Figure 1

In the presence of PEG, the ion current through the alpha-Hemolysin channel exhibits stepwise transitions between completely open and partially blocked states. The amplitude, frequency, and duration of these transient steps depend on the polymer molecular weight. (A) $400\mu \mathrm{M}$ PEG600, time averaging $5\mu \mathrm{s}$; (B) $40\mu \mathrm{M}$ PEG1500, time averaging $30\mu \mathrm{s}$; (C) $2\mu \mathrm{M}$ PEG3000, time averaging $100\mu \mathrm{s}$; (D) $20\mu \mathrm{M}$ PEG8000, time averaging $100\mu \mathrm{s}$. Note the change in the time scale from (A) to (B). Bilayer lipid membranes of 20 pF capacitance were formed from diphytanoyl phosphatidylcholine (Avanti Polar Lipids, Alabaster, AL, USA) at a room temperature of $23\pm 2°\mathrm{C}$ as previously described [4]; membrane-bathing aqueous solution contained 4 M KCl at pH 7.5 buffered by 5 mM Tris with PEG (polyethylene glycol standard, Fluka, GmbH, Germany) added from the “trans” side of the membrane; alpha-Hemolysin was added from the “cis” side of the membrane; applied voltage was 40 mV, negative from the side of protein addition.Reuse & Permissions

Figure 2

The relative conductance of the channel in the polymer-blocked state shows that the blocking efficiency of PEG increases with its molecular weight saturating to about a 95% level for PEG4000 and larger. The solid line is a fit by Eq. (1) to the data for the smaller PEG600, PEG1000, and PEG1500. The volume fraction $\varphi$ is set proportional to the polymer molecular weight with a common adjustable parameter described in the text.Reuse & Permissions

Figure 3

The on rate coefficient of polymer capture by the channel is a surprisingly weak function of polymer molecular weight. (A) The time between successive blockages is distributed exponentially as shown for $40\mu \mathrm{M}$ PEG1500. Both linear (left) and logarithmic [15] (right) plots show excellent fits to a single exponential (solid lines) with characteristic time ${\tau }_{\mathrm{on}}=13\mathrm{ms}$. (The total number of events was 1470.) In the case of a logarithmic time axis, the histogram is logarithmically binned. Therefore, the single exponential fitting curve is $\propto t\exp (-t/{\tau }_{\mathrm{on}})$; it exhibits a maximum at $t={\tau }_{\mathrm{on}}$. (B) Results of the on rate analysis for all polymers. Data for at least 3 single channels reconstituted in separate experiments were used for the plot. Inset: The average time between successive blockages is inversely proportional to the PEG1000 concentration.Reuse & Permissions

Figure 4

The residence time of a captured polymer vs polymer molecular weight shows a crossover. (A) The residence time distributions demonstrate statistically significant deviations from a single exponential (solid lines) with characteristic time 1.6 ms in both linear (left) and logarithmic (right) plots. Both longer and shorter time tails are present. Statistics were collected from the same track of 1470 events as in Fig. 3a. (B) The data for all polymers obtained through the single exponential fitting illustrated in (A). The decreasing portion of the curve corresponds to polymers that are too large to be accommodated by the channel. We hypothesize that the free part of the polymer molecule acts as an entropic spring.Reuse & Permissions