A single-molecule barcoding system using nanoslits for DNA analysis

Jo et al. 10.1073/pnas.0611151104.

Supporting Information

Files in this Data Supplement:

SI Figure 6
SI Materials and Methods
SI Movie 1




Fig. 6. Unity-based maps from barcoding with Nb.BbvCI for three BACs: BAC79 [four intervals: 83.9 (3.6), 7.0 (1.7), 16.4 (2.4), and 6.4 (1.9) kb], BAC150 [four intervals: 38.3 (4.2), 33.2 (4.2), 25.8 (3.8), and 19.6 (3.5) kb], and BAC614 [five intervals: 17.0 (2.7), 21.5 (2.9), 4.2 (1.6), 25.7 (3.2), and 14.1 (2.6) kb]; values in parentheses are the standard deviations in kb. The diagonal line portrays an r2 value of 1. Error bars show standard deviations on the means; numbers of molecules used in this analysis were 16 (BAC79), 55 (BAC150), and 31 (BAC614). Integrated fluorescence intensities of each interval were calculated by integrating the fluorescence intensity of the DNA backbone between two nicked positions marked by red punctates as shown in Fig. 5 and described in SI Materials and Methods.





SI Movie 1

Movie 1. Shown are the loading and stretching of T4 DNA (Fig. 4, molecule 1) within 100-nm ´1-mm nanoslits. DNA elongation reached equilibration after the electric field was removed. This movie was has been temporally scaled 25% (runs four times faster than real time) to enable file compression.





SI Materials and Methods

Soft Lithography and Fabrication of Master Wafer. Fabrication of PDMS devices followed standard rapid prototyping procedures (1). For these devices, a master wafer of nanoslits was dry-etched instead of having a photoresist template built on the wafer due to a gradual erosion of the SU-8 photoresist pattern (100 nm height) with the repetition of replica molding. An etched silicon wafer master has higher mechanical stability relative to an SU-8 patterned one although an SU-8 photoresist was used as a durable master pattern for the microchannels. A mask of 750 nm x 5 mm arrays was created using e-beam lithography (University of Wisconsin Center for Nanotechnology). This mask was then used in the first cycle of photolithography where a negative photoresist (SU-8 2000.5) was spin-coated onto silicon wafers creating arrays of 1-mm-wide, 5-mm-long slits. The slit width was controlled by the exposure time due to light diffraction. The wafer was etched 100 nm deep using a Unaxis 790 RIE (Unaxis Wafer Processing, St. Petersburg, FL) with a CF4 at 10 mTorr for 8 min and cleaned using piranha solution (80% H2SO4 and 20% H2O2) to lift off the photoresist layer. The height of the nanoslits (100 nm high) was measured by an alpha step profilameter, and the width (1 mm wide) was measured by scanning electron microscopy. After fabrication of the nanoslits, a microchannel array (3 mm high, 100 mm wide, and 10 mm long) was overlaid on the nanopatterned wafer using the negative photoresist SU-8 2005 in the second cycle of photolithography. Vapor deposition of tridecafluoro- 1,1,2,2- tetrahydro octyl-tricholoro silane was used for silanization of the patterned wafer to promote PDMS release (1, 2).

PDMS Preparation. Replica molded patterns were produced by curing PDMS [poly(dimethylsiloxane), Sylgard 184, Dow Corning, Midland, MI] onto a photolithographic negative master. More specifically, the basic steps involved thorough mixing of the polymer components with a catalyst at a 10 to 1 ratio. The mixture was then poured onto the silicon wafer master and allowed to cure at 65°C for 24 h, which was critical for stable PDMS nanostructures. To make the surface hydrophilic, oxygen plasma treatment was used which renders any exposed PDMS surface into hydrophilic silica (O2 pressure ~0.67 millibars; load coil power 100 W; 36 s; Technics Plasma GMBH 440, Florence, KY). Since PDMS surfaces were reactive immediately following plasma treatment, these plasma-treated devices were stored in high-purity water for 24 h. Afterward, the PDMS devices were washed with 50 ml of 0.5 M EDTA (pH 8.5) for 15 min with sonication for extraction of Pt2+ ions, used to catalyze PDMS polymerization but whose presence diminishes the fluorescence of YOYO-1 stained DNA because high concentration of cations displace YOYO-1 dyes from DNA backbones (3). Then, the PDMS devices were thoroughly washed with high-purity water by sonicating three times in 50 ml water for 15 min each and were stored in high-purity water before use. Finally, the PDMS devices were mounted on cleaned glass surfaces after drying as previously described (4).

Microscopy and Image Processing. DNA molecules were identified using segmentation of neighboring pixels, then filtered by integrated fluorescence intensity histograms to identify single molecules (Fig. 3). For FRET, two emission filters were used: an emission filter for the green channel (XF3086) and another for the red channel (XF3076; Omega Optical, Inc., Brattleboro, VT). The green channel acquired DNA backbones stained with YOYO-1 (491 nm, absorption; 509 nm, emission), while the red FRET imaged Alexa Fluor 647 (650 nm, absorption; 665 nm emission) punctates. For analysis, images from the green channel and red channel were overlapped and corresponding punctate positions were determined against background (green channel bleed-through) using integrated fluorescence intensity profiles.

Unity-Based Mapping. Early optical maps were constructed using a "unity-based" approach (5-7) where integrated fluorescence intensity or apparent length was used for estimation of restriction fragment masses. Such measurements were performed on a per fragment basis then normalized by total fluorescence intensity (or mass) of the entire molecule, so that the apportionment of fragment fluorescence intensities (or masses) sums to 1.0. The prevailing assumption is that molecules are not broken, and this generally holds for molecules below ≈180 kb (8, 9). Final maps were then constructed by averaging fluorescence intensity measurements of like restriction fragments over a number of molecules allowing estimation of precision expressed as a standard deviation. Final fragment sizes, in kb, were made based on independent sequence or map information, or used an internal standard such as a cloning arm of known size (6). Here, we adapted this approach by measuring the integrated fluorescence intensity of intervals, demarcated by fluorescent punctates. A plot (Fig. 6) shows our relative sizing errors, calculated as rErr (%) = [(measured - expected)/expected] ´ 100 (%); where measured refers to the averaged values of the same intervals, on different molecules, multiplied by the known size of a BAC clone (kb), here obtained from sequence information (G. Plunkett III, unpublished data). Expected refers to interval sizes calculated using sequence data, as an in silico map using the same nicking restriction enzyme. The average absolute relative error is calculated from the rErr (%), by taking its absolute value, averaged over all interval measurements.

1. Effenhauser CS, Bruin GJM, Paulus A, Ehrat M (1997) Anal Chem 69:3451-3457.

2. Duffy DC, McDonald JC, Schueller OJA, Whitesides GM. (1998) Anal Chem 70:4974-4984.

3. Eriksson M, Karlsson HJ, Westman G, Akerman B (2003) Nucleic Acids Res 31:6235-42.

4. Dimalanta ET, Lim A, Runnheim R, Lamers C, Churas C, Forrest DK, de Pablo JJ, Graham MD, Coppersmith SN, Goldstein S, et al. (2004) Anal Chem 76:5293-301.

5. Jing J, Reed J, Huang J, Hu X, Clarke V, Edington J, Housman D, Anantharaman TS, Huff EJ, Mishra B, et al. (1998) Proc Natl Acad Sci USA 95:8046-8051.

6. Meng X, Benson K, Chada K, Huff JE, Schwartz DC. (1995) Nat Genet 9:432-438.

7. Schwartz DC, Li X, Hernandez LI, Ramnarain SP, Huff EJ, Wang YK (1993) Science 262:110-4.

8. Cai W, Aburatani H, Vincent P. Stanton J, Housman DE, Wang Y-K & Schwartz DC. (1995) Proc Natl Acad Sci USA 92:5164-5168.

9. Cai WW, Jing JP, Irvin B, Ohler L, Rose E, Shizuya H, Kim UJ, Simon M, Anantharaman T, Mishra B, et al. (1998) Proc Natl Acad Sci USA 95:3390-3395.

This Article

  1. Published online before print February 12, 2007, doi: 10.1073/pnas.0611151104 PNAS February 20, 2007 vol. 104 no. 8 2673-2678
  1. AbstractFree
  2. Figures Only
  3. Full Text
  4. Full Text (PDF)
  5. » Supporting Information
About Our New Site Design >>
Link: Advanced Search

This Week's Issue

  1. August 19, 2008, 105 (33)
  1. Current Issue
  1. Alert me to new issues of PNAS

Most