Selective aluminum passivation for targeted immobilization of single DNA polymerase molecules in zero-mode waveguide nanostructures

Korlach et al. 10.1073/pnas.0710982105.

Supporting Information

Files in this Data Supplement:

SI Text
SI Figure 6
SI Figure 7
SI Movie 1
SI Table 2
SI Figure 8
SI Figure 9
SI Figure 10

SI Figure 6

Fig. 6. Passivation of aluminum surfaces from protein adsorption by PVPA deposition, assayed by ellipsometry. Aluminum (Al) or fused silica (SiO₂) surfaces were untreated or treated with PVPA as described in Materials and Methods. Passivation was assayed by adsorption of neutravidin (A) or 40-nm neutravidin-coated latex beads (B) followed by washing and subsequent measurement of the ellipsometric thickness.

SI Figure 7

Fig. 7. Passivation of aluminum surfaces from DNA polymerase adsorption by PVPA deposition. (A) Mixed material patterned chips containing 0.5-mm aluminum squares (Al) on fused silica (SiO₂) were untreated (Upper) or treated (Lower) with PVPA as described in Materials and Methods. Passivation was assayed by f29 DNA polymerase adsorption, followed by rolling-circle DNA synthesis and visualization using SybrGold DNA stain. (Left) The entire chip. (Scale bars, 1 mm.) (Right) Wide-field epifluorescence microscopy images of the boundary regions of the two surface materials. (Scale bars, 10 mm.) The average density of DNA on PVPA-treated Al surfaces was 0.12 ± 0.04/mm² (n = 4 chips).

SI Figure 8

Fig. 8. Generation of DNA length standards. (A) Agarose gel of rolling circle DNA extension reactions of different durations, using the same conditions as performed on ZMW arrays. (B) Molecular weight marker positions and band center positions were extracted to determine the average size of the DNA products for each time point. (C) DNA length as a function of reaction time, yielding an average speed of DNA synthesis of 150 bases per min under these conditions. (D) Electrophoretic mobility of linearized single-stranded DNA of known lengths (M13mp18 = 7,249 bp; FX174 = 5,386 bp).

SI Figure 9

Fig. 9. Generation of DNA brightness vs. length of standard curve. (A) Fluorescence microscopy images (Left) and brightness histograms (Right) from three representative time points from the DNA length samples shown in SI Fig. 8. The DNA products from the different time points were immobilized on separate PVPA-treated aluminum chips, the fluorescence intensity of the DNA molecule was integrated and the peak position of each histogram was determined. (Scale bars, 10 mm.) (B) Standard curve, correlating DNA length from the agarose gel (SI Fig. 8) to average fluorescence brightness from the molecular intensity analysis shown in A. A linear fit to the data (r² = 0.99) was used to generate the bottom x axis of Fig. 5, with an associated average error of DNA length assignment of ± 500 bases (dashed lines, 95% confidence interval).

SI Figure 10

Fig. 10. Single molecule ZMW occupancy as a function of overall occupancy, assuming Poisson statistics governing polymerase deposition (SI Eq. 4), which was used to derive single DNA polymerase occupancies shown in Fig. 4B (gray circles).

SI Movie 1

Movie 1. Real-time, fluorescence microscopy video of a ZMW array containing long, fluorescently labeled DNA products from f29 DNA polymerase-mediated rolling-circle DNA synthesis, imaged from the top- (solution) side of the array. The DNA exits the confined volume of the ZMW and diffuses more freely in the open solution above the ZMW, while tethered to the ZMW floor by the polymerase. The ZMWs in this array are arranged in lines (Inset transmission image; line spacing, 3.1 μm; ZMW spacing, 1.1 μm). (Scale bar: 10 μm.)

SI Text

Nanostructure Fabrication. Mixed material chips. Patterns of 0.5 mm aluminum squares were manufactured on 100 mm fused silica wafers of 175 mm thickness (Mark Optics, Santa Ana, CA). The wafers were first cleaned using an RCA-1 Bath (Ammonium Hydroxide, Hydrogen Peroxide and Water in a 1:1:5 vol/vol ratio) heated at 75°C. The wafers were then dried and further cleaned using a 10 min oxygen plasma clean. The surface of the wafer was then primed with hexamethyldisilazane in a YES oven (Yield Engineering Systems, Livermore, CA). The wafer was further coated with a layer of Shipley 1813 photoresist (Microchem) for a final thickness of »1.4 mm. The resist was then baked at 115°C for 2 min on a hotplate. The wafers were exposed on a g-line stepper using a mask with the pattern design.

The resist was then developed in MIF-300 developer (Microchem), then rinsed and dried. A brief descum was performed in a Glenn 1000 Plasma asher (Yield Engineering Systems). Metallization was performed in a thermal evaporator using a tungsten crucible, a crystal monitor was used to monitor the thickness of the deposited aluminum (final thickness 100 nm). The pattern was then lifted off in a bath of 1165 resist stripper (Microchem) under sonication. It took »3 min for the entire pattern to be lifted off, at which point the wafers were rinsed, dried and diced.

ZMW array chips. A detailed description of ZMW fabrication will be published elsewhere (Foquet et al., submitted). Briefly, patterning of wafers was performed using electron beam lithography. The ZMWs were exposed in a negative resist, leaving pillars of resist on the surface after development. Aluminum was deposited onto the array pillars by thermal evaporation. The resist was then dissolved under sonication by 1165 photoresist stripper (Shipley), leaving the ZMW structures behind. ZMW size determinations were carried out by scanning electron microscopy (SEM), with errors in diameter measurements of ±8 nm standard deviation.

