Hermanson G. Bioconjugate Techniques, Second Edition

Подождите немного. Документ загружается.

640 15. Buckyballs, Fullerenes, and Carbon Nanotubes

ratio is so high for nanotubes, bundles often are observed to be knotted masses, which are very

difﬁ cult to unravel. The best that can be done is to disrupt the bundle and form a dispersion of

the nanotubes in a solvent medium.

Some of the better solvents for pure SWNTs are the amide-containing ones, like DMF or

N-methylpyrrolidone, but they still do not permit full dissolution, just dispersion (Boul et al .,

1999; Liu et al., 1999). The addition of surfactants to carbon nanotube suspensions can aid in

their solubilization, and even permit their complete dispersion in aqueous solution. The hydro-

phobic tails of surfactant molecules adsorb onto the surface of the carbon nanotube, while the

hydrophilic parts permit interaction with the surrounding polar solvent medium.

2.2. Nanotube Functionalization

An important route to solubilization of carbon nanotubes is to functionalize their surface

to form groups that are more soluble in the desired solvent environment. It has been shown

that acid treatment of nanotube bundles, particularly with HCl or HNO

at elevated temper-

atures, opens up the aggregate structure, reduces nanotube length, and facilitates dispersion

(An et al., 2004; Kordás et al., 2006). Nitric acid treatment oxidizes the nanotubes at the

defect sites of the outer graphene sheet, especially at the open ends (Hirsch, 2002; Álvaro et al .,

2004), and creates carbonyl, carboxyl, and hydroxyl groups, which aid in their solubility in

polar solvents.

Such carbonyls may be further oxidized using potassium permanganate (KMnO

) and per-

chloric acid (HClO

) to convert all of these groups into carboxylic acids. Once functionalized

in this manner, the nanotubes can be fully dispersed in aqueous systems. Kordás et al. (2006)

used these derivatives to print nanotube patterns on paper or polymer surfaces to create con-

ductive patterns for potential use in electronic circuitry. The carboxylates also may be used as

conjugation sites to link other ligands or proteins to the nanotube surface using a carbodiimide

reaction as previously discussed (Section 1, this chapter; Chapter 2, Section 1.11; Chapter 3,

Section 1).

Untreated carbon nanotubes nonspeciﬁ cally adsorb protein in an irreversible manner

much like the noncovalent adsorption of protein onto hydrophobic surfaces (Chen et al .,

2003). It is essential, therefore, to modify the surfaces of nanotubes to prevent hydropho-

bic binding of biomolecules, especially if the resultant conjugates are to be used in biological

assays or systems. Two main strategies have been used to make nanotubes biocompatible: (1)

the noncovalent modiﬁ cation of the graphene surface with amphipathic molecules, which have

functional groups that can be used for conjugation or (2) the covalent modiﬁ cation of the outer

nanotube surface to promote hydrophilicity and create functional groups for coupling other

molecules.

2.3. Detergent or Lipid Modiﬁ cation of Carbon Nanotubes

Detergents have been used for simple solubilization of SWNTs in aqueous solution. Ionic deter-

gents such as SDS will coat the nanotube surface and expose the negatively charged sulfonate

groups to the surrounding aqueous environment, thus allowing SWNT dispersion in aqueous

environments. Similarly, nonionic detergents such as Triton X-100 will coat the tubes and

present their hydrophilic groups to the aqueous phase. Fischer et al. found that coating SWNTs

with sodium dodecylbenzene sulfonate resulted in the best solubilization properties with long-

term stability of the nanotubes in aqueous buffers (Zhou et al., 2004). The detergent molecules

don’t merely lie in a random pattern on the nanotube surface; they coat the SWNT in a series

of half-micelle structures, which create small knobs along its length ( Figure 15.12 ).

