SERS and Computational Studies on MicroRNA Chains Adsorbed on Silver Surfaces

Maurizio Muniz-Miranda,* Cristina Gellini, Marco Pagliai, Massimo Innocenti,Pier Remigio Salvi, and Vincenzo SchettinoDipartimento di Chimica, UniVersita di Firenze, Via della Lastruccia 3, 50019 Sesto Fiorentino, Italy

ReceiVed: April 13, 2010; ReVised Manuscript ReceiVed: July 12, 2010

Surface-enhanced Raman scattering (SERS) of adenine-containing microRNA chains has been obtained byadsorption on roughened silver substrates. The spectral features of all of these samples appear dominated bythe bands of adenine. By comparison with the SERS spectra of adenine and adenosine obtained on the samesubstrates, along with DFT calculations on the interaction sites of adenine and adenosine with silver, inferencesare discussed about the structural arrangement of the microRNA chains with respect to the metal surface.This approach gives suitable guidelines in order to investigate the adsorption of complex biomolecules onmetal substrates.

1. Introduction

The adsorption of biomolecules like nucleic acids and proteinson metal surfaces is the subject of high interest for basicmolecular biology1 and for applications in clinical medicine andnanotechnology.2,3 The adsorption processes and the metal/biomolecule interactions can be studied with the help ofvibrational spectroscopy, which provides information on boththe dynamic and structural properties of the adsorbates. How-ever, ordinary Raman spectroscopy is usually unable to sharethese requirements for biological samples, due to low scatteringintensity and overlap with fluorescence. These inconveniencescan be effectively overcome by resorting to the surface enhancedRaman scattering (SERS) technique,4-6 where huge enhance-ments of adsorbate Raman signals are coupled with a drasticquenching of fluorescence. As the analyte adheres to surfacesof metals like Ag, Au, and Cu with nanoscale roughness, twomechanisms usually contribute to the Raman enhancement.7 Theelectromagnetic effect, responsible for enhancement factorshigher than 104, arises from the resonance of the exciting andRaman scattering radiations with the plasmon resonance bandof the metal electrons localized at the nanostructured surface.The chemical effect, on the other hand, is related to polarizabilitychanges when metal/molecule complexes are formed, producingenhancements up to 102. SERS spectroscopy can reach muchhigher enhancement factors (1014-1015) in single moleculeexperiments.8,9 As a consequence, this experimental methodcombines a fluorescence-like sensitivity with the molecularrecognition peculiar to the vibrational spectroscopy.

In this context, SERS spectroscopy has emerged as a powerfultechnique for the characterization of biological materials, inparticular single- and double-stranded DNA oligomers, takingadvantage of the development and use of reproducible metalsubstrates.10-12

Following these considerations, in the present study, SERSspectra are reported for microRNA sequences of adenine,adenine/uracil, and adenine/guanine/cytosine (hereafter calledAAA, AUA, and AGC, respectively, see Figure 1) adsorbedon roughened silver substrates, and compared with thosecorresponding to adenine and adenosine adsorbed on bothroughened Ag plates and Ag colloidal nanoparticles. The SERS

spectra of the microRNA chains as well as that of adenosineare due essentially to the presence of adenine. In order tocharacterize the molecular site interacting with the metal surface,the adsorbate has been modeled by means of the densityfunctional theory (DFT) as a complex formed by one Ag+ ionand one adenine (or adenosine) molecule. This approach hasbeen successfully proposed by our group for other adsorbedmolecules.13-15 The computational results constitute a set of dataupon which structural inferences about the microRNA adsorp-tion sites may be worked out.

2. Experimental Section

2.1. Roughening Procedure. Ag plates were polished withsuccessively finer grades of alumina powder down to 0.3 mm(Buehler Micropolish II), mixed with water distilled twice, thefirst from mineral water, the second with addition of alkalinepermanganate to the first purified fraction, discarding heads forboth operations. The plates were kept in a stirred solution of30 mM thiourea (Fluka) and 20 mM Fe(NO3)3 ·9(H2O) for 30 s,producing etching of silver with homogeneous roughness.16,17

2.2. Ligand Adsorption on Roughened Ag Plate. Adenine(Sigma, g99% purity) and adenosine (Sigma, g99% purity)* To whom correspondence should be addressed. E-mail: muniz@unifi.it.

