AngewandteChemie

Gold NanorodsDOI: 10.1002/anie.201107304

Quantitative Replacement of Cetyl TrimethylammoniumBromide by Cationic Thiol Ligands on the Surface ofGold Nanorods and Their Extremely Large Uptake byCancer Cells**Leonid Vigderman, Pramit Manna, and Eugene R. Zubarev*

.AngewandteCommunications

636 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2012, 51, 636 –641

Over the past decade, gold nanorods (Au NRs) have receivedbroad attention as possible therapeutic and diagnostic agentsbecause of their anisotropic physical properties.[1–8] However,as-synthesized Au NRs based on the most common syntheticapproach, the seed-mediated synthesis,[9,10] are not directlyuseable for these applications. They are synthesized in aconcentrated cetyl trimethylammonium bromide (CTAB)solution and are noncovalently coated with a CTABbilayer.[11] For CTAB-capped rods to remain soluble, theconcentration of free CTAB in solution must remain above acertain value and there is a constant dynamic exchange ofCTAB molecules between the solution and NR surfaces.[12]

This feature limits the usefulness of these rods for manybiological applications because free CTAB is known to behighly cytotoxic.[13, 14] Various strategies have been developedto functionalize gold nanorods with a variety of differentligands to reduce their cytotoxicity.[12, 15–20] Further reportshave dealt with the problems of stability and biocompatibilityof Au NRs by using various polymer shells,[21] polyelectro-lytes,[22, 23] peptides,[24] surfactants,[25] and lipids[14, 26] to modifythe Au NR surface. Partial replacement of CTAB wasqualitatively confirmed in many of these cases, but the exactsurface composition of nanorods and the amount of residualCTAB was either unknown or not possible to determine. Forthat reason, there is always some ambiguity in interpretationof in vitro and in vivo experiments involving such nanorods.[27]

Beyond just biocompatibility and low toxicity, there hasbeen great interest in understanding the cellular uptake of AuNRs.[28–32] For example, if CTAB is replaced by nonionicpolyethylene glycol, the cytotoxicity is substantially reduced,but the cellular uptake drops as much as 94 %, and virtuallyno nanorods can enter the cells.[30] When polyelectrolytes arerandomly wrapped around the CTAB bilayer, the uptakeincreases and the average number of rods that enter each cellranges from hundreds[31] to one hundred thousand.[29] How-ever, even polyelectrolyte-coated Au NRs are taken by cellsin picogram quantities, which is only 0.1% of a typical cellmass. This small weight fraction may explain a somewhatlimited success of photothermal oblation of tumor cells,[33]

which is strongly dependent on the actual number of nano-rods present inside the cells. Therefore, an obvious goal is todesign stable and nontoxic (CTAB free) nanorods that couldenter cancer cells in much larger quantities.

Here we report a strategy for complete exchange ofCTAB for its thiolated analogue (16-mercaptohexadecyl)tri-methylammonium bromide (MTAB) and we directly deter-mine the chemical composition of the surface coating.Through 1H NMR analysis, we are able to prove that CTABis fully replaced by a covalent MTAB monolayer. By

combining thermogravimetric analysis (TGA) with TEMsize analysis of the nanorods, we are able to accuratelydetermine the packing density of the self-assembled thiolmonolayer on the surface of Au NRs. Cytotoxicity of thesenovel nanorods was studied by MTT assay on breast cancercells (MCF-7), which were imaged by correlated optical andscanning electron microscopy (SEM). Extremely largenumber of nanorods (about 2 � 106 NRs per cell) werefound inside the viable cells as confirmed by transmissionelectron microscopy (TEM) and quantified by inductivelycoupled plasma-optical emission spectrometry (ICP-OES).

Several reports have shown that even with systems thathave successfully functionalized gold nanorods, CTAB is stillobserved to be present on the NRs.[28] We hypothesized thatwe could use a molecule with a structure very similar toCTAB to completely exchange the CTAB, but with the abilityto bind more strongly to the NR surface. Thus, we chose tosynthesize a thiolated CTAB analogue (MTAB) whosesynthesis is shown in Scheme 1. Commercially available1,16-hexadecanediol was converted to a corresponding dibro-mide under standard bromination conditions, which wasfollowed by the synthesis of its monothioester. Cleavage ofthe acetyl group was carried out in anhydrous methanol usingin situ generated hydrogen chloride. The resulting 16-bromo-1-hexadecanethiol was subjected to exhaustive methylation toyield the final product (see the Supporting Information).Importantly, the purified MTAB compound was found to bewater-soluble, which offers an opportunity to perform ligandexchange in aqueous media. The MTAB ligand contains anentire CTAB moiety, but installs a pendant thiol group to beable to strongly anchor it to the gold surface. The cross-section of this molecule is small enough to form a compactmonolayer on the surface of nanorods, thus providing theirsolution stability.

