J. Phys. Chem. B 2008, 112, 6703–6707
6703
Surface-Enhanced Raman Spectroscopic Studies of Coactivator-Associated Arginine Methyltransferase 1 G. V. Pavan Kumar,† Ruthrotha Selvi,‡ A. Hari Kishore,‡ Tapas K. Kundu,‡ and Chandrabhas Narayana*,† Light Scattering Laboratory, Chemistry and Physics of Material Unit, Jawaharlal Nehru Center for AdVanced Scientific Research, Jakkur, Bangalore, -560064, India, Transcription and Disease Laboratory, Molecular Biology and Genetics Unit, Jawaharlal Nehru Center for AdVanced Scientific Research, Jakkur, Bangalore, -560064, India ReceiVed: December 9, 2007; ReVised Manuscript ReceiVed: March 7, 2008
We report, for the first time, the surface-enhanced Raman spectra of an important enzyme, coactivator-associated arginine methyltransferase 1 (CARM1), involved in various biological activities such as tumor suppressor function and stem cell differentiation. We have employed surface-enhanced Raman scattering (SERS) to obtain insight into the structural details of CARM1 by adsorbing it to silver (Ag) nanoparticles. The enzyme retains its activity even after its adsorption onto Ag nanoparticles. We observe strong SERS modes arising from amide vibrations and aromatic ring amino acids. The SERS spectra revealed amide I bands at 1637 cm-1 and 1666 cm-1, which arise as a result of the R helix of the protein and the polypeptide backbone vibration of a random coil, respectively. In order to confirm the amide vibrations, we have performed SERS on deuterated CARM1, which exhibits a clear red shift in amide band positions. The SERS spectra may provide useful information, which could be harnessed to study the functional interactions of CARM1 with small molecule modulators. Introduction Chromatin-modifying enzymes have emerged as key regulators of the various cellular processes, including transcription, replication, and repair.1 These enzymes, apart from catalyzing the post-translational modifications of histones and nonhistone proteins, also serve as important transcriptional coactivators (e.g., p300 and coactivator-associated arginine methyltransferase (CARM1)). They are components of large protein complexes and thereby regulate gene expression at various levels, helping in the maintenance of chromatin dynamicity.2 Protein arginine methyltransferases (PRMTs) are a highly conserved group of proteins catalyzing the transfer of methyl residues from the donor S-adenosyl methionine (SAM) to the arginine residues.3 So far, 11 members of this group have been identified (PRMT 1-11). The coactivator-associated arginine methyltransferase CARM1/PRMT4 is a type I arginine methyltransferase catalyzing the formation of asymmetric dimethyl arginine on the target protein. It was initially identified by a yeast two-hybrid screening for interacting partners of p160 coactivators.4 CARM1 acts on multiple substrates; both histones and nonhistone proteins. Histone H3 has two important CARM1 methylation sites: H3R17 and H3R26. Apart from these sites, there are other reported in vitro sites as well, which include H3R2, and few methylation at arginine residues at the C terminus of histone H3. CARM1 is a glucocorticoid receptor interacting protein (GRIP1). It is a secondary coactivator of several nuclear receptors, including estrogen, glucocorticoid, thyroid, and androgen receptors.5–7 It has been shown to interact with other histone acetyltransferases such as p300/CBP and PRMT1 to bring about cooperative transcriptional activation of tumor * Corresponding author. E-mail:
[email protected]. † Chemistry and Physics of Material Unit. ‡ Molecular Biology and Genetics Unit.