Figure 1. Chemical structures of adenine and adenosine, along withschematic representations of microRNA chains.

J. Phys. Chem. C 2010, 114, 13730–1373513730

10.1021/jp103304r 2010 American Chemical SocietyPublished on Web 07/27/2010

were used as received. MicroRNA with adenine and uracil(AUA) was purchased from Sigma (purity: desalted); microR-NAs (purity: desalted) with only adenine (AAA) and withadenine, guanine, and cytosine (AGC) were kindly providedby Invitrogen srl (Italy). The Ag plates were immersed in 10-3

M adenine or adenosine water solution. After 48 h, the sampleswere accurately washed with current water and dried at roomtemperature and the relative SERS spectra were measured, asdescribed in a following subsection. Other roughened Ag plates,freshly prepared according to the chemical treatment above-described, were immersed for 2 days in water solutions ofmicroRNAs (10-5 M concentration) and dried at roomtemperature.

2.3. Preparation of Ag Colloidal Particles. Silver colloidswere prepared following Creighton’s procedure,18 by addingsilver nitrate (Aldrich, purity 99.998%) to an aqueous solutionof excess sodium borohydride (Aldrich, purity 99%) as areducing agent. The addition of adenine or adenosine (10-4 M)induced aggregation of the silver nanoparticles, which weredeposited and dried on glass to be examined by the microRamaninstrument.

2.4. Ag(I)/Adenine Complex Preparation. 5 × 10-2 MAgNO3 aqueous solution was dropwise added to hot 2 × 10-2

M adenine aqueous solution, obtaining a white precipitate, whichwas accurately washed and filtered at room temperature. Theelementary analysis indicated the formation of a 1:1 Ag(I)/adenine coordination compound with the simplest formula[AgNO3 (adenine) 3 H2O]: Ag weight 30% (expected 30%);N/C weight ratio 1.36 (expected 1.43); H/C weight ratio 0.18(expected 0.18). The presence of nitrate as a counterion wasascertained in the Raman spectrum by the occurrence of theNO3

- symmetric stretching mode at 1044 cm-1.2.5. Raman Measurements. The SERS spectra of adenine,

adenosine, and microRNAs were measured using a RenishawRM2000 microRaman apparatus, coupled with a diode lasersource emitting at 785 nm. Sample irradiation was accomplishedusing the ×50 microscope objective of a Leica MicroscopeDMLM. The beam power was ∼3 mW, and the laser spot sizewas adjusted between 1 and 3 µm. Raman scattering was filteredby a double holographic Notch filters system and collected byan air cooled CCD detector. The acquisition time for eachmeasurement was 10 s. All spectra were calibrated with respectto a silicon wafer at 520 cm-1. The same Raman apparatus wasemployed to measure the normal Raman spectra of adenine,adenosine, and Ag(I)/adenine complex as crystalline powders.

2.6. Computational Details. DFT calculations have beenperformed by means of the Gaussian 03 suite of programs19 foradenine, adenosine, and Ag+/adenine model complexes with theBecke three-parameter hybrid functional combined with the Lee-Yang-Parr correlation functional (B3LYP) and the LanL2DZbasis set. The validity of the computational approach for theAg+/adenine complexes was verified by the satisfactory simula-tion of the normal Raman spectrum of adenine, concerning bothfrequency positions and relative intensities. In particular, themean error (∑N |νcalc - νobs |/N, where N represents the numberof the observed frequencies compared with the calculated ones)is significantly lower than those previously obtained, as shownin Table S1 (Supporting Information). Figure S2 (SupportingInformation) shows the satisfactory simulation of the Ramanspectrum of adenine, and Figure S3 (Supporting Information)highlights the vibrational assignment of the prominent Ramanbands of adenine on the basis of the Cartesian displacementsof the corresponding calculated normal modes.

Also the simulated Raman spectrum of adenosine matchesthe experimental one, as shown in Figure S4 (SupportingInformation), by considering the suitable constraints betweenadenine and ribose. With two constraints for the ribose ring,the optimized structure is nearly identical to the molecularstructure of adenosine in the crystal lattice (see SupportingInformation, Figure S5). Analogous DFT calculations (namely,by using the same basis set) were performed also for adenosineinteracting with N1, N3, or N7 nitrogen atoms. All structureswere optimized with a very tight convergence criterion, and theharmonic frequencies were calculated with an improved gridfor the integral calculation, INTEGRAL(GRID ) 199974).

