Bridging the electrochemical and biological worldswith hybrid nanocomposites

David A. Glesne

EnergyandEnvironmentalScienceandTechnologyDivision,ArgonneNationalLaboratory,9700SCassAvenue,Argonne, IL60439,USA

Recent discoveries arising from a combination of the

biological, physical, chemical and materials sciences

have resulted in the invention of numerous hybrid mol-

ecules that possess strengths inherent to each individ-

ual discipline. Nanocomposites that link biological

molecules to inorganic moieties have led to a family of

new reagents with unique capabilities for cellular ima-

ging and macromolecule detection. A recent report has

extended the applications of these hybrid molecules

from their use as detection and scaffolding reagents

into the realm of a biologically functional molecule.

The desire to combine the innate specificity inherent tobiological systems with the energetic capabilities of thechemical and physical domain has long stimulated inter-disciplinary research between these communities. Forexample, potent toxins, such as ricin, have been covalentlylinked to antibodies in the hope of utilizing the specificityof the antibody–ligand interaction to target the toxin to adistinct cellular population [1]. Also, in photodynamictherapy, radiosensitive agents are selectively taken up bytumors allowing a targeted dose of radiation to affect asubpopulation of cells [2]. Although previously designedhybrid molecules have shown some degree of biologicaluse, they all have certain limitations. For example,systemic breakdown of toxin-laden antibodies releasessome free poison, which leads to non-targeted cellularexposure.

An ideal molecule would combine the biological speci-ficity of nucleic acids or proteins with a functional groupthat was inert until activation by either physical orchemical means, on which a process could be selectivelycatalyzed. Recently published research by Paunesku andcolleagues [3] might have lead to the identification of justsuch a multifunctional molecule.

Biological applications of hybrid nanocomposites

Nanoparticles covalently linked to nucleic acids have had abrief but exciting history as detection reagents andscaffolding matrices. For example, gold nanoparticlestagged to nucleic acids [4] imbue array detection technol-ogies with sensitivity orders of a degree higher thantraditional flourescent detection [5]. Cell labeling appli-cations are currently being developed using cadmium andselenium nanocomposites linked to biologicals in whichthe unique properties of the attached semiconductor, orquantum dot, can be exploited for fluorescently excited

visualization purposes [6,7]. However, Paunesku andcolleagues are among the first to attempt to use semi-conductors to perform a biologically catalytic reaction.They chose titanium dioxide (TiO2) nanoparticles as anactivatable molecule; TiO2 possesses several attributesthat could make it a molecule of choice in biologicalsystems. First, titanium is relatively biologically inert andtolerated well in vivo. Second, TiO2 acts as a semiconduc-tor and, particularly when present as a nanoparticle, canact as a miniaturized electrochemical cell. When this metaloxide semiconductor is linked to an organic modifier andexposed to incident energy greater than the band-gap,semiconduction can occur through both the TiO2 and theattached modifier [8]. Paunesku et al. reasoned thatattachment of such a nanocomposite to a nucleic acidmight lead to a multifunctional molecule possessing thespecificity inherent to Watson–Crick base pairing coupledto controlled energy release from a visible light-reactivephotocatalyst. But would such a modified nucleic acidparticipate in standard nucleic acid hybridization andenzymatic reactions? Preliminary evidence suggests thatthe answer is affirmative. Nanocomposite oligonucleotideshybridized to their cognate antisense targets display atime-dependent cleavage of the nucleic acid moietyfollowing photoactivation with visible light. In a secondin vitro approach, the investigators determined thatenzymatic reactions are indeed possible because nano-composite oligonucleotides served as successful primersfor polymerase chain reaction (PCR) amplification, follow-ing which the resultant products can be cleaved byphotoactivation.

The long-term goal of these experiments is to use thesenanocomposites in an in vivo context, but how can they bedetected in a cellular setting? The investigators relied onthe nascent technology of hard X-ray flourescence micro-probe analysis [9]. When elements are excited by X-rays ofhigher energy than their K shell electron binding energythey radiate fluorescence energy specific to their outershell electronic state. The energy of this Ka and Kb

fluorescence is unique to each element. By focusingmonochromatic X-rays from an undulator source usingFresnel zone plate optics to a sub-micron spot and rasterscanning a cell, spectra from each point can be collectedusing an energy-dispersive germanium detector, allowingelemental distributions and concentrations to be mappedwithin a single cell. With an incident beam energy of,10 keV, such an approach enables detection of allelements with atomic numbers between 15 and 30Corresponding author: David A. Glesne (glesne@anl.gov).

Update TRENDS in Biotechnology Vol.21 No.9 September 2003 379

http://tibtec.trends.com

(phosphorous to zinc). The detection of titanium (atomicnumber 22) in an energy detection area free fromfluorescent signals from other biologically inherentmetals, is ideally suited to such an approach. The authorsused standard DNA transfection techniques to introducethe nanocomposite nucleic acids into rat PC12 cells [3].Examination of transfected cells via the aforementionedscanning X-ray microprobe analysis demonstrated thatbetween 20% and 50% of cells successfully incorporatedthe nanocomposites. More importantly, these compositeswere predominantly localized to the nucleus of these cells.The nucleic acid used within the composite was a sequencecorresponding to 18S rRNA genomic DNA, a segment ofthe genome that resides within the nucleolus in thenucleus. Interestingly, the titanium signal derived fromthe scans of transfected nuclei localized to a sub-structuresuggestive of the nucleolus, implying that specific base-pair targeting might be possible using these modifiedoligonucleotides.

