Gold Conjugate Protocol Guide

1. Introduction

Our Gold Conjugates are made to the highest possible specification and will give the best possible results in all applications when correctly used. This article is designed to help achieve the best results by giving attention to specimen preparation, choice of gold conjugate and incubation conditions with different specimen types. It is not comprehensive. Many books and articles have been written describing the extensive techniques of immunogold labelling. This article will, however, endeavour to describe the fundamental factors which influence the result that can be obtained, suggest a number of commonly used protocols, and point to solutions to common problems.

2. Technical Assistance

In addition to the information given here, our Technical Services Department is always ready to give help, guidance, analysis and troubleshooting suggestions to all our customers. Please feel welcome to contact us directly or through our world-wide distributors for discussion of your applications.

3. Specimen Preparation

Correct specimen preparation procedures are absolutely crucial for optimal labelling of antigens in cells and tissues. Methods commonly used for ultrastructural preservation in the EM or LM must usually be modified to ensure that antigens are not only retained but also accessible for labelling. This often involves compromise with the structure but careful selection of preparation methods can yield excellent combinations of structural detail and immunochemical labelling. Several different specimen preparation methods have been described in the literature and are comprehensively covered in the texts listed at the end of this article. It would not be possible to summarise all of them in this short text. However the fundamental steps for each method are described here and protocols which follow indicate simple methods for post embedding labelling as well as preembedding labelling techniques with tissues as well as a simple approach to labelling cell surface antigens in cell suspensions.

3.1 Fixation

Cells and tissues may be fixed for subsequent examination or may be sometimes labelled in the unfixed state. Fixatives either denature tissue proteins by coagulation (eg acetone or methanol) or by forming additive cross linking compounds (e.g. aldehydes), or both. The resulting complexes inevitably differ from the unfixed protein in both their chemical and antigenic profiles. Each tissue requires its own fixation protocol. For example, too much cross linking in a tissue with high protein density may mask many antigens. On the other hand, a loose tissue with low protein content may disintegrate without adequate fixation and antigens may simply be washed out. For many tissues the best compromise is a mixture of formaldehyde (eg 2-4%) for rapid stabilising with low cross linking, and weak glutaraldehyde (e.g. 0.1%) for greater structural preservation.

For cytological investigation a precipitating or coagulating fix such as acetone or methanol may be preferred. Formal acetone has also been used for fixing cell preparations. In some studies simple air drying may allow enough antigenic preservation for immunolabelling. Post fixation for electron microscopy has mostly involved the use of osmium tetroxide in order to preserve membrane components and provide image contrast. The introduction of osmium into tissue is, however, not always desirable and more recently tannic acid has been suggested as an alternative (1). Whatever method of fixation is selected it must serve the dual function of retaining the essential structural and antigenic components of the tissue without introducing any material which may interfere with the labelling. In some cases the introduction of heavy metals such as osmium or uranium into the tissue may cause increased non specific labelling and must be treated with caution.

Reference
1. M A Berryman et al (1992) “Effects of tannic acid on antigenicity and membrane contrast in ultrastructural
immunocytochemistry” J Histochem Cytochem 40, 6, 845-857.

3.2 Washing

Thorough washing of the tissue following fixation is extremely important. It may sometimes be necessary to wash for at least as long as the tissue has been fixed in order to remove excess aldehyde or other fixation residues which may cause non specific labelling. In some cases quenching the tissue with ammonium chloride is performed to neutralise aldehyde groups. The wash is best performed in a buffer having a tonicity similar to the natural tissue.

3.3 Embedding

Tissues may be embedded in different types of resin for EM applications. In either case the embedding should allow good preservation of the antigens of the antigens without sacrifice of structural information. Generally there are two types of resin, epoxy resins (eg Araldite, Epon) which have an aromatic ring structure and are strongly cross linked, and acrylic resins (eg UNICRYL – link to product) which have a lower cross linking. Epoxy resins are hydrophobic whereas acrylic resins may be hydrophobic or hydrophilic.

The best structural preservation and stability is provided by epoxy resins while the best immunolabelling is achieved with acrylic resins. This is because acrylic resins, especially UNICRYL, cut in such a way as to reveal the proteins at the section surface and they also wet more easily, thus giving greater accessibility to the antibodies during subsequent incubations.

For optimal results a compromise must be reached and the usual choice is a polar (ie hydrophilic) resin with moderate cross linking. UNICRYL gives excellent immunolabelling characteristics together with a high degree of stability and structural preservation. UNICRYL allows the embedding procedure to be performed at low temperatures. This ensures that vital components are not extracted during dehydration and that no excessive temperature rise occurs during polymerisation which may damage the tissue antigens.

For LM applications tissues may be embedded in acrylic resins such as UNICRYL as described above, or in paraffin wax. Resin embedded sections provide very high resolution of labelling but only allow labelling of those proteins at the surface. Wax embedded sections permit the whole depth of the section to be labelled when de-waxed. Where high temperatures must be avoided cryosections can sometimes be used for immunolabelling for LM studies.

3.4 Sectioning

Sections should be mounted on nickel or gold grids for EM investigations. Copper grids introduce unwanted background into the immunolabelling. For post embedding labelling section thickness is unimportant since the antibodies do not penetrate the resin sections. Sections of 100nm thickness should be adequate. While epoxy resin sections float easily and flatten onto the surface of water, acrylic resin sections require a more careful approach. Flattening is best achieved by cutting at the right speed with sharp knives, diamond if possible. They may be mounted on Formvar coated grids or bare grids. The user should be aware that Formvar and epoxy resin contains high levels of sulphur which produces non specific attraction of gold particles. It is also important to remember that it is the proteins at the section surface which will be exposed to subsequent solutions for labelling and that internal protein will not be accessed in resin sections. In this respect it should also be noted that the proteins on the section surface will be exposed to water, not buffer, during the sectioning procedure and that this may affect the subsequent labelling to some degree.

