Fueling protein–DNA interactions inside porous nanocontainers

Cisse et al. 10.1073/pnas.0610673104.

Supporting Information

Files in this Data Supplement:

SI Figure 3
SI Figure 4
SI Figure 5

SI Figure 3A
SI Figure 3B

Fig. 3. Partial duplex DNA (pdDNA) encapsulation. Imaging solution contains 1 mM ATP. (A) Low FRET dwell time histogram, with single exponential fit, for pdDNA encapsulated in 200 nm diameter vesicles. (B) High FRET dwell time, with single exponential fit, for pdDNA in 200 nm diameter vesicle. (B Inset) Rebinding frequencies for encapsulated pdDNA and ssDNA under the same condition.

The pdDNA, with 18 nucleotides in the double strand and (dT)₁₉ in the overhang (5'-Cy5- GCC TCG CTG CCG TCG CCA-3' and the complementary sequence has the overhang and Cy3 at the 3'-end), was encapsulated under the same condition as the ssDNA described in the text. The pdDNA was labeled such that a filament formation on the overhang can be detected [similar to the surface tethering assay in Joo et al. (1) but without biotin]. In the presence of ATP a repetitive filament rebinding was observed also at frequency orders of magnitude greater than in the surface tethering assays, suggesting that the frequent rebinding is not specific to ssDNA. The dwell times of high and low FRET states, for 77 vesicles, are presented in SI Fig. 3 fitted to a single exponential decay.

Compared to ssDNA, the observed rebinding frequency for the encapsulated pdDNA is about three times lower (SI Fig. 3B Inset), perhaps due to the presence of possible additional binding sites (in the duplex region) beside the single strand overhang (the only site where binding is detectable in our assay).

1. Joo C, McKinney SA, Nakamura M, Rasnik I, Myong S, Ha T (2006) Cell 126:515-527.

SI Figure 4

Fig. 4. Heterogeneity plot for ssDNA encapsulated in 200-nm diameter vesicles. Each black point represents the average dwell times for one vesicle. The red point represents the average dwell times for all of the vesicles as determined by the histograms (see Fig. 1 C and D).

Vesicles were randomly selected from the set analyzed in Fig. 1. For each vesicle, the average low and high FRET dwell times were calculated and plotted in SI Fig. 4to illustrate the intervesicle variability. In this case, we suspect the variability to be due to differences in the number of encapsulated RecA and in actual vesicle sizes (within 10% variation from the selected extrusion diameter as determined using dynamic light scattering, data not shown).

SI Figure 5A
SI Figure 5B
SI Figure 5C

Fig. 5. ssDNA inside 100-nm diameter vesicles. Imaging solution contains 1 mM ATP. (A) Low FRET dwell time histogram with single exponential fit. (B) High FRET dwell time histogram with single exponential fit. (Inset) Rebinding frequencies for ssDNA in 200- and 100-nm diameter vesicles. (C) Single-molecule trace of one ssDNA inside the 100-nm diameter vesicle.

For 1 ssDNA and 7.5 RecA inside a 100-nm-diameter vesicle, the corresponding local concentrations are 3.2 mM for ssDNA and 24 mM for RecA. To obtain a reasonable encapsulation yield we should mix the protein and DNA at concentrations comparable to the expected local concentration inside the vesicle.

However, because the desired RecA concentration is close to our protein stock, we instead use 400 nM ssDNA (and 3 mM RecA) and rely on the probability that after encapsulation (in 100 nm diameter) a few of the vesicles may still have 1 ssDNA and enough RecA to form a stable filament. The yield of such vesicles varied widely from 2% to 7% (at best).

In SI Fig. 5, we present a comprehensive analysis with 30 of these vesicles (100 nm in diameter) for qualitative comparison to the 200-nm vesicle data presented in the main text.

The low FRET dwell time (filament bound, SI Fig. 5A) is comparable to that of the 200 nm vesicles, while the high FRET dwell time (filament away from ssDNA, SI Fig. 5B) was reduced resulting in higher rebinding frequency (SI Fig. 5B Inset). The increased frequency is consistent with a more constrained diffusion in the smaller vesicle, but it is presently unclear what exactly should govern the scale of this increase.