Controlling the Translocation of Single-Stranded DNA through α

Dec 22, 2008 - Translocation of single-stranded DNA through R-hemolysin (R-HL) channels is investigated in glycerol/water mixtures containing 1 M KCl...
0 downloads 0 Views 287KB Size
Langmuir 2009, 25, 1233-1237

1233

Controlling the Translocation of Single-Stranded DNA through r-Hemolysin Ion Channels Using Viscosity Ryuji Kawano,* Anna E. P. Schibel, Christopher Cauley, and Henry S. White* Department of Chemistry, UniVersity of Utah, 315S. 1400E, Salt Lake City, Utah 84112 ReceiVed October 25, 2008 Translocation of single-stranded DNA through R-hemolysin (R-HL) channels is investigated in glycerol/water mixtures containing 1 M KCl. Experiments using glass nanopore membranes as the lipid bilayer support demonstrate that the translocation velocities of poly(deoxyadenylic acid), poly(deoxycytidylic acid), and poly(deoxythymidylic acid) 50-mers are decreased by a factor of ∼20 in a 63/37 (vol %) glycerol/water mixture, relative to aqueous solutions. The ion conductance of R-HL and the entry rate of the polynucleotides into the protein channel also decrease with increasing viscosity. Precise control of translocation parameters by adjusting viscosity provides a potential means to improve sequencing methods based on ion channel recordings.

Introduction Analyzing the fluctuations in ionic current as single-strand DNA (ss-DNA) or RNA translocate through an R-hemolysin (R-HL) ion channel, reconstituted in an electrically insulating lipid bilayer, is being actively investigated as a inexpensive means to determining the nucleic acid sequence.1-8 The electrophoretically driven translocation of ss-DNA through a solitary R-HL channel, Figure 1, is readily detected using ion channel recording methods. As ss-DNA translocates through the pore, the ion channel current decreases to ∼90% of the open channel value; the duration of the translocation (τt) is a measure of the length of the ss-DNA. Ideally, the current vs time trace recorded during the translocation of an individual ss-DNA molecule will exhibit four distinct levels, each level corresponding to one of the four bases (adenine (A), thymine (T), guanine (G), and cytosine (C)). As the biopolymer translocates the channel, the electrical readout of the four current levels provides the nucleotide sequence. However, the small difference in blockade currents associated with the different nucleotides ( poly-(dT)50 > poly(dC)50) is same in both aqueous and glycerol solutions. There are two possible factors that contribute to determining the translocation event. First, poly(nucleic acids) generally form helical structures via stacking interactions which favor parallel orientation of adjacent bases.23 Thus, the rate of base unstacking such that ss-DNA can translocate the channel may influence values of τt. Second, the interactions between the bases and amino acid residues of the protein wall are also important when ssDNA passes though the pore. Glycerol or glycerol/water mixed solution may destabilize the DNA stacking structure.24,25 However, the tendencies of τt of these three DNA in the aqueous and glycerol solutions are similar. The translocation rates of a single nucleotide of each DNA in the 63 vol % glycerol solution (21) Henrickson, S.; Misakian, M.; Robertson, B.; Kasianowicz, J. Phys. ReV. Lett. 2000, 85, 3057–3060. (22) Zhang, X.; Leddy, J; Bard, A. J. J. Am. Chem. Soc. 1985, 107, 3719– 3721. (23) Cantor, C. R.; Schimmel, P. R. Biophysical Chemistry; Freeman: New York, 1980. (24) Bonner, G.; Klibanov, A. M. Biotechnol. Bioeng. 2000, 68, 339–344. (25) Ansari, A.; Kuznetsov, S. V. J. Phys. Chem. B 2005, 109, 12982–12989.

Figure 6. (a) Translocation time and (b) association rate constant of poly-(dA)50, poly-(dC)50, and poly-(dT)50 in aqueous and 63 vol % glycerol/H2O solutions containing 1.0 M KCl, 10 mM Tris, 1 mM EDTA buffer. All measurements were performed at -120 mV (cis side negative) at 24 ( 1 °C. Table 1. Translocation Rates of Single Nucleotides through the Protein Channel in Aqueous and 63/37 vol % Glycerol/H2O Solutions, Containing 1.0 M KCl, 10 mM Tris, 1 mM EDTA Buffer at -120 mV Bias Voltage (cis side negative) translocation rate at 1 mPa · s (µs base-1) poly-(dA)50 poly-(dC)50 poly-(dT)50

3.2 ( 0.9 1.5( 0.7 2.2 ( 1.0

translocation rate at 18.5 mPa · s (µs base-1) 66 ( 20 22 ( 10 62 ( 28

range from 15 to 28 times (including 30 to 45% distributions) slower than that of the aqueous solution, as listed in Table 1. These values roughly correlate with bulk viscosity. Figure 6b shows the association rate constant of the poly(nucleotides) in the aqueous buffer and the 63 vol % glycerol solution. In the case of the aqueous solution, ka values of all three DNA molecules are about 5 to 7 × 105 M-1 s-1. This indicates that the collision frequency of the three different DNA molecules with R-HL are nearly similar. The ka of poly-(dT)50 in the glycerol solution is smaller than that of poly-(dA)50 and poly-(dC)50. The ka values of poly-(dA)50 and poly-(dC)50 are mostly same despite different base structure. All translocation events were carried out from cis to trans side. ss-DNA must pass sequentially through from the cis solution compartment to the protein vestibule, the inner constriction, β-barrel, and finally emerges into the trans solution compartment (see Figure 1). Butler et al. have recently reported that poly-(dT)50 resides longer in the protein vestibule region compared to the other homopolymer DNA as it threads the channel.26 In conclusion, we have demonstrated that the translocation of poly(deoxynucleotides) through a R-HL channel is strongly (26) Bulter, T. Z.; Gundlach, J. H.; Troll, M. Biophys. J. 2007, 93, 3229–3240.

Translocation of Single-Stranded DNA

dependent on the electrolyte viscosity. The viscosity can be increased up to 18.5 mPa · s by addition of glycerol, while maintaining a sufficiently high KCl concentration (1.0 M) for ion channel recordings. The ionic conductance of the R-HL channel, the DNA translocation velocity, and rate constant of translocation events each decrease with increasing viscosity as glycerol is added to the solution. The translocation speed of poly-(dA)50, poly-(dC)50, and poly-(dT)50 through the protein are 66 ( 20, 22 ( 10, and 62 ( 28 µs per base, respectively, ∼20 times slower than that in the aqueous electrolyte. However, the advantage gained by this decrease in velocity for improving SNR in sequencing methods is offset partially by the 10-fold decrease in channel conductivity. One potential means to improve upon this situation is to recognize that the conductivity is largely determined by the mobilities and concentrations of ions within the R-HL channel, while the velocity of a biopolymer may be

Langmuir, Vol. 25, No. 2, 2009 1237

influenced by viscous forces acting on the polymer chain extending outside of the channel. Thus, it may be possible to identify chemical additives that increase the bulk solution viscosity but partition only slightly into the normally H2O-filled and highconductance R-HL channel. Investigations along these lines are currently underway in our laboratory. Acknowledgment. This research was supported by the Defense Advanced Research Project Agency (FA9550-06-C-00C) and by the National Science Foundation (CHE-0616505). Additionally, R.K. gratefully acknowledges financial support from the Japan Society for the Promotion of Science. Supporting Information Available: Conductance-time trace and histogram of of poly-(dA)50 translocation events. This material is available free of charge via the Internet at http://pubs.acs.org. LA803556P