This exposure should be kept as short as possible to avoid denaturing the proteins. For similar reasons sections cut from resin embedded tissue will not necessarily have an infinite life time when stored at ambient conditions. Freshly cut sections are likely to be more reliable. To avoid leaching of proteins in water some investigators have cut sections onto buffer immediately before incubating them with antibodies. If a tissue block is sectioned then it may often be observed that sections from the outer region of the block will have better structural integrity and lower labelling intensity than sections from the centre of the block. This is related to the penetration of the fixative and the degree of cross linking and antigen retention. When comparing results between incubations it is always good to use adjacent tissue sections.

For LM sections embedded in resin the above considerations are also true. For wax embedded sections, however, the thicker the section the greater the labelling intensity after dewaxing. With LM a section mounting securely on glass slides is of great importance to withstand further incubations. Slides should be first coated with BIOBOND (link to product code) to eliminate variations in glass composition from different sources and to provide a firm section adhesive without introducing non-specific background. Lysine should be avoided at all costs as a tissue adhesive since it attracts gold very strongly. Chrome alum also produces significant background.

3.5 Etching

If the immunoreactivity is weak or nonexistent some attempt may be made to unmask antigens from resin embedded sections for LM or EM in order to make them more accessible to the antibodies. This masking of antibodies may be due to cross linking by the aldehyde fixation, chemical interaction with the resin, covering with other cross linked proteins, osmication, covering of epitopes with resin, etc. When using etching techniques it is important to avoid causing further damage to the epitopes while optimally removing the resin or digesting the surrounding proteins. In general etching is not advised because of the difficulty of controlling the reaction. For acrylic resins it is almost totally ineffective anyway.

3.5.1 Methods of Etching

(a) For sections of tissue post fixed with osmium tetroxide, treat sections with 1-4% hydrogen peroxide solution for 5-10 min at room temperature. This is an oxidising reaction which removes reduced osmium metal and makes the sections more hydrophilic. Wash sections thoroughly in water before performing normal incubations.
(b) Treat sections of osmium post fixed tissue with saturated sodium metaperiodate (0.1% diluted in distilled water) for 3-5 min at room temperature. This acts in a similar way to hydrogen peroxide (a).
(c) Treat epoxy sections with a saturated solution of sodium hydroxide in ethanol or methanol for 30 min.at room temperature. This removes epoxy resin from sections. It is difficult to control for EM sections.
(d) Treat acrylic sections with 70-100% ethanol. This partially dissolves the methacrylate bonds and removes resin slowly. Acrylics feel soapy when treated with alcohol for this reason.
(e) Treat acrylic resins with weak hydrochloric acid (0.01 – 0.1%) for 10 min.(on gold grids only). This gradually removes the resin by interfering with the ester bonds. It is difficult to control without affecting antigenicity.
(f) Treat sections with 0.1% trypsin in TBS for 30 min at 37C. This unmasks antigens from unreactive precursors and digests surrounding epitopes. Useful for treating heavily fixed tissue (eg after glutaraldehyde). It is difficult to control selective digestion without interfering with specific epitopes.

4. Incubations

4.1 Choice of Gold Conjugate

Gold conjugates may be used directly or indirectly to label antigens in cells and tissues. In the direct method the gold is conjugated to the primary antibody, whether monoclonal or polyclonal. This allows a single incubation to be performed and provides the simplest detection system. In the indirect method a primary unlabelled antibody is applied to the specimen to locate the antigen. This is followed by a gold labelled secondary antibody that detects the primary antibody.

This gold labelled secondary antibody is almost always an affinity purified polyclonal. In this way an amplification of the signal is achieved, often up to 10 x compared with a direct incubation, according to the particle size. If, for example, the primary antibody is from mouse (a monoclonal) then the second gold labelled antibody will be Goat anti-mouse.

The indirect method for immunolabelling is the most common for studying antigens in cells and tissues and avoids the need to label every primary antibody. It provides a universal method for detecting any primary antibody from the same primary species, ie all rabbit antibodies may be detected by Goat anti-rabbit gold conjugates. The indirect method is longer than the direct method because of the need for two or three separate incubations but is the more widely used. The protocols described below use the indirect method exclusively.

The choice of gold particle size depends upon the magnification which is to be used in the EM. In principle smaller gold particles produce a higher labelling intensity due to their greater concentration in solution and the lower steric hindrance produced. For example an IgG may be approximately 8nm in length. Gold particles of 1nm or 5nm produce very little steric hindrance to the antibody activity, while larger particle sizes reduce the potential labelling intensity due to their sheer size. Larger particle sizes, however, are more easily seen at lower magnifications. For those users undecided about which particle size to choose for preliminary investigations, 10nm gold conjugates are preferred since these may be clearly seen over a wide range of magnifications in the EM.

For ultra high sensitivity the very smallest, 1nm, gold conjugates may be preferred since the number of gold particles attaching to the tissue will be very high. This requires that extensive washing following the incubations is necessary. These particles are invisible, however, at magnifications below 500000 x and will need to be silver enhanced. Silver enhancing allows the particles to grow to any size suitable, the reaction stopped simply by washing in water. Silver enhancing procedures for EM applications are described below and are identical to those for LM use.

For LM studies, where silver enhancing is required, the choice is between 1nm and 5nm gold conjugates. 1nm conjugates provide very intense labelling and can penetrate cell membranes during pre-embedding routines. Very thorough washing is required to prevent unwanted background following the incubations.

