Reference Work Entry

Encyclopedic Reference of Genomics and Proteomics in Molecular Medicine

pp 1506-1509

Protein Interaction Analysis: Chemical Cross‐Linking

  • Owen W. NadeauAffiliated withDepartment of Biochemistry and Molecular Biology, Kansas University Medical Center, Kansas City Email author 


Protein conjugation; bioconjugation; protein chemical modification; covalent coupling


Protein interaction analysis is one of the main fields of functional proteomics, covering an array of traditional protein chemical and newly developing techniques that center primarily either on identifying components or determining mechanisms of protein‐protein interactions in both short‐lived and long‐lived protein complexes. The stability of a complex is determined, in part, by the surface contact area or protein interface formed between interacting proteins. Surface contact areas are generally unique for interacting pairs of proteins, with some proteins containing several protein‐binding domains that permit multiple contacts in large protein complexes. Chemical cross‐linking, an extension of protein chemical modification, is an established technique used to join two or more of the proteins in a complex covalently using multi‐functional chemical compounds, termed cross‐linking reagents. These reagents generally contain two or more functional groups that target reactive amino acid side chains on adjacent peptides located within or in close proximity to the protein interface of a protein complex. The architecture and chemical potential of the protein interface are important determinants in complex formation and have a direct impact on the type of cross‐linking reagent that can be used successfully to probe the interaction.


Chemical cross‐linking provides essentially a snapshot of binding events or other physiological processes that require the interaction of two or more proteins. Cross‐linking reagents generally sample only a few of the possible interactions that can occur throughout surface contact regions between two interacting proteins; such regions may vary in area from approximately 500–5000 Å2. Although cross‐linking provides information for only a fraction of the potential contacts that are possible between proteins in a complex, it can be used to determine the spatial organization of proteins in a complex, to measure the distance between interacting surface residues and to detect either small structural rearrangements or global conformational changes in proteins. Resurgence in the use of chemical cross‐linking has occurred with the advent of recent advances in mass spectrometric techniques and the development of large protein data bases; the combined power of these methodologies significantly increases the potential for determining protein‐protein interactions with increasingly smaller amounts of starting material.

Proteins as Chemical Reactants

Numerous physiological processes are mediated through protein‐protein interactions, including enzyme catalysis, muscle contraction, receptor‐ligand binding, antibody‐antigen immunocomplex formation and many others. Both structural and chemical factors (hydrogen bonding, hydrophobic and electrostatic forces) that facilitate protein‐protein interactions also influence chemical cross‐linking of protein complexes (1). The chemical nature of the protein is another important determinant in bioconjugation reactions. Proteins, which are also considered to be reactants in cross‐linking reactions, are linear polymers containing different combinations of amino acids, each with side chains of differing chemistries varying in charge, shape, size and polarity. Cross‐linking generally occurs between appropriate nucleophilic amino acid side chains (Table 1) at distinct sites on the protein where both limited steric constraints and the chemical environment complement the size and chemistry of the cross‐linking reagent. Intramolecular cross‐linking occurs only when proximal residues on one protein target are covalently linked. Chemical coupling of amino acid side chains on adjacent proteins in a complex, termed intermolecular cross‐linking, can occur either at sites forming the interface or at sites in close proximity to the contact surface. In some protein complexes, over 30% of the surface area of a monomer can be subsumed by the protein interface. This comprises numerous contact sites with varying hydrophobic and/or hydrophilic potentials, partially influencing the type of cross‐linking reagent that can access the sites. Protein interfaces also contain water‐filled cavities, which can accommodate hydrophilic (polar) cross‐linkers of appropriate size. Many enzymes have substrate‐binding sites or hydrophobic patches that can bind hydrophobic reagents. Phosphorylase kinase, a regulatory enzyme complex containing sixteen subunits, has several high affinity binding sites for the hydrophobic cross‐linker, phenylenedimaleimide (Table 1), that are in close proximity to the protein interfaces between its regulatory α and β and catalytic γ subunits (2).
Table 1


Structure and Name


Characteristic Solubility

Reactive Groups


Homo‐ bifunctional



2.m‐Maleimidobenzoyl‐N‐hydroxy‐ succinimide ester

Hetero‐ bifunctional


Maleimide/ N‐hydroxysuccinimide ester

Sulfhydryl/ amine
3. Formaldehyde

Mono‐ functional



Broad/ amine
4. N‐5‐Azido‐2‐nitrobenzoyloxy‐ succinimide

Hetero‐ bifunctional: Photo‐reactive


N‐hydroxysuccinimide ester/azide

5. 1‐Ethyl‐3‐[3‐dimethylaminopropyl] carbodiimide hydrochloride






Environment of Chemical Cross‐Linking

Although the chemical environment of a protein has a direct impact on its interactions with potential partners, the complexity of its environment determines the level of information that can be gained in a cross‐linking experiment (3). Cross‐linking either in vitro or in vivo forms the basis for the two most basic approaches to cross‐linking.

In the in vivo approach, proteins of interest are cross‐linked directly within cells or tissues. Hydrophobic cross‐linkers, such as formaldehyde (Table 1), which are able to penetrate cell membranes, are generally required for this approach. The primary advantage of cross‐linking in vivo is that it has the potential to couple proteins in their native environment and is less likely to generate non‐specific conjugates for a given group of interacting proteins, particularly if reagents with short cross‐linking spans are used. The major disadvantage of this technique however, is that it is very difficult to control conditions of cross‐linking within the cell or target any specific protein or protein complex. Once the cross‐linker is introduced into the cell, it is free to interact with numerous potential targets, which complicates interpretation of the data.