Stabilization of gold nanorods is generally difficultbecause of their small surface-to-volume ratio, and finding acompound that can directly exchange with CTAB is furthercomplicated by the high-density positive charge and theamphiphilic nature of the CTAB bilayer (Figure 1). In thepast, researchers have exchanged only the tips of goldnanorods because of relatively weak CTAB binding thereand have noted that it is much stronger on the sides.[34–36]

Furthermore, as the CTAB on the sides of the rods begins toexchange, it appears that the bilayer structure breaks downand rods tend to aggregate prematurely. However, choosing a

Scheme 1. Synthesis of (16-mercaptohexadecyl)trimethylammoniumbromide (MTAB; NBS= N-bromosuccinimide, THF = tetrahydrofuran).

[*] L. Vigderman, Dr. P. Manna, Prof. E. R. ZubarevDepartment of Chemistry, Rice UniversityHouston, TX 77005 (USA)E-mail: zubarev@rice.eduHomepage: http://www.owlnet.rice.edu/~zubarev/

[**] This work was supported by the NSF (grant numbers DMR-0547399and DMR-1105878).

Supporting information for this article is available on the WWWunder http://dx.doi.org/10.1002/anie.201107304.

AngewandteChemie

637Angew. Chem. Int. Ed. 2012, 51, 636 –641 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org

cationic ligand with a chemical structure similar to nativeCTAB allows for the direct exchange in water to proceedsmoothly as shown schematically in Figure 1. The noncova-lent bilayer of CTAB is replaced with a compact thiolatemonolayer of MTAB which is covalently anchored to thesurface through gold–sulfur bonds. UV/Vis absorbance spec-troscopy is often used to assess the stability of gold nanorodsin solution as the intensity, the peak shape, and the peakposition of the longitudinal plasmon resonance (LSPR) aresensitive to any aggregation that may occur. The UV/Visspectrum (see Figure S1A in the Supporting Information)shows the typical absorbance spectrum for Au NRs anddemonstrates what happens when excess CTAB is removedfrom solution through multiple rounds of centrifugationfollowed by dispersion in pure water. As expected, after justfour rounds, the NRs have lost all solubility as evidenced bythe lack of any absorbance. In stark contrast, when MTAB-modified NRs are purified by the same technique, there is noappreciable change in the absorbance spectrum even afterfive rounds of centrifugation and dispersion (see Figure S1Bin the Supporting Information), indicating their muchimproved solution stability while also completely purifyingthe product.

Additional experiments show that these rods can becompletely dried and kept in the solid state indefinitelywithout losing their water solubility. Surprisingly, a standardlyophilization technique can be applied to aqueous solution ofMTAB NRs to produce a fluffy dark brown powder. Fig-ure 2a shows a photograph of 15 mg of lyophilized Au NRsthat occupy an area of several square centimeters. Mostimportantly, when a small amount of this powder is placed onthe surface of pure water, rapid dissolution takes place asmanifested by the concentration swirls and a continuousdiffusion of the colored solute (Figure 2b). The nanorodsdissolve in a matter of seconds without any heating orsonication. It is sufficient to flip the vial only once to form ahomogeneous solution which does not contain any freeorganic molecules. The solution remains intact for at leastseveral months as confirmed by UV/Vis analysis. This dataclearly shows that a dense monolayer of small cationic thiolMTAB is capable of stabilizing large metallic particles in purewater and preventing their flocculation through entropicallydriven depletion interactions.[37] This data also showes for the

first time that the presence of free surfactant in the solution ofgold nanorods is not necessary as long as their surface iscovalently functionalized by a dense organic shell.

Another indirect method that has often been used tocharacterize Au NR complexes is the zeta potential measure-ment, which can qualitatively describe the charge surroundinga nanoparticle. The zeta potential of the MTAB NRs wasfound to be + 55 mV, demonstrating the cationic structures asexpected from the high concentration of quaternary ammo-nium groups positioned around the NR surface. However, thezeta potential is unable to distinguish between CTAB andMTAB on the surface of the rods given that they both havethe same cationic headgroup exposed to the solution and so isuseful only as a secondary confirmation of structure. Wechose 1H NMR spectroscopy to analyze the organic compo-nent of the MTAB NRs which should be able to differentiatebetween the CTAB and MTAB molecules. We used KCN todissolve the gold NR core and release any surface-boundorganic material into solution[39] after having rigorouslypurified the functionalized NRs. Because of the low surface-to-volume ratio as well as high density of the gold core, weperformed an oxidative dissolution of 105 mg of Au NRs inone milliliter of D2O to get enough organic material for asufficiently strong 1H NMR signal. The dissolution proceededvery slowly, taking approximately two weeks to dissolve all ofthe gold, suggesting a high packing density for the MTABsurface functionality.[40] On the contrary, CTAB-coated NRsdissolved in about one hour under similar conditions. Thedissolution procedure caused the organic component toprecipitate from the D2O solution, apparently because ofthe formation of a Au/CN/MTAB complex. The NMRspectrum of the D2O supernatant showed no signals, con-firming that the entire organic component had precipitatedfrom the solution. The NMR spectrum of this precipitate

Figure 1. Exchange of the CTAB bilayer for the MTAB thiol monolayer.