4.2 Incubation Buffer (Buffer A)

A typical incubating buffer suitable for both EM and LM incubations, which provides the components for eliminating the sources of background, may be as follows:
PBS or TBS
+ 1% Normal serum (eg goat serum if using Goat anti-rabbit IgG:gold) + 0.1% Tween 20
+ 1% BSA
+ 0.1% sodium azide adjusted to pH8.2.

This is referred to as buffer A in the following text and is typical of the buffer required for each step of the incubation. The PBS or TBS provides the tonicity for antibody reaction. Antibodies will not work in the absence of adequate salt. The normal serum covers tissue sites which may otherwise attract the gold labelled antibodies non-specifically. The BSA blocks other “sticky” tissue components which would otherwise attract the antibodies non-specifically. The Tween 20 (link to product code) reduces the possibility of hydrophobic attraction between tissue components and the gold particles. The pH of 8.2 is often preferred to reduce background caused by charge attraction of gold particles to positively charged tissue components. At higher pH values there is less positive charge on proteins in the tissue and thus less attraction for negatively charged gold particles. This is usually the main reason for non specific background and is cured simply by raising the pH of the incubating buffer.

In order to achieve the highest possible signal with the lowest possible non specific background it is important to be aware of all the factors in both specimen preparation and in subsequent incubations which can affect these results. For the indirect, two step method of immunolabelling attention must be given to the quality of the antibodies, the necessary dilution of each one, the buffers used, and the incubation protocol. Wherever possible the method should be kept as simple as possible and each step or component must be questioned as to its value. There is no point adding many ingredients to a buffer in order to avoid background if no such background exists. It may actually cause the introduction of unwanted background.

4.3 Dilutions

When establishing a new protocol it is necessary to determine the optimum concentrations of both the primary and the secondary gold labelled antibodies. This is achieved by first incubating separate sections with various dilutions of the primary over an appropriate range (eg 1/100 – 1/10,000). The various primary incubations are then followed by second incubations, using a constant dilution, such as 1/100, of the gold labelled second antibody. The dilution of the primary antibody which gives the best signal/background ratio is thus determined. The procedure is then repeated, this time using the chosen primary antibody dilution and varying the gold labelled secondary antibody dilution over a range of, say, 1/10 antibodies will be determined.

4.4 Incubation procedure

The normal steps to be taken in incubation – 1/200. In this way the best dilution of both protocols are as follows:
(i) Blocking (where necessary)
(ii) Primary antibody
(iii) Washing
(iv) Secondary antibody (gold labelled)
(v) Washing
(vi) Silver enhancing (where required)
(vii) Counterstaining (where required)

These are discussed in general below and in the specific protocols which follow.

(i) Blocking
Non specific reactive sites on tissues and cell surfaces may need blocking before antibodies are applied to the specimens. This can reduce potential non-specific attraction of the primary or secondary antibodies to the tissue components or the resin material. Blocking reagents are often used before the primary antibody step and can be added to the buffer in each step also. They include:

(a) BSA (link to product code) covers non specific “sticky” sites of tissue that would otherwise attract proteins in general. BSA also masks aldehyde groups which may still be present in the tissue. Typically BSA blocking uses solutions of up to 10%. The BSA solution should be freshly made and microfiltered to avoid clumps of protein attaching to the section and producing unwanted visual contamination.

(b) Normal serum of the second antibody species (e.g. normal goat serum when using Goat anti-rabbit: gold conjugates). This overcomes the possible attraction of tissue components for goat antibodies per se. Normal (ie non immune) serum should also be microfiltered before use and should be heat inactivated. It is often used up to 5% in blocking procedures. Unfortunately normal goat serum is relatively impure and can add unwanted impurities to the incubation buffer. It should only be used if necessary. The cross reactivity between the normal serum and the primary or secondary antibodies may not always be well defined. As with all other components in the incubation procedure, blocking reagents should only be used if necessary. Otherwise they may introduce unknown signals into the final immunolabelling. It is best to avoid using blocking procedures until it is found necessary to do so.

(ii) Primary antibody
The primary antibody should be of the highest titre and of the highest specific purity to allow the highest possible dilution. Cross reactivity must be low, especially with the sample tissue. A low quality primary antibody is the greatest cause of low specific signal and high background during the labelling procedure. The buffer composition and the pH is important for the incubations. A normal TBS or PBS buffer is usually chosen with suitable additions to maintain a low background as described above. Buffer A is typical. The causes of non specific background are described below with their remedies. The dilution of the primary antibody is discussed above. The specimen must be washed thoroughly with buffer between incubations.

(iii) Washing
The tissue section is thoroughly washed in the same buffer used (eg buffer A) to dilute the primary antibody. Thorough washing is needed to remove all non specifically attached primary antibodies.

(iv) Secondary antibody
A high quality secondary antibody is essential to label the primary specifically and with low background. BBI gold conjugates are affinity purified and of the highest quality. The conjugates contain no free antibody, all antibodies being adsorbed securely to the surface of the gold particles. BBI gold conjugates need not and should not be centrifuged. Their high titre and purity means that for incubation they may be highly diluted (typically 1/10 – 1/400) in the typical incubating buffer shown above (i.e. same as for the primary antibody), so reducing non-specific background whilst achieving a high signal intensity.