The optimized geometries correspond to true energy minima,as revealed by the lack of imaginary values in the calculationof the vibrational modes.

The Raman activities (Ai) calculated with the Gaussian 03program were converted to relative Raman intensities (Ii) usingthe following relationship derived from the basic theory ofRaman scattering20-22

where ν0 is the exciting frequency (in cm-1 units), νi is thevibrational frequency (in cm-1 units) of the ith normal mode,h, c, and k are fundamental constants, and f is a suitably chosencommon normalization factor for all peak intensities.

The calculated spectra were reported by assigning to eachnormal mode a Lorentzian shape with a 5 cm-1 full width athalf-maximum.

3. Results and Discussion

3.1. SERS Results. Ag plates, roughened according to theprocedure described in the Experimental Section, are employedfor the adsorption of adenine-containing microRNAs, whosesequences are reported in Figure 1. The spectral investigation,performed by microRaman measurements and 785 nm laserexcitation, has allowed obtaining suitable SERS spectra fromthese silver substrates, without significant interference offluorescence. Also, adenine and adenosine (see Figure 1) areused as ligands, in order to relate their SERS bands to those ofthe microRNA chains. In Figure 2, the SERS spectrum ofadenine adsorbed on a roughened silver plate is compared withthat obtained from Ag nanoparticles, exhibiting quite similarspectroscopic features, although with a lower S/N ratio, andindicating the same interaction with silver. The SERS spectraobtained from different substrates, which closely correspond tothe normal Raman spectrum (NRS) of the Ag(I)/adeninecomplex, are dominated by the ring breathing band at 735 cm-1,markedly upshifted with respect to that of NRS of adenine at722 cm-1.

Since adenine in RNA chains is replaced by adenosine, alsothe SERS spectra of adenosine adsorbed on both a roughenedsilver plate and Ag nanoparticles have been obtained andcompared in Figure 3 with the NRS of solid adenosine. It maybe seen that adenosine exhibits a number of bands closely relatedto those of adenine (722, 1247, 1332, and 1479 cm-1) and, inaddition, bands due to vibrational modes more localized on theribose ring (762, 844, 1507, and 1574 cm-1). The SERS spectraof adenosine from different substrates are quite similar, confirm-ing also for adenosine the same interaction with silver. Thestrongest SERS band of adenosine occurs at 731 cm-1, corre-sponding to that observed in the SERS of adenine at 735 cm-1,while the vibrational modes localized on the ribose ring are

Ii ) f(ν0 - νi)4Ai/νi[1 - exp(-hcνi/kT)] (1)

SERS and Computational Studies on MicroRNA Chains J. Phys. Chem. C, Vol. 114, No. 32, 2010 13731

totally absent. This indicates that adenosine binds to silver inan adsorption arrangement similar to adenine, leading tosignificant enhancements only for modes localized on theadenine moiety, and suggests that our Raman studies onmicroRNA samples must be essentially focused on the detectionof the marker band at 731 cm-1. In fact, the SERS spectra ofall three microRNAs (Figure 4) are dominated by the 731 cm-1

band, so that the chains are reasonably bound to the Ag substratethrough this nucleobase.

In the SERS spectrum of AAA, other bands occur at 630,1267, 1330, and 1460 cm-1, in close proximity to those ofadsorbed adenine (see Figure 2). For the AUA chain, whereadenine and uracil nucleotides alternate, additional weak bands

are observed at 790 and 1235 cm-1, whereas, for the AGC chain,where adenine and guanine are present with a smaller percentof cytosine, a further weak peak is found at 675 cm-1.Comparison with reported data11,23 suggests for the latter bandsof AUA and AGC an assignment to Raman modes of uraciland guanine, respectively. In saturated water solution of uracilat neutral pH and for the polycrystalline powder, two intensepeaks, 784/790 cm-1 and 1235/1236 cm-1, occur in the NRS.23

Therefore, the AUA peaks, located at 790 and 1235 cm-1, areattributable to uracil not directly bound to the substrate,confirming that the chain interacts with the substrate throughadenine. Further, in agreement with the assignment of the breathingband of guanine at 678 cm-1 in the NRS of the SA20N2oligonucleotide sequence (containing guanine),11 the sameassignment is here proposed for the 675 cm-1 SERS band ofAGC. The absence of spectral features related to cytosine24 inthe SERS spectrum of AGC agrees with the fact that thisnucleobase is a minor component of the chain not directlyinvolved in the chemisorption.