suppressor p53 responsive genes.8 CARM1 has also been shown to be a positive regulator of cyclin E19 and NF-κB promoter activity.10 Apart from these important transcriptional coactivation functions, it is also known to play important roles in various cellular processes through its ability to methylate nonhistone substrates. CARM1 has been implicated in muscle development,11 T-cell development,12 stem cell differentiation,13 RNA processing and transcription,14 and tumorigenesis as well.15 CARM1 is a 608 amino acid-protein with distinct functional and structural domains.16 Figure 1 shows a schematic diagram of CARM1. CARM1 catalyzes bisubstrate reaction involving the transfer of the methyl group from the donor SAM to the substrate, arginine on proteins. The enzymatic activity of the protein is regulated by three domains: the two substrate binding domains (the SAM or AdoMet binding domain (145-250 aa) and the dimer interface (155-230, 313-332)) and the arginine binding pocket (258-267 aa). There is a dimer interface from 165-230, 313-332 amino acids. Apart from this domain organization, the enzyme also has been structurally demarcated into the following domains: the C-terminal activation domain (500-608 aa), the GRIP1 binding domain (120-460 aa), the homooligomerization domain (120-460 aa), the methyltransferase domain (120-500 aa), and the coactivator domain, which is the full length protein (1-608 aa). Unlike, the other PRMTs, it does not recognize the glycine arginine-rich (GAR) motifs; presumably it prefers to recognize the XXPRX or XXRPX motif. The enzyme activity has been shown to be regulated by phosphorylation in a negative manner.17 However, not much structural information on the protein is available as such. The crystal structure of mouse CARM1 domains has been attempted, and very few details have been obtained.18 In order to understand the protein small molecule interactions of such an important protein, surface enhanced Raman spectroscopy was employed.
10.1021/jp711594z CCC: $40.75 2008 American Chemical Society Published on Web 05/07/2008
6704 J. Phys. Chem. B, Vol. 112, No. 21, 2008
Kumar et al.
Figure 1. Structural functional domains of CARM1/PRMT4. Schematic diagram showing the structural and fuctional domain organization of CARM1/PRMT4: 1-608 represents the amino acids in the protein. Structurally, the protein has three distinct regions: the AdoMet binding pocket (145-250), the dimer interface (155-230, 313-332) and the arginine binding pocket (258-267). The functional domains are, as indicated, the methyltransferase domain (120-500), the homooligomerization domain (120-460), the GRIP1 binding domain (120-460), the activation domain (500-608) and the full length protein (1-500), which is required for the coactivator function.
Raman spectroscopy has been a complimentary technique to X-ray crystallography to study the structure of protein. The major hurdle in such an exercise has been the Raman signals. Upon adsorption of molecules on coinage metallic nanosurfaces exhibiting atomic scale roughness, the Raman signal intensity of the molecule is enhanced by several orders of magnitude. This phenomenon is called SERS.19 In recent years, SERS has emerged as an effective tool for ultratrace analysis of molecules.20 It is bestowed with single-molecule sensitivity21 with chemical imaging capabilities, which make it a potent tool for biomolecular detection. Of late, SERS has been used to study different aspects of biology,22–29 with a considerable amount of success. It has also emerged as an efficient tool to probe and detect proteins that are yet to be crystallized.27 SERS can probe the secondary structure of the protein in solution phase and hence could be very important in studying the properties of the protein in a biochemical environment. Human CARM1 protein belongs to a category of proteins whose crystal structure is yet to be revealed. Thus, the present study gains importance from both the structure-function relationship of the enzyme as well as the interaction study with modulators. This could also be a valuable tool for exploring the post translational modifications of the enzyme and thereby