(v) Washing
The tissue section must again be thoroughly washed with the same buffer to remove unwanted non-specific gold particles. After washing in buffer the section must also be thoroughly washed in distilled water to remove all traces of salt and loose protein before it is allowed to dry. During the incubations, a greater sensitivity may be achieved by constantly flushing or agitating the tissue section in the solutions. This may be achieved on a gentle shaker platform or by immersing the EM section in a small tube with the incubating solutions and placing it on a roller bed, so bringing fresh solution continuously to the tissue surface proteins as would be the case in vivo. If the antibody solutions are tolerant the incubations may also be performed at 37C to mimic in vivo conditions further. If floating sections on the surface of droplets of antibody, it is important to prevent any antibody solutions from drying on either side of the tissue section. Drying can produce non specific aggregations of gold particles across the section surface.

(vi) Silver Enhancing
Silver enhancing may be used for EM studies where the gold particles are too small to be seen in the available microscope or to grow them from any original size to a size that gives a more comprehensive view of labelling at lower magnifications. The method is quite easy to perform for any original gold particle size chosen. In order to ensure that the gold labels are not removed from the section it is common to fix the section for 10 minutes in 1% glutaraldehyde in water before silver enhancing. Using the Silver Enhancing Kit for LM and EM (linl to product code SEKL) a drop of enhancer and initiator are mixed together on a clean petri dish and the gold labelled EM section placed on the surface or even immersed in the droplet. The time of enhancement must be judged empirically but approximately 5-10 minutes is usually sufficient to grow from 1nm to 5nm, or from 5nm to 20nm. It is temperature dependent. The enhancement process is stopped by washing in distilled water and the grids may be re-enhanced after examining in the EM until the chosen size is achieved. Provided the sections are not exposed to an electron beam, silver enhanced sections may be further immunolabelled in a subsequent incubation procedure. This may provide another method of double labelling. For LM studies silver enhancing is a necessary but straightforward procedure.

(vii) Counterstaining
Sections which have been immunogold labelled may also be counterstained by the usual methods of uranyl acetate and lead citrate. Again, the use of glutaraldehyde will reduce the loss of gold particles from the section surface.

5. Specific Immunolabelling Protocols

The following protocols are provided as a guide for immunolabelling different tissue and cell samples in the EM and LM. Many others exist and may be found in the literature. These protocols serve to indicate general approaches to achieving maximum sensitivity and minimum background.

5.1 Post embedding labelling tissues for EM

(a) Place the grid, section downwards, on a 50ul droplet of 10% heat inactivated normal serum (of the gold labelled animal species, eg for Goat anti-rabbit:gold use normal goat serum) in TBS or PBS buffer and incubate for 10 min. For Protein A, Protein G or Protein A/G gold conjugates omit this step.
(b) Without washing, gently blot off excess and transfer the grid to the surface of a 50ul droplet of primary antibody diluted in buffer A (above). Incubate for 30 min. to overnight.
(c) Transfer the grid to a series of 5 or more 50ul droplets of buffer A, leaving 5 min. on each, to remove unbound antibody.
(d) Transfer the grid to two successive 50ul droplets of gold conjugate diluted in the appropriate buffer A. The first drop will remove excess buffer and the second drop will provide the incubation. Incubate for 1 – 4 hours.
(e) Transfer the grid to a series of 50ul droplets of distilled water (5 x 2 minutes or more) to remove unbound gold conjugate.

Optional silver enhancement:
(f) If required place the grid on a drop of freshly mixed silver enhancing solution (link to product code SEKL) for a period of 5 – 10 minutes. Determine the time experimentally beforehand by enhancing gold particles which have been allowed to dry on a Formvar coated grid.
(g) Wash thoroughly in distilled water to stop the reaction.
(h) Counterstain if required with uranyl acetate and lead citrate as normal. Frozen sections may be stained with osmium tetroxide vapour. Wash and examine in the electron microscope.

5.2 Pre-embedding labelling tissues for EM

(a) Dissect the tissue to the smallest possible size (eg 0.5mm) for handling and good penetration.
(b) Lightly fix tissue in 2-4% paraformaldehyde + 0.1% glutaraldehyde for 30-60 min. depending on tissue size.
(c) Wash tissue very thoroughly in PBS (eg 30-60 min) to remove excess aldehyde groups.
(d) Optionally permeabilise tissue with 0.1% Tween 20 or Triton-X 100 for 15-30 min.
(e) Wash tissue very thoroughly in buffer A for 1 hour.
(f) Incubate tissue with 20 x volume of primary antibody, diluted in buffer A, for 1-4 hours while agitating.
(g) Wash in buffer A very thoroughly, ie for 1 hour while agitating, to remove excess primary antibody.
(h) Incubate tissue with gold conjugate suitably diluted in buffer A for 1-4 hours while agitating. For most applications only 1nm gold conjugates will penetrate the cell membranes. 1nm gold conjugates are usually diluted to 1/100 – 1/400 for incubations.
(i) Wash tissue very thoroughly in buffer A (30 min) and then in PBS for 30 min. while agitating to remove excess gold conjugate.
(j) Fix tissue in 1% glutaraldehyde for 10min to strengthen antibody binding to the tissue.
(k) Wash thoroughly in water for 30 min. while agitating to remove excess aldehyde and PBS.

Optional silver enhancing:
(l) Immerse the tissue in fresh silver enhancing solution (link to product SEKL) for 5-15 min. Experiment with a number of tissue blocks to determine the time required.
(m) Wash thoroughly in distilled water to stop enhancing.
(n) Continue through to dehydration and resin embedding for EM or LM investigations.