Figure 2. a) Photograph of lyophilized powder of MTAB-functionalizedAu NRs and b) its spontaneous dissolution in pure water. The imageswere taken consecutively within intervals of five second.

.AngewandteCommunications

638 www.angewandte.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2012, 51, 636 –641

dissolved in deuterated methanol, as well as that of pureCTAB and MTAB thiol, is shown in Figure 3. Comparing thespectra of CTAB and MTAB, it is clear that the isolated signalfrom the terminal methyl group of CTAB at 0.91 ppm is notpresent in the MTAB spectrum, making it a convenientmarker for the presence of any CTAB in solution. Analysis ofthe NMR spectrum of the dissolved NRs (spectrum C inFigure 3) clearly shows that there is no signal correspondingto this methyl group, proving that no CTAB remained on thesurface of the NRs after the exchange and the followingpurification. Further analysis of the NMR spectrum of thedissolved NRs shows the formation of the expected disulfideof MTAB as evidenced by the triplet at 2.75 ppm correspond-ing to the a-disulfide protons (instead of 2.5 ppm for a-thiolprotons). Importantly, the NMR technique is capable ofdetecting small organic molecules at concentration close to10�5

m, whereas the concentration of our solution was 10�2m.

This implies that even if there were some residual CTABundetectable by NMR spectroscopy, it would not exceed0.1%. Therefore, for all practical purposes one can concludethat the method described hear offers an opportunity toquantitatively replace CTAB by its thiolated analogue.

Because the exact chemical composition of the NRsurface agents was determined, it was possible to use TGAto further characterize the MTAB NRs. TGA analysis is ahighly useful analytical technique for hybrid inorganic–organic nanostructures because it allows us to accuratelyquantify the weight percentage of a nanostructure that isorganic compared to inorganic core.[39, 41, 42] For our study,more than 100 mg of MTAB NRs were synthesized andfunctionalized to be used for TGA and NMR measurements,which is quite a large amount when it comes to Au NRsynthesis, requiring a synthesis on the four liter scale.However, even this amount of material is difficult to handle,forming an almost unusable thin film when dried normally. Tosolve this problem, the MTAB NRs were lyophilized from anaqueous solution, forming a low density powder which couldbe easily handled, as mentioned previously (Figure 2). TGAanalysis of the lyophilized material (see Figure S2 in the

Supporting Information) shows a weight loss of 5.3% in therange between 200 and 450 8C corresponding to the percent-age of organic material in the structure. Since the NMRanalysis proved that no residual CTAB was left, this weightmust be entirely because of MTAB ligands covalentlyattached to the surface of rods.

We then performed a careful TEM analysis by measuringover 300 nanorods, which determined that the average lengthand width of MTAB NRs were 41.88 and 9.87 nm, respec-tively. With this information, we were able to calculate agrafting density of about 3.7 moleculesnm�2 for an MTABmonolayer and 5013 MTAB molecules residing on each AuNR (see the Supporting Information for detailed calcula-tions). This is close to the grafting density of 4.5 mole-culesnm�2 for neutral alkanethiols on flat gold substrates[43]

and may be slightly lower because of repulsive forces betweenthe positively charged headgroups. We believe this is the firstproof of a dense self-assembled thiol monolayer on goldnanorods with a calculated grafting density. Estimation of theextent of thiol functionalization has been carried out onDNA-functionalized Au NRs by measuring fluorescenceintensity, but such structures contained only about 40 DNAmolecules per nanorod[17] as compared to over 5000 MTABmolecules.

With the characterization complete, we performed cyto-toxicity and cell uptake experiments to see if the MTAB NRscould be useful for biological applications. We first measuredtheir in vitro cytotoxicity compared to regular CTAB-cappedNRs using the standard MTT assay on MCF-7 breast cancercells (see Figure S3 in the Supporting Information). Thesestudies reaffirmed the cytotoxicity of regular CTAB-cappedgold nanorods and showed the decreased cytotoxicity of thethiolated rods, which are not cytotoxic even up to concen-trations of 0.1 gL�1. This finding supports previous assertionsthat surface charge may not be a key factor for thecytotoxicity of nanoparticle systems.[31] Although previousreports have shown that it is mostly free CTAB thatcontributes to the cytotoxicity of CTAB NRs in the shortterm,[31] it is unknown what may happen to surface-boundCTAB in the longer term in in vivo experiments. Our systemhas the advantage of having quantitative CTAB replacementsuch that this is no longer an issue. Low cytotoxicity isparticularly useful if the NRs can be taken up by cellsefficiently. To study the uptake, MCF-7 cells were treated witha 20 mg mL�1 solution of MTAB NRs in Eagle’s minimumessential medium for 24 h. The cells were subsequently rinsedmultiple times with PBS solution to remove the excess of NRsthat did not enter the cells. After that, the cells weretrypsinized, redispersed into medium, and carefully platedonto a new culture slide. This approach allowed us to create aclear background without any free nanorods, which iscritically important for differentiating between the internal-ized particles and those which are not associated with thecells. The dark field optical image (Figure 4A) clearly shows alarge uptake of MTAB NRs and demonstrates their highscattering efficiency. It also appears that NRs are clusteredtogether inside the cells rather than being evenly distributedthroughout their interior, suggesting that the MTAB NRsenter the cells through an endosomal pathway. To perform a

Figure 3. 1H NMR spectra in deuterated methanol of CTAB (A), pureMTAB (B), and the organic product released upon oxidative dissolu-tion of MTAB NRs (C).