Remaining questions and future prospects

Many issues need to be resolved before any intracellularcatalytic role for these hybrid nanocomposites can becontemplated. What type of break (single stranded ordouble stranded? same DNA strand or opposite DNAstrand?) is induced by transfer of the electrochemicalcharge? Is there any sequence context specificity of theinduced break? Can the nature of the induced break bemodified by either the size of the nanocomposite or thenature of the organic linker? In this study, the investi-gators document only a single strand break using,45-angstrom diameter semiconductors ligated to thenucleic acids through a dopamine linkage [3]. In vitro,DNA is capable of electron transfer reactions with a strongpreference for localizing and accumulating charge densityat GG dinucleotides [10,11], leading to preferentialoxidation at these sites. Whether similar hole-transfereffects and base context-specific accumulation occur in anin vivo environment was not investigated. How specificallyare the oligonucleotides interacting with their genomictarget? Using unmodified DNA bases in their syntheticoligonucleotides, Paunesku et al. present evidence thatsuggests some level of targeting towards the nucleoluswith rRNA sequences but it is unclear whether any actualbase-pairing has occurred [3]. Enhanced binding to acognate target could be achieved either by using oligonu-cleotides capable of triplex formation [12] or by modifyingthe bases to incorporate a peptide nucleic acid (PNA)backbone more capable of strand displacement than astandard unmodified oligonucleotide [13].

Lessons learned and strategies employed in the use oftriplex forming oligonucleotides (TFOs) might be informa-tive as to potential future design and uses of titaniumdioxide nanoparticles. For example, TFOs have beenconjugated to the DNA interchalator psoralen withsuccessful mutation of a target sequence following UV-irradiation [14]. TiO2 nanocomposites might be even moreeffective in such gene manipulation approaches becausenot only can photocatalysis using non-mutagenic visiblelight irradiation be used but also there are fewer toxicityconcerns regarding intracellular breakdown of the

composite because free TiO2 is relatively inert comparedwith psoralen.

If both intracellular specificity and effective cleavagecan be demonstrated, a wide range of promising appli-cations awaits these molecules. For example, current geneknockout or knockdown strategies allow for little temporalcontrol of the inactivation process: all cells that take up thereagent are subject to the inactivation. Because theactivation of TiO2 nanocomposites can be controlled byexposure of specific cells to the photoactivator, however,cell fate studies and temporally controlled inactivationstrategies could now be used. This capacity could bepowerfully exploited in a system such as the nematodeCaenorhabditis elegans in which precise cell fate of all cellsis known. Not only could inactivation of a targeted geneproduct within a specific cell be accomplished but also,more importantly, timing the exposure to the photoacti-vator could control the timing of this inactivation. And ofcourse, long term, there could be the potential for specificmutation repair in cell populations using the appropriateoligonucleotide in a gene therapy approach. In addition, itremains to be seen what other biological materials TiO2

nanocomposites can be attached to, perhaps enzymes orantibodies, and how such novel hybrids might be used.Although mostly speculative at this point, this ground-breaking study not only holds great promise for futurebiomedical intervention but also demonstrates that theconvergence of physical, chemical, and biologicalapproaches can result in the construction of powerfulmultifunctional tools.

References

1 Reiter, Y. and Pastan, I. (1998) Recombinant Fv immunotoxins and Fvfragments as novel agents for cancer therapy and diagnosis. TrendsBiotechnol. 16, 513–520

2 Ackroyd, R. et al. (2001) The history of photodetection and photo-dynamic therapy. Photochem. Photobiol. 74, 656–669

3 Paunesku, T. et al. (2003) Biology of TiO2-oligonucleotide nanocompo-sites. Nature Mater. 2, 343–346

4 Storhoff, J.J. et al. (1998) One-pot colorimetric differentiation ofpolynucleotides with single base imperfections using gold nanoparticleprobes. J. Am. Chem. Sci. 120, 1959–1964

5 Taton, T.A. et al. (2000) Scanometric DNA array detection withnanoparticle probes. Science 289, 1757–1760

6 Bruchez, M. et al. (1998) Semiconductor nanocrystals as fluorescentbiological labels. Science 281, 2013–2016

7 Chan, W.C.W. and Nie, S. (1998) Quantum dot bioconjugates forultrasensitive noniotopic detection. Science 281, 2016–2018

8 Rajh, T. et al. (1999) Improving optical and charge separationproperties of nanocrystalline TiO2 by surface modification withvitamin C. J. Phys. Chem. B 103, 3515–3519

9 Lai, B. et al. (2002) Report on the workshop of biological applications ofX-ray microbeams. Intl. J. Rad. Bio. 78, 749–752

10 Meggers, E. et al. (1998) Sequence dependent long range holetransport in DNA. J. Am. Chem. Sci. 120, 12950–12955

11 Lewis, F.D. et al. (2000) Direct measurement of hole transportdynamics in DNA. Nature 406, 51–53

12 Vasquez, K.M. and Wilson, J.H. (1998) Triplex-directed modification ofgenes and gene activity. Trends Biochem. Sci. 23, 4–9

13 Faruqi, A.F. et al. (1998) Peptide nucleic acid-targeted mutagenesis ofa chromosomal gene in mouse cells. Proc. Natl. Acad. Sci. U. S. A. 95,1398–1403

14 Majumdar, A. et al. (1998) Targeted gene knockout mediated by triplehelix forming oligonucleotides. Nat. Genet. 20, 212–214

0167-7799/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.doi:10.1016/S0167-7799(03)00192-6

Update TRENDS in Biotechnology Vol.21 No.9 September 2003380

http://tibtec.trends.com