3.3. MicroRNA Interaction Sites. The adsorption of adenineon metal surfaces has been discussed in much detail over theyears;25-35 in particular, this nucleobase was reported to adsorbon silver surfaces Via different nitrogen atoms of the two rings(N1, N3, and N7 atoms) or through the external amino group,in either a side-on or a flat-on orientation. Undoubtedly, thesecontroversial conclusions were related to the variety of interac-tion sites and to the different tautomeric forms of this molecule.In fact, adenine is known to exist in two forms in solution, withthe 9-H tautomer dominant over the 7-H tautomer.36-38 On theother hand, DFT calculations on surface complexes with adeninebound to one silver atom suggested an interaction through theN3 nitrogen,32 consistent also with that recently proposed foradenine on gold.35 The authors of ref 31, however, recognizedthat other model systems involving silver clusters or silver ionscould provide a more precise picture of the SERS spectrumconcerning both frequency shifts and relative intensities.

Here, the silver/adenine interaction geometry is examinedmainly in relation to the structural changes possibly occurringwhen adenine enters as a component of the microRNAsequences. In this respect, the Raman spectrum of the Ag(I)/adenine complex (Figure 2B) appears almost identical to theSERS of adenine (Figure 2C), apart from the symmetricstretching band of the NO3

- counterion at 1044 cm-1, indicatingthat the interaction of adenine with the silver surface resemblesthat of the Ag(I) coordination compound. This point providesuseful information: (i) adenine interacts with silver as a neutral

Figure 2. SERS spectra of adenine adsorbed on a Ag plate (C) andon Ag nanoparticles (D), compared with the normal Raman spectra ofadenine (A) and Ag(I)/adenine complex (B) as polycrystalline powders.The asterisk refers to the NO3

- stretching band of the Ag(I)/adeninecomplex. Intensities in arbitrary units. Excitation: 785 nm laser line.

Figure 3. SERS spectra of adenosine adsorbed on a Ag plate (B) andon Ag nanoparticles (C), compared with the normal Raman spectrumof adenosine (A) as polycrystalline powder. Intensities in arbitrary units.Excitation: 785 nm laser line.

Figure 4. SERS spectra of microRNAs adsorbed on Ag plates.Intensities in arbitrary units. Excitation: 785 nm laser line.

13732 J. Phys. Chem. C, Vol. 114, No. 32, 2010 Muniz-Miranda et al.

molecule, not in anionic form; (ii) the adsorption of adenine onsilver can be suitably modeled like a molecule bound to onesilver ion, as in the Ag(I) coordination compound. For thisreason, the present DFT calculations have been performed onadenine molecule and on Ag+/adenine complexes with one silverion linked alternatively to N1, N3, or N7. Eight minimastructures, a1, a2, b, c1, c2, d, e1, and e2, have been found(Figure 5), by considering adenine also in the 7-H tautomericform. The energies of the different Ag+/adenine complexes arereported in Table 1. Not surprisingly, the lowest minimumcorresponds to the tautomeric structure d, where the moleculeis bound to the silver ion through N3 and (in a weaker way)N9. The second most stable structures are a1 and b, where themolecule interacts with silver Via one nitrogen atom of thepyrimidinic ring. The Ag+/adenine complex involving the amino

group has been excluded from our calculations, since thestructure migrates toward a1 during the optimization step.Considering now the vibrational results, the most strikingexperimental difference between NRS and SERS of adenine isrelative to the ring breathing mode: the NRS band at 722 cm-1

undergoes a significant upshift, ∼13 cm-1, when adenine isadsorbed on silver, along with a marked increase of the relativeintensity, as shown in Figure 2. The calculated upshifts for theb and d complexes are 18 and 15 cm-1, respectively. The basicindication from the vibrational calculations is therefore that thebreathing mode frequency considerably upshifts only when Ag+

