Quantized Plasmon Quenching Dips Nanospectroscopy via Plasmon Resonance Energy Transfer Gang Logan Liu1, 2, 3, 4, Yi-Tao Long1,4, Yeonho Choi1, Taewook Kang1 and Luke P. Lee1, 2, 5 1
Biomolecular Nanotechnology Center, Berkeley Sensor & Actuator Center, Department of Bioengineering, University of California-Berkeley, Berkeley, CA 94720-1762 2 Joint Graduate Group in Bioengineering, University of California at San Francisco and Berkeley, 1700 4th Street, San Francisco, CA 94158 -2330 3 Present Address: Center for Micro and Nano Technology, Lawrence Livermore National Laboratory, P. O. Box 808, L-223, Livermore, CA 94551 4 Equally contributed to this work. 5 To whom correspondence should be addressed. E-mail:
[email protected]
Prof. Luke P. Lee Department of Bioengineering University of California-Berkeley 485 Evans Hall Berkeley, CA 94720-1762 (510) 642-5855
[email protected]
1
Abstract We report the first observation of quantized plasmon quenching dips in the resonant Rayleigh scattering spectrum by plasmon resonance energy transfer (PRET) from a single nanoplasmonic particle to adsorbed metalloproteins cytochome c. PRET permits to develop an innovative label-free biomolecular absorption nanospectroscopy with sub-100 nm spatial resolution and ultrahigh molecular sensitivity. The multiplexed nanoplasmonic PRET probes may also allow in-vivo nanospectroscopic molecular imaging of metalloproteins in living cells.
2
Optical absorption spectroscopy at visible wavelength range is a common analytical method in chemistry and biology; however, the low sensitivity and spatial resolution of this technique prohibit its application in microscopic or nanoscopic biomolecular analysis and in-vivo cellular/molecular imaging. In this communication, we demonstrated a novel nanobiotechnology tool to bring new applications to visible absorption spectroscopy using PRET. By constructing a hybrid nanoparticle-biomolecule PRET system, the sensitivity of absorption spectroscopy was improved by several orders of magnitude, approaching hundreds of molecules. By shrinking the detection site from a typical 1-cm-pathlength cuvette to a sub-100 nm nanoparticle, the spatial resolution of absorption spectroscopy was increased to nanometer precision. Nanoparticle plasmon resonance1 is a free-electron oscillation spatially confined within the physical boundary of metallic nanoparticles. It is conjectured that the plasmon resonance energy can be transferred to chemical or biological molecules adsorbed on metallic nanostructures
2, 3
.
Although not observed
previously, PRET is postulated to account for surface enhanced Raman scattering
4, 5
, fluorescence 6 and luminescence on single nanoparticles. For the
first time, we observed the direct quantized plasmon quenching dips in the spectra of resonant Rayleigh scattering light from single nanoplasmonic particles due to the direct quantized plasmon resonance energy transfer from the particle to adsorbed cytochrome c molecules on its surface. The overlap of the electronic resonance peak positions of biomolecules with the plasmon resonance peak of the metallic nanoparticle generated distinguishable spectral quenching dips on
3
the Rayleigh scattering spectrum of the single nanoparticle, which allowed ultrasensitive nanoscopic absorption spectroscopy. Cytochrome c, a metalloprotein in cellular mitochondria membrane, acts as the charge transfer mediator 7 and plays a crucial role in bioenergy generation, metabolism, and cell apoptosis. Unlike many other proteins, cytochrome c has several optical absorption peaks in visible range around 550 nm coinciding with the 30 nm gold nanoparticle plasmon resonance, and more importantly it is a natural energy acceptor with electron tunneling channels 8. Similar to the donoracceptor energy matching in Fluorescent (or Förster) Resonance Energy Transfer (FRET) between two fluorophores, the critical matching of the localized resonating plasmon energy Ep in gold nanoparticles with the electron transition energy from ground to excited state Ee – Eg in cytochrome c molecules permits the PRET process. The quantized energy is likely transferred through the dipoledipole interaction between the resonating plasmon dipole in nanoparticle and the biomolecular dipole. Previous work on surface plasmon-mediated FRET process 9