Cross‐linking in vitro permits greater control over the conditions of cross‐linking for a given protein; in addition, the number of potential interacting partners can be adjusted, as well as the concentration of reactants, pH, ionic strength, temperature, time of cross‐linking and the type of cross‐linkers used. Thus both non‐polar and water‐soluble cross‐linkers can be used to probe hydrophobic and hydrophilic sites respectively on target proteins. As opposed to the in vivo method, the level of purity for many proteins of interest can be manipulated, effectively reducing the number of potential conjugates that can be formed, so that cross‐linking patterns can be more easily interpreted. The primary disadvantage of the in vitro technique is that proteins are modified under non‐physiological conditions. Detergents and other reagents used to disrupt cells can have a deleterious affect on protein function, altering both intrinsic intermolecular interactions and extrinsic near‐neighbor interactions (3). Moreover, many proteins are compartmentalized in cells, potentially limiting their accessibility to other cellular proteins, which are targeted by non‐specific cross‐linking only under non‐physiological conditions.

Cross‐Linking Reagents

Two comprehensive treatises by Wong and Hermanson cover both the mechanisms and actions of most of the cross‐linking reagents that are commercially available (4, 5). Several basic types of cross‐linkers will be covered herein. Cross‐linking reagents can be separated into two basic classes on the basis of their solubility in either organic solvents and/or water and are termed hydrophobic or hydrophilic cross‐linkers, respectively. Hydrophobic cross‐linkers are often introduced into aqueous environments with carrier solvents such as acetonitrile or acetone. The most commonly used cross‐linkers are bifunctional reagents (Table 1); these compounds contain two reactive groups, separated by a spacer group. The reactive groups react with two neighboring amino acid side chains, covalently joining them at a distance defined primarily by the length of the spacer group. Both the geometry of cross‐linking and the solubility of the reagent are determined, in part, by the spacer group. Cleavable cross‐linkers possess modified spacer groups that can be readily cleaved by oxidizing reagents, reducing reagents or bases. Homobifunctional and heterobifunctional cross‐linkers are bifunctional reagents that have identical and different reactive groups, respectively. Photoactivatible cross‐linkers are heterobifunctional reagents that combine both chemically active and photoreactive groups, permitting an extra layer of control in the cross‐linking process. For example, the protein of interest can first be labeled with the chemically reactive group, purified to remove excess cross‐linker and then cross‐linked to a subsequently added target by exposing the complex to activating wavelengths of light.

Zero‐length cross‐linkers primarily activate the carboxyl‐containing side chains of aspartate and glutamate, forming reactive intermediates that are targeted by proximal ɛ‐amine side chains of lysine. Ultimately, the two amino acids are coupled directly through an amide bond, without any residual, intervening atoms from the cross‐linking reagent. The use of formaldehyde, a reagent traditionally used for fixing tissues, approximates zero‐length cross‐linking by forming a methylene bridge (‐CH2‐) between two proximal lysine (or other nucleophilic) side chains, covalently joining reactants at distances between 2 to 3 Å. Other amino acid side chains are also cross‐linked by this reagent, including those of tyrosine, histidine and arginine.

Detection of Cross‐Linked Proteins

Cross‐linking is not a stand‐alone technique. In order to detect the formation of a conjugate and determine the identity of its integral components, cross‐linking must be coupled with methods that enable visualization, purification and detection of cross‐linked proteins. Rapid screening of cross‐linked proteins is commonly carried out using polyacrylamide gel electrophoresis (PAGE), which is a technique used to separate proteins on the basis of their size. Two‐dimensional polyacrylamide gel electrophoresis techniques, which are capable of resolving proteins based on their charge and size, are employed for analysis of conjugates formed in more complex mixtures of proteins. The proteins are visualized by staining in polyacrylamide gels. There are many traditional forms of chromatography that can be used to purify cross‐linked proteins from complex mixtures of proteins, including size exclusion chromatography, affinity chromatography, hydrophobic interaction chromatography and high performance liquid chromatography (HPLC). These methods often require large quantities (nanomolar amounts) of starting material to provide enough of the purified conjugate of interest for further analysis. A more popular method for analyzing conjugates combines 2D‐PAGE and mass spectrometric techniques. Cross‐linked proteins are first resolved and visualized by 2D‐PAGE. Appropriate samples are then excised from the gel and digested directly in the gel matrix with proteases or chemical reagents that cleave proteins after specific amino acids, generating mixtures of peptides that are characteristic of the interacting proteins. The peptide digests are then analyzed by mass spectrometric techniques on the basis of their charge to mass ratios. The masses measured for peptides in the digest are then compared against those masses predicted for families of peptides that result from the cleavage of potential protein candidates with a specific cleavage reagent. Integral proteins that are cross‐linked in the isolated complex are identified from best matches generated by an appropriate predictive program; sites of cross‐linking on specific peptide stretches between interacting partners are determined by similar methods. There are many Web sites (ExPASy, SWISS‐PROT, NCBI and other data bases) that contain search engines and predictive programs for these forms of analyses