussion After formation of stable pores in the lipid membrane, peptides were added to the cis chamber and events were recorded. At these
concentrations, a complete profile could be accumulated in approximately 1 h. Typical traces for PrP(143–169) with and without antibody are shown in Fig. 2. The histograms of blockade times are given in the Supplementary material. The histogram of blockade current for PrP(143–169) is shown in Fig. 3A, and the event parameters are listed in Table 1. PrP(143–169) is unusual in that it gives three peaks, with each peak being roughly Gaussian. As shown previously, the peak centered at 30 pA is due to bumping events, whereas the peaks 50 and 80 pA are translocation events. The simplest explanation is that the peaks represent either N- or C-terminal entry to the pore or two different conformations. Currently, these possibilities cannot be distinguished. On the addition of antibody M2188 (Fig. 3B) at a ratio of 2:1 (peptide/immunoglobulin G [IgG]), the number of events is reduced and no clear translocation peak can be discerned. The antibody by itself (see Fig. 5C) also gives rise to a significant number of events that were subtracted from the event profile for the complex (Fig. 3C). In this case, the average number of translocation events is reduced even further, although the event profile in the bumping region appears to be more complex. In Fig. 3C, the appearance of a negative
Table 1 Event parameters for PrP(143–169), PrP(143–169)/M2188 complex, PrP(168–178), and PrP(168–178)/M2188 complex. Compound PrP(143–169) PrP(143–169)/M2188 2:1 PrP(168–178) PrP(168–178)/M2188 4:1
I1 (pA) 28.6 (0.1) 32.5 (0.6) – –
I2 (pA)
I3 (pA)
48.9 (0.3) – – –
84.4 (1.7) – 74.1 (0.1) 72.1 (0.2)
T1 (ms)
T2 (ms)
T3 (ms)
A1
A2
A3
W1
W2
W3
0.156 (0.016) 0.018 (0.005) – –
0.255 (0.017) – – –
0.415 (0.039) – 0.188 (0.030) 0.271 (0.026)
15.6 3.7 – –
9.6 – – –
4.3 – 12.0 13.5
8.5 11.7 – –
9.6 – – –
15.0 – 7.6 8.5
Note. Standard deviations are in parentheses. I, current blockade; T, time of blockade; A, number of events per pore per minute; W, peak width at half-height. The subscripts 1 and 2, 3 refer to bumping and translocation events, respectively.
39
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41 Table 2 Event parameters for bPrP, bPrP/M2188 complex, and M2188. Compound
I1 (pA)
I2 (pA)
T1 (ms)
T2 (ms)
A1
A2
W1
W2
bPrP
26.9 (0.3) 25.8 (0.3) 23.5 (0.0)
65.1 (0.8) –
0.022 (0.002) 0.190 (0.057) 0.120 (0.015)
0.149 (0.019) –
3.7
3.4
12.2
20.0
3.5
–
8.0
–
–
6.6
–
2.1
–
bPrP/M2188 2:1 M2188
–
Note. Standard deviations are in parentheses. I, current blockade; T, time of blockade; A, number of events per pore per minute; W, peak width at half-height. The subscripts 1 and 2 refer to bumping and translocation events, respectively.
Fig. 4. Numbers of translocation events as a function of antibody: ratio for the two peptides. To conserve antibody, only single experiments were performed for each ratio. In the absence of antibody, the reproducibility is estimated to be ±20%.
bumping peak after subtraction suggests that the complex gives rise to fewer bumping events than does the sum of the peptide and antibody alone. This is not unreasonable given that complex formation reduces the number of molecular species. In addition, the complex may diffuse to the pore more slowly than the antibody alone, again giving rise to fewer bumping events. Alternatively, the
individual molecules may interfere with each other’s access to the pore even in the absence of significant binding. As a control, the same analysis was repeated with PrP(168–178), which does not contain the epitope for antibody M2188 (Figs. 3D–F). It is clear from the event profiles that no significant interaction can be detected, although there appears to be a small increase in the translocation time (Table 1). Intermediate ratios of antibody/peptide were also studied, and the results are summarized in Fig. 4. No significant change in the number of events is detected with PrP(168–178), whereas the total number of events is reduced to approximately 12% at the highest ratio of antibody with PrP(143–169). Experiments with the full-length bPrP are shown in Fig. 5A, and the event parameters are listed in Table 2. The translocation peak centered at 65 pA is much broader than that for the peptide; this may be caused by different orientations of the protein as it enters the pore or different modes of unfolding as it passes through the pore. The ratio of translocations/bumping events is approximately
Fig. 5. Current blockade histograms for bPrP (A), bPrP with M2188 at a 2:1 ratio (B), M2188 alone (C), and subtraction of panel C from panel B (D).