the regulation. Experimental Details Expression and Purification of Human (HeLa) Core Histones. Human core histones were purified from a HeLa nuclear pellet. The pellet (5 mL) was homogenized with a blender in 40 mL of buffer A (0.1 M potassium phosphate [pH 6.7], 0.1 mM EDTA, 10% glycerol, 0.1 mM phenylmethylsulfonyl fluoride, 0.1 mM dithiothreitol [DTT]) containing 0.63 M sodium chloride and centrifuged at 16 000 rpm at 4 °C. The supernatant was incubated with 18 mL of previously swollen Biogel-HTP resin (DNA grade; Bio-Rad) for 3 h. The resin was then packed into an Econo column (Bio-Rad) and washed overnight with buffer A containing 0.63 M NaCl to remove any contaminating HAT activity. The bound core histones were eluted with buffer A containing 2 M NaCl and dialyzed against
BC100 buffer (20 mM Tris-HCl [pH 7.9], 100 mM KCl, 20% glycerol, 0.1 mM DTT) for 3 h. Expression and Purification of Full-Length Recombinant Human CARM1/PRMT4. FLAG epitope-tagged recombinant CARM1/PRMT4 was expressed and purified by infection of Sf21 cells with recombinant baculovirus followed by affinity chromatography of a whole-cell extract on M2 agarose (as shown in Figure 2). Briefly, a whole-cell extract was mixed with previously equilibrated M2 agarose beads in BC400 buffer (20 mM Tris-HCl [pH 7.9], 400 mM KCl, 20% glycerol, 0.01 mM dithiothreitol [DTT]). After being washed with the same buffer, the bound protein was eluted by pH elution with 0.1 M glycine, pH 3.5, and neutralization with Tris-NaCl buffer (0.5 M Tris, pH 7.4, 1.5 M NaCl). For performing SERS, the protein was dialyzed against Tris-NaCl buffer extensively so as to remove glycine. Histone Methyltransferase (HMTase) Assay. The indicated amount of highly purified HeLa core histones along with CARM1 (20 ng) alone or adsorbed onto silver nanoparticles were incubated in the HMTase buffer containing 20 mM Tris, pH 8.0, 4 mM EDTA, pH 8.0, 200 mM NaCl, and 1 µL of 15 Ci/mmol 3H SAM (Amersham), in a final reaction volume of 30 µL at 30 °C for 30 min. The reaction was stopped by quenching on ice for 5 min before blotting onto P81 (Whatman) filter paper. The radioactive counts were recorded on a Wallac 1409 Liquid Scintillation counter. The visualization of the radiolabeled histones was done by resolving the histones on 15% SDS-polyacrylamide gel followed by processing for fluorography (see Figure 2 A). SERS Measurements. Ag nanoparticles were prepared by the Lee and Meisel method.30 The average diameter of the nanoparticle was around 50 nm, and had a plasmon band centered at 426 nm. The protein in buffer solution was mixed with nanoparticles in a ratio of 10:70 by volume. After 5 min, the mixture was drop-coated over a microscope slide and place under a water immersion objective lens of a custom built Raman microscope, whose details can be found elsewhere.31 A 532 nm, frequency-double Nd:YAG laser was used as the Raman
SERS Study of CARM1 Enzyme
J. Phys. Chem. B, Vol. 112, No. 21, 2008 6705
Figure 2. Effect of silver nanoparticles on CARM1 activity. (A) Purified HeLa Core histones resolved on 15% SDS-polyacrylamide gel. (B) Full length FLAG-tagged CARM1/PRMT4 protein used in the experiment: CARM1 was purified from recombinant baculovirus-infected Sf-21 cells by immunoaffinity and analyzed on 12% SDS-polyacrylamide gel and visualized by coomassie blue staining. Lane 1, protein molecular weight marker; lane 2, purified CARM1. (C) Filterbinding Assay: HMTase assay was performed with CARM1 (20 ng) by using highly purified HeLa corehistones (600 ng). Lane 1, corehistones without enzyme; lane 2, histones with enzyme; lane 3, histones with CARM1 incubated with silver nanoparticles (1:5). The results are the mean values with the error bars (SD) of duplicate reactions. (D) Fluorography: HMTase assay was performed with CARM1 (20 ng) by using highly purified HeLa core histones (3 µg). Lane 1, corehistones without enzyme; lane 2, histones with enzyme; lane 3, histones with CARM1 incubated with silver nanoparticles (1:5). The radiolabeled histones were resolved on 15% SDS-polyacrylamide gel and visualized by coomassie blue staining followed by fluorography and autoradiography.
