N A N O P R O B E S     E - N E W S

Vol. 6, No. 12          December 13, 2005

Little is known regarding the applicability of EnzMet for immunophenotypic detection of protein using immunohistochemistry, but the superior resolution evidenced for in situ hybridization via discrete metallographic deposits offers the potential for enhancing high-resolution immunophenotyping. Using high-complexity tissue microarrays (TMAs), 88 common solid tumors were evaluated by enzyme metallography (EnzMet) using an automated slide staining system (Ventana Medical Systems); targets were chosen to assess the ability of EnzMet to specifically localize encoded antigens in the nucleus (estrogen receptor), cytoplasm (cytokeratins), and cytoplasmic membrane (HER2) in TMAs.

For the comparison of DAB IHC with EnzMet IH, target antigens with established cellular immunophenotypic distributions were used in order to evaluate each cellular compartment. Specifically, the cell membrane antigen, HER2 (Ventana), the cytoplasmic antigen cytokeratin (Roche Molecular), and the nuclear antigen estrogen receptor (ER; LabVision) were chosen for the comparison of enzyme metallography-based immunohistochemistry (IHC) with conventional HRP-DAB IHC. All procedural steps were conducted on the Ventana Benchmark automated immunostainer. For HER2 scoring, the FDA HercepTest scoring criteria were used (0, 1+, 2+, 3+). For ER immunostaining, the percentage of tumor cells positive was assessed using visual estimation. Cytokeratin staining was recorded as positive or negative if any tumor cells were positive for cytokeratin in the core. Results were compared with conventional IHC diaminobenzidine (DAB) immunostaining. A schematic of the process, and examples of ER staining with DAB and enzyme metallography, is shown below:

HER2 was positive in only the breast carcinomas in this series (3/15 cases). The intensity of staining for all three antigens evaluated was comparable for breast tumors as well as carcinomas of kidney, colon, and prostate. However, the quality of staining in EnzMet IHC preparations was much sharper, and the stain deposits were better defined, showing a more punctate appearance, than that found with DAB. Full concordance was found between the EnzMet and conventional IHC results. The EnzMet reaction product was dense and sharply defined, did not appreciably diffuse, and provided excellent high-resolution differentiation of cellular compartments in paraffin sections for the nuclear, cytoplasmic, and cell membrane-localized antigens evaluated. The higher density of elemental silver deposited during enzyme metallography permitted evaluation of core immunophenotypes at a relatively low magnification, without the need for oil immersion, allowing more tissue to be screened in an efficient manner. In addition, the signal is stable and provides a permanent record. Concordance data is shown below:

Data are given as cores positive/total cores.
*Epithelial tumor cells negative; occasional stromal cells positive by both DAB and EnzMet IHC.

Reference:

Tubbs R.; Pettay J.; Powell R.; Hicks D. G.; Roche P.; Powell W.; Grogan T., and Hainfeld, J. F.: High-resolution immunophenotyping of subcellular compartments in tissue microarrays by enzyme metallography. Appl. Immunohistochem. Mol. Morphol., 13, 371-375 (2005).

• Abstract (courtesy of the PubMed (Medline)).

More information:

• Enzyme Metallography for correlative labeling (our Microscopy & Microanalysis 2004 abstract)
• Enzyme Metallography - a New Staining Method (our Microscopy & Microanalysis 2002 abstract)
• Enzyme Metallography for in situ hybridization
• Product application special report: Nanogold in situ hybridization
• Other product and research applications

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Alternatives for Antibody Labeling

Although we recommend using Monomaleimido Nanogold® to label at a hinge thiol site, this is not the only option for antibody labeling. Recently, we have found that antibody IgG labeled using Mono-Sulfo-NHS-Nanogold gives labeling that is just as sensitive and specific. Mono-Sulfo-NHS-Nanogold reacts with primary amines: because every protein contains at least one primary amine, you can usually label directly with Mono-Sulfo-NHS-Nanogold without needing the prior preparation, such as reducing the hinge disulfides, that is often required for Monomaleimido Nanogold labeling.