40
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41
1:2. On the addition of antibody, no clear translocation peak can be seen and the ratio of translocation/bumping events decreases to approximately 1:10. Compared with the PrP peptides, the total number of events per pore per minute is much less, and so subtraction of the control with antibody alone (Fig. 5C) is more important. The antibody alone shows a sharp bumping peak at approximately 24 pA and a few events with blockade currents above 40 pA. It seems very unlikely that the latter represent translocation of the antibody because it is an order of magnitude larger than the pore and contains multiple, very stable b-sheet domains with several disulfide bonds. Alternatively, the apparent translocations may be due to impurities within the antibody preparation. (Small molecule contaminants may be a particular problem because they will diffuse more rapidly to the pore and, thus, give rise to a greater proportion of events than their corresponding molar concentration.) After subtraction (Fig. 5D), the average number of events with a blockade current above 30 pA is insignificant, but there is now a large negative bumping peak. Again the simplest explanation is that there are fewer molecular species and the complex diffuses to the pore more slowly than either the bPrP or the antibody. It is perhaps surprising that the translocation time (T) for bPrP is less than that for the peptide PrP(143–169) (Table 1). However, T for maltose binding protein of 370 residues is also in the range of 0.2 ms, so the larger value for the peptide may be due to strong interaction with the pore [45]. In addition, there are other values of T that are difficult to understand; for example, the bumping events for PrP(143–169) and bPrP with and without antibody (Tables 1 and 2). However, all four events are due to different structures or complexes, and currently the relationship between T and structure is not well understood; thus, predicting T values is not possible. In any event, the major focus of this article is the change in the number of events on adding antibody, not changes in event times. In conclusion, we have demonstrated that complex formation between antibodies and prion peptides and proteins can be detected by nanopore analysis. In principle, a nanopore can detect a single molecule; thus, this work represents the first step towards the development of a prion detector. The nanopore will provide the sensitivity and antibodies will provide the specificity to distinguish between PrPC and PrPSC. Acknowledgment Funding was provided by the Natural Science and Engineering Research Council of Canada (NSERC) and Saskatchewan Agriculture.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ab.2009.08.028. References [1] S.B. Prusiner, Neurodegenerative diseases and prions, N. Engl. J. Med. 344 (2001) 1516–1526. [2] N.J. Cobb, W.K. Surewicz, Prion diseases and their biochemical mechanisms, Biochemistry 48 (2009) 2574–2585. [3] J. Collinge, A.R. Clarke, A general model of prion strains and their pathogenicity, Science 318 (2007) 930–936. [4] P. Chien, J.S. Weissman, A.H. DePace, Emerging principles of conformationbased prion inheritance, Annu. Rev. Biochem. 73 (2004) 617–656. [5] B.J. Bennion, M.L. DeMarco, V. Daggett, Preventing misfolding of the prion protein by trimethylamine N-oxide, Biochemistry 43 (2004) 12955–12963. [6] M. Morillas, D.L. Vanik, W.K. Surewicz, On the mechanism of a-helix to b-sheet transition in the recombinant prion protein, Biochemistry 40 (2001) 6982– 6987.
[7] V.N. Uversky, C.J. Oldfield, A.K. Dunker, Intrinsically disordered proteins in human diseases: Introducing the D2 concept, Annu. Rev. Biophys. 37 (2008) 215–246. [8] F. Chiti, C.F. Dobson, Protein misfolding, functional amyloid, and human disease, Annu. Rev. Biochem. 75 (2006) 333–366. [9] E. Gaggelli, H. Kozlowski, D. Valensin, G. Valensin, Copper homeostasis and neurodegenerative disorders (Alzheimer’s, prion, and Parkinson’s diseases and amyotrophic lateral sclerosis), Chem. Rev. 106 (2006) 1995–2044. [10] P.T. Lansbury, H.A. Lashuel, A century-old debate on protein aggregation and neurodegeneration enters the clinic, Nature 443 (2006) 774–779. [11] G.P. Saborio, B. Permanne, C. Soto, Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding, Nature 411 (2001) 810– 813. [12] J. Castilla, D. Gonzalez-Romero, P. Saá, R. Morales, J. De Castro, C. Soto, Crossing the species barrier by PrPSc replication in vitro generates unique infectious prions, Cell 134 (2008) 757–768. [13] H. Seeger, M. Heikenwalder, N. Zeller, J. Kranich, P. Schwarz, A. Gaspert, B. Seifert, G. Miele, A. Aguzzi, Coincident scrapie infection and nephritis lead to urinary prion excretion, Science 310 (2005) 324–326. [14] L. Sage, Detecting prions in blood, Anal. Chem. 78 (2006) 14. [15] V. Grinkevich, N. Issaeva, S. Hossain, A. Pramanik, G. Selivannova, PrPSc in mammary glands of sheep affected by scrapie and mastitis, Nat. Med. 11 (2005) 1137–1138. [16] C. Soto, Diagnosing prion diseases: needs, challenges, and hopes, Nat. Rev. Microbiol. 