excitation source. The typical signal accumulation time was around 40 to 60 s. The Raman spectra shown in the figures were smoothed using a standard five-point fast Fourier transform (FFT) filtering technique. Results and Discussion Effect of Nanoparticle Adsorption on CARM1 Activity. The methyltransferase activity of recombinant CARM1 enzyme was assayed by filter binding and fluorography using core histone substrate. In order to record the SERS spectra of CARM1, the enzyme is adsorbed onto a silver nanoparticle. Hence it is essential to find out whether the activity of the enzyme is unaffected by the presence of nanoparticles. This was confirmed using the filter binding and fluorography shown in Figure 2. The enzyme activity remained unaltered upon adsorption onto the nanoparticles (compare Figure 2 C, lanes 2 and 3). Surface-Enhanced Raman Spectra of CARM1. Raman spectra of macromolecules such as protein are complex in nature. The spectrum would be an averaged signal of the whole protein, which would have contributions from various amino acids. However, in case of SERS, only specific modes of vibration are enhanced, depending upon the orientation and distance of the molecule from the metallic surface. The surface selection rule32,33 implies that the vibrational modes of a molecule perpendicular to the metallic nanosurface are enhanced to the greatest extent, whereas those modes parallel to the surface are not enhanced. This could give information about the orientation
of a molecule on a metallic nanosurface by employing SERS.34 Also, it is important to consider the electromagnetic enhancement factor,20 which scales as 1/r,12 where r is the distance between the molecule and the nanoparticle surface. This is an important factor, especially while studying proteins, because the interaction between the nanoparticle surface and the protein will be mainly confined to those amino acids that are at the surface of the protein. Also, it has been previously observed that dominant bands in SERS of proteins arise from aromatic amino acids like tryptophan (Trp), tyrosine (Tyr), phenylalanine (Phe) and histidine (His). This is due to a strong interaction between the π electrons in aromatic rings and the metallic surfaces.35,36 In the present case, the CARM1 has 56 aromatic amino acids (Tryptophan-5, Tyrosine-23, Phenylalanine-28). Therefore, we can expect SERS of CARM1 to be dominated by the aromatic amino acid modes. Figure 3 shows the SERS of CARM1 recorded in solution phase. It can be observed that the signalto-noise ratio of the spectra is high. As usual, we carried out the SERS of the buffer and Raman spectra of the Ag nanoparticle (see Supporting Information (SI)), which do not show any Raman peaks associated with the SERS spectra shown in Figure 3. This confirms that the protein spectrum is indeed that of CARM1. The complete band assignment of the SERS spectra of CARM1 is shown in Table 1, and some of the significant modes present in the SERS are discussed below. Amide Vibration. Amide I vibration mainly arises as a result of CdO stretching with a small contribution from N-H
6706 J. Phys. Chem. B, Vol. 112, No. 21, 2008
Figure 3. SERS spectra of CARM1 protein. λ ) 532 nm; signal accumulation time ) 40 s.
TABLE 1: SERS Band Assignment of CARM1 Proteina Raman shift (cm-1)
band assignment
1666m 1637vs 1588m 1532s 1466s 1383vs 1351vs 1270w 1197vs 1145w 1089w 1040w 981w 927s 819s 734m 616m 567m 437w
amide I (random coil) amide I (R helix) Trp, Tyr and /or Phe (ν8a) amide II and/or Trp δ(CH2) ν (COO-) Trp and/or δ (CH) amide III Tyr and/or Phe (ν9a) νas(CRCN) ν(CRN) Gly Met ν(C-COO-) Tyr and/or νas(C-S-C) Trp or His Phe (ν6b) Trp Pro
a m ) medium, vs ) very strong, s ) strong, w ) weak; ν ) symmetric stretch, νas ) asymmetric stretch, δ ) deformation.