, surface plasmon resonance shift due to redox molecules
10
and bulk optical
extinction spectroscopy of nanoplasmonic particle clusters with conjugated resonant molecules
11
also suggest the possibility of such dipole-dipole
interactions. The setup configuration in our experiment is shown in Fig. 1a. The typical scattering spectrum of a single nanoplasmonic particle is shown in Fig. 1b and the absorption spectrum of bulk biomolecule solutions is shown in Fig. 1c. The plasmon energy quenching of nanoparticle due to PRET is represented as “spectral dips” in the single nanoparticle scattering spectrum (Fig. 1d) where the
4
positions of “spectral dips” match the molecular absorption peak positions (Fig. 1c). Because PRET is a direct energy transfer process and thus more efficient and faster than optical energy absorption 6, the absorption spectral peaks of the conjugated cytochrome c molecules on single gold nanoparticles can be detected with a simple optical system, which is otherwise impossible using conventional visible absorption spectroscopic methods. In comparison with the visible scattering spectrum of gold nanoparticles coated with only cysteamine (no absorption in the visible region) (Fig. 2a), the raw scattering spectra of gold nanoparticles conjugated with reduced (Fig. 2b) and oxidized cytochrome c (Fig. 2c) showed not only a scattering peak (plasmon resonance peak) but also distinctive dips next to it. The absorption spectra of cytochrome c bulk solution are shown in Fig. 2d. The spectral dips on the nanoparticle scattering spectrum can be deconvoluted and correspond to the visible absorption peaks of the reduced and oxidized cytochrome c molecules (Figs. 2e and 2f). Processed scattering spectra have matched absorption peaks of reduced cytochrome c around 525 nm and 550 nm, and oxidized cytochrome c at 530 nm in comparison with the bulk measurements (Fig. 2d). In order to further confirm the spectral dips are from cytochrome c, we used the real time electrochemical technique to cyclically modulate the potential of the gold nanoparticles from - 500 mV to 500 mV vs. Ag/AgCl electrode. The PRET intensity at 550 nm showed the cyclic alteration indicating the repeating conversion between oxidized and reduced cytochrome c (Supplementary Fig. 1 online). According to the calculation, the dramatic spectral dips were not a result
5
of the direct optical absorption of cytochrome c molecules (Supplementary Methods online) but the unique PRET process. The plasmon resonance of gold and silver nanoparticles conjugated with various biomolecules
12, 13
has been studied. Due to the mismatch between the
typical biomolecular electronic resonance modes in ultraviolet wavelengths and nanoparticle plasmon resonance modes in visible wavelengths, only the shift of plasmon resonance peak is observed.
In our case, the energy matching
condition in PRET was confirmed by three negative control experiments and simulation.
In the first control experiment, we conjugated a 30 nm gold
nanoparticle with synthetic peptides which have no absorption peaks in the wavelength range of the plasmon resonance.
As expected, the scattering
spectrum of this peptide conjugated system showed only the scattering peak (Fig. 2g). In the second control experiment using a large gold nanoparticle cluster which has a plasmon resonance wavelength beyond 650 nm, the conjugated cytochrome c absorption peaks were essentially undetected, including the 525 nm and 550 nm peaks for reduced cytochrome c (Fig. 2h).
