Immobilization theory

The first step in the interaction analysis is the immobilization of one of the interactants on the sensor chip surface. This immobilization can be permanent in the form of a covalent bond or transient by means of capturing.

The suitability of an immobilization technique should be determined based on the ligand type (protein, sugar, DNA, low molecular mass-substance), the analyte to be used (small or large interactants) and the purpose of the study (specificity, concentration, affinity, kinetics). In addition, the ligand must retain its biological activity after immobilization (1),(2),(3). Information about the ligand size, pI, amino acid composition, pH stability and possible sites for oriented coupling is beneficial in determining the immobilization method.

Covalent coupling chemistries

Several covalent coupling chemistries are available to immobilize the ligand depending on the available reactive groups. Three well-established procedures are the use of amine (-NH₂), thiol (-SH₂) and aldehyde (-COOH) coupling chemistries. Covalent coupling is stable and, in general, does not require any modification of the ligand. The immobilization level is easily controlled and the ligand consumption is low.

However, the standard covalent immobilization techniques are prone to random orientation of the ligand to the dextran surface. This random orientation can block binding sites and thereby reduce the number of available binding sites. Furthermore, free reactive groups close to the binding site can be obscured, reducing or eliminating the affinity for the analyte (4). In some cases, the low pH or the blocking agent used in the immobilization can inactivate the ligand. The standard covalent immobilization techniques can convert even a homogeneous ligand into a heterogeneous one, which can make the results more difficult to analyse.

Two main solutions to the random orientation are available:

• Unidirectional immobilization
• Affinity capturing

Unidirectional immobilization

Unidirectional immobilization is used to reduce or eliminate induced heterogeneity by the random coupling chemistries. For instance, biotinylation of antibodies via the carbohydrate groups will give a high degree of unidirectional immobilized molecules (5). Antibodies can also be digested with Papain to form F(ab')2 fragments containing the two Fab fragments connected by the disulphide bond. Reducing this bond gives the single Fab' fragments that can be bound unidirectionally and covalently to the sensor chip using thiol chemistry (4). Although both examples above are for antibodies, they are applicable to most ligands. Look for suitable reactive groups and reaction protocols in the literature and the tables below. When the ligand of interest is to be modified, it should be purified to one form to obtain the most homogeneous immobilization. It is advisable to check if the protein is still functional after modification (6).

Affinity capturing systems

A different approach is to use capturing antibodies or systems to put the ligand in a specific orientation. The affinity capturing system used must have sufficient affinity to the ligand in order to make a stable complex. Some of the systems are completely regenerable but then the ligand consumption is high and the ligand capturing is not always stable. In addition, the capturing system must not interfere with the function or binding site of the ligand. An additional advantage of affinity capturing is that the procedure does not require highly purified ligands because the capturing is analogous to affinity purification (7).

Most affinity capturing systems consist of a covalent bound antibody to the ligand of interest. Generally, the capturing antibody is immobilized with the amine coupling procedure. This will also create capturing antibodies, which are non-functional. In general, this is not of great concern because enough functional antibodies will be present. The functional groups on the ligand can be for example the GST, myc, MBP or the FLAG-tag. For a more thorough discussion see the publication of Lichty et al (8).

A slightly different system involves the modification of the sensor chip to capture special tags. For instance, with the introduction of six histidine residues, it is possible to capture a protein with the NTA group (see NTA sensor chip). Modifying the sensor chip surface with a hydrophobic compound makes it possible to capture vesicles. The immobilized vesicles serve as a platform into which you can insert proteins with hydrophobic regions, which are otherwise difficult to immobilize (see L1 sensor chip).

Choosing an immobilization method

Although the kinetics is in principle not influenced by which protein is immobilized as a ligand, there are some issues to take into account before immobilizing a protein.

As already pointed out, the immobilization can alter the ligand because it will be covalently bound to the surface. Affinity capture systems can overcome this problem, providing that the capture site is sufficiently far away from the kinetic binding site. The advantage of immobilizing an antibody is that once regeneration conditions are established there is a stable and reusable surface. Capturing a protein with an antibody uses substantially more ligand, which can be problem. However, on the other hand, it is possible to capture the ligand of interest directly from culture media or other sources.

The suitability of an immobilization method depends mostly on the nature of the ligand. Amine coupling is the most generally applicable coupling chemistry and is should be the first choice to consider. Most macromolecules contain amine groups, which can be used in amine coupling. Amine coupling is less suitable in situations involving any of the following: acidic ligands (pI < 3.5), ligands where an amine is in the active site, and molecules possessing several amine groups.

The suitability of thiol coupling depends mostly on the availability of thiol groups on the ligand. However, it is relatively easy to introduce thiol groups on the ligand. Thiol chemistry is more robust than amine chemistry and therefore the coupling conditions are less critical. Thiol coupling cannot be used with strong reducing conditions, because the disulphide bond is unstable under such conditions.

Aldehyde coupling is the best choice in specific cases. For instance, it works well with polysaccharides and glycoconjugates which have cis-diols and sialic acids. These groups are oxidized relatively easily to aldehydes. Aldehyde groups are spontaneously reactive towards amine groups over a wide pH range, and their reactivity is retained for a long period of time under ambient conditions (10).

(Strept)avidin-biotin coupling is the best choice when the amine or thiol coupling is unsatisfactory or unsuitable. Nucleic acids, polysaccharides and glycoconjugates are relatively easily biotinylated using a variety of reagents and functional groups. Acidic and acid-sensitive proteins can be coupled at neutral pH with high efficiency.