bending. Amide II and amide III vibrations are due to NH2 scissoring and C-N stretching with N-H bending, respectively. There is a 1:1 correspondence between the secondary structure of the protein and the amide vibrations.37 The normal, nonresonant Raman spectra of a protein exhibits amide I and amide III vibration, whereas amide II vibration is generally a Raman inactive mode.37 However, in SERS, the amide II vibration is also active because of the surface selection rule.33 In CARM1, we observe all three amide vibrations. Interestingly, we observe two distinguishable amide I modes, which correspond to two different parts of the secondary structure of proteins. At 1666 cm-1, we observe a band corresponding to a random coil of the protein, and at 1637 cm-1, an intense band arises due to the R-helix. We observe a strong band at 1532 cm-1 and a weak band at 1270 cm-1, which were assigned to amide II and amide III vibrations, respectively. It should be noted here that amide II and III bands have overlap with bands arising from other amino acids. All these modes of vibration, especially the amide I, can be used as marker bands that are sensitive to secondary structural changes in protein. In order to confirm amide vibrations, we have performed SERS of the protein in D2O. To perform the SERS experiments
Kumar et al.
Figure 4. SERS spectra of CARM1 protein in H2O compared to D2O. λ ) 532 nm; signal accumulation time ) 40 s. The arrows in red indicate red shifting of amide modes. The spectrum was vertically shifted for clarity.
in D2O, both the nanoparticles and the CARM1 protein were centrifuged (to remove the H2O content) and were dispersed in pure D2O. After incubating the protein in D2O for sometime, they were mixed together in the proportion mentioned in the experimental section. Upon deuteration of protein, a significant number of hydrogen atoms will be replaced by the heavier deuterium. Assuming that the molecular vibrations are a simple harmonic motion, a heavier atom would produce a lower frequency of vibration. Therefore, we expect the vibrational modes associated with groups containing exchangable hydrogen (such as the amides in this case) to decrease in frequency as a result of the change in their reduced mass. Figure 4 compares the SERS of CARM1 in H2O and D2O. We observed a decrease of ∼6 cm-1 for amide I and 17 cm-1 for amide II modes. It is to be noted here that the intensities of the amide I and II modes also show a large change after deuteriation (see Figure 4). Amide I mode is a combination of 80% CdO stretching, 10% C-N stretching, and 10% N-H bending, whereas amide II mode is 60% N-H bending and 40% C-N stretching. Unlike Raman, the selection rules of SERS depends on the orientation of the vibrating molecules in relation to the metal surface, and results in changes in the intensity of the various vibrational modes. Deuteration rarely changes the conformation of the protein, but it might affect the hydrogen bonding characteristics of the protein or peptide. This would further introduce a change in the orientation of the amide groups with respect to the Ag metal surface, resulting in a change in the intensities of the modes. It should also be noted that the H-O-D bending vibration overlaps with the amide II region and could also contribute to the change in the intensity of the amide II band. All the above arguments authenticate the assignment of amide bands. Aromatic Amino Acid Vibrations. The Raman and SERS spectra of aromatic amino acids have been previously studied in detail.35 The SERS