Usually, the recommended way to make Nanogold-Fab' conjugates is to label at a hinge thiol using Monomaleimido Nanogold. To make Fab' from an IgG molecule, the IgG must be digested to remove the Fc portion while leaving the hinge disulfide bonds intact. You can use two enzymes, pepsin or ficin, and the procedures are given in the references below:

• Parham, P.: On the fragmentation of monoclonal IgG1, IgG2a, and IgG2b from BALB/c mice. J. Immunology, 131, 2895-2902 (1983).
  • Abstract (courtesy of the Journal of Immunology).
• Mariani, M.; Camagna, M.; Tarditi, L., and Seccamani, E.: A new enzymatic method to obtain high-yield F(ab)2 suitable for clinical use from mouse IgGl. Molecular Immunology, 28, 69- (1991).
  • Abstract (courtesy of the PubMed (Medline)).

However, digesting IgG to F(ab')₂ only works for certain classes of IgG. Pepsin will only digest IgG1 and IgG2a, while ficin produces F(ab')₂ only from IgG1. Even for these subtypes, digestion can be difficult, and frequently provides low yields of products. If you have an IgG2b, or you encounter problems with the digestion, an alternative is to use papain to prepare Fab fragments. Although these no longer contain the hinge disulfide and so cannot be labeled with Monomaleimido Nanogold, they do contain at least one primary amine (the N-terminal amine and any lysine residues), and therefore they can still be labeled using Mono-Sulfo-NHS-Nanogold. The options for labeling are shown below:

My labeling is low - what went wrong?

While the factors that can affect each labeling reaction are different, there are some general factors that you should bear in mind when planning or conducting a Nanogold labeling reaction. The following suggestions may help to improve your labeling efficiency:

• Check the stoichiometric ratio of Nanogold : molecule to be labeled - and make sure that this is expressed in terms of the number of molecules. For example, if you have a solution of a plasmid that contains 0.1 mg/mL, and the molecular weight of the plasmid is 10,000, then you have 10 nmol (nanomoles) per mL. Therefore, in order to ensure that you put at least one Nanogold per plasmid, you will need at least 10 nmol of Nanogold. If you are working with a smaller amount of Nanogold (for example, one 6 nmol vial), then you need to reduce the amount of molecule to be labeled to maintain a ratio of Nanogold : biomolecule that ensures successful labeling.

  You also need to consider which component you need in excess, an issue which has been explored previously in detail. Generally, you should use an excess of the component that is more easily separated from the conjugate. Since we usually recommend gel filtration as a separation method, this usually means that you use an excess of the smaller of the two reagents: if you are labeling a molecule that is larger than Nanogold, use an excess of Nanogold, and if you are labeling a molecule that is smaller than Nanogold, use an excess of that molecule. Nanogold is about 2.5 - 2.7 nm in diameter (including its protective layer of ligands); this is about the same size as an 8,000 - 10,000 MW protein, so you might expect to use excess Nanogold to label a biomolecule of 10,000 or higher molecular weight, and excess biomolecule if it has a molecular weight of 5,000 or less.

• Ensure that the Nanogold is well dissolved before adding to the molecule to be labeled. The best way to ensure solution is to dissolve in a small amount of dimethyl sulfoxide (DMSO) before adding water (10 - 20% of final volume). DMSO is a highly effective solvent for Nanogold, and will ensure that it is fully dissolved and able to react.

• Make sure that the pH of your reaction mixture is optimized for labeling. Monomaleimido Nanogold will react most selectively with thiols at pH 6.0 - 7.0; at lower pH values reaction is slower, while at higher pH values, maleimides begin t react with amines as well as thiols and the reaction may become less specific. Mono-Sulfo-NHS-Nanogold reacts best in the pH range 7.5 to 8.2; below pH 7.5 reaction is unreliable, and above pH 8.2, hydrolysis becomes rapid.

  Some fine-tuning is possible within this pH range which can help to favor labeling at one site over another. Usually, the N-terminal amine has a slightly higher pKa than that of a lysine residue, and therefore using a pH towards the lower end of the suggested range would favor labeling at the N-terminal amine over lysine residues. However, other factors, in particular accessibility of the different sites and steric hindrance, will also help determine which sites are labeled.