In the third control
experiment, we conjugated dielectric polystyrene (PS) nanoparticles with cytochrome c. The plasmon quenching spectral dips cannot be found even though the dielectric nanoparticle without plasmon resonance also scatters light, which indicated the presence of excited free electrons is necessary for the PRET process (Fig. 2i). Like the dependence of the FRET efficiency on the spectral overlap between donor and acceptor, the PRET efficiency was also dependent on the extent of the spectral overlap. The better the spectral overlap was, the
6
higher PRET efficiency and plasmon quenching intensity can be observed (Fig. 3). We controlled the average surface density of cytochrome c molecules on an individual gold nanoparticle by the molar concentration ratio used in the conjugation process. The surface coverage was ~ 150 molecule/particles according to electrochemical measurement. We found similar surface coverage by using surface thiol exchange method (Supplementary Methods online). We measured the scattering spectrum of many individual 30 nm nanoparticles and extracted the reduced cytochrome c visible absorption peaks. Due to the nonuniformity of the surface molecule numbers on each particle and nanoparticle plasmon resonance wavelength, the plasmon quenching or cytochrome c absorption peak intensity showed variations from particle to particle; whereas the spectral measurement on each individual nanoparticle was repeatable and stable (Supplementary Fig. 2 online), and no photochemical changes are observed. Besides of using gold nanoparticle and cytochrome c as the PRET pair, we also studied the PRET effect in the case of Hemoglobin molecules conjugating on single silver nanoparticle and we observed the plasmon quenching dips corresponding to the Soret band (~ 407 nm) of Hemoglobin (Supplementary Fig. 3 online). Although only the PRET in the visible wavelength range was observed here due to the optical properties of gold and sliver nanoparticles and the studied metalloprotein molecules, the PRET process at ultraviolet and near infrared range could be envisioned by using metallic nanoparticles with different properties (i.e. size, shape, free electron density, etc) and ultraviolet or near
7
infrared plasmon resonance wavelengths. Similarly to FRET, the PRET efficiency could be dependent on the distance between electronic resonance biomolecules and nanoplasmonic particles, their relative orientation More
experimental
characterization
work
and
14
and light polarizations. theoretical
simulation
(Supplementary Fig. 4 online) are underway. We also note recently the nanoparticle plasmon resonance quenching by fluorescent quantum dots was reported
15
, which further supports our discovery. The PRET-based ultrasensitive
biomolecular absorption spectroscopy on single metallic nanoparticle could be used for molecular imaging such as genetic analysis of small copies of latent nucleotides, activity measurements of small numbers of functional cancer biomarker proteins, and rapid detection of little biological toxin, pathogen and virus molecules. Additionally PRET could be applied in intracellular biomolecule conjugated nanoparticle sensors to detect localized in vivo electron transfer, oxygen concentration and pH value changes in living cells with nanoscale spatial resolutions.
8
Author Contributions G.L.L., ‘discover the new finding’, ‘conceive the idea’, and ‘perform the measurements’; Y.L., ‘conceive
the idea’, ‘initiated experiment’, ‘perform the
measurements’ and ‘process the data’; Y.C., ‘repeat the experiments’ and ‘design additional experiments’; T.K., ‘improve the conjugation chemistry’ and ‘repeat the experiments’; L.P.L., ‘conceive the idea’, ‘design the experiments’ and ‘advise other authors’.
References 1. C. F. Bohren & D. R. Huffman, Absorption and Scattering of Light by Small Particles 335-336 (Wiley, New York, 1998). 2. M. Moskovits, Rev. Mod. Phys. 57, 783,1985. P. Mulvaney, Langmuir 12, 788, 1996. D. A. Schultz, Curr. Opin. Biotechnol 14, 13, 2003. 3. J. R. Lombardi, R. L. Birke, T. H. Lu & J. Xu, J. Chem. Phys. 84, 4174, 1986. 4. S. Nie & S. R. Emory, Science 275, 1102, 1997. 5. M. Futamata, Y. Maruyama, & M. Ishikawa, J. Phys. Chem. B 108, 13119, 2004. 6. P. Das, & H. Metiu, J. Phys. Chem. 89, 4680, 1985. 7. D. S. Wuttke, M. J. Bjerrum, J. R. Winkler & H. B. Gray, Science 256, 1007, 1992. 8. C. Lange & C. Hunte, Proc. Natl. Acad. Sci. USA 99, 2800, 2002. 9. P. Andrew & W. L. Barnes, Science 306, 1002, 2004. 10. S. Boussaad, J. Pean, & N. J. Tao, Anal. Chem. 72, 222-226 (2000)
9
11. A. J. Haes, S. Zou, J. Zhao, G. C. Schatz & R. P. Van Duyne, J. Am. Chem. Soc. 128, 10905, 2006 12. G. L. Liu et al. Nature Nano. 1, 47, 2006. 13. G. Rascheke et al. Nano Lett. 3, 935, 2003. 14. T. Ambjornsson, G. Mukhopadhyay, S. P. Apell & M. Kall, Phys. Rev. B 73, 085412 (2006). 15. M. J. Romero, J. van de Lagemaat, I. Mora-Sero, G. Rumbles & M. M. AlJasslm, Nano Lett. 6, 2833-2837 (2006).