spectra of well-studied proteins,36 such as lysozyme, bovine serum albumin (BSA), IgG, and cytochrome C, are dominant in aromatic amino acid vibrations. In the case of p300, which is a human transcriptional coactivator protein, we have observed a similar behavior.27 Hence we expect the SERS spectrum of CARM 1 to contain a number of modes arising from aromatic amino acids (see Table 1). Significant contribution comes from different substituted ring modes of Tyr (1588, 1197, and 819 cm-1), Phe (1588, 1197, and 616 cm-1), and Trp (1588 and 1351 cm-1). It was interesting to observe that the phenyl “ring-breathing” vibration, ν12, which yields an intense band in the normal Raman spectra of proteins, was
SERS Study of CARM1 Enzyme completely absent from the SERS spectra of CARM1. This kind of behavior has previously been observed in SERS of proteins such as lysozyme, OXT, SSI, STI,36 and p300.27 Aliphatic Side-Chain Vibrations. It is a common feature in SERS spectra of proteins to exhibit strong modes around 1380 and 930 cm-1, which correspond to symmetric stretching of COO- and the stretching vibration of C_-COO-, respectively. These bands arise as a result of the adsorption of carboxylate groups pertaining to amino acids Asp, Glu, and/or C terminal groups. We observe the asymmetric stretching mode of CRCN and the symmetric stretching mode of CRCN at 1145 and 1089 cm-1, respectively. The deformation modes of CH2 and CH were observed at 1466 and 1351 cm-1, respectively. Both these modes have a strong overlap with vibrations of aromatic amino acids. Conclusion SERS has been performed on CARM1 protein in the solution phase for the first time. The spectrum is dominated by modes arising from aromatic amino acids. The amide vibrations were observed, and their values were in good agreement with other well-studied proteins. The amide modes were further confirmed by performing SERS of CARM1 in D2O. A clear red shift in amide bands upon deuterating the protein substantiated our band assignment. In the absence of the crystal structure of CARM1, the SERS spectra can give some insight into the secondary structure of the protein. The amide bands in SERS spectrum can be further harnessed as marker bands to study specific secondary structural changes of CARM1 upon its interaction with drug molecules. This would have implications in therapeutical studies of the protein, especially in understanding drug-protein interactions. Acknowledgment. C.N. would like to thank JNCASR for extending the funding under the Inter-Unit Collaborative Project. T.K.K. would like to thank the Department of Biotechnology, Government of India, for financial assistance for this work. C.N. and T.K.K. would like to acknowledge Prof. Dipak Dasgupta of Saha Institute of Nuclear Physics, Kolkata, for the fruitful discussions during the preparation of this manuscript. Supporting Information Available: Raman spectra of silver nanoparticles, CARM1, and SERS of buffer, which is compared with the SERS of CARM1. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Pal, S.; Sif, S. J. Cell. Physiol. 2007, 213, 306–315. (2) Wysocka, J.; Allis, C. D.; Coonrod, S. Front. Biosci. 2006, 11, 344–355. (3) Bedford, M. T.; Richard, S. Mol. Cell 2005, 18, 263–272. (4) Chen, D.; Ma, M.; Hong, H.; Koh, S. S.; Huang, S. M.; Schurter, B. T.; Aswad, D. W.; Stallcup, M. R. Science 1999, 284, 2174–2177. (5) Stallcup, M. R.; Kim, J. H.; Teyssier, C.; Lee, Y. H.; Ma, H.; Chen, D. J. Steroid Biochem. Mol. Biol. 2003, 85, 139–145.
J. Phys. Chem. B, Vol. 112, No. 21, 2008 6707 (6) Chen, D.; Huang, S. M.; Stallcup, M. R. J. Biol. Chem. 2000, 275, 40810–40816. (7) Kang, Z.; Ja¨nne, O. A.; Palvimo, J. J. Mol. Endocrinol. 2004, 18, 2633–2648. (8) An, W.; Kim, J.; Roeder, R. G. Cell 2004, 117, 735–748. (9) El Messaoudi, S.; Fabbrizio, E.; Rodriguez, C.; Chuchana, P.; Fauquier, L.; Cheng, D.; Theillet, C.; Vandel, L.; Bedford, M. T.; Sardet, C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 13351–13356. (10) Covic, M.; Hassa, P. O.; Saccani, S.; Buerki, C.; Meier, N. I.; Lombardi, C.; Imhof, R.; Bedford, M. T.; Natoli, G.; Hottiger, M. O. EMBO J. 2005, 24, 85–96. (11) Chen, S. L.; Loffler, K. A.; Chen, D.; Stallcup, M. R.; Muscat, G. E. O. J. Biol. Chem. 2002, 277, 4324–4333. (12) Kim, J.; Lee, J.; Yadav, N.; Wu, Q.; Carter, C.; Richard, S.; Richie, E.; Bedford, M. T. J. Biol. Chem. 2004, 279, 25339–25344. (13) Torres-Padilla, M. E.; Parfitt, D. E.; Kouzarides, T.; Zernicka-Goetz, M. Nature 2007, 445, 214–218. (14) Cheng, D.; Cote, J.; Shaaban, S.; Bedford, M. T. Mol. Cell 2007, 25, 71–83. (15) Majumder, S.; Liu, Y.; Ford III, O. H.; Mohler, J. L.; Whang, Y. E. Prostate 2006, 66, 1292–1301. (16) Teyssier, C.; Chen, D.; Stallcup, M. R. J. Biol. Chem. 2002, 277, 46066–46072. (17) Higashimoto, K.; Kuhn, P.; Desai, D.; Cheng, X.; Xu, W. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 12318–12323. (18) Troffer-Charlier, N.; Cura, V.; Hassenboehler, P.; Moras, D.; Cavarelli, J. Acta Crystallogr., Sect. F: Struct. Biol. Cryst. Commun. 2007, 63, 330–333. (19) Moskovits, M. ReV. Mod. Phys. 1985, 57, 783. (20) Kneipp, K.; Kneipp, H.; Itzkan, I.; Dasari, R. R.; Feld, M. S. J. Phys.: Condens. Matter 2002, 14, R597-R624. (21) Kneipp, K.; Wang, Y.; Kneipp, H.; Perelman, L. T.; Itzkan, I.; Dasari, R. R.; Feld, M. S. Phys. ReV. Lett. 1997, 78, 1667–1670. (22) Moore, B. D.; Stevenson, L.; Watt, A.; Flitsch, S.; Turner, N. J.; Cassidy, C.; Graham, D. Nat. Biotechnol. 2004, 22, 1133–1138. (23) Graham, D.; Mallinder, B. J.; Smith, W. E. Angew. Chem., Int. Ed. 2000, 39, 1061–1063. (24) Fabriciova, G.; Sanchez-Cortes, S.; Garcia-Ramos, J. V.; Miskovsky, P. Biopolymers 2004, 74, 125–130. (25) Streltsov, S.; Oleinikov, V.; Ermishov, M.; Mochalov, K.; Sukhanova, A.; Nechipurenko, Y.; Grokhovsky, S.; Zhuze, A.; Pluot, M.; Nabiev, I. Biopolymers 2003, 72, 442–454. (26) Ermishov, M.; Sukhanova, A.; Kryukov, E.; Grokhovsky, S.; Zhuze, A.; Oleinikov, V.; Jardillier, J. C.; Nabiev, I. Biopolymers 2000, 57, 272– 281. (27) Kumar, G. V. P.; Reddy, B. A. A.; Arif, M.; Kundu, T. K.; Narayana, C. J. Phys. Chem. B 2006, 110, 16787–16792. (28) Etchegoin, P.; Liem, H.; Maher, R. C.; Cohen, L. F.; Brown, R. J. C.; Milton, M. J. T.; Gallop, J. C. Chem. Phys. Lett. 2003, 367, 223– 229. (29) Kneipp, K.; Haka, A. S.; Kneipp, H.; Badizadegan, K.; Yoshizawa, N.; Boone, C.; Shafer-Peltier, K. E.; Motz, J. T.; Dasari, R. R.; Feld, M. S. Appl. Spectrosc. 2002, 56, 150–154. (30) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 86, 3391–3395. (31) Kumar, G. V. P.; Narayana, C. Curr. Sci., in press. (32) Moskovits, M. J. Chem. Phys. 1982, 77, 4408–4416. (33) Moskovits, M.; Suh, J. S. J. Phys. Chem. 1984, 88, 5526–5530. (34) Campion, A.; Kambhampati, P. Chem. Soc. ReV. 1998, 27, 241– 250. (35) Podstawka, E.; Ozaki, Y.; Proniewicz, L. M. Appl. Spectrosc. 2004, 58, 570–580. (36) Podstawka, E.; Ozaki, Y.; Proniewicz, L. M. Appl. Spectrosc. 2004, 58, 1147–1156. (37) Mayo, D. W.; Miller, F. A.; Hannah, R. W. Course Notes on the Interpretation of Infrared and Raman Spectra; Wiley: Hoboken, NJ, 2004; Chapter 8, p 205.
JP711594Z