2 (2004) 809–819. [17] A.F. Hill, M. Zeidler, J. Eronsid, J. Collinge, Diagnosis of new variant Creutzfeldt– Jakob disease by tonsil biopsy, Lancet 349 (1997) 99–100. [18] B. Caughey, G.S. Baron, Prions and their partners in crime, Nature 443 (2006) 803–810. [19] E. Paramithiotis, M. Pinard, T. Lawton, S. Laboissier, V.L. Leathers, W. Zou, L.A. Estey, J. Lamontagne, M.T. Lehto, L.H. Kondejewski, G.P. Francoeur, M. Papadopoulos, A. Haghighat, S.J. Spatz, M. Head, R. Will, J. Ironside, K. O’Rourke, Q. Tonelli, H.C. Ledebur, A. Chakrabartty, N.R. Cashman, A prion protein epitope selective for the pathologically misfolded conformation, Nat. Med. 9 (2003) 893–899. [20] B. Caughey, Probing for prions: recognizing misfolded PrP, Nat. Med. 9 (2003) 819–820. [21] M.H. Groschup, S. Harmeyer, E. Pfaff, Antigenic features of prion proteins of sheep and of other mammalian species, J. Immunol. Methods 207 (1997) 89– 101. [22] S. Demart, J.G. Fournier, C. Creminon, Y. Frobert, F. Lamoury, D. Marce, C. Lasmézas, D. Dormont, J. Grassi, J.P. Deslys, New insight into abnormal prion protein using monoclonal antibodies, Biochem. Biophys. Res. Commun. 265 (1999) 652–657. [23] R.J. Kascsak, R. Fersko, D. Pugiano, R. Rubenstein, R.I. Carp, Immunodiagnosis of prion disease, Immunol. Invest. 26 (1997) 259–268. [24] C. Korth, B. Stierli, P. Streit, M. Moser, O. Schaller, R. Fischer, W. SchulzSchaeffer, H. Kretzschmar, A. Raeber, U. Braun, F. Ehrensperger, S. Hornemann, R. Glockshuber, R. Riek, M. Billeter, K. Wüthrich, B. Oesch, Prion (PrPSc)-specific epitope defined by a monoclonal antibody, Nature 390 (1997) 74–77. [25] V. Cˇurin Šerbec, M. Bresjanac, M. Popovic´, K. Pretnar Hartman, V. Galvani, R. Rupreht, M. Cˇernilec, T. Vranac, I. Hafner, R. Jerala, Monoclonal antibody against a peptide of human prion protein discriminates between Creutzfeld– Jacob’s disease-affected and normal brain tissue, J. Biol. Chem. 279 (2004) 3694–3698. [26] W.Q. Zou, J. Zheng, D.M. Gray, P. Gambetti, S.G. Chen, Antibody to DNA detects scrapie but not normal prion protein, Proc. Natl. Acad. Sci. USA 101 (2004) 1380–1385. [27] O. Andrievskaia, H. McRae, C. Elmgren, H. Huang, A. Balachandran, K. Nielsen, Generation of antibodies against bovine recombinant prion protein in various strains of mice, Clin. Vaccine Immunol. 13 (2006) 98–105. [28] M. Polymenidou, R. Moos, M. Scott, C. Sigurdson, Y. Shi, B. Yajima, I. HafnerBratkovi, R. Jerala, S. Hornemann, K. Wuthrich, A. Bellon, M. Vey, G. Garen, M.N.G. James, N. Kav, A. Aguzzi, The POM monoclonals: a comprehensive set of antibodies to non-overlapping prion protein epitopes, PLoS One 3 (2008) e3872. [29] S.V. Antonyuk, C.R. Trevitt, R.W. Strange, G.S. Jackson, D. Sangar, M. Batchelor, S. Cooper, C. Fraser, S. Jones, T. Georgiou, A. Khalili-Shirazi, A.R. Clarke, S.S. Hasnain, J. Collinge, Crystal structure of human prion protein bound to a therapeutic antibody, Proc. Natl. Acad. Sci. USA 106 (2009) 2554–2558. [30] F.L. Heppner, C. Musahl, I. Arrighi, M.A. Klein, T. Rülicke, B. Oesch, R.M. Zinkernagel, U. Kalinke, A. Aguzzi, Prevention of scrapie pathogenesis by transgenic expression of anti-prion protein antibodies, Science 294 (2001) 178–182. [31] T.C. Sutherland, Y. Long, R. Stefureac, I. Bediako-Amoa, H. Kraatz, J.S. Lee, Structure of peptides investigated by nanopore analysis, Nano Lett. 4 (2004) 1273–1277. [32] L. Movileanu, J.P. Schmittschmitt, J.M. Scholtz, H. Bayley, Interactions of peptides with a protein pore, Biophys. J. 89 (2005) 1030–1045. [33] M.M. Mohammad, L. Movileanu, Excursion of a single polypeptide into a protein pore: simple physics, but complicated biology, Eur. Biophys. J. 37 (2008) 913–925. [34] Q. Zhao, D.A. Jayawardhana, D. Wang, X. Guan, Study of peptide transport through engineered protein channels, J. Phys. Chem. B 113 (2009) 3572– 3578.
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41 [35] L. Movileanu, Squeezing a single polypeptide through a nanopore, Soft Matter 4 (2008) 925–931. [36] L. Movileanu, Interrogating single proteins through nanopores: challenges and opportunities, Trends Biotechnol. 27 (2006) 333–341. [37] A.J. Wolfe, M.M. Mohammad, S. Cheley, H. Bayley, L. Movileanu, Catalyzing the translocation of polypeptides through attractive interactions, J. Am. Chem. Soc. 129 (2007) 14034–14041. [38] R. Stefureac, Y. Long, H. Kraatz, P. Howard, J.S. Lee, Transport of a-helical peptides through a-hemolysin and aerolysin pores, Biochemistry 45 (2006) 9172–9179. [39] C.P. Goodrich, C. Kirmizialtin, B.M. Huyghues-Despointes, A. Zhu, J. Scholtz, D.E. Makarov, L. Movileanu, Single-molecule electrophoresis of b-hairpin peptides by electrical recordings and Langevin dynamics simulations, J. Phys. Chem. B 111 (2007) 3332–3335. [40] R. Stefureac, J.S. Lee, Nanopore analysis of the folding of zinc fingers, Small 4 (2008) 1646–1650.
41
[41] C. Dekker, Solid-state nanopores, Nat. Nanotechnol. 2 (2007) 209–215. [42] Z. Siwy, L. Trofin, P. Kohli, L.A. Baker, C. Trautmann, C.R. Martin, Protein biosensors based on biofunctionalized conical gold nanotubes, J. Am. Chem. Soc. 127 (2005) 5000–5001. [43] N. Hurt, H. Wang, M. Akeson, K.R. Lieberman, Specific nucleotide binding and rebinding to individual DNA polymerase complexes captured on a nanopore, J. Am. Chem. Soc. 131 (2009) 3772–3778. [44] N. Han, M. Creus, G. Schurmann, V. Linder, T.R. Ward, N.F. de Roos, U. Staufer, Label-free detection of single protein molecules and protein– protein interactions using synthetic nanopores, Anal. Chem. 80 (2008) 4651–4658. [45] G. Oukhaled, J. Mathé, A.L. Biance, L. Bacri, J.M. Betton, D. Lairez, J. Pelta, L. Auvray, Unfolding of proteins and long transient conformations detected by single nanopore recording, Phys. Rev. Lett. 98 (2007) 158101–158104. [46] R. Stefureac, L. Waldner, P. Howard, J.S. Lee, Nanopore analysis of a small 86residue protein, Small 4 (2008) 59–63.
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Nanopore detection of antibody prion interactions Claudia Avis Madampage a, Olga Andrievskaia b, Jeremy S. Lee a,* a b
Department of Biochemistry, University of Saskatchewan, Saskatoon, Sask., Canada S7N 5E5 Canadian Food Inspection Agency, Nepean, Ont., Canada K1A 0Y9
a r t i c l e
i n f o
Article history: Received 26 June 2009 Available online 21 August 2009 Keywords: Prions Transmissible spongiform encephalopathy Nanopore a-Hemolysin Electrochemical detection Monoclonal antibodies
a b s t r a c t In nanopore analysis, peptides and proteins can be detected by the change in current when single molecules interact with an a-hemolysin pore embedded in a lipid membrane. A prion peptide, PrP(143–169), can readily translocate through the pore, but on the addition of monoclonal antibody M2188, which binds the peptide, the number of translocations is reduced because the complex is too large to translocate. At a peptide-to-immunoglobulin G (IgG) ratio of 2:1, only bumping events were observed. The event profile of a control peptide that does not bind the antibody was unchanged. Similarly, the presence of the antibody prevents translocation of the full-length prion protein. Because a nanopore can detect a single molecule, these experiments represent an important first step towards the development of a sensitive prion detector. Ó 2009 Elsevier Inc. All rights reserved.
The prion protein has been implicated in a number of neurodegenerative diseases called transmissible spongiform encephalopathies (TSEs)1 [1,2]. These include scrapie in sheep, mad cow disease in cattle, chronic wasting disease in elk and deer, and Creutzfeldt–Jacob disease in humans [3,4]. In normal tissue, the prion protein is soluble and protease sensitive and the structure is mostly a-helical (PrPC). In the disease state, the protein aggregates and is insoluble, becomes protease resistant, and adopts a mostly b-sheet structure (PrPSC) [5,6]. The aggregated PrPSC eventually leads to neuronal apoptosis similar to other protein misfolding diseases such as Alzheimer’s and Parkinson’s diseases [7–10]. In contrast to the latter, prion diseases are transmissible because PrPSC can cause the conversion of PrPC to more PrPSC in an autocatalytic reaction [2,3,11,12]. Although the incidence of the disease in humans is low, there is evidence that it can be transmitted by consuming infected meat from livestock [1]. In addition, there is no reliable premortem test, and so the discovery of a single sick animal has led to the mandatory slaughter of the whole herd. Thus, the development of a simple diagnostic test for TSE before the occurrence of neurological symptoms is imperative. Although the concentration of PrPSC is highest in the brain, it is also found in lymphoid tissue as well as in blood, urine, saliva, and milk [13–16]. Thus, a noninvasive diagnostic test is possible, but it would need to be extremely sensitive because it is estimated that
* Corresponding author. Fax: +1 306 966 4390. E-mail address: [email protected] (J.S. Lee). 1 Abbreviations used: TSE, transmissible spongiform encephalopathy; PrPC, normal cellular prion protein; PrPSC, scrapie isoform of prion protein; bPrP, recombinant bovine prion protein; IgG, immunoglobulin G. 0003-2697/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2009.08.028
the detection limit would need to be on the order of 104 molecules per milliliter of blood during the presymptomatic phase [17]. The test would also need to be extremely specific because PrPC is present on the surface of many cell types [18]. For this reason, several groups have explored the use of antibodies that can distinguish between PrPC and PrPSC based on the exposure of different epitopes in the two forms [19–28]. One promising epitope is VYYRP, residues 171–175 in the bovine sequence, which is buried in PrPC but is available as part of an exposed b-sheet in PrPSC [19]. There is also evidence that antibodies can be used therapeutically by blocking conversion or helping to clear PrPSC from infected cells [29,30]. Here we describe the use of nanopore analysis to investigate the interaction between antibodies and prion peptides and proteins. Briefly, a single pore, such as the toxin a-hemolysin, is inserted into a lipid membrane separating two chambers filled with buffer (Fig. 1). When a voltage is applied across the pore, a constant current will result due to the passage of ions through the pore. When a large molecule diffuses close to the pore (called a bumping event) or translocates through the pore (called a translocation event), a reduction in the current, I, will be recorded for a time, T. Previous work with peptides has shown that the parameters I and T are dependent on many factors such as length, charge, hydrophobicity, and even the dipole moment of the peptide [31–38]. In addition, conformational changes can be detected [39,40]. For example, in the absence of metal ions Zn finger peptides can readily translocate, but on binding Zn(II) the peptide folds into a rigid conformation that is now too large to pass through the pore [40]. Similarly, larger molecular complexes using solid-state pores have been studied. For example, when
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41
37
Fig. 1. Antibody binding to a peptide will prevent translocation. (The antibody is not drawn to scale).
an antibody Fab fragment binds to its target, the current blockade increases and so the free and bound forms can be distinguished [41–43]. In addition, an assay for the nanopore detection of hormone/antibody complexes that compares well with other immunoassays has been described [44]. Recently, it has been demonstrated that proteins can translocate the a-hemolysin pore, especially in the presence of denaturing agents that aid in protein unfolding [45,46]. Here we provide evidence that PrPC can also translocate even though it contains a disulfide bond. It has been shown that both b-hairpins and disulfide-linked a-helical hairpins can translocate, and so a single disulfide bond is not sufficient to prevent translocation (Ref. [39] and unpublished results). However, as shown in Fig. 1, the translocation events are eliminated on binding antibody; thus, antibody specificity can still be investigated.
Materials and methods The following peptides were purchased from Sigma (St. Louis, MO, USA) with a purity greater than 98%: Prp(143–169): SAMSRPLIHFGNDYEDRYYRENMYRYP (the epitope for antibody M2188 is underlined) (net charge +1). PrP(168–178): YPNQVYYRPMD (net charge 0). The recombinant bovine prion protein (bPrP, net charge +8), residues 25–242 with a C-terminal His5 tag, and monoclonal antibody M2188 were expressed and purified as described previously [27]. The peptides were dissolved in 1 M KCl and 10 mM KPi buffer (pH 7.8) at 5 mg/ml, and bPrP was dissolved in 10 mM Tris–HCl buffer (pH 7.8) at 1 mg/ml.
Fig. 2. Current traces for PrP(143–169) with antibody (A) and without antibody (B). Typical bumping and translocation events are indicated. In the presence of antibody, the event frequency is decreased and those events that do occur are mostly bumping.
38
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41
Fig. 3. Current blockade histograms for the peptides: (A) PrP(143–169); (B) PrP(143–169) with M2188, 2:1 ratio; (C) subtraction of M2188 from panel B; (D) PrP(168–178); (E) PrP(168–178) with M2188, 4:1 ratio; (F) subtraction of M2188 from panel E.
Nanopore analysis was performed as described in detail previously with an a-hemolysin pore that self-assembles into a lipid membrane [31,38,40]. The electrolyte solution was 1.0 M KCl and 10 mM KPi buffer (pH 7.8) for the peptides or 10 mM Tris–HCl buffer (pH 7.8) for bPrP. Peptide (5 ll) or bPrP (20 ll) was incubated with varying concentrations of antibody for 1 h. Once stable pore insertion was detected (fewer than three pores), these solutions were added to a 1.5-ml cis chamber proximal to the aperture. The experiments were carried out at 22 ± 1 °C with an applied potential of 100 mV at a bandwidth of 10 KHz. The blockade current populations were obtained by fitting the blockade current distribution with the Gaussian function. The lifetime data were obtained by fitting each blockade duration distribution with a single exponential function [31,38,40].
Results and discussion After formation of stable pores in the lipid membrane, peptides were added to the cis chamber and events were recorded. At these
concentrations, a complete profile could be accumulated in approximately 1 h. Typical traces for PrP(143–169) with and without antibody are shown in Fig. 2. The histograms of blockade times are given in the Supplementary material. The histogram of blockade current for PrP(143–169) is shown in Fig. 3A, and the event parameters are listed in Table 1. PrP(143–169) is unusual in that it gives three peaks, with each peak being roughly Gaussian. As shown previously, the peak centered at 30 pA is due to bumping events, whereas the peaks 50 and 80 pA are translocation events. The simplest explanation is that the peaks represent either N- or C-terminal entry to the pore or two different conformations. Currently, these possibilities cannot be distinguished. On the addition of antibody M2188 (Fig. 3B) at a ratio of 2:1 (peptide/immunoglobulin G [IgG]), the number of events is reduced and no clear translocation peak can be discerned. The antibody by itself (see Fig. 5C) also gives rise to a significant number of events that were subtracted from the event profile for the complex (Fig. 3C). In this case, the average number of translocation events is reduced even further, although the event profile in the bumping region appears to be more complex. In Fig. 3C, the appearance of a negative
Table 1 Event parameters for PrP(143–169), PrP(143–169)/M2188 complex, PrP(168–178), and PrP(168–178)/M2188 complex. Compound PrP(143–169) PrP(143–169)/M2188 2:1 PrP(168–178) PrP(168–178)/M2188 4:1
I1 (pA) 28.6 (0.1) 32.5 (0.6) – –
I2 (pA)
I3 (pA)
48.9 (0.3) – – –
84.4 (1.7) – 74.1 (0.1) 72.1 (0.2)
T1 (ms)
T2 (ms)
T3 (ms)
A1
A2
A3
W1
W2
W3
0.156 (0.016) 0.018 (0.005) – –
0.255 (0.017) – – –
0.415 (0.039) – 0.188 (0.030) 0.271 (0.026)
15.6 3.7 – –
9.6 – – –
4.3 – 12.0 13.5
8.5 11.7 – –
9.6 – – –
15.0 – 7.6 8.5
Note. Standard deviations are in parentheses. I, current blockade; T, time of blockade; A, number of events per pore per minute; W, peak width at half-height. The subscripts 1 and 2, 3 refer to bumping and translocation events, respectively.