AngewandteChemie

639Angew. Chem. Int. Ed. 2012, 51, 636 –641 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org

correlated SEM imaging of the exact same group of cells, wefixed the sample with glutaraldehyde and further cross-linkedthe cells with osmium tetroxide. This enabled us to preventthe collapse of cells under high vacuum conditions requiredfor the SEM. The SEM imaging correlated with the opticalexperiments (Figure 4B) shows areas of increased brightnessinside the cells because of clusters of internalized MTABNRs. The locations of the NR clusters match up well with thebright spots seen in the optical micrograph, although there aresome morphological changes that occurred during the fixationand dehydration processes. Most importantly, SEM imagingreveals that there are no MTAB NRs on the surface of thecells, which would be impossible to confirm by the opticalmicroscopy alone. In contrast, the visualization of nanorodsresiding in thicker cell areas is more efficient by opticalmicroscopy, displaying their strong potential as imagingcontrast agents.

To further characterize the cellular uptake, we performedtransmission electron microscopy (TEM) imaging of cellstreated with the MTAB NRs as shown in Figure 5. UnlikeSEM and optical microscopy, TEM images of the microtomed

cross-sections reveal precise location and spatial distributionof individual nanorods. We prepared 75 nm thick sections thatwere cut through the center of cells. The TEM image of theentire cell cross-section (Figure 5A) shows an extremely largeamount of NRs, which appear as dark particles. The volume ofthis cross-sectional sample is approximately 0.5% of the totalcell volume (20 mm in diameter), which means that the totalnumber of nanorods inside the cell is approximately 200 timesgreater than that present in this image (about 10 000 NRs).Thus, even on the basis of TEM one can roughly estimate thatthere are at least two million nanorods per cell. If confirmedby quantitative methods, it would be significantly larger thanthe numbers reported in the literature for nanorods of similarsize, which range from several hundred[31] to one-hundred-fifty thousand NRs per cell.[29] Even at low magnification onecan see that NRs do not enter the nucleus and are clustered inendosomes. Higher magnification of regions outlined by theboxes (Figure 5B and C, respectively) clearly shows individ-ual nanorods that do not appear to be free in the cyto-plasm.[29, 44–46] The amount of uptake was compared to anotherpopular NR system, pegylated nanorods (PEG NRs), whichare known to resist cellular uptake and thus form a goodnegative control for our system. To quantify the averagenumber of NRs taken by cells, we performed ICP-OESanalysis on dissolved cells which had been treated with bothMTAB NRs and their pegylated analogues. Based on theseresults, while almost no PEG NRs were taken up (less than1%), about 40% of MTAB NRs in solution were taken up bythe cells. This corresponds to a level of about 2.17 millionMTAB NRs per cell that have been treated with a solution ofnanorods (see the Supporting Information for detailedcalculations). Such a great amount of metallic gold increasesthe mass of cancer cell by 0.13 ng, which is approximately

Figure 4. A) Dark field optical micrograph and B) SEM image of cellstreated with MTAB NRs.

Figure 5. TEM images of microtomed MCF-7 cancer cells treated withthe MTAB NRs (75 nm thick). A) View of an entire cell cross-section.B) Magnified image of the area outlined by the black box in panel A;C) Magnified image of the area outlined by the dashed box in panel B.

.AngewandteCommunications

640 www.angewandte.org � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2012, 51, 636 –641

13% of a typical cell mass (1 ng). It is particularly remarkablethat the cells retain their viability and continue to proliferatewith such a significant amount of gold nanostructures insidethem. These results show that the MTAB NRs have a very lowcytotoxicity and suggest that they may be promising forfurther biomedical applications such as drug/gene deliveryand photothermal therapy.

In conclusion, we have synthesized highly stable, func-tionalized gold nanorods as an alternative to CTAB-cappedNRs through a direct quantitative exchange with a thiolatedCTAB analogue. The MTAB NRs showed a highly increasedstability to multiple rounds of purification compared toCTAB NRs. We were able to quantitatively prove thecomplete removal of CTAB through the use of 1H NMRspectroscopy, and thus determined the exact composition ofthe organic component of the hybrid nanostructures. We alsosuccessfully performed a TGA analysis to measure the exactorganic versus inorganic composition of hybrid nanostruc-tures and determined a grafting density for the MTAB thiol,proving the formation of a compact self-assembled monolayeron gold nanorods. The MTAB NRs were not only found to benontoxic, but were also efficiently taken up by cancer cellsin vitro in very large amounts as shown by a comprehensivecombination of optical microscopy, SEM, TEM, and ICP-OES data.