DNA Polymerase Generation. A mutant form of f29 DNA polymerase, f29^N62D, exhibiting reduced 3'-5' exonuclease activity (1, 2), was cloned into pET41 (Invitrogen, Carlsbad, CA), providing N-terminal Histidine (His) and GST tags for purification and enhanced surface activity upon fused silica physisorption, respectively. The enzyme was overproduced in Escherichia coli. Cells were disrupted using lysozyme (chicken egg white, Sigma-Aldrich) in buffer B (50 mM Tris-HCl, pH 7.5, 7 mM 2-mercaptoethanol, 5% glycerol) containing 0.2 M NaCl; and sonication. DNA was degraded by DNase I (bovine pancreas, Sigma-Aldrich). Cell debris was removed by centrifugation for 30 min at 15,000 g. Supernatant was adjusted to 0.5 M NaCl for purification on a 1-ml HisTrap FF column (Ni-resin; GE Healthcare) equilibrated with buffer B containing 0.5 M NaCl. His-tagged polymerase retained on the HisTrap column was washed with at least 50 column volumes, first using buffer B containing 1M NaCl, 0.2% Tween-20, and 20 mM imidazole, followed by buffer B containing 0.5 M NaCl and 50 mM imidazole. The polymerase was eluted with buffer B containing 300 mM imidazole. The pooled "HisTrap" fraction was adjusted to 0.2 M NaCl using buffer C (50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 7 mM 2-mercaptoethanol, 5% glycerol). The sample was loaded to a Heparin-Sepharose CL-6B (10 ml; GE Healthcare) equilibrated in buffer C. Polymerase was eluted with buffer C and a gradient of 0-1 M NaCl. The pooled "Heparin" fractions were concentrated using Centricon YM-50 (Millipore), and adjusted to the final storage buffer (50 mM Tris-HCl, pH 7.5, 0.2 M NaCl, 1 mM EDTA, 7 mM 2-mercaptoethanol, 50% glycerol).

Protein fractions were analyzed by SDS/PAGE (10% or 15% acrylamide) in all purification steps. The purified polymerase was at least 97% homogenous. Protein concentration was determined both by measuring the absorbance at 280 nm, and by the Bradford method using known amount of BSA (Bio-Rad).

DNA Template Generation. Oligonucleotides were obtained from Integrated DNA Technologies. The template sequence was 5'-cctatctacatcaccccaatcaccataacataccactttGacaataacacactctcaacaccacctacca-3', with the single guanine site highlighted. 5 mM DNA was circularized using 300 units of CircLigase (Epicentre Biotechnologies, Madison, WI) in 63 ml reaction solution containing 1´ CircLigase buffer supplied by the manufacturer, 0.1 mM ATP and 2.5 mM MnCl₂. Ligation reactions were carried out for 1 h at 60°C, followed by enzyme inactivation for 10' at 80°C. Unligated DNA was removed by addition of 60 units exonuclease I and 600 units of exonuclease III (USB), incubation for 30 min at 37°C and enzyme inactivation for 15 min at 80°C. Primer annealing was carried out by addition of NaCl to 120 mM and primer equimolar to template (5'-ggtgatgtagataggtggtaggtggtgtt-3'), incubation for 2 min at 80°C, followed by slow cooling to room temperature over ~30 min, followed by QIAprep spin column purification (Qiagen). DNA concentration of the eluted fractions was determined by measuring absorbance at 260 nm.

Fluorescence Intensity vs. DNA Length Determination. DNA synthesis reactions as described in Materials & Methods were incubated for various durations in the dark at room temperature. The reactions were stopped by addition of EDTA to 10 mM and the samples were split. One half was run on 0.7% agarose gels, DNA sizes were obtained from the molecular size dsDNA marker. The validity of the extracted lengths against the double-stranded marker bands was confirmed by comparison to linear single-stranded DNA of known lengths [(SI Fig. 8D, M13mp18 and FX174 single stranded DNA (NEB)], linearized by annealing of a primer spanning a restriction site and subsequent digestion by a restriction enzyme). The second half was used for deposition and fluorescence microscopy imaging, using the same instrumental conditions as used for ZMW arrays, on blank, PVPA-treated aluminum chips (SI Fig. 9). Centroids of fluorescence intensity histograms of the labeled DNA were plotted against the DNA length to obtain a standard curve for correlating DNA brightness to length, which was used to generate the bottom x axis in Fig. 5.

Single Molecule ZMW Occupancy Determination. The percentage of single active polymerases synthesizing DNA in the ZMW arrays (gray circles, Fig. 4B) was obtained assuming Poisson-distributed polymerase adsorption onto the bottom surface of ZMWs. At average polymerase occupancy, l, the fraction of ZMWs occupied by k active polymerases is thereby given as:

. [1]

The fraction of empty (k = 0) and singly occupied (k = 1) ZMWs is:

[2]

and

, [3]

respectively.

Solving (2) for l and substituting into (3) yields an expression for singly occupied ZMWs as a function of empty ZMWs in the array:

, [4]

The empty fraction, f(0,l), was obtained from the relative size of the zero peak in the histograms shown by example in Fig. 4A, converted to ZMW occupancies, 1-(f(0,l)). Over a relatively wide range of ZMW occupancies (40-85%), the fraction of single molecule occupancy stays relatively constant at 30% or higher (SI Fig. 10), with a maximum of 37% (for 63% total occupancy).

1. de Vega M, Lazaro JM, Salas M, Blanco L (1996) EMBO J 15:1182-92.

2. Blanco L, Salas M (1996) J Biol Chem 271:8509-12.