39
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41 Table 2 Event parameters for bPrP, bPrP/M2188 complex, and M2188. Compound
I1 (pA)
I2 (pA)
T1 (ms)
T2 (ms)
A1
A2
W1
W2
bPrP
26.9 (0.3) 25.8 (0.3) 23.5 (0.0)
65.1 (0.8) –
0.022 (0.002) 0.190 (0.057) 0.120 (0.015)
0.149 (0.019) –
3.7
3.4
12.2
20.0
3.5
–
8.0
–
–
6.6
–
2.1
–
bPrP/M2188 2:1 M2188
–
Note. Standard deviations are in parentheses. I, current blockade; T, time of blockade; A, number of events per pore per minute; W, peak width at half-height. The subscripts 1 and 2 refer to bumping and translocation events, respectively.
Fig. 4. Numbers of translocation events as a function of antibody: ratio for the two peptides. To conserve antibody, only single experiments were performed for each ratio. In the absence of antibody, the reproducibility is estimated to be ±20%.
bumping peak after subtraction suggests that the complex gives rise to fewer bumping events than does the sum of the peptide and antibody alone. This is not unreasonable given that complex formation reduces the number of molecular species. In addition, the complex may diffuse to the pore more slowly than the antibody alone, again giving rise to fewer bumping events. Alternatively, the
individual molecules may interfere with each other’s access to the pore even in the absence of significant binding. As a control, the same analysis was repeated with PrP(168–178), which does not contain the epitope for antibody M2188 (Figs. 3D–F). It is clear from the event profiles that no significant interaction can be detected, although there appears to be a small increase in the translocation time (Table 1). Intermediate ratios of antibody/peptide were also studied, and the results are summarized in Fig. 4. No significant change in the number of events is detected with PrP(168–178), whereas the total number of events is reduced to approximately 12% at the highest ratio of antibody with PrP(143–169). Experiments with the full-length bPrP are shown in Fig. 5A, and the event parameters are listed in Table 2. The translocation peak centered at 65 pA is much broader than that for the peptide; this may be caused by different orientations of the protein as it enters the pore or different modes of unfolding as it passes through the pore. The ratio of translocations/bumping events is approximately
Fig. 5. Current blockade histograms for bPrP (A), bPrP with M2188 at a 2:1 ratio (B), M2188 alone (C), and subtraction of panel C from panel B (D).
40
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41
1:2. On the addition of antibody, no clear translocation peak can be seen and the ratio of translocation/bumping events decreases to approximately 1:10. Compared with the PrP peptides, the total number of events per pore per minute is much less, and so subtraction of the control with antibody alone (Fig. 5C) is more important. The antibody alone shows a sharp bumping peak at approximately 24 pA and a few events with blockade currents above 40 pA. It seems very unlikely that the latter represent translocation of the antibody because it is an order of magnitude larger than the pore and contains multiple, very stable b-sheet domains with several disulfide bonds. Alternatively, the apparent translocations may be due to impurities within the antibody preparation. (Small molecule contaminants may be a particular problem because they will diffuse more rapidly to the pore and, thus, give rise to a greater proportion of events than their corresponding molar concentration.) After subtraction (Fig. 5D), the average number of events with a blockade current above 30 pA is insignificant, but there is now a large negative bumping peak. Again the simplest explanation is that there are fewer molecular species and the complex diffuses to the pore more slowly than either the bPrP or the antibody. It is perhaps surprising that the translocation time (T) for bPrP is less than that for the peptide PrP(143–169) (Table 1). However, T for maltose binding protein of 370 residues is also in the range of 0.2 ms, so the larger value for the peptide may be due to strong interaction with the pore [45]. In addition, there are other values of T that are difficult to understand; for example, the bumping events for PrP(143–169) and bPrP with and without antibody (Tables 1 and 2). However, all four events are due to different structures or complexes, and currently the relationship between T and structure is not well understood; thus, predicting T values is not possible. In any event, the major focus of this article is the change in the number of events on adding antibody, not changes in event times. In conclusion, we have demonstrated that complex formation between antibodies and prion peptides and proteins can be detected by nanopore analysis. In principle, a nanopore can detect a single molecule; thus, this work represents the first step towards the development of a prion detector. The nanopore will provide the sensitivity and antibodies will provide the specificity to distinguish between PrPC and PrPSC. Acknowledgment Funding was provided by the Natural Science and Engineering Research Council of Canada (NSERC) and Saskatchewan Agriculture.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ab.2009.08.028. References [1] S.B. Prusiner, Neurodegenerative diseases and prions, N. Engl. J. Med. 344 (2001) 1516–1526. [2] N.J. Cobb, W.K. Surewicz, Prion diseases and their biochemical mechanisms, Biochemistry 48 (2009) 2574–2585. [3] J. Collinge, A.R. Clarke, A general model of prion strains and their pathogenicity, Science 318 (2007) 930–936. [4] P. Chien, J.S. Weissman, A.H. DePace, Emerging principles of conformationbased prion inheritance, Annu. Rev. Biochem. 73 (2004) 617–656. [5] B.J. Bennion, M.L. DeMarco, V. Daggett, Preventing misfolding of the prion protein by trimethylamine N-oxide, Biochemistry 43 (2004) 12955–12963. [6] M. Morillas, D.L. Vanik, W.K. Surewicz, On the mechanism of a-helix to b-sheet transition in the recombinant prion protein, Biochemistry 40 (2001) 6982– 6987.