is bound to N3. The calculated frequency shifts of Table 1 makethis point particularly evident. Moreover, by considering theCartesian displacements of this mode in the N1, N3, or N7complexes (Figure 6), the nitrogen atom bound to silver movessignificantly from the equilibrium geometry toward the metalonly in the b complex. A similar behavior has been observedfor the d structure. This justifies the intensity increase of thecorresponding SERS band on the basis of the surface selectionrules,38 for which totally symmetric vibrations showing largeshifts of the equilibrium position in the direction normal to themetal surface should be greatly enhanced. In fact, for an edge-on adsorption through N3, the motion of the nitrogen atom withits electronic density induces a large polarizability change alongthe direction normal to the silver surface and, consequently,strongly enhances the SERS band. The SERS bands observedat higher frequencies (1273, 1331, 1372, and 1459 cm-1)correspond, instead, to normal modes not appreciably involvingthe N3 motion. In conclusion, the DFT calculations of the Ag+/adenine complexes point to an interaction of adenine with silverVia N3, described by either b or d geometries. The SERSfrequencies are satisfactorily calculated by means of both modelsystems, as shown in Table 2. The d model, where adenineinteracts with silver in the 7-H tautomeric form, is moreplausible on the basis of the strong stabilization of this complex(Table 1) and the good matching of the simulated SERS profilewith the experimental one, as shown in Figure 7. Actually, thering breathing mode in the Ag+ complex appears stronglyenhanced with respect to the corresponding mode of the freemolecule, as experimentally observed.

Figure 5. Optimized structures of Ag+/adenine complexes obtainedby means of DFT calculations.

TABLE 1: Calculated Energies, ∆E, in cm-1 and kJ ·mol-1

of Ag+/Adenine Complexes Shown in Figure 5 with Respectto the Lowest Energy Complex (d)a

∆E

model complex cm-1 kJ ·mol-1 νbreathb ∆νbreath

c r(Ag · · ·Nx)

a1 2092 25.03 712 2 r(Ag · · ·N1) ) 2.167a2 3862 46.20 713 3 r(Ag · · ·N1) ) 2.324

r(Ag · · ·N10) ) 2.468b 2057 24.61 728 18 r(Ag · · ·N3) ) 2.158c1 3862 46.20 714 3 r(Ag · · ·N7) ) 2.156c2 2678 32.04 701 -9 r(Ag · · ·N7) ) 2.304

r(Ag · · ·N10) ) 2.400d 725 15 r(Ag · · ·N3) ) 2.223

r(Ag · · ·N9) ) 2.808e1 6741 80.64 710 0 r(Ag · · ·N1) ) 2.171e2 7569 90.55 714 4 r(Ag · · ·N1) ) 2.264

r(Ag · · ·N10) ) 2.582

a The corresponding breathing mode frequencies, νbreath (cm-1),the frequency shifts with respect to that of adenine, ∆νbreath (cm-1),and the distances between the Ag+ ion and the closer nitrogenatoms of adenine, r(Ag · · ·Nx) (Å), are reported. b The breathing modefrequency of adenine is 710 cm-1, according to our DFTcalculations. c The experimental shift of the SERS mode of adenineadsorbed on a silver plate with respect to the corresponding Ramanmode is 13 cm-1.

Figure 6. Cartesian displacements of the calculated breathing modesof Ag+/adenine complexes with interaction sites located at N1, N3,and N7, corresponding to the SERS band of adenine observed at 735cm-1. Hydrogen atoms are omitted for the sake of simplicity.

SERS and Computational Studies on MicroRNA Chains J. Phys. Chem. C, Vol. 114, No. 32, 2010 13733

The b and d structures can be used as a basis for thediscussion on the interaction sites of microRNAs, whose SERSspectra are dominated by the adenine bands (see Figure 4). Onthe other hand, since in microRNAs the N9 site is occupied bya ribose group, as well as in adenosine, the preferred adsorptionstructure of adenine in these nucleobase chains corresponds tomodel b, by interaction with silver only through the N3 nitrogen.To verify this conclusion, DFT calculations have been performedon adenosine linked to metal via N1, N3, or N7. In spite of thesimplicity of the model systems, where the metal surface ismodeled by one silver ion, the simulated SERS profileslegitimate the indication of N3 as the most probable interactionsite. As shown in Figure 8, only the Ag+/adenosine complexwith N3 interaction is able to satisfactorily reproduce theobserved strong enhancement of the ring breathing modeobserved at 731 cm-1.