5.3 Pre-embedding labelling of cell surface antigens for EM

(a) Isolate cells from tissue with trypsinisation or separate blood cells by gradient centrifugation. Spin to a soft pellet and resuspend in subsequent incubation solutions.
(b) Prefix cells in 0.1% glutaraldehyde or 2-4% paraformaldehyde in PBS depending on antigen sensitivity.
(c) Wash 3 times in PBS containing 50mM glycine to quench aldehydes.
(d) Block cells for 30 min. in PBS + 1% BSA + 10% normal serum (second antibody species).
(e) Wash twice with PBS + 1% BSA.
(f) Incubate with primary antibody diluted in buffer A for 1 hour (typically 1/10-1/100 for a monoclonal)
(g) Wash twice in buffer A.
(h) Incubate in second gold labelled antibody diluted in buffer A for 1 hour (typically 1/10 – 1/100).
(i) Wash twice in PBS
(j) Fix in 1% glutaraldehyde in PBS for 15 min.
(k) Wash twice in distilled water.

Optional silver enhancement:
(l) Incubate in fresh silver enhancing solution (link to product code SEKL) for 5-10 min. This is determined experimentally.
(m) Wash thoroughly in distilled water to stop the enhancement and to remove silver ions from the cells.
(n) Spin to a firm pellet, continue through to dehydration and embedding for EM or LM investigations. Counterstain as normal.

Reference
C Ferrari, et al (1989) “Preembedding immunogold staining of cell surface antigens performed on suspended cells and tissue sections” in Hayat MA, Colloidal Gold, Vol 2.

5.4 Double labelling of EM sections

(i) Labelling both sides of the section
In this method the first antigen is labelled on the upper side and the second antigen on the lower side of the section. Use uncoated grids to mount sections. Perform all incubations by floating sections on droplets to avoid contaminating the upper side.
(a) Follow steps 5.1 (a – e) above using the first primary antibody.
(b) Optionally coat the reacted side of the section with carbon evaporation.
(c) Turn the section over and repeat steps in 5.1 (a – e) using the second primary antibody but also using a different gold particle size for the second gold labelled antibody. Typically this means using 5nm for the first side and 15nm for the second side. Because two different sides of the section are incubated independently the two primary antibodies may be raised in the same animal species.

Optional silver enhancement:
(d) If required either side of the section may be silver enhanced. Place the grid on a drop of freshly mixed silver enhancing solution (link to product SEKL) for a period of 5-10 minutes. Determine the time experimentally. Alternatively, the same particle size may be used on both sides of the section and one side enhanced with silver to produce a different final size for double labelling.
(e) Wash thoroughly in distilled water to stop the reaction.
(f) Counterstain if required with uranyl acetate and lead citrate as normal. Wash and examine in the electron microscope.

(ii) Labelling the same side of the section
In this method both antigens are labelled on the same side of the section using different particle sizes. The tissue section may be mounted on a Formvar coated grid for the incubations.
(a) Follow steps 5.1 (a – e) above using the first primary antibody (eg from mouse).
(b) Repeat steps 5.1 (a – e) above but using a second primary antibody from a different species (eg from rabbit) with the appropriate gold labelled antibody (eg Goat anti-rabbit).

Optional silver enhancement:
(c) The section may be silver enhanced after step (a) above in order to grow the particles to any size required before performing the second incubation procedure. Follow steps 5.1 (f – g). Do not expose the section to the electron beam before the second immunolabelling procedure since this will denature all proteins.
Counterstain:
(d) Counterstain if required with uranyl acetate and lead citrate as normal. Frozen sections may be stained with osmium tetroxide vapour. Wash and examine in the electron microscope.

5.5 Immunolabelling following in situ hybridisation for EM

Methods for performing in situ hybridisation of nucleic acids in embedded EM tissue sections are widely reported in the literature and no attempt is made here to repeat them. The user is encouraged to read a good text book on in situ hybridisation techniques before venturing into immunolabelling of target probes. The steps for in situ hybridisation are: specimen preparation; probe labelling; prehybridisation; hybridisation with a specific labelled probe (labelled with biotin or digoxigenin), followed by antibody incubations. If biotin labelled probes have been employed then Goat anti-biotin or streptavidin gold conjugates are used for detection. If digoxigenin labelled probes are used then Sheep anti-digoxigenin gold conjugates are employed.

(a) Perform the in situ hybridisation steps according to a suitable protocol and finish with stringency washing procedures to remove unbound probe from the section.
(b) Place the grid, section downwards, on a 50ul droplet of 10% heat inactivated normal serum (of the gold labelled animal species, eg for Goat anti-biotin:gold use normal goat serum; for Sheep anti-digoxigenin use normal sheep serum; for streptavidin use nothing) in TBS or PBS buffer and incubate for 10 min.
(c) Transfer the grid to a 50ul droplet of gold conjugate diluted in buffer A. Incubatefor 1 – 4 hours.
(d) Transfer the grid to a series of 50ul droplets of distilled water (5 x 2 minutes or more) to remove unbound gold conjugate.

Optional silver enhancement:
(e) If required place the grid on a drop of freshly mixed silver enhancing solution (link to product SEKL) for a period of 5 – 10 minutes. Determine the time experimentally.
(f) Wash thoroughly in distilled water to stop the reaction.
(g) Counterstain if required with uranyl acetate and lead citrate as normal.