Received: October 17, 2011Published online: November 15, 2011

.Keywords: cancer cells · cellular uptake · gold · nanorods ·surface chemistry

[1] N. J. Durr, T. Larson, D. K. Smith, B. A. Korgel, K. Sokolov, A.Ben-Yakar, Nano Lett. 2007, 7, 941 – 945.

[2] X. Huang, I. H. El-Sayed, W. Qian, M. A. El-Sayed, J. Am.Chem. Soc. 2006, 128, 2115 – 2120.

[3] C. Yu, J. Irudayaraj, Anal. Chem. 2007, 79, 572 – 579.[4] X. Liu, Q. Dai, L. Austin, J. Coutts, G. Knowles, J. Zou, H. Chen,

Q. Huo, J. Am. Chem. Soc. 2008, 130, 2780 – 2782.[5] R. Guo, L. Zhang, H. Qian, R. Li, X. Jiang, B. Liu, Langmuir

2010, 26, 5428 – 5434.[6] T. Kawano, Y. Niidome, T. Mori, Y. Katayama, T. Niidome,

Bioconjugate Chem. 2009, 20, 209 – 212.[7] H. Wang, T. B. Huff, D. A. Zweifel, W. He, P. S. Low, A. Wei,

J. X. Cheng, Proc. Natl. Acad. Sci. USA 2005, 102, 15752 – 15756.[8] a) L. Zhao, X. Pang, R. Adhikary, J. Petrich, M. Jeffries-EL, Z.

Lin, Adv. Mater. 2011, 23, 2844 – 2849; b) L. Zhao, X. Pang, R.Adhikary, J. Petrich, Z. Lin, Angew. Chem. 2011, 123, 4044 –4048; Angew. Chem. Int. Ed. 2011, 50, 3958 – 3962.

[9] B. Nikoobakht, M. A. El-Sayed, Chem. Mater. 2003, 15, 1957 –1962.

[10] N. R. Jana, L. Gearheart, C. J. Murphy, J. Phys. Chem. B 2001,105, 4065 – 4067.

[11] B. Nikoobakht, M. A. El-Sayed, Langmuir 2001, 17, 6368 – 6374.[12] C. J. Orendorff, T. M. Alam, D. Y. Sasaki, B. C. Bunker, J. A.

Voigt, ACS Nano 2009, 3, 971 – 983.[13] R. Cortesi, E. Esposito, E. Menegatti, R. Gambari, C. Nastruzzi,

Int. J. Pharm. 1996, 139, 69 – 78.[14] H. Takahashi, Y. Niidome, T. Niidome, K. Kaneko, H. Kawasaki,

S. Yamada, Langmuir 2006, 22, 2 – 5.

[15] H. W. Liao, J. H. Hafner, Chem. Mater. 2005, 17, 4636 – 4641.[16] S. K. Basiruddin, A. Saha, N. Pradhan, N. R. Jana, Langmuir

2010, 26, 7475 – 7481.[17] A. Wijaya, K. Hamad-Schifferli, Langmuir 2008, 24, 9966 – 9969.[18] A. Gole, C. J. Murphy, Chem. Mater. 2005, 17, 1325 – 1330.[19] D. Gentili, G. Ori, F. M. Comes, Chem. Commun. 2009, 5874 –

5876.[20] I. E. Sendroiu, M. E. Warner, R. M. Corn, Langmuir 2009, 25,

11282 – 11284.[21] G. Prencipe, S. M. Tabakman, K. Welsher, Z. Liu, A. P. Good-

win, L. Zhang, J. Henry, H. Dai, J. Am. Chem. Soc. 2009, 131,4783 – 4787.

[22] A. P. Leonov, J. Zheng, J. D. Clogston, S. T. Stern, A. K. Patri, A.Wei, ACS Nano 2008, 2, 2481 – 2488.

[23] H. C. Huang, S. Barua, D. B. Kay, K. Rege, ACS Nano 2009, 3,2941 – 2952.

[24] N. Chanda, R. Shukla, K. V. Katti, R. Kannan, Nano Lett. 2009,9, 1798 – 1805.

[25] A. M. Alkilany, P. K. Nagaria, M. D. Wyatt, C. J. Murphy,Langmuir 2010, 26, 9328 – 9333.

[26] S. E. Lee, D. Y. Sasaki, T. D. Perroud, D. Yoo, K. D. Patel, L. P.Lee, J. Am. Chem. Soc. 2009, 131, 14066 – 14074.

[27] N. Khlebtsov, L. Dykman, Chem. Soc. Rev. 2011, 40, 1647 – 1671.[28] A. K. Oyelere, P. C. Chen, X. Huang, I. H. El-Sayed, M. A. El-

Sayed, Bioconjugate Chem. 2007, 18, 1490 – 1497.[29] T. S. Hauck, A. A. Ghazani, W. C. Chan, Small 2008, 4, 153 – 159.[30] H. J. Parab, H. M. Chen, T. C. Lai, J. H. Huang, P. H. Chen, R. S.