The dominance of the adenine bands, observed in the SERSspectra of microRNAs, was previously observed in the SERSof adenine-containing DNA oligomers adsorbed on Au sub-strates and attributed to the higher SERS cross section of thisnucleobase.11 In the present study, the interaction of microRNAsto silver Via the N3 nitrogen atom of adenine can be reasonablyproposed on the basis of the SERS and DFT results for adenineand adenosine. This fact is consistent with a stronger chemicalinteraction of the adenine nucleobase with the metal surface,because the pyrimidinic ring is directly involved in the bondwith silver. Unlike other DNA/RNA nucleobases (thymine,

uracil, guanine, cytosine), the pyrimidinic ring of adenine hasaromatic character, favoring a stronger metal/molecule chemicalinteraction and, as a consequence, a stronger SERS enhance-ment. In this respect, it is interesting to note that the quasi-isoenergetic a1 and b structures (Table 1) describe the possibleinteractions of adenine by means of the N1 or N3 nitrogen atomsof the pyrimidinic ring, which are more efficient than thoseinvolving the N7 nitrogen atom of the five-membered ring. Thisdisplays the importance of involving the delocalized π electronsof the aromatic ring in the chemical bond with silver, whichcould be used as a useful guideline for investigating theadsorption processes of these biomolecules.

Finally, some qualitative considerations can be advancedabout the binding of adenine to the silver substrate in themicroRNA chains. Reference is made to AAA oligomers forwhich several sets of structural data have been reported.40-44 Itis in fact known that oligoAAAs exist in single-strandconformations40,41 and polyA (polyriboadenylic acid) as a single-strand flexible helix,42 both structures being stabilized by basestacking. A model has been described for polyA in more detail42

as a single helix with a pitch height of ∼25 Å and nine

TABLE 2: Experimental SERS Frequencies (cm-1) of Adenine, Adenosine, and AAA MicroRNA Compared with ThoseCalculated (cm-1) for the b and d Complex Models

SERS calculated

adenine Ag plate adenosine Ag plate AAA Ag plate b model d model Sa assignment

1459 1460 1460 1466 1470 A′ ring stretching + H bending1372 1365 1397 1384 A′ H bending + ring stretching1331 1328 1330 1354 1346 A′ H bending + ring stretching1273 1269 1267 1250 1253 A′ H bending + ring stretching735 731 731 728 725 A′ ring breathing632 628 630 610 600 A′ ring bending570 566 543 553 A′ ring bending

a Symmetry species (Cs group).

Figure 7. Simulated SERS spectrum (B) of adenine (d complexmodel), compared with the simulated normal Raman spectrum (A) andwith the observed SERS spectrum (C).

Figure 8. Simulated SERS spectra of adenosine by interaction of theN1, N3, or N7 nitrogen atom with one silver ion, compared with theSERS spectrum of adenosine (upper panel).

13734 J. Phys. Chem. C, Vol. 114, No. 32, 2010 Muniz-Miranda et al.

nucleotides per turn, on the basis of the structure of thetrinucleoside diphosphate ApApA.44 A similar helicoidal ar-rangement can be reasonably proposed for the microRNAchains. The occurrence of the ring breathing mode of adenineat the same frequency (731 cm-1) for all microRNAs suggeststhat the same base is involved in the interaction with the metal,probably the initial adenine in the nucleobase sequences.

4. Conclusions

SERS-active silver substrates have been prepared by chemicalroughening and employed for the adsorption of adenine,adenosine, and adenine-containing microRNA chains. TheseSERS platforms, which provide quite similar spectroscopicfindings in comparison with the silver colloidal nanoparticles,are easy to fabricate, stable in time, and reusable after successivesurface treatment. Moreover, the adsorption process merelyconsists of plunging a roughened silver plate into the ligandsolution, which may then be retrieved. These advantages, withrespect to other SERS substrates as silver colloids, widelycompensate a lower S/N ratio. They, however, exhibit quitesatisfactory SERS efficiency: an approximate evaluation of theeffective quantity of ligand observed by a microRaman mea-surement, by comparing the UV absorption of the solutionbefore and after adsorption on silver, results for AAA microRNAin a very small amount, ∼10-17 mol.