More information:

• Guide to Gold Cluster Labeling
• Technical help: Nanogold labeling reagents
• Nanogold: Plan your Labeling for Best Results
• Labeling Other Functional Groups: Hydroxyls
• Labeling Other Functional Groups: Aldehydes and Ketones
• Labeling Other Functional Groups: Carboxylic Acids

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

NTA-Ni(II)-Nanogold® Reveals Multienzyme Complex Structure

• The nitrilotriacetic acid - Ni(II) chelate is much smaller than an antibody or protein, so it brings the gold nearer to the target and labeling resolution is higher.

• Because it is so small, NTA-Ni(II)-Nanogold is better able to penetrate into specimens and access restricted sites within them. It also perturbs biological structures and processes less than a larger antibody probe.

• NTA-Ni(II)-Nanogold is made with a modified gold particle, with very high solubility and stability. At 1.8 nm in size, is readily visualized by electron microscopy.

• Binding constants for Ni(II)-NTA are very high due to the chelate effect of multiple histidine binding and multiple Ni(II)-NTA functionalization. Dissociation constants are estimated to be between 10^-7 to 10^-13 M^-1. For many applications, this provides binding strengths comparable to antibodies.

These properties make this probe ideal for labeling sites within macromolecular complexes with very high resolution, and several applications of this type have been described. Wolfe and colleagues provide another illustration in the current Journal of Biological Chemistry. Its small size also means that it is useful for identifying and localizing components of protein complexes or other macromolecular assemblies without disrupting the structure.

The process of protein synthesis is performed by many cellular polypeptides acting in concert within protein complexes, structures that contain multiple proteins with interrelated functions. In multicellular eukaryotes, one of these assemblies is a multienzyme complex composed of eight proteins that have aminoacyl-tRNA synthetase activities as well as three non-synthetase proteins (p43, p38, and p18) with diverse functions. The authors used electron microscopy and three-dimensional reconstruction to elucidate the arrangement of proteins and tRNA substrates within this core multisynthetase complex, and used both NTA-Ni(II)-Nanogold and Monoamino-Nanogold labeling to identify different components of the complex. A stable cell line has been produced that incorporates hexahistidine-labeled p43 into the multisynthetase complex.

For the tRNA binding experiments, the core multisynthetase complex was isolated from K562 cells; for samples used in localization of p43, a new isolation method using lysine-agarose chromatography was developed: approximately 1 g (wet weight) of control or transfected 293F cells was suspended in hypotonic lysis buffer containing 10mM Hepes, pH 7.2, 5mM beta-mercaptoethanol, 1 mM N_alpha-p-tosyl-L-arginine methyl ester hydrochloride, 1 mM phenylmethylsulfonyl fluoride, and protease inhibitor-EDTA mixture (Roche Applied Science). After gentle Dounce homogenization one-fifth volume of 5 x column buffer (250 mM Hepes, 25 mM Mg Acetate, 2.5 mM EDTA, 5mM dithiothreitol, 50% glycerol, pH 7.2) was added. The lysate was centrifuged (10 min at 13,000 x g) to remove insoluble cell debris, then separated using a 2-ml lysine-Sepharose 4B column pre-equilibrated with 1 x column buffer. The column was successively washed with three column volumes of 25 mM lysine in 1 x column buffer then one column volume each of 1 x column buffer alone, 1mg/ml yeast tRNA (Sigma) in 1 x column buffer, and 2 mg/ml tRNA in 1 x column buffer. Partially purified multisynthetase complex was then eluted with 10 mg/ml tRNA in 1 x column buffer. For specific isolation of multisynthetase complex containing p43-His6, lysate was loaded onto a 1-ml HIS-Select nickel affinity column (Sigma) that had been pre-equilibrated with 1 x column buffer. After washing with 3 column volumes of the same buffer, protein was eluted with a gradient of 0 to 200 mM imidazole in column buffer.