Acknowledgements This work was supported by a grant (code #: 05K1501-02810) from ‘Center for Nanostructured Materials Technology’ under ‘21st Century Frontier R&D Programs’ of the Ministry of Science and Technology, Korea.
Supplementary Items Supplementary Figure 1 Cyclic voltammogram of Cytochome c (Cyt c) on gold nanoparticles immobilized on a modified ITO surface. Supplementary Figure 2 Time-lapse measurement of scattering spectra of a single gold nanoparticle conjugated with reduced Cytochrome c molecules. Supplementary Figure 3 The PRET spectra for hemoglobin on silver nanoparticles. Supplementary Figure 4 Simulation of nanoparticle plasmon resonance coupling to a single Cytochrome c molecule. Supplementary Methods
10
Figure Captions Fig.
1 Schematic diagram of quantized plasmon quenching dips
Nanospectroscopy via PRET (a), Experimental system configuration. (b), Representative Rayleigh scattering spectrum of bare gold nanoparticles.
(c),
Representative
(d),
absorption
spectra
of
biomolecule
bulk
solutions.
Representative quantized plasmon quenching dips in the Rayleigh scattering spectrum of biomolecule conjugated gold nanoparticles.
Fig. 2 Experimental results of PRET from single gold nanoparticle to conjugated cytochrome c molecules. The Rayleigh scattering spectrum of a single gold nanoparticle coated with (a), only cysteamine coating, (b), cysteamine linker and reduced cytochrome c and c, cyteamine and oxidized cytochrome c.
The Rayleigh scattering spectrum was obtained using 1 sec
integration time. (d), The bulk visible absorption spectra of oxidized (blue solid line) and reduced (red solid line) 8 µM cytochrome c using a conventional UV-vis absorption spectroscopy. (e), The fitting curve for the spectrum in b. Black open circle: raw data, Green solid line: fitting curve of the raw data, Yellow solid line: Lorentzian scattering curve of bare gold nanoparticle, Red solid line: processed absorption spectra for the reduced conjugated cytochrome c by subtracting yellow curve from the green curve. (f), The fitting curve for the spectrum in (c). Blue solid line: processed absorption spectra for the oxidized conjugated cytochrome c by subtracting yellow curve from the green curve. (g), PRET spectra of gold nanoparticles coated with Cys-(Gly-Hyp-Pro)6 peptide. (h), PRET
11
spectra of cyteamine-cytochrome c on larger gold nanoparticles. (i), The control data of the PRET for 40 nm polystyrene bead coated by cytochrome c. The scale bars in the inset images stand for 2 μm.
Fig. 3 The PRET spectra for 3 representative gold nanoparticles conjugated with reduced cytochrome c molecules. The nanoparticle plasmon resonance wavelengths and the intensities of PRET-induced plasmon quenching dips vary from particle to particle in (a)-(c); however the plasmon quenching peak positions are consistent. Open circle: raw data, Green solid line: fitting curve, Red solid line: Lorentzian scattering curve of bare gold nanoparticle, Blue solid line: processed absorption spectra for the reduced conjugated cytochrome c by subtracting red curve from the green curve. The scale bars in the inset images stand for 2 μm.
12
