Research ArticleNANOTECHNOLOGY

Photonic-plasmonic hybrid single-molecule nanosensor measures the effect of fluorescent labels on DNA-protein dynamics

See allHide authors and affiliations

Science Advances  26 May 2017:
Vol. 3, no. 5, e1602991
DOI: 10.1126/sciadv.1602991
  • Fig. 1 Photonic-plasmonic hybrid antenna-in-a-nanocavity.

    (A) Illustration of the biosensing system, consisting of a silicon photonic chip for biosensing and a polydimethylsiloxane (PDMS) microfluidic chip for sample delivery. (B) Scanning electron microscope image (SEM) of the silicon photonic chip shows the multiplexed photonic crystal nanobeam cavities (zoomed SEM inset) connected by waveguiding components to the edge of the chip for input/output coupling. (C) SEM image of the photonic crystal nanobeam cavity, with a single 50-nm-diameter gold particle located in the central grating of the nanocavity, thus forming an antenna-in-a-nanocavity architecture. (D) Illustration of the biofunctionalized gold nanoparticle. XPA proteins are immobilized to a self-assembled monolayer of 11-mercaptoundecanoic acid (11-MUA) on gold and interact with a double-stranded DNA (dsDNA). (E) Dimensionless factor Qη/Embedded Image is compared among different microphotonic and nanophotonic systems (for details, see table S1). Solid and hollow circles denote measurements performed in liquid and air, respectively. Our antenna-in-a-nanocavity system is indicated by the red arrow. PhC, photonic crystal cavity; SPR, surface plasmon resonance. (F) Resonance shift of 440 pm and Q-factor drop from 105 to 8.2 × 103 are the indications of trapping a gold nanoparticle. a.u., arbitrary units. (G) Electromagnetic field distribution of a bare photonic crystal nanobeam cavity without the gold nanoparticle. The cavity mode spans at wavelength cubed scale. (H) Electromagnetic field distribution of the antenna-in-a-nanocavity system. The hybrid mode is strongly localized in the gap region at the nanoparticle-silicon interface. Inset shows the zoomed-in field distribution at the gold nanoparticle (orange hemisphere). (I) Temperature increase distribution. The maximum temperature rise is ~0.2°C under the experimental condition: a power of 5 μW through the silicon waveguide.

  • Fig. 2 No charge transfer from gold to molecule.

    (A and B) PDOS of Lys-MUA (A) and Lys-MUA-Au (B) solved using DFT (see Materials and Methods). Red dashed line indicates the Fermi level, which shifts to lower energy in the presence of gold. (C) Energy diagram near the Fermi level for Lys-MUA-Au. The HOMOs and LUMOs (lowest unoccupied molecular orbitals) are shown. The Fermi level (red dashed line) is 2.2 eV above HOMO state of MUA. (D and E) Electron density of HOMO for Lys-MUA (D) and Lys-MUA-Au (E).

  • Fig. 3 Single-molecule DNA-XPA dynamics.

    (A and B) Real-time binding dynamics of the mismatched dsDNA and XPA in the standard binding buffer, measured by tracking the resonances of the antenna-in-a-nanocavity system. The dsDNA concentration in the microfluidic channel is 10 nM. The binding kinetics are fitted from the event histogram (fig. S5): kon = 0.20 ± 0.04 nM−1 s−1 and koff = 5.0 ± 1.1 s−1. (C and D) Resonance signals tracked in real time for the normal dsDNA and XPA as control. The mismatched dsDNA exhibits much longer residence time on the binding state than the normal dsDNA. (E to G) The solvent-accessible surface potential (φ = kBT/e) of a mismatched dsDNA (E), a normal dsDNA (F), and an XPA protein (G) obtained from MD simulation. Mismatched dsDNA has a significantly higher surface potential, with an abnormal twist at the mismatched site (indicated by the arrow). Blue-colored regions on the XPA protein are positively charged domains, which bind to the negatively charged dsDNA.

  • Fig. 4 DNA-XPA dynamics at different ionic strengths.

    (A to D) Real-time binding dynamics of the mismatched dsDNA and XPA, measured at different ionic strengths (IS). The dsDNA concentration in the microfluidic channel is 10 nM for all. (E) koff values obtained at different ionic strengths are consistent with the surface potentials calculated from MD simulation over a range of 7 logs.

  • Fig. 5 Fluorescent labeling weakens DNA-XPA interaction.

    (A and B) Real-time binding dynamics of the FITC-labeled mismatched dsDNA and XPA under the standard binding buffer condition, measured from the resonance shifts of the antenna-in-a-nanocavity. The dsDNA concentration in the microfluidic channel is 10 nM. The binding rates are fitted from the event histogram: kon = 0.15 ± 0.04 nM−1 s−1 and koff = 10.3 ± 2.4 s−1. (D to F) Real-time data for single GFP-labeled dsDNA and XPA protein in the standard buffer (D and E). DNA concentration, 10 nM; kon = 0.05 ± 0.01 nM−1 s−1; koff = 21.3 ± 5.3 s−1. (C and F) The solvent-accessible surface potentials (φ = kBT/e) of a FITC-labeled mismatched dsDNA (C) and GFP-labeled mismatched dsDNA (F). (G) kon values obtained from the resonance shift measurements scale well with the diffusion constants obtained from the dynamic light scattering ensemble measurements, indicating that the association rates are limited by the diffusion process. (H) koff values obtained from the resonance shift measurements agree well with the surface potentials obtained from MD simulations, indicating that fluorescent labels redistribute the surface charges and decrease the electrostatic interactions. (I) Resonance shifts agree well with the molecular masses of the DNA, FITC-DNA, and GFP-DNA complexes, indicating no simultaneous multimolecule binding events or aggregations.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/3/5/e1602991/DC1

    Supplementary Text

    fig. S1. Cavity resonance measurement setup.

    fig. S2. Electrostatic calculation of nanoplasmonic enhancement.

    fig. S3. Baseline drift.

    fig. S4. Step fitting algorithm.

    fig. S5. kon and koff fitting.

    fig. S6. Concentration dependence of DNA-XPA interaction.

    table S1. Comparison of Q-factors and mode volumes (V).

    References (5969)

  • Supplementary Materials

    This PDF file includes:

    • Supplementary Text
    • fig. S1. Cavity resonance measurement setup.
    • fig. S2. Electrostatic calculation of nanoplasmonic enhancement.
    • fig. S3. Baseline drift.
    • fig. S4. Step fitting algorithm.
    • fig. S5. kon and koff fitting.
    • fig. S6. Concentration dependence of DNA-XPA interaction.
    • table S1. Comparison of Q-factors and mode volumes (V).
    • References (59–69)

    Download PDF

    Files in this Data Supplement:

Navigate This Article