Detergents also have been exploited in the noncovalent modiﬁ cation of carbon nanotubes

by using modiﬁ ed detergents containing a coupled afﬁ nity ligand, which is linked to the

hydrophilic part of the detergent molecule. Detergents that contain both a hydrophobic por-

tion and a hydrophilic part with at least one terminal functional group for conjugation can be

used in this process. The hydrophobic tail of many detergents will strongly adsorb to the nano-

tube’s outer graphene cylinder leaving the hydrophilic portions pointing outward and available

for conjugation, if they contain an appropriate functional group. The result is a hydrophilic

surface that is completely masked to prevent nonspeciﬁ c protein binding. This approach to

nanotube functionalization also leaves the chemical structure of the graphene cylinder unaf-

fected, thus avoiding defects that could alter its electronic properties.

As an example of this strategy, Chen et al. (2003) used an activated Tween 20 detergent to

coat SWNTs for subsequent conjugation to biotin, protein A, and U1A antigen. Tween 20 is

a poly-oxyethylene sorbitan monolaurate compound with 20 ethylene oxide units, 1 sorbitol

unit, and 1 lauric acid group esteriﬁ ed as the hydrophobic tail. The fatty acid group avidly

adsorbs to the carbon nanotube surface, leaving the three PEG arms, each containing terminal

2. Carbon Nanotubes 641

Figure 15.12 Detergent molecules can be used to solubilize carbon nanotubes by adsorption onto the surface

through hydrophobic interactions and create half-micelle structures with the hydrophilic head groups facing

outward into the aqueous environment.

642 15. Buckyballs, Fullerenes, and Carbon Nanotubes

Figure 15.13 Tween 20 can be activated with CDI using its hydroxyl groups to create an amine-reactive imida-

zole carbamate intermediate that then can be used to coat a carbon nanotube. The result is an activated nano-

tube that can be used to couple proteins and other amine-containing molecules.

hydroxyl groups, sticking out from the coating to create an extremely biocompatible construct

(Figure 15.13 ). The hydroxyl groups on the Tween molecules can be activated for coupling to

biomolecules using many of the methods discussed for hydroxylic particles in Chapter 14. For

instance, activation of Tween 20 with carbonyldiimidazole (CDI) or disuccinimidyl carbonate

(DSC) in a nonaqueous solution provides a reactive derivative suitable for coupling to amine-

containing molecules, such as proteins.

The following protocol for the activation of Tween 20 with CDI and its subsequent use

in modifying a carbon nanotube and coupling an afﬁ nity ligand is based on the method of

Chen et al . (2003).

Protocol

Activation of Tween 20

1. Dissolve 5 mg of Tween 20 in 25 ml of dry DMSO with stirring.

2. Add 4 g of CDI with mixing and react for 1 hour at room temperature.

3. Precipitate the CDI-activated Tween 20 by the addition of ethyl ether, and then isolate

the precipitate using centrifugation or ﬁ ltration. Redissolve the precipitated product in

DMSO and repeat the precipitation process two more times to insure removal of excess

reactants and reaction by-products. After the ﬁ nal precipitation, dry the isolated precipi-

tate overnight in vacuo to remove remaining solvent.

Coating Nanotubes with CDI-Activated Tween 20

4. Suspend the SWNTs in 1 percent (w/w) CDI-activated Tween 20 solution in water using

sonication and allow the detergent molecules to bind for 30 minutes at room temperature.

5. Quickly remove excess detergent by ﬁ ltration on a 0.2 m ﬁ lter and washing the modi-

ﬁ ed nanotubes with water.

Coupling of Activated Tween 20 to an Amine-Containing Ligand

6. Dissolve an amine-containing ligand in 0.1 M sodium carbonate, pH 9.5. If coupling a

small ligand, such as a biotin-PEG-amine compound (Chapter 18), then use a concentra-

tion of about 5–10 mM in the carbonate buffer. For proteins, concentrations of 10 nM

to 1  M can be used with success.

7. Resuspend the activated nanotubes in the ligand-containing carbonate buffer and react

with mixing overnight at room temperature or 4 ° C (e.g., for sensitive proteins).