Liu, M. Hsiao, C. H. Chen, D. P. Tsai, Y. K. Hwu, J. Phys. Chem.C 2009, 113, 7574 – 7578.

[31] A. M. Alkilany, P. K. Nagaria, C. R. Hexel, T. J. Shaw, C. J.Murphy, M. D. Wyatt, Small 2009, 5, 701 – 708.

[32] T. B. Huff, M. N. Hansen, Y. Zhao, J. X. Cheng, A. Wei,Langmuir 2007, 23, 1596 – 1599.

[33] a) L. C. Kennedy, L. R. Bickford, N. A. Lewinski, A. J. Cough-lin, Y. Hu, E. S. Day, J. L. West, R. A. Drezek, Small 2011, 7,169 – 183; b) R. Choi, J. Yang, J. Choi, E.-K. Lim, E. Kim, J.-S.Suh, Y.-M. Huh, S. Haam, Langmuir 2009, 25, 17520 – 17 527.

[34] Z. Nie, D. Fava, E. Kumacheva, S. Zou, G. C. Walker, M.Rubinstein, Nat. Mater. 2007, 6, 609 – 614.

[35] K. K. Caswell, J. N. Wilson, U. H. Bunz, C. J. Murphy, J. Am.Chem. Soc. 2003, 125, 13914 – 13915.

[36] M. Sethi, G. Joung, M. R. Knecht, Langmuir 2009, 25, 1572 –1581.

[37] a) R. Tuinier, J. Rieger, C. G. de Kruif, Adv. Colloid InterfaceSci. 2003, 103, 1 – 31; b) M. C. LeMieux, Y.-H. Lin, P. D. Cuong,H.-S. Ahn, E. R. Zubarev, V. V. Tsukruk, Adv. Funct. Mater.2005, 15, 1529 – 1540.

[38] R. H. Terrill et al, J. Am. Chem. Soc. 1995, 117, 12537 – 12548.[39] S. S. Agasti, C. C. You, P. Arumugam, V. M. Rotello, J. Mater.

Chem. 2008, 18, 70 – 73.[40] A. C. Templeton, M. J. Hostetler, C. T. Kraft, R. W. Murray, J.

Am. Chem. Soc. 1998, 120, 1906 – 1911.[41] a) B. P. Khanal, E. R. Zubarev, Angew. Chem. 2009, 121, 7020 –

7023; Angew. Chem. Int. Ed. 2009, 48, 6888 – 6891; b) W.-S.Chang, L. S. Slaughter, B. P. Khanal, P. Manna, E. R. Zubarev, S.Link, Nano Lett. 2009, 9, 1152 – 1157.

[42] Z. Li, P. Huang, X. Zhang, J. Lin, S. Yang, B. Liu, F. Gao, P. Xi, Q.Ren, D. Cui, Mol. Pharm. 2010, 7, 94 – 104.

[43] L. H. Dubois, R. G. Nuzzo, Annu. Rev. Phys. Chem. 1992, 43,437 – 463.

[44] R. Shukla, V. Bansal, M. Chaudhary, A. Basu, R. R. Bhonde, M.Sastry, Langmuir 2005, 21, 10644 – 10654.

[45] A. Verma, O. Uzun, Y. Hu, Y. Hu, H. S. Han, N. Watson, S. Chen,D. J. Irvine, F. Stellacci, Nat. Mater. 2008, 7, 588 – 595.

[46] P. Nativo, I. A. Prior, M. Brust, ACS Nano 2008, 2, 1639 – 1644.

AngewandteChemie

641Angew. Chem. Int. Ed. 2012, 51, 636 –641 � 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org

Supporting Information

� Wiley-VCH 2011

69451 Weinheim, Germany

Quantitative Replacement of Cetyl Trimethylammonium Bromide byCationic Thiol Ligands on the Surface of Gold Nanorods and TheirExtremely Large Uptake by Cancer Cells**Leonid Vigderman, Pramit Manna, and Eugene R. Zubarev*

anie_201107304_sm_miscellaneous_information.pdf

1

Supporting Information

Synthesis of MTAB. Synthesis of 1,16-dibromohexadecane, 1. A 50 mL solution of

triphenylphosphine (3.93 g, 15 mmol) in anhydrous THF was added to a stirred solution of N-

bromosuccinamide (2.67 g, 15 mmol) in 50 mL THF at 0 °C. Upon vigorous stirring, a solution of

hexadecane-1,16-diol (1 g, 3.9 mmol) in 25 mL THF was slowly added to the mixture of NBS and

Ph3P. The resulting solution was warmed to room temperature and then heated at 60 °C for 3.5 hours.