After evaluation of a variety of methods for tRNA labeling, it was found that modification of the 5'-end could be accomplished consistently and efficiently with highest retention of specific binding to the appropriate polypeptide within the aminoacyl-tRNA synthetase complex. In a typical reaction, 50 micrograms of tRNA in50 microliters of 0.1 M imidazole, pH 6, and 5 mg of N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride were mixed with 6 nmol of monoamino-Nanogold. After 2 h at room temperature on a rotating wheel, excess labeling reagent was removed by gel filtration, and the tRNA was collected by precipitation at -20°C with the addition of 0.1 volume of 3 M sodium acetate, pH 5, and 3 volumes of cold 100% ethanol. Extent of labeling was calculated using A_420nm for the Nanogold and A_260nm for tRNA; labeling efficiency was routinely 8090%. This conjugate was then used to covalently label multisynthetase complex, via 3'-end oxidation of the tRNAs and reaction with protein in the presence of mild reductant. Oxidized tRNAs were prepared by incubation in 3mM sodium periodate for 10 minutes at room temperature in the dark, then collected by ethanol precipitation and stored at -20°C in 10 mM sodium acetate, pH 5.5, 10 mM magnesium acetate. Reaction mixtures for tRNA linkage to multisynthetase complex contained 115 pmol of protein, 1525 molar excess of oxidized tRNA and 5 micromolar sodium cyanoborohydride, in 5mM Hepes, pH 7.2, with 5% glycerol. Reactions were performed for 20 minutes at 30°C. For electron microscopy, nonspecific binding was minimized by the addition of NaCl to a concentration of 0.1 M, and removal of excess tRNA by HPLC gel filtration (BioSep SPC5 column, Phenomenex). Derivatized complex was eluted isocratically with 25 mM Hepes, pH 7.2, 100 mM NaCl.

Labeling of the multisynthetase complex containing p43-His6 for electron microscopy was conducted by incubating samples from control and transfected cells for 15 minutes with 50 microliters of NTA-Ni(II)-Nanogold, filtered through a 0.45-micron polyvinylidene difluoride membrane (Millipore) and excess labeling reagent removed by HPLC size fractionation: samples of 300400 microliters were applied to a Biosep-Sec-S 4000 column (Phenomenex) equilibrated with 100 mM NaCl, 25 mM Hepes, pH 7.2. The same buffer was used for isocratic elution of protein at a flow rate of 0.35 ml/min with collection of ~100-microliter fractions. Absorbance was simultaneously monitored at 260, 280, and 420 nm.

Samples for electron microscopy were negatively stained with NanoVan (methylamine vanadate). Using a LEO 912 microscope at 100 kV with the inline energy filter, micrographs were taken on Kodak SO-163 film with minimum dose focusing at nominal magnification of 50,000 x and an approximate defocus of 2 microns. The negatives were digitized on an Agfa Duoscan flatbed scanner at an optical resolution that provided 4.01 Å/pixel on the image scale. Three-dimensional structures of multisynthetase complexes were calculated using SPIDER software: for each new structure, completely new data sets were obtained except for 3520 images that were previously used for the initial calculation of the structure of the human cytoplasmic multisynthetase complex from K562 cells. All data sets were obtained by manual particle selection; the number of images selected per micrograph ranged from ~150 to 250.

Binding of unfractionated tRNA established that these molecules are widely distributed on the exterior of the structure. Binding of gold-labeled tRNALeu and examination of the difference map of three-dimensional reconstructions of control and Nanogold-tRNALeu-labeled multisynthetase complex showed two distinct sites of new density, and placed leucyl-tRNA synthetase and the bifunctional glutamyl-/prolyl-tRNA synthetase at the base of this asymmetric "V"-shaped particle. Using a gold-labeled nickel-nitrilotriacetic acid probe, and comparing the density differences between labeled and unlabeled complexes with earlier electron microsopic data, the polypeptides of the p43 dimer were located along one face of the particle. The results of this and previous studies may be combined into an initial three-dimensional working model of the multisynthetase complex. This is the first conceptualization of how the protein constituents and tRNA substrates are arrayed within the structural confines of this multiprotein assembly.

Reference:

Wolfe, C. L.; Warrington, J. A.; Treadwell, L., and Norcum, M. T.: A three-dimensional working model of the multienzyme complex of aminoacyl-tRNA synthetases based on electron microscopic placements of tRNA and proteins. J. Biol. Chem., 280, 38870-38878 (2005).

• Abstract (courtesy of the Journal of Biological Chemistry).