8. Wash the coupled nanotubes with water or a suitable buffer using ﬁ ltration and ﬁ nally

store them in buffer containing a preservative at 4 ° C.

In a similar approach to the noncovalent modiﬁ cation of carbon nanotubes with detergent

molecules, Kam et al. (2005) used phospholipid derivatives to coat SWNTs for photo-

therapeutic agents against tumor cells in vivo. A phospholipid containing a PEG-NH

group

was used to couple folic acid as an afﬁ nity ligand (using EDC to form an amide bond), which

preferentially can be taken up by cancer cells. SWNTs were modiﬁ ed by the lipid derivative in

aqueous solution using sonication for 1 hour and centrifuged to remove insoluble material. The

aqueous fraction contained modiﬁ ed nanotubes, which contained the surface adsorbed lipid

derivative. After the modiﬁ ed SWNTs were incubated with cells, the solution was irradiated

2. Carbon Nanotubes 643

644 15. Buckyballs, Fullerenes, and Carbon Nanotubes

using an 808 nm laser. The nanotubes absorb light in this region of the spectrum and heat up

to the point of causing cell death.

2.4. Pyrene Modiﬁ cation of Carbon Nanotubes

Another method for the noncovalent modiﬁ cation of carbon nanotubes involves the interaction

of pyrene derivatives with the graphene sidewalls, presumably due to -stacking (Nakashima

et al., 2002) ( Figure 15.14 ). This interaction forms tight complexes that completely coat the

nanotube surface, and if the pyrene contains a hydrophilic portion, it can impart water solubil-

ity to the SWNT, as well. The additional presence of reactive groups or functional groups on a

pyrene side chain permits conjugation to afﬁ nity ligands for biological applications.

Nakashima et al. (2002) found that a pyrene derivative containing a single side chain termi-

nating in a positively charged quaternary amine imparted water solubility to SWNTs treated

with the compound. The application of other pyrene derivatives can be done to contribute

both water solubility and an appropriate functionality for bioconjugation. Examples of pyrene

derivatives that are suitable for the noncovalent modiﬁ cation of carbon nanotubes include

those that have a single modiﬁ cation to the pyrene ring structure. Derivatives that contain mul-

tiple modiﬁ cations off the pyrene group may affect its interaction potential for the SWNTs

and should be avoided. For instance, water-soluble poly-sulfonated derivatives of pyrene, such

Figure 15.14 The NHS ester of a pyrene butyric acid derivative can be used to modify a carbon nanotube by

adsorption of its rings onto the surface of the tube. The NHS ester groups then can be used to couple amine-

containing molecules to form amide bonds.

as the Cascade Blue ﬂ uorescent dyes described in Chapter 9, Section 5, should not be used,

because the pyrene structure is too hydrophilic to associate with the nanotube surface due to

the three negative charges contributed by the sulfonic acids.

Some commercially available pyrene compounds that may be used to functionalize a car-

bon nanotube by this method include 1-pyrenebutyric acid and 1-aminopyrene (from Acros or

Aldrich) as well as N-(1-pyrenyl)maleimide, 2-(1-pyrenyl)ethyl chloroformate, 1-pyrenebutyric

acid N-hydroxysuccinimide ester, 1-pyrenecarboxaldehyde, and 1-pyreneacetic acid (from

Aldrich). Each of these compounds provides a single site of derivatization off the basic pyrene

rings to contain either a functional group or reactive group for coupling ligands. Molecules

modiﬁ ed with these pyrene derivatives may be used to treat a carbon nanotube to form a stable

noncovalent complex.

