Preparation of glass flow cells for single-molecule measurements of probe-target hybridization rates

Unless otherwise noted, all chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Borosilicate coverslips were placed in ceramic racks (Thomas Scientific, Swedesboro, NJ), cleaned with piranha solution (a mixture of three parts concentrated H₂SO₄ and one part 30% H₂O₂, which is extremely corrosive and explosive) twice for 30 min, rinsed copiously with deionized water, and hydroxylated with 0.5 M KOH solution for 1 hr with sonication. The coverslips were then rinsed copiously with deionized water, rinsed twice by dipping in acetone (Chromasolv-Plus) for 10 s, and placed into a 3% solution of aminopropyltriethoxylane (APTES, Thermo Fischer Scientific, Waltham, MA) in acetone for 45 min, with gentle shaking. APTES-treated coverslips were then rinsed copiously with deionized water, and blow-dried with nitrogen. 50 μl drops of solution of 10% PEG-succinimidyl valerate (PEG-SVA, Mw = 5,000, Laysan) and 0.05% biotin-PEG-SVA (Mw = 5,000, Laysan) in 0.45M K₂SO₄ and 0.1M NaHCO₃ (pH 9.0) was squeezed between pairs of coverslips, and the coverslip pairs were incubated with PEG at 30°C for 30 min. PEG-treated coverslips were then rinsed copiously with deionized water, blow-dried with nitrogen, and the unreacted amine groups were end-capped by squeezing a 50 μl drop of solution of 2 mg sulfo-succinimidyl acetate (Pierce) in 0.1M NaHCO₃ (Pierce) for 10 min. The PEG-treated and amine-capped coverslips were rinsed copiously with deionized water, blow-dried with nitrogen, and stored dry at −80°C. (We found that end-capping of unreacted APTES amine groups was essential to eliminate non-specific interactions of fluorescently labeled probes with the glass surfaces at high probe concentrations.) An imaging flow cell containing seven reaction channels (volume of each channel: ∼25 μl) and side injection ports was constructed using two PEG-treated coverslips and double-sided adhesive tape (3M VHB 4095). The flow cell was mounted on the microscope, and the inner surface was decorated with the 2.8 μm magnetic beads for active stage stabilization (∼1–3 beads per 100 × 100 micron field of view) and functionalized with streptavidin (by incubating 5 μg/ml solution of streptavidin in phosphate-buffered saline for 30 s in a channel immediately before use).