The SERS spectra of microRNAs, where the bands of adenineappear dominant, have been analyzed by comparison with thoseof adenine, adenosine, and the Ag(I)/adenine coordination com-pound. A detailed DFT study on the possible silver/adenine surfacecomplexes has been performed. This investigation, which pointsto an interaction of adenine in the 7-H tautomeric form with thesilver surface through the N3 nitrogen atom of the pyrimidinic ring,sheds a new light on the “vexata quaestio” concerning theadsorption of adenine on metal. Our conclusions are based not onlyon the stability of the surface complexes and the agreement betweencalculated and observed wavenumbers but also on the verysatisfactory simulation of both the frequency shift and the intensityincrease for the marker SERS band of adenine observed around730 cm-1. The present study allows also suggesting reliableconsiderations about the adsorption on metal and the structuralarrangement of complex systems like RNA/DNA polynucleotidechains, along with an acceptable explanation for the predominanceof the adenine bands in the SERS spectra of these biomolecules,essentially based on the involvement of aromatic π electrons inthe interaction with the metal surface.

Acknowledgment. The authors gratefully thank the ItalianMinistero dell’Universita e Ricerca for the financial support andDr. Barry Howes (University of Firenze), who kindly providedus microRNA samples.

Supporting Information Available: Table showing vibra-tional frequencies and figures showing Raman spectra, Cartesiandisplacements, and optimized structures. This material is avail-able free of charge via the Internet at http://pubs.acs.org.

References and Notes

(1) Davis, L. G. Basic Methods in Molecular Biology; Appleton &Lange: Noralk, CT, 1994.

(2) Pierstorff, E.; Ho, D. J. Nanosci. Nanotechnol. 2007, 7, 2940.(3) Herne, T. M.; Ahern, A.; Garrell, R. L. J. Am. Chem. Soc. 1991,

113, 846.(4) Chang, R. K.; Furtak, T. E. Surface-Enhanced Raman Scattering;

Plenum Press: New York, 1981.(5) Otto, A. In Light Scattering in Solids; Cardona, M., Guntherodt,

G., Eds.; Springer-Verlag: Berlin, 1984; Vol. IV, p 289.

(6) Campion, A.; Kambhampati, P. Chem. Soc. ReV. 1998, 27, 241.(7) Aroca, R. Surface-enhanced Vibrational Spectroscopy; Wiley &

Sons: 2006.(8) Kneipp, K.; Wang, Y.; Kneipp, H.; Perelman, L. T.; Itzkan, I.;

Dasari, R. R.; Feld, M. S. Phys. ReV. Lett. 1997, 78, 1667.(9) Nie, S.; Emory, S. R. Science 1997, 275, 1102.

(10) Green, M.; Liu, F.-M.; Cohen, L.; Kollensperger, P.; Cass, T.Faraday Discuss. 2006, 132, 269.

(11) Barhoumi, A.; Zhang, D.; Tam, F.; Halas, N. J. J. Am. Chem. Soc.2008, 130, 5523.

(12) Braun, G. J.; Lee, S. J.; Dante, M.; Nguyen, T.-Q.; Moskovits, M.;Reich, N. J. Am. Chem. Soc. 2007, 129, 6378.

(13) Muniz-Miranda, M.; Pergolese, B.; Bigotto, A. Vib. Spectrosc. 2007,43, 97.

(14) Cardini, G.; Muniz-Miranda, M. J. Phys. Chem. B 2002, 106, 6875.(15) Pagliai, M.; Bellucci, L.; Muniz-Miranda, M.; Cardini, G.; Schettino,

V. Phys. Chem. Chem. Phys. 2006, 8, 171.(16) Geissler, M.; Wolf, H.; Stutz, R.; Delmarche, E.; Grummt, U. W.;

Michel, B.; Bietsch, A. Langmuir 2003, 19, 6301.(17) Carla, F.; Innocenti, M.; Loglio, F.; Muniz-Miranda, M.; Salvi,

P. R.; Gellini, C.; Cavallini, M.; Felici, R.; Lastraioli, E.; Salvietti, E.;Foresti, M. L. Mater. Sci. Semicond. Process. 2009, 12, 21.