The pyrene derivatives containing a carboxylate group, chloroformate, aldehyde, or an NHS

ester can be used to couple to amine-containing ligands, including proteins. The maleimide-

pyrene derivative may be used to couple with thiol-containing ligands, while the amine–pyrene

compound may be used to conjugate with carboxylate-containing ligands. Also available are

(1-pyrenyl)butyric acid hydrazide and pyrene-1-isothiocyanate from Molecular BioSciences,

which react with aldehydes and amines, respectively. Once a ligand is conjugated to a pyrene

derivative, the complex may be incubated with a carbon nanotube to produce the ﬁ nal non-

covalent complex. Since the initial modiﬁ cation is done on a water-insoluble nanotube, it is

best to do the primary coating of the pyrene derivative in an organic solvent, such as DMF. It

then is desirable to make the nanotube water-soluble by linking a hydrophilic spacer arm to

the pyrene–nanotube complex. If a hydrophilic spacer is built into the resultant pyrene conju-

gate, such as the use of a short-chain PEG compound, then the resultant SWNT complex will

be completely water-soluble (for example, see Figure 15.15 ). The PEG spacer chosen for this

purpose should contain a terminal functional group for coupling to another molecule. At this

point, the water-soluble complex can be reacted with a protein or other afﬁ nity ligand in aque-

ous buffer to make the desired bioconjugate. This multi-step process will result in a biocom-

patible carbon nanotube that retains its electronic properties, is water-soluble, and has added

ﬂ uorescent properties due to the pyrene molecules coating its surface.

Maehashi et al. (2007) used pyrene adsorption to make carbon nanotubes labeled with

DNA aptamers and incorporated them into a ﬁ eld effect transistor constructed to produce a

label-free biosensor. The biosensor could measure the concentration of IgE in samples down

to 250 pM, as the antibody molecules bound to the aptamers on the nanotubes. Felekis and

Tagmatarchis (2005) used a positively charged pyrene compound to prepare water-soluble

SWNTs and then electrostatically adsorb porphyrin rings to study electron transfer interac-

tions. Pyrene derivatives also have been used successfully to add a chromophore to carbon

nanotubes using covalent coupling to an oxidized SWNT (Álvaro et al., 2004). In this case, the

pyrene ring structure was not used to adsorb directly to the nanotube surface, but a side-chain

functional group was used to link it covalently to modiﬁ ed SWNTs.

2.5. Modiﬁ cation of Carbon Nanotubes by Cycloaddition

The covalent methods previously discussed for fullerene modiﬁ cation using cycloaddition

reactions also can be applied to carbon nanotubes. This strategy results in chemically link-

ing molecules to the graphene rings on the outer surface of the cylinder, resulting in stable

2. Carbon Nanotubes 645

646 15. Buckyballs, Fullerenes, and Carbon Nanotubes

conjugates that can be designed to include hydrophilic groups for water solubilization.

Georgakilas et al. (2002) describe the use of a 1,3-dipolar cycloaddition process to carbon nan-

otubes with azomethine ylides, generated by condensation of an amino acid derivative and an

aldehyde. The reaction occurs in organic solvent at high temperature over a time period of sev-

eral days.

Typically, SWNTs are suspended in DMF using sonication and the aldehyde and glycine

derivatives are added to the mixture with stirring. The -amine derivative of glycine can include

hydrophilic spacers to make the resultant nanotube water-soluble as well as include protected

functional groups to couple afﬁ nity ligands after deprotection (Kurz et al., 1998). The aldehyde

also can include R groups that add water solubility or functionality to the nanotube. A combi-

nation of a glycine derivative with an aldehyde derivative can result in both hydrophilicity and

a functional group to conjugate ligands ( Figure 15.16 ).

Figure 15.15 An aldehyde derivative of pyrene can be used to couple a hydrophilic amino-PEG-carboxylate

spacer by reductive amination. The resultant derivative then can be used to coat a carbon nanotube through

pyrene ring adsorption and result in a water-soluble derivative containing terminal carboxylates for coupling

amine-containing ligands.

Felekis and Tagmatarchis (2005) used this cycloaddition process to prepare SWNT deriva-

tives possessing photoactive components, such as the addition of ferrocene groups. They used a

short PEG-type spacer on the glycine to impart water solubility at the same time.

Singh et al. (2006) also used cycloaddition to prepare carbon nanotubes containing indium

labeled diethylenetriamine pentaacetic acid (DTPA) derivatives ( Figure 15.17 ). In the initial

modiﬁ cation, a SWNT was derivatized to contain a primary amine at the end of a short PEG

spacer. The resultant water-soluble nanotube then was reacted with DTPA to create a metal

chelating group at the end of the chain. Subsequent loading of the chelate with

111

In created a

radionuclide–SWNT complex for in vivo biodistribution studies.

Figure 15.16 Some modiﬁ cation methods that are useful for fullerenes also can be used with carbon nanotubes.

The reaction of an N-glycine compound with an aldehyde derivative can result in cycloaddition products, which

create pyrrolidine modiﬁ cations on the nanotube surface.

2. Carbon Nanotubes 647

648 15. Buckyballs, Fullerenes, and Carbon Nanotubes

Figure 15.17 An amino-PEG-pyrrolidine derivative of carbon nanotubes can be used to couple metal chelating

groups, such as DTPA. Subsequent coordination of

111

In results in an indium chelate that can be used for imag-

ing applications.

The demonstration that the 1,3-dipolar cycloaddition process with azomethine ylides works

with nanotubes implies that similar reactions developed for use with fullerenes also may

be successful with carbon nanotubes. In particular, the cyclopropanation reactions discussed

previously for the modiﬁ cation of C

, likely will work for derivatization of SWNTs and

MWNTs (Zakharian et al ., 2005).

649

Mass spectrometry has become one of the most important tools for analyzing proteins in com-

plex biological samples. The ability to separate proteins and peptides in high resolution has made

possible the simultaneous identiﬁ cation of hundreds of proteins within samples (for reviews, see

Gingras et al., 2005; Hamdan and Righetti, 2005; Siuzdak, 2006). Proteins can be analyzed for

their presence or compared between samples for their relative expression level. One cell popula-

tion treated with a drug candidate, for instance, can be compared by mass spec to another sample

as control to assess the affect of the drug on expression levels of certain proteins.

There are several ways that proteins can be analyzed using mass spec. Whole proteins can

be separated using electrospray ionization (ESI) technique or by using matrix-assisted laser

desorption/ionization (MALDI). Both of these methods inject intact proteins into the mass

spectrometer, ionize them, and separate the resultant charged components by their individual

mass/charge ratios. These methods work well for small and medium sized proteins in sam-

ples of low complexity, but analysis of larger proteins or highly complex samples is difﬁ cult.

Alternatively, proteins ﬁ rst can be proteolytically digested using an enzyme such as trypsin and

then the peptides analyzed by mass spec. This proteolysis method is more universally appli-

cable, because analysis of peptide fragments allows mass spec separation to be done for all

proteins regardless of their original intact mass prior to digestion. The peptides typically are

subjected next to a higher energy secondary mass spec separation that fragments them into

their component amino acids, which then can be identiﬁ ed by their masses. Samples then are

analyzed by correlation of the peptide sequences to online databases of mass spec information,

which can identify the protein that each peptide came from.

However, the interpretation of mass spec data on whole samples can be daunting, especially

when analyzing proteolytically digested samples, which results in many times more species to

analyze per protein than intact proteins. In order to reduce the complexity of sample analy-

sis, a number of techniques have been developed to fractionate the proteome prior to mass

spec separation. For instance, two-dimensional electrophoresis can separate proteins both by

charge and by molecular weight and allow picking of only certain spots for subsequent mass

spec analysis. However, two-dimensional electrophoresis is severely limited in its sensitivity for

picking up low or medium copy proteins (Gygi et al., 2000). Alternatively, afﬁ nity separations

on resins or surfaces can be done to capture only those proteins having certain epitopes or

chemical characteristics, such as post-translational modiﬁ cations. In addition, nanoliter HPLC

Mass Tags and Isotope Tags