5.6 Immunolabelling of paraffin embedded sections for LM

(a) Dewax sections in xylene and bring to water through ethyl alcohol.
(b) Equilibrate the tissue section in PBS for 30 min.
(c) Place on the section a 50ul droplet of 10% heat inactivated normal serum (of the gold labelled animal species, e.g. for Goat anti-rabbit:gold use normal goat serum) in TBS or PBS buffer and incubate for 30 min. For Protein A, Protein G or Protein A/G gold conjugates omit this step.
(d) Without washing, gently blot off excess and replace with 50ul of primary antibody diluted in buffer A (above). Incubate for 1 hour to overnight. Cover with a dish to avoid evaporation.
(e) Wash the slide thoroughly by immersion in buffer A for 30 min. while agitating.
(f) Place on the section a 50ul droplet of gold conjugate (1nm or 5nm) diluted in buffer A. Incubate for 1 – 4 hours.
(g) Wash the slide thoroughly by immersion in water for 15 min. by agitation.
(h) Place on the section 50ul of 1% glutaraldehyde in water for 10 min. to fix the conjugate in place. Wash thoroughly in distilled water for 10 min. by agitation.
(i) Place on the section 50ul of freshly mixed silver enhancing solution (link to product SEKL) for a period of 5 – 10 minutes. Determine the time experimentally during the enhancement procedure by observing in a light microscope. Do not expose to prolonged excessive light.
(j) Wash thoroughly in distilled water to stop the reaction.
(k) Counterstain using normal histological stains.
(l) Mount the section in BIOMOUNT (link to product code) to prevent fading of the signal produced by some other types of mount.
The silver enhancement step will convert the invisible gold to first a brown stain similar to a peroxidase stain and then to an intense black stain that will, under correct incubation procedures, be specific and localised with very good resolution.

5.7 Pre-embedding labelling of tissues for LM

(a) Follow the steps described in 5.2 (a – m).
(b) Dehydrate the tissue and embed as usual for LM studies using either resin or paraffin wax.
(c) Section and counterstain as required.

5.8 Labelling of cell surface antigens for LM
(a) Isolate cells from tissue with trypsinisation or separate blood cells by gradient centrifugation.
(b) Wash cells in PBS + 1% BSA and suspend to 10 x 109 cells / litre in PBS + 5% BSA.
(c) Prepare a cytocentrifuge preparation or make a smear from the buffy coat onto a glass slide coated with BIOBOND (link to product code).
(d) Air dry overnight.
(e) Fix for 30-90 secs with phosphate buffered 9.25% formol and 45% acetone at pH 6.6.
(f) Rinse the slide in PBS.
(g) Incubate for 10 min. in PBS + 5% non-fat milk or 0.01% casein + 5% normal serum (of the gold conjugate) + required concentration of the gold conjugate.
(h) Rinse with PBS.
(i) Post fix the cells with buffered formal acetone for 2 min. at room temperature.
(j) Wash the slides thoroughly in distilled water.
(k) Silver enhance while monitoring under the LM.
(l) Rinse with distilled water to stop enhancement.
(m) Counterstain with May-Grunwald-Giemsa stain.
(n) Mount in BIOMOUNT (link to product code) to prevent fading of the silver stain. (o) View in the LM using bright field and epipolarised light.

References
1. M de Waele et al (1991) “Leukaemia and lymphoma immunophenotyping in cell smears with immunogold silver staining” Am J Clin Path 96, 3, 351-359.
2. M De Waele (1991) “Colloidal gold as an immunocytochemical marker for the light microscope detection of leukocyte cell surface antigens” PhD Thesis, University of Brussels.

5.9 Immunolabelling following in situ hybridisation for LM

Methods for performing in situ hybridisation of nucleic acids in embedded EM tissue sections are widely reported in the literature and no attempt is made here to repeat them. The steps for in situ hybridisation are: specimen preparation; probe labelling; prehybridisation; hybridisation with a specific labelled probe (labelled with biotin or digoxigenin), followed by antibody incubations. If biotin labelled probes have been employed then Goat anti-biotin or streptavidin gold conjugates are used for detection. If digoxigenin labelled probes are used then Sheep anti-digoxigenin gold conjugates are employed.

(a) Perform the in situ hybridisation steps according to a suitable protocol and finish with stringency washing procedures to remove unbound probe from the section.
(b) Place on the section a 50ul droplet of 10% heat inactivated normal serum (of the gold labelled animal species, eg for Goat anti-biotin:gold use normal goat serum; for Sheep anti-digoxigenin:gold use normal sheep serum; for streptavidin use nothing) in TBS or PBS buffer and incubate for 10 min.
(c) Replace with a 50ul droplet of gold conjugate diluted in buffer A. Incubate for 1-4 hours.
(d) Wash the slide thoroughly with distilled water for 30 min. to remove unbound gold conjugate.
(e) Fix with 1% glutaraldehyde in water for 10 min. to bind the gold conjugate. Wash thoroughly in distilled water for 10 min while agitating to remove aldehyde groups.
(f) Place on the slide a 50ul drop of freshly mixed silver enhancing solution (SEKL) from BBI for a period of 5-10 minutes. Determine the time experimentally while observing under the LM. Avoid the use of excessive light intensity for prolonged periods.
(g) Wash thoroughly in distilled water to stop the reaction.
(h) Counterstain with normal histological stains.
(i) Mount the section in BIOMOUNT (link to product code) to avoid the fading of the silver enhancement.

6. Trouble shooting (microscopy)

Optimisation of incubating conditions will give the maximum signal with the minimum background. Some of the reasons for poor signals and high background in both EM and LM incubations are given here. This list is not exhaustive since so much depends upon the actual protocols used for preparing tissue and incubating sections.