THF was removed by rotary evaporation and the residue was re-crystallized form ethanol to obtain 1.1

g of 2 as white powder (70 % isolated yield). 1H NMR (CDCl3, 400 MHz): δ 1.26-1.46 (m, 24 H), 1.85

(q, 4H), 3.41 (t, 4H).

Synthesis of 16-bromo-1-hexadecanethioacetate, 2. One gram (2.60 mmol) of 1 was dissolved in 40

mL methanol in a three-neck flask and the solution was degassed for one hour. Separately, 124 mg of

sodium methoxide was dissolved in 12 mL dry, ice-cold methanol and mixed with 204 mg (2.6 mmol)

of thioacetic acid. This mixture was transferred into a funnel. The solution in the three-neck flask was

refluxed under argon atmosphere and the content in the funnel was slowly added to the solution over

the course of 4 hours. After the reaction was complete, the content of the flask was cooled to room

temperature and methanol was removed under reduced pressure. The yellow oil was further purified by

column chromatography (20 % ethyl acetate in hexane) to obtain 480 mg of 3 (50 % yield).

Synthesis of 16-bromo-1-hexadecanethiol 3. To a stirred solution of 2 (400 mg, 1.05 mmol) in 10 mL

of methanol, 4 mL of acetyl chloride was added drop-wise and the reaction mixture was kept at 50 °C

for 4 hours. After the reaction was complete, 200 mL of CH2Cl2 was added to the reaction mixture and

excess acetyl chloride and HCl were removed by multiple extractions with DI water. The mixture was

2

dried over sodium sulfate. Methylene chloride was evaporated under reduced pressure to obtain 284

mg of 3 as colorless oil (80 % isolated yield). 1H NMR (CDCl3, 400 MHz): δ 1.26-1.46 (m, 25H), 1.60

(m, 2H), 1.85 (q, 2H), 2.52 (q, 2H), 3.41 (t, 2H).

Synthesis of 16-mercaptohexadecyltrimethylammonium bromide, MTAB. To a solution of 3 (284 mg,

0.85 mmol) in 5 mL of ethyl acetate, 3 mL of 4.2 M ethanolic solution of trimethylamine was added.

The mixture was vigorously stirred under argon for 4 days. The resulting white precipitate was filtered

and the washed several times with ethyl acetate to remove excess trimethylamine. The residue was

vacuum dried to obtain 270 mg of MTAB (80 % isolated yield). 1H NMR (CDCl3, 400 MHz): δ 1.26-

1.46 (m, 25H), 1.60 (m, 2H), 1.85 (m, 2H), 2.52 (q, 2H), 3.5 (s, 9H), 3.55-3.7 (m, 2H).

Gold nanorods synthesis. Gold nanorods were synthesized according to procedure described

elsewhere.[41] Briefly, a seed solution was created by adding 60 µL of a 0.1 M NaBH4 solution to 10

ml of 5.0x10-4 M HAuCl4 in 0.1 M CTAB solution upon rigorous stirring. Separately, the growth

solution was prepared by adding 2 mL of 0.1 M AgNO3 to 2 liters of 5.0x10-4 M HAuCl4 in 0.1 M

CTAB solution. To this solution was added 11.5 mL of 0.1 M ascorbic acid solution and hand-stirred

until the solution became clear, followed immediately by the addition of 3.2 mL of seed solution. A

dark brown color indicating the presence of gold nanorods was visible after about 30 min and the

solution was allowed to sit overnight before functionalization.

Gold nanorods functionalization with MTAB thiol. Four liters of Au NRs solution was

concentrated to about 9 mL using centrifugation at 13,000 rpm. The concentrated NRs solution was

purified by two cycles of centrifuging at 13,000 rpm, removing the supernatant and redispersing the

precipitate in pure water. To this solution was added 60 mg of MTAB and the solution was stirred for

3

2 days. The MTAB-NRs were purified by five cycles of centrifugation as described earlier and finally

dispersed in 4 mL of water to a concentration of 15 mg/mL as determined by ICP-OES measurements.

Figure S1. UV-vis absorbance spectrum of CTAB-NRs (A) and MTAB-NRs (B) after purification steps by

high-speed centrifugation and redispersion into pure water.

Analysis of MTAB-NRs. For TGA analysis (TA Q-600 TGA/DSC), MTAB-NRs were first

lyophilized to obtain a useable powder. 1 mL of 15 mg/mL NR solution was placed into an Eppendorf

tube, flash frozen with liquid nitrogen, and lyophilized. In the TGA measurement, the temperature was

held at 120 °C for 1 hour to remove any residual water and then ramped at 10 °C/min to 850 °C. For

1H-NMR analysis (Bruker 400 MHz), 1.5 mL of 15 mg/mL NR solution was transferred into D2O by

two rounds of centrifugation at 13,000 rpm and redispersion into D2O. To the D2O solution was added

50 mg KCN and the solution stirred for two days until the solution was colorless and a white

precipitate had formed. The precipitate was isolated by centrifugation and dissolved in deuterated

methanol. UV-Vis analysis was carried out on a Varian Cary 500 UV-Vis-NIR Spectrophotometer.

Zeta Potential measurements were carried out on a Malvern Instruments Zetasizer Nano-ZS.

4

Figure S2. TGA curve of MTAB-NRs. Residual water is lost at 100 °C. Main weight loss of about 5.3 %

from organic component occurs from 200 to 450 °C.

Calculation of MTAB binding parameters from TGA data. From TEM size analysis of ~200 NRs,

average length and width are 41.88 nm and 9.87 nm, respectively. Assume NR = cylindrical middle

with radius (r) and height (h) plus 2 spherical caps with base radius (r) and height (a). Based on TEM,

the spherical caps extend for about 2 nm on each side so h = 37.88 nm, r = is 4.935 nm, and a = 2 nm.

Thus, the surface area of the rod = π(r2 + a2) + 2πrh = 1352.6 nm2 and the volume =( π/3) a (3r2 + a2) +

πr2h = 3059.6 nm3. Mass of Au per NR = 3059.6 nm3 × 59 (atoms/nm3) × 197 = 3.56·107 Da. Mass

MTAB = 0.053 × (mass Au per NR + mass MTAB). Mass MTAB = 1.99·106 Da. Number of MTAB

per NR = 1.99·106 Da / 397 (Da/molecule) = 5,013 MTAB molecules per NR. Molecular footprint =

1352.6 nm2 / 5,013 molecules = 0.270 nm2 per molecule of MTAB or 3.7 MTAB molecules per nm2.

5

Figure S3. Cytotoxicity plot based on MTT analysis, which shows the percent survival of MCF-7 cells when

treated with different concentrations of CTAB-NRs (blue) and MTAB-NRs (red).

Calculation of amount of NR uptaken per cell. 500,000 cells were treated with 2 mL of medium

containing 80 µg/mL MTAB-NR in medium. From ICP-OES, the total amount of up-taken gold was

determined to be 40% of 160 µg or 64 µg Au / 500,000 cells = 1.28·10-4 (g Au/cell). Given the

density of gold and the previously calculated volume per NR of 3060 nm3 × 1.93·10-14 (g Au/nm3) =

5.91·10-11 (g Au/NR). So, 1.28·10-4 (g Au/cell) / 5.91·10-11 (g Au/NR) = 2.17·106 NR per cell.

TEM imaging of MTAB-NR uptake into cells. MCF-7 cells were plated into a 6-well plate at

500,000 cells per well and treated with 2 mL of a 20 µg/mL solution of NRs per well in medium for 24

6

hours. Cells were washed several times with PBS and fixed with a solution containing 3%

glutaraldehyde plus 2% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.3, for 1 hour. After

fixation, the samples were washed and treated with 0.1% Millipore-filtered cacodylate buffered tannic

acid, postfixed with 1% buffered osmium tetroxide for 30 min, and stained en bloc with 1% Millipore-

filtered uranyl acetate. The samples were dehydrated in increasing concentrations of ethanol,

infiltrated, and embedded in LX-112 medium. The samples were polymerized in a 70 oC oven for 2

days. Ultrathin sections were cut in a Leica Ultracut, stained with uranyl acetate and lead citrate in a

Leica EM Stainer, and examined in a JEM 1010 transmission electron microscope (JEOL) at an

accelerating voltage of 80 kV. Digital images were obtained using AMT Imaging System (Advanced

Microscopy Techniques Corp).

Cytotoxicity and cell uptake. MCF-7 cells were cultured in Eagles Minimum Essential Medium with

non-essential amino acids, 2 mM L-glutamine, 1 mM sodium pyruvate, and fetal bovine serum to a

final concentration of 10 %. MCF-7 cells were plated into a 96-well plate at 15,000 cells per well and

6 wells per each concentration. Cells were treated with an appropriate amount MTAB-NRs or CTAB-

NRs in medium for 24 hours, followed by the addition of 25 µL of a 5 mg/mL MTT solution in PBS.

After incubating for 4 hours, the medium was removed, the cells were lysed with DMSO, and the

absorbance measured on a plate reader at 570 nm. For cell uptake imaging experiments, MCF-7 cells

were plated into a 6-well plate at 500,000 cells per well and treated with 2 ml of a 20 µg/mL solution

of NRs in medium for 24 hours, trypsinized, and then plated onto a culture slide. For quantification of

NR uptake by ICP-OES, cells were treated with 2 mL of an 80 µg/mL solution of NRs in medium for

24 hours, washed several times with PBS, trypsinized, pelleted, dissolved with aqua regia, and the gold

concentration measured by ICP. For SEM imaging cells were first fixed by using 2.5 % glutaraldehyde

in PBS for an hour followed by 0.4 % osmium tetroxide solution. Next, cells were dehydrated by

7

consecutive treatment with 25 %, 50 %, 70 %, 95 %, and 100 % ethanol followed by 50 % and 100%

hexamethyldisilazane and allowed to dry overnight.

Figure S4. TEM image of MTAB-functionalized gold nanorods.