Hainfeld, J. F.; Liu, W.; Halsey, C. M. R.; Freimuth, P., and Powell, R. D.: Ni-NTA-Gold Clusters Target His-Tagged Proteins. J. Struct. Biol., 127, 185-198 (1999).

• Abstract (courtesy of Science Direct)

More information:

• NTA-Ni(II)-Nanogold: catalog information
• NTA-Ni(II)-Nanogold: product information and use protocols
• Monoamino Nanogold: product information and use protocols
• Application Note: RNA Labeling with Nanogold
• Technical Help: NTA-Ni(II)-Nanogold
• Other References - NTA-Ni(II)-Nanogold
• 5 nm Gold-Ni-NTA binds His Tags: our 2005 paper (from Microscopy and Microanalysis 2005)

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Undecagold and Image Analysis for Cryo-EM

To delineate the boundary of EF-G within the ALC, the authors tagged an amino acid residue near the tip of the G0 domain of EF-G with undecagold, which was then visualized by three-dimensional cryo-electron microscopy (cryo-EM). All of the three naturally occurring Cys residues from the wild-type EF-G were substituted (C114D, C266A, and C398S); subsequently, a Cys residue was introduced at position 209A. 10-fold molar excess of Monomaleimido-undecagold was reacted with the purified EF-G-209C protein; to increase the labeling efficiency, the sample was concentrated by ultrafiltration on a Centricon-3 (Millipore), then incubated for 1 hour at room temperature, and held overnight at 4°C for completion of the reaction. The reaction mixture was purified by HPLC, using a TSK G2000SW (Tosh Biosciences) column. The peaks showing high absorbance at both 280 nm and 420 nm were pooled as EF-G-209C-UG/NG. For 100% labeling, the ratio of A280 to A420 would be ~8.2; it was ~7.8 in our experiments, suggesting the presence of ~95% labeled EF-G in the pooled fractions (see Figure S1); the purity of UG-labeled EF-G-209C was most critical in the cryo-EM visualization of the UG mass. The labeled protein was concentrated using a Centricon-30 concentrator, and the buffer was changed to ribosome binding buffer containing 20 mM HEPES-KOH (pH 7.5), 6 mM MgCl₂, 150 mM NH₄Cl, 2 mM spermidine, and 0.4 mM spermine for subsequent reactions. The undecagold-labeled EF-G was biologically active, as verified in a ribosome-dependent GTPase assay.

Complexes of the 70S ribosome with undecagold-labeled or unlabeled EF-G-209C were prepared, and the extent of binding of EF-G-209C-UG to the 70S ribosome in all complexes was similar to the binding obtained for unlabeled-EF-G-209C, as assessed by SDS-PAGE with silver staining. Cryo-EM grids were prepared following standard procedures, and EM data collected under low-dose conditions on a Philips Tecnai F20 field-emission-gun electron microscope at 200 kV, magnification 50,7603. Cryo-EM data were collected at between 0.7 and 3.5 mm under focus; however, the majority of the data were acquired at close-to-focus settings, between 0.7 and 1.7 mm under focus. The micrographs were digitized with a step size of 14 mm on a Zeiss/Imaging scanner, corresponding to 2.76 Å on the object scale. The three-dimensional reconstructions were calculated using the 3D projection alignment procedure. Each data set was subdivided into defocus groups and analyzed with SPIDER software to obtain CTF-corrected 3D cryo-EM maps (Penczek et al., 1997). X-ray solution scattering data for the ribosome were used for Fourier amplitude correction. 3D maps were also computed by exclusively using the subsets of the data that had been collected at close-to-focus settings (between 0.7 and 1.7 mm under focus), in order to produce more sharply defined peaks for the electron-dense heavy metal clusters.

Two distinct positions for the undecagold, observed in the GTP-state and GDP-state cryo-EM maps of the ribosome bound EF-G, allowed the determination of the movement of the labeled amino acid. Molecular analyses of the cryo-EM maps show that three structural components, the N-terminal domain of ribosomal protein L11, the C-terminal domain of ribosomal protein L7/L12, and the G0 domain of EF-G, participate in formation of the ALC. It was also apparent that both EF-G and the ribosomal protein L7/L12 undergo large conformational changes to form the ALC.