6.1 Poor labelling

(a) Antigen absent or destroyed by preparative procedures. Use a different procedure (eg fixes, cryosections, resins). Check that the temperature of the tissue does not rise too high during embedding or polymerisation.
(b) Antigen present in genuinely low amounts. Use longer incubation times and more concentrated primary antibody.
(c) Antibody poor due to low titre, age, poor storage, wrong dilution, excessive freezing and thawing, low avidity, wrong antibody, no antibody. Change and check the antibody. Check the positive controls.
(d) Primary antibody too concentrated, producing the hook effect. Use greater dilutions (1/100 – 1/10,000).
(e) Wrong gold conjugate. Check and repeat the incubation.
(f) pH of buffer is wrong (excessive alkali or acid). Check solutions.
(g) Insufficient salt content in buffer to allow antibody interactions. Salt content should be at least 50mM.
(h) Section not exposed to solutions (ie wrong way up) if on a plastic film.
(i) Labelling not visible due to heavy stains (eg for 5nm particles). Reduce counterstaining. Use higher magnification (eg 200,000x for 5nm, 100,000x for 10nm, 80,000x for 15nm, 50,000x for 20nm).
(j) Silver enhancement procedure is incomplete (stain is light brown or absent) Use longer enhancement times. Check positive controls.
(k) Counterstaining is masking the silver stain. Use less counterstain.
(l) Silver stain fades after some time. Remove coverslip with xylene and wash in developer. Remount in BIOMOUNT.

6.2 Excessive background labelling

(a) Ionic concentration too low in solutions. Use increased salt concentration (up to 2.5%). Add BSA or normal goat serum (not for Protein A, G, A/G) to approximately 1% in incubation solutions.
(b) Inadequate washing between incubations. Thoroughly wash (hours if necessary).
(c) Non-specific charge attraction of antibody. Raise the pH of the buffer to pH8.2-8.5 to reduce the positive charge on tissue proteins.
(d) Hydrophobic attraction of gold particles to tissue components. Increase the Tween 20 content to 1 – 2%.
(e) Non specific attraction of second antibody to tissue components. Include higher levels of normal serum (e.g. Goat serum for Goat anti-rabbit:gold) in the buffer, up to 5%.
(f) Impurities in the normal serum used. Omit the serum and observe the results.
(g) High levels of sulphur attracting gold particles in the tissue or embedding resin. Change to acrylic resins or include up to 0.01% gelatin in the gold conjugate for that step only.
(h) Free aldehyde groups in the tissue. Reduce by floating sections on 0.5M ammonium chloride for 1 hour before incubations.
(i) Primary antibody concentration too high. Dilute by orders of magnitude. Wherever possible use affinity purified antibodies.
(j) Gold conjugate concentration too high. Dilute further.
(k) Inadequately fixed tissue. Necrotic and damaged cells will also stain non-specifically. Improve fixation conditions and use smaller pieces of tissue. Check the positive control.
(l) Osmium tetroxide fixation may introduce excess charge into tissue. Omit osmium or wash thoroughly after fixing.
(m) Excessive exposure to silver enhancing solutions, especially for low signal. Reduce the development time.
(n) Excessive exposure to light during silver enhancement. Reduce the lightintensity.
(o) Spontaneous precipitation of silver due to impurities in glassware and the water. Use ultraclean glassware and distilled/ deionised water of high purity.
(p) Spontaneous precipitation of silver onto endogenous metals in tissue (eg Zn, Fe, Cd, etc). Carefully check the negative controls.
(q) Non specific attraction of gold to lysine when used as a tissue binding agent on slides. Avoid the use of lysine and use BIOBOND instead.

6.3 Clustering

Clustering described here should not be confused with amplification produced with indirect labelling. In the amplification process each primary antibody may have attached several gold labelled secondary antibodies. This will lead to small groups of gold particles, perhaps 2-4, which surround single antigen targets in the tissue section.

For LM Grade and BL Grade conjugates clustering is of less importance since the particles are not seen individually in the LM sections or on blotted membranes. Nevertheless excessive clustering during manufacture is strictly avoided since this would otherwise lead to long term instability. Excessive clustering may sometimes be seen on EM sections for the following reasons.
(a) Clumped primary antibody. Use fresh antisera. Microfilter if possible to remove bacteria.
(b) Contamination of the gold conjugate. Caused by lysine or other protein additives, by thimerosal or mercaptoethanol. Conjugate stored at extreme pH. Conjugate overheated. Store at 4C in separate aliquots in the original buffer supplied until required for use. Do not freeze and thaw frequently. Do not place pipettes into a single vial frequently.

6.4 Buffer Recipes

The basic PBS and TBS buffers are given concentrations of Tween 20, NaCl, or BSA
(a) Tris buffered saline
Add 100ml distilled water here. These are supplemented with higher as required.
+ 0.242g (20mM) Tris (tris-hydroxymethyl-aminomethane)
+ 0.13g (20mM) NaN3 (preservative)
+ 0.9g (225mM) NaCl
• Adjust to pH 8.2 (or other pH as necessary) with 0.1N NaOH or 0.1% HCl and microfilter.
(b) Phosphate buffered saline
Add 100ml distilled water
+ 0.148g Na2HPO4
+ 0.043g KH2PO4
+ 0.72g NaCl
+ 0.13g NaN3 • Adjust to pH with 0.1N NaOH or 0.1N HCl and microfilter.

7. Immunoblotting of nucleic acids and proteins (BL Grade)

Blotting applications using Gold Conjugates include the demonstration of antigens, antibodies, proteins, macromolecules, total DNA/RNA, and DNA/RNA fragments and gene sequences transferred to an immobilising matrix. The procedures for transfer of these proteins and nucleotide sequences follow the well established methods of Western and Southern blotting, or direct dot blotting.

For Southern or Northern blots, for example, individual fragments of DNA or RNA, as produced by restriction enzyme digestion, are separated by gel electrophoresis and then transferred to an inert support such as nylon or nitrocellulose. The immobilised DNA/RNA fragments can then be analysed by in situ hybridisation using a suitable biotinylated cDNA probe.

For Western blotting, protein dispersions from one or two dimensional gels, or from dot blots of proteins, are immobilised onto a nitrocellulose sheet and the whole sheet, or part of it, is incubated in the appropriate primary antibody.

In all cases, Gold Conjugates of 1nm or 20nm particle diameter are used to visualise the immobilised materials following incubation with an appropriate primary antibody or probe. The 1nm gold conjugates give much higher labelling intensity but must be silver enhanced to be visible. The 20nm gold particles are directly visible on the membrane. The 1nm conjugate also requires extensive washing to remove unbound particles from the membrane. Because of the inexpensive and highly concentrated nature of the BL Grade conjugates, large volumes may be used at high dilutions. Southern blots from DNA gels are best immobilised on nylon based membranes where binding is more efficient. Proteins bind more effectively onto nitrocellulose. With nylon membranes, however, it is necessary to block unoccupied sites more extensively after blotting to avoid background staining. A simple protocol for staining blotted protein or DNA is given below. Blotted proteins, antibodies or antigens may be examined by the simple two step indirect labelling method. DNA/RNA fragments, labelled with biotinylated probes may be detected by using a gold labelled antibody to biotin.

7.1 Immunostaining of DNA gel transfers (Southern blots)

(a) Perform a normal silver stain or whole DNA stain on a narrow strip of gel or duplicate gel to gauge the total DNA pattern. Alternatively use GENOGOLD to provide a total nucleic acid stain on blotted membranes.
(b) Transfer the DNA/RNA pattern to the immobilising matrix by established blotting procedures and perform hybridisation experiments as appropriate with biotinylated probes.
(c) Wash the membrane in PBS.
(d) Block the membrane with 10% BSA to fill unoccupied sites. Incubate the membrane in diluted gold conjugate (eg Goat anti-biotin:gold for biotinylated probes) for 1-24 hours while agitating. Whole sheets may be incubated in flat plastic bags. High dilutions of antibody will give lower non specific background. Dilutions may be made in TBS or PBS supplemented with additions to reduce background (see 4.2).
(e) Wash the membrane thoroughly in buffer with several changes to remove unbound antibody. Gold conjugates of 20nm will be visible as red bands. The 1nm conjugates will require silver enhancing.
(f) Wash the membrane very thoroughly in distilled water to remove all non specific gold and salts.
(g) Immerse the membrane in fresh silver enhancing solution (Silver Enhancing Kit for Blotting, SEKB (link to product code)) and agitate for 10-20 min. Stop the reaction by washing in water before the membrane turns grey. The DNA bands will appear as black bands.

7.2 Immunostaining of protein gel transfers (Western blots)

(a) Perform a Coomassie blue stain on a strip of the gel to determine the whole protein staining pattern. Alternatively use PROTOGOLD (link to product) on a separate blotted membrane to give a total protein stain.
(b) Transfer the protein pattern to the immobilising matrix by established blotting procedures.
(c) Wash the membrane thoroughly in PBS.
(d) Block the unoccupied sites by immersing the membrane in 10% BSA for 30 min.
(e) Transfer the membrane to the primary antibody diluted in buffer A (see 4.2). Incubate for 1 hour while agitating.
(f) Wash the membrane very thoroughly in buffer A while agitating for 30 min.
(g) Immerse the membrane in gold labelled second antibody diluted in buffer A while agitating. Dilution of the antibody should be determined as in 4.3. Incubate for 1 hour.
(h) Wash the membrane very thoroughly in buffer A for 30 min. and then in distilled water for 30 min. while agitating. Gold conjugates of 20nm will appear visible directly as red stains. For 1nm gold conjugates silver enhancing is necessary.
(i) Immerse the membrane in fresh silver enhancing solution and agitate for 10-20 min. Wash thoroughly in water to stop the reaction. The protein bands will appear as black lines.

7.3 Trouble shooting (membranes)

All of the trouble shooting indicators listed above in section 6 apply to membrane incubations. In addition the following should be considered for membrane incubations.
(i) Poor labelling
(a) Check transfer system is working.
(b) High background is masking specific staining.
(ii) Excessive background
(a) Inadequate blocking of transfer membrane (especially nylon based membranes). Saturate for longer times at higher temperatures. Use higher concentrations of BSA in the gold conjugate. Include 0.1% gelatin in the gold conjugate to reduce binding to sulphur. Wash the membrane very thoroughly before the incubations.
(b) Spontaneous precipitation of silver due to impure water used for washing membrane. Use only clean distilled water in clean trays with no contamination. Use a totally negative control simultaneously.

(c) Inadequate washing of membrane following the gold incubation. Lengthen the washing procedure in water to remove all salts.

8. Controls

It is of great importance to perform controls during immunogold labelling procedures in order to confirm the specific labelling in sections and membranes. In addition controls will determine the causes of poor performance of individual reagents or weakness in any part of the entire procedure. Some are included here:
(a) Omit the first antibody. This will determine if the second gold labelled antibody is labelling non-specifically.
(b) Replace the primary antibody with non-immune serum, preferably from the same animal before immunisation. This will determine if the primary antibody is specific.
(c) Replace the primary antibody with an inappropriate antibody from another species. This will also test the specificity of the primary antibody and of the gold conjugate.
(d) Absorb the primary antibody with its antigen (if available) by addition of approximately 1nM/l for 1 hour before the incubation. This will also test the specificity of the primary.
(e) Use a different gold conjugate directed towards a different animal species. This will determine if the tissue has non specific attraction for gold particles.
(f) Use a known negative tissue.
Where possible a known positive control should also be included in the series to demonstrate the validity of the primary antibody and of the method.