(18) Creighton, J. A.; Blatchford, C. G.; Albrecht, M. G. J. Chem. Soc.,Faraday Trans. 2 1979, 75, 790.

(19) , Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople,J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.

(20) Polavarapu, P. L. J. Phys. Chem. 1990, 94, 8106.(21) Keresztury, G.; Holly, S.; Varga, J.; Besenyei, G.; Wang, A. Y.;

Durig, J. R. Spectrochim. Acta, Part A 1993, 49, 2007.(22) Keresztury, G. In Handbook of Vibrational Spectroscopy; Chalmers,

J. M., Griffiths, P. R., Eds.; Wiley & Sons: Chichester, U.K., 2002; Vol. 1, p 71.(23) Giese, B.; McNaughton, D. J. Phys. Chem. B 2002, 106, 1461.(24) Sanchez-Cortes, S.; Garcıa-Ramos, J. V. Surf. Sci. 2001, 473, 133.(25) Koglin, E.; Sequaris, J. M.; Valenta, P. J. Mol. Struct. 1980, 60, 421.(26) Watanabe, T.; Kawanami, O.; Katoh, H.; Honda, K.; Nishimura,

Y.; Tsuboi, M. Surf. Sci. 1985, 158, 341.(27) Kim, S. K.; Joo, T. H.; Suh, S. W.; Kim, M. S. J. Raman Spectrosc.

1986, 17, 381.(28) Otto, C.; Van den Tweel, T. J. J.; de Mul, F. F. M.; Greve, J. J.

Raman Spectrosc. 1986, 17, 289.(29) Otto, C.; de Mul, F. F. M.; Huizingz, A.; Greve, J. J. Phys. Chem.

1988, 92, 1239.(30) Itoh, K.; Minami, K.; Tsujino, T.; Kim, M. J. Phys. Chem. 1991,

95, 1339.(31) Giese, B.; McNaughton, D. J. J. Phys. Chem. B 2002, 106, 101.(32) Watanabe, H.; Ishida, Y.; Hayazawa, N.; Inouye, Y.; Kawata, S.

Phys. ReV. B 2004, 69, 155418.(33) Li, J.; Fang, Y. Spectrochim. Acta, Part A 2007, 66, 994.(34) Feng, F.; Zhi, G.; Jia, H. S.; Cheng, L.; Tian, Y. T.; Li, X. J.

Nanotechnology 2009, 20, 295501.(35) Kundu, J. O.; Neumann, O.; Janesko, B. G.; Zhang, D.; Lal, S.;

Barhoumi, A.; Scuseria, G. E.; Halas, N. J. J. Phys. Chem. C 2009, 113, 14390.(36) Callis, P. R. Annu. ReV. Phys. Chem. 1983, 34, 329.(37) Wiorkiewicz-Kuczera, J.; Karplus, M. J. Am. Chem. Soc. 1990,

112, 5324.(38) Shugar, D.; Psoda, A. In Landolt-Børnstein - New Series:Biophysics

of Nucleic Acids; Saenger, W., Ed.; Springer: Berlin, 1990; Vol. VII/1d, p 308.(39) Creighton, J. A. In Spectroscopy of Surfaces; Clark, R. J. H., Hester,

R. E., Eds.; Wiley: Chichester, U.K., 1988; pp 37-85.(40) Brahms, J.; Michelson, A. M.; Van Holde, K. E. J. Mol. Biol. 1966,

15, 467.(41) Holcomb, D. N.; Tinoco, I., Jr. Biopolymers 1965, 3, 121.(42) Poland, D.; Vournakis, J. N.; Scheraga, H. A. Biopolymers 1966, 4, 223.(43) Saenger, W.; Riecke, J.; Suck, D. J. Mol. Biol. 1975, 93, 529.(44) Suck, D.; Manor, P. C.; Saenger, W. Acta Crystallogr., Sect. B

1976, 32, 1727.

JP103304R

SERS and Computational Studies on MicroRNA Chains J. Phys. Chem. C, Vol. 114, No. 32, 2010 13735