[7] V.N. Uversky, C.J. Oldfield, A.K. Dunker, Intrinsically disordered proteins in human diseases: Introducing the D2 concept, Annu. Rev. Biophys. 37 (2008) 215–246. [8] F. Chiti, C.F. Dobson, Protein misfolding, functional amyloid, and human disease, Annu. Rev. Biochem. 75 (2006) 333–366. [9] E. Gaggelli, H. Kozlowski, D. Valensin, G. Valensin, Copper homeostasis and neurodegenerative disorders (Alzheimer’s, prion, and Parkinson’s diseases and amyotrophic lateral sclerosis), Chem. Rev. 106 (2006) 1995–2044. [10] P.T. Lansbury, H.A. Lashuel, A century-old debate on protein aggregation and neurodegeneration enters the clinic, Nature 443 (2006) 774–779. [11] G.P. Saborio, B. Permanne, C. Soto, Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding, Nature 411 (2001) 810– 813. [12] J. Castilla, D. Gonzalez-Romero, P. Saá, R. Morales, J. De Castro, C. Soto, Crossing the species barrier by PrPSc replication in vitro generates unique infectious prions, Cell 134 (2008) 757–768. [13] H. Seeger, M. Heikenwalder, N. Zeller, J. Kranich, P. Schwarz, A. Gaspert, B. Seifert, G. Miele, A. Aguzzi, Coincident scrapie infection and nephritis lead to urinary prion excretion, Science 310 (2005) 324–326. [14] L. Sage, Detecting prions in blood, Anal. Chem. 78 (2006) 14. [15] V. Grinkevich, N. Issaeva, S. Hossain, A. Pramanik, G. Selivannova, PrPSc in mammary glands of sheep affected by scrapie and mastitis, Nat. Med. 11 (2005) 1137–1138. [16] C. Soto, Diagnosing prion diseases: needs, challenges, and hopes, Nat. Rev. Microbiol. 2 (2004) 809–819. [17] A.F. Hill, M. Zeidler, J. Eronsid, J. Collinge, Diagnosis of new variant Creutzfeldt– Jakob disease by tonsil biopsy, Lancet 349 (1997) 99–100. [18] B. Caughey, G.S. Baron, Prions and their partners in crime, Nature 443 (2006) 803–810. [19] E. Paramithiotis, M. Pinard, T. Lawton, S. Laboissier, V.L. Leathers, W. Zou, L.A. Estey, J. Lamontagne, M.T. Lehto, L.H. Kondejewski, G.P. Francoeur, M. Papadopoulos, A. Haghighat, S.J. Spatz, M. Head, R. Will, J. Ironside, K. O’Rourke, Q. Tonelli, H.C. Ledebur, A. Chakrabartty, N.R. Cashman, A prion protein epitope selective for the pathologically misfolded conformation, Nat. Med. 9 (2003) 893–899. [20] B. Caughey, Probing for prions: recognizing misfolded PrP, Nat. Med. 9 (2003) 819–820. [21] M.H. Groschup, S. Harmeyer, E. Pfaff, Antigenic features of prion proteins of sheep and of other mammalian species, J. Immunol. Methods 207 (1997) 89– 101. [22] S. Demart, J.G. Fournier, C. Creminon, Y. Frobert, F. Lamoury, D. Marce, C. Lasmézas, D. Dormont, J. Grassi, J.P. Deslys, New insight into abnormal prion protein using monoclonal antibodies, Biochem. Biophys. Res. Commun. 265 (1999) 652–657. [23] R.J. Kascsak, R. Fersko, D. Pugiano, R. Rubenstein, R.I. Carp, Immunodiagnosis of prion disease, Immunol. Invest. 26 (1997) 259–268. [24] C. Korth, B. Stierli, P. Streit, M. Moser, O. Schaller, R. Fischer, W. SchulzSchaeffer, H. Kretzschmar, A. Raeber, U. Braun, F. Ehrensperger, S. Hornemann, R. Glockshuber, R. Riek, M. Billeter, K. Wüthrich, B. Oesch, Prion (PrPSc)-specific epitope defined by a monoclonal antibody, Nature 390 (1997) 74–77. [25] V. Cˇurin Šerbec, M. Bresjanac, M. Popovic´, K. Pretnar Hartman, V. Galvani, R. Rupreht, M. Cˇernilec, T. Vranac, I. Hafner, R. Jerala, Monoclonal antibody against a peptide of human prion protein discriminates between Creutzfeld– Jacob’s disease-affected and normal brain tissue, J. Biol. Chem. 279 (2004) 3694–3698. [26] W.Q. Zou, J. Zheng, D.M. Gray, P. Gambetti, S.G. Chen, Antibody to DNA detects scrapie but not normal prion protein, Proc. Natl. Acad. Sci. USA 101 (2004) 1380–1385. [27] O. Andrievskaia, H. McRae, C. Elmgren, H. Huang, A. Balachandran, K. Nielsen, Generation of antibodies against bovine recombinant prion protein in various strains of mice, Clin. Vaccine Immunol. 13 (2006) 98–105. [28] M. Polymenidou, R. Moos, M. Scott, C. Sigurdson, Y. Shi, B. Yajima, I. HafnerBratkovi, R. Jerala, S. Hornemann, K. Wuthrich, A. Bellon, M. Vey, G. Garen, M.N.G. James, N. Kav, A. Aguzzi, The POM monoclonals: a comprehensive set of antibodies to non-overlapping prion protein epitopes, PLoS One 3 (2008) e3872. [29] S.V. Antonyuk, C.R. Trevitt, R.W. Strange, G.S. Jackson, D. Sangar, M. Batchelor, S. Cooper, C. Fraser, S. Jones, T. Georgiou, A. Khalili-Shirazi, A.R. Clarke, S.S. Hasnain, J. Collinge, Crystal structure of human prion protein bound to a therapeutic antibody, Proc. Natl. Acad. Sci. USA 106 (2009) 2554–2558. [30] F.L. Heppner, C. Musahl, I. Arrighi, M.A. Klein, T. Rülicke, B. Oesch, R.M. Zinkernagel, U. Kalinke, A. Aguzzi, Prevention of scrapie pathogenesis by transgenic expression of anti-prion protein antibodies, Science 294 (2001) 178–182. [31] T.C. Sutherland, Y. Long, R. Stefureac, I. Bediako-Amoa, H. Kraatz, J.S. Lee, Structure of peptides investigated by nanopore analysis, Nano Lett. 4 (2004) 1273–1277. [32] L. Movileanu, J.P. Schmittschmitt, J.M. Scholtz, H. Bayley, Interactions of peptides with a protein pore, Biophys. J. 89 (2005) 1030–1045. [33] M.M. Mohammad, L. Movileanu, Excursion of a single polypeptide into a protein pore: simple physics, but complicated biology, Eur. Biophys. J. 37 (2008) 913–925. [34] Q. Zhao, D.A. Jayawardhana, D. Wang, X. Guan, Study of peptide transport through engineered protein channels, J. Phys. Chem. B 113 (2009) 3572– 3578.
Nanopore detection of antibody prion interactions / C.A. Madampage et al. / Anal. Biochem. 396 (2010) 36–41 [35] L. Movileanu, Squeezing a single polypeptide through a nanopore, Soft Matter 4 (2008) 925–931. [36] L. Movileanu, Interrogating single proteins through nanopores: challenges and opportunities, Trends Biotechnol. 27 (2006) 333–341. [37] A.J. Wolfe, M.M. Mohammad, S. Cheley, H. Bayley, L. Movileanu, Catalyzing the translocation of polypeptides through attractive interactions, J. Am. Chem. Soc. 129 (2007) 14034–14041. [38] R. Stefureac, Y. Long, H. Kraatz, P. Howard, J.S. Lee, Transport of a-helical peptides through a-hemolysin and aerolysin pores, Biochemistry 45 (2006) 9172–9179. [39] C.P. Goodrich, C. Kirmizialtin, B.M. Huyghues-Despointes, A. Zhu, J. Scholtz, D.E. Makarov, L. Movileanu, Single-molecule electrophoresis of b-hairpin peptides by electrical recordings and Langevin dynamics simulations, J. Phys. Chem. B 111 (2007) 3332–3335. [40] R. Stefureac, J.S. Lee, Nanopore analysis of the folding of zinc fingers, Small 4 (2008) 1646–1650.
41
[41] C. Dekker, Solid-state nanopores, Nat. Nanotechnol. 2 (2007) 209–215. [42] Z. Siwy, L. Trofin, P. Kohli, L.A. Baker, C. Trautmann, C.R. Martin, Protein biosensors based on biofunctionalized conical gold nanotubes, J. Am. Chem. Soc. 127 (2005) 5000–5001. [43] N. Hurt, H. Wang, M. Akeson, K.R. Lieberman, Specific nucleotide binding and rebinding to individual DNA polymerase complexes captured on a nanopore, J. Am. Chem. Soc. 131 (2009) 3772–3778. [44] N. Han, M. Creus, G. Schurmann, V. Linder, T.R. Ward, N.F. de Roos, U. Staufer, Label-free detection of single protein molecules and protein– protein interactions using synthetic nanopores, Anal. Chem. 80 (2008) 4651–4658. [45] G. Oukhaled, J. Mathé, A.L. Biance, L. Bacri, J.M. Betton, D. Lairez, J. Pelta, L. Auvray, Unfolding of proteins and long transient conformations detected by single nanopore recording, Phys. Rev. Lett. 98 (2007) 158101–158104. [46] R. Stefureac, L. Waldner, P. Howard, J.S. Lee, Nanopore analysis of a small 86residue protein, Small 4 (2008) 59–63.