Reference:

Datta, P. P.; Sharma, M. R.; Qi, L.; Frank, J., and Agrawal, R. K.: Interaction of the G' Domain of Elongation Factor G and the C-Terminal Domain of Ribosomal Protein L7/L12 during Translocation as Revealed by Cryo-EM. Mol. Cell., 20, 723-731 (2005).

• Abstract (courtesy of Science Direct).

More information:

• Undecagold - Catalog information
• Product information - Maleimido Undecagold
• Undecagold - Technical Help
• Undecagold - References, sorted by Application

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Other Recent Publications

Reference:

Lesniak, W.; Bielinska, AU.; Sun, K.; Janczak, KW.; Shi, X.; Baker, JR Jr., and Balogh, LP.: Silver/dendrimer nanocomposites as biomarkers: fabrication, characterization, in vitro toxicity, and intracellular detection. Nano Lett., 5, 2123-2130 (2005).

• Abstract (courtesy of the PubMed (Medline)).

Englebienne and Hoonacker describe a simple synthetic procedure to encapsulate colloidal gold nanoparticles by electrostatic adsorption with the water-soluble conductive polymer, poly(aniline-2-carboxylic acid) in the current Journal of Colloid and Interface Science. 27 nm colloidal gold nanoparticles were produced by the reduction under boiling of a hydrogen tetrachloroaurate aqueous solution with sodium citrate. The poly(aniline-2-carboxylic acid) (PANI-COOH) solution was synthesized by oxidizing a solution of 2-aminobenzoic acid in 1 M HCl with ammonium persulfate and ferric chloride. Encapsulation entailed first the determination of the suitable pH for adsorption of the polymer on the gold colloid, by incubating with a fixed saturating amount of aqueous PANI-COOH solution, then challenging with 1 M sodium chloride to agglutinate the gold particles that were not stabilized by a polymer layer. Next, the amount of polymer solution required for full stabilization was determined by incubation with increasing quantities of polymer solution at optimalpH until challenging with NaCl produced no change in maximum absorption wavelength (indicating no agglutination). The optimized encapsulation process was then scaled up to a larger volume (100 ml) and the resulting composite nanoparticle solution buffered at pH 9 using 50 mM borate. The composite nanoparticles are stable in aqueous buffer and retain the respective optical reactivity of the gold colloid to refractive index increases, and of the conductive polymer to pH changes and oxidoreduction. The new composite was found to displays significant enhancements in photonic performance compared to the individual components, which may result from electronic interplay between the two materials in the hybrid structure. This may enable its use for biosensing applications.

Reference:

Englebienne, P., and Van Hoonacker, A.: Goldconductive polymer nanoparticles: A hybrid material with enhanced photonic reactivity to environmental stimuli. J. Coll. Interface Sci., 292, 445-454 (2005).

• Abstract (courtesy of Science Direct).

Mercogliano and Derosier want to eliminate gold particle conjugation altogether - through the development of clonable heavy atom labels where the gold particle is prepared on the label. To investigate this concept, the authors reacted metallothionein, a small metal-binding protein, with aurothiomalate, an anti-arthritic gold compound. Electrospray ionization and matrix-assisted laser desorption/ionization (MALDI) mass spectrometry measurements indicated a distribution of gold atoms bound to individual metallothionein molecules, but unlike previous reports, these data show gold binding occurred as the addition of single atoms without retention of additional ligands; in addition, under certain conditions, MALDI spectra show gold binding ratios of greater than 1 : 1 with the cysteine residues of metallothionein, perhaps hinting at a gold-binding mechanism similar to gold nanocluster formation. Metallothioneingold complexes visualized in the TEM show a range of sizes from about 1 to 5 nm, similar to those used as current TEM labels. This suggests that metallothionein has potential as a clonable TEM label, in which the gold cluster is grown on the label, circumventing the problems associated with attaching gold clusters.

Reference:

Mercogliano, C. P., and Derosier, D. J.: Gold Nanocluster Formation using Metallothionein: Mass Spectrometry and Electron Microscopy. J. Mol. Biol., 355, 211-223 (2006).

• Abstract (courtesy of Science Direct)

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .