Labeling with Gold Nanoparticles:
Improving Labeling Reactions:
Tips for the Best Labeling of Your Molecule
How well a particular molecule labels is affected by several factors. The sections below discuss the effects of properties of the Nanogold® and undecagold labels themselves; however, the properties of thebiomolecule and the labeling site can also affect labeling. It is a good idea to find out as much as possible about the properties of the molecule before doing the labeling reaction. In particular, solubility properties are important: a substrate with low solubility tends to form gold conjugates with low solubility.
In this section:
The reagent should be completely dissolved; however, it is sometimes difficult to redissolve lyophilized labeling reagents in water. The best method is to dissolve it in a small amount (10 % or less of the final volume) of dimethylsulfoxide (DMSO) or isopropanol; however, dimethlformamide (DMF) should be avoided since it can precipitate and degrade the gold reagents. If your biomolecule is soluble only in DMF, you should consider using DMA (N,N-dimethylacetamide) instead: this has very similar solvating properties, but it also effectively solubilizes the gold nanoparticles without the problems associated with DMF.
Nanogold® and undecagold labeling reagents are supplied lyophilized from solutions of the buffers in which we usually use them. Maleimido reagents are supplied lyophilized from 0.02 M sodium phosphate buffer with 150 mM sodium chloride and 1 mM EDTA, pH 6.5; Mono-sulfo-NHS- reagents are lyophilized from 0.02 m HEPES - sodium hydroxide, pH 7.5; and Monoamino- reagents are supplied pure. Nanogold® reagents are always lyophilized from a solution containing 30 nmol/mL; therefore, when you purchase a 30 nmol vial and resuspend the reagent in 1 mL of water, the buffer will be restored to this concentration. Undecagold reagents are always lyophilized from a 50 nmol/mL solution; dissolving a 50 nmol vial in 1 mL of water will restore the original buffer concentration.
Most other buffers may be used without problems; the important consideration is to avoid buffers which can react with the labeling reagents. Do not use Sulfo-NHS- gold reagents in TRIS, or maleimido reagents in the presence of thiols, since these will react and prevent labeling.
The smaller the reaction volume, the higher the labeling will be. This is particularly true with small amounts of gold nanoparticle labels. We find that for many molecules, labeling is higher if the reaction mixture is concentrated as soon as the gold nanoparticle reagent is added: use a membrane centrifuge concentrator, such as the Centricon, with a molecular weight cut-off below the smallest component of your system (for example, if you are Nanogold®-labeling a larger protein, Centricon-10 would be appropriate; for undecagold-labeling a 4 kDa peptide, Centricon-3 would be appropriate).
Labeling efficiencies can also be improved by gentle agitation of the mixture after addition of the nanoparticle reagent, although you should be careful not to generate air bubbles which can denature proteins or promote oxidation of thiol groups. Reaction is usually close to complete in one hour at room temperature: therefore, gentle agitation at room temperature for one hour followed by refrigeration overnight is usually recommended for the best labeling.
An important goal with gold nanoparticle labeling is isolating the labeled conjugate from unbound gold nanoparticles or from unlabeled biomolecules. Product separations are usually be gel filtration HPLC, since the gold nanoparticles are insufficiently dense for the pelleting and resuspension procedures used with colloidal gold conjugates. Since molecules with a large size or molecular weight difference are much more easily separated by this method, you should plan your reaction using an excess of the smaller component, whether this is the gold nanoparticle or the biomolecule, and aim to leave as little as possible of the larger component unreacted at the end. For example, if you are labeling a 200 kDa protein use a substantial excess (ten-fold or twelve-fold) of the gold nanoparticles: this is because the excess gold nanoparticles are easily removed, but separating unlabeled from labeled protein would be very difficult. Conversely, if you are labeling a short peptide or a drug, use a substantial excess of the peptide or drug, because it is much easier to remove the excess small molecule than to separate bound from unbound nanoparticles.
When the gold nanoparticle and the molecule to be labeled are more similar in size it is more difficult to predict which component is more easily separated: In these cases, a smaller excess (such as two-fold or four-fold) of the slightly smaller component should be used. The molecular weights of two of our most popular gold nanoparticles are:
- Nanogold®: 15,000
- Undecagold: 5,000
However, because of the highdensity of the gold core, their retention times are often slightly longer than these MW values might suggest.
Maleimido- reagents react at pH values of about 6 - 7.5. At higher pH maleimides begin to react with amines and be hydrolyzed by water; we usually conduct maleimide labeling at pH 6.5. A significant problem is oxidation of thiols. This is thought to be catalyzed by trace quantities of transition metals and atmospheric oxygen, and can be minimized by following these suggestions:
- EDTA is necessary if the thiol sites in your molecule are easily oxidised to disulfides (for example, hinge thiols in reduced IgG molecules). At least 1 mM of EDTA should be present in such solutions. Some of these oxidation reactions are catalyzed by trace amounts of transition metals inthe presence of atmospheric oxygen; these can also be minimized by
- Exclude as much air as possible: do reductions of IgG or F(ab')2 in sealed tubes, and pick a size which is only slightly larger than your reaction volume. Make sure all buffers are degassed immediately before use.
- Make sure agitation of the thiolated sample does not generate air bubbles.
Because the amino- groups in your molecule can be protonated (and rendered less reactive), pH, both of the bulk solution and locally around the specific functional group you wish to label, is important. The pKa of the N-terminal amine of proteins and peptides is often slightly lower than that of the side-chain amines; therefore, under the right pH conditions, it will be preferentially labeled. We supply Mono-Sulfo-NHS- gold nanoparticle reagents lyophilized from pH 7.5 buffer, which we find results in consistent labeling with a 1 : 1 ratio of gold nanoparticle : protein in the product. However, other workers have found that pH 8 or pH 8.2 give highest labeling; if you obtain low labeling at pH 7.5, try pH 8 or 8.2.
In some cases, the charge properties of the molecule hinder labeling with amine-reactive gold nanoparticles even when the pH of the surrounding solution is correct. For example, in our initial experiments, amino-derivatized oligonucleotides sometimes labeled poorly with some formulations of gold nanoparticles under conditions in which proteins were effectively labeled. This might be because the charges associated with the phosphate backbone keep the amine protonated to higher pH values; it is also possible that the ngative charge associated with the phosphate groups repels the negatively charged sulfo-NHS group. The formulation of Nanogold has since been modified for higher solubility and stability, and successful labeling of oligonucleotides with this reagent has been demonstrated. However, if you are labeling highly charged or ionized molecules, consider incorporating a thiol (for example, by enzymatically introducing a cysteine) instead of an amine and labeling this with Monomaleimido-gold.
- Braig, K., Simon, M., Furuya, F., Hainfeld, J.F., and Horwich, A.L. A polypeptide bound by the chaperonin groEL is localized within a central cavity. Proc. Natl. Acad.Sci., 90, 3978-3982 (1993).
- Gregori, L., Hainfeld, J. F., Simon, M. N., and Goldgaber, D. Binding of amyloid beta protein to the 20S proteasome. J. Biol. Chem., 272, 58-62 (1997).
- Hainfeld, J.F. Undecagold-antibody method. In Colloidal Gold: Principles Methods, and Applications., M.A. Hayat (Ed.), San Diego, Academic Press; Vol. 1, pp. 413-429 (1989).
- Hainfeld, J.F. Gold cluster-labelled antibodies. Nature, 333, 281-282 (1988).
- Ribrioux, S., Kleymann, G., Haase, W., Heitmann, K., Ostermeier, C., and Michel, H. Use of Nanogold- and Fluorescent-labeled Antibody Fv Fragments in Immunocytochemistry. J. Histochem. Cytochem., 44, 207-213 (1996).
- Segond von Banchet, G., and Heppelman, B.: Non-radioactive localization of substance P binding sites in rat brain and spinal cord using peptides labeled with 1.4 nm gold particles; J. Histochem. Cytochem., 43, 821 (1995).
- Yang, Y. S.; Datta, A.; Hainfeld, J. F.; Furuya, F. R.; Wall, J. S., and Frey, P. A.: Mapping the lipoyl groups of the pyruvate dehydrogenase complex by use of gold cluster labels and scanning transmission electron microscopy. Biochemistry, 16;33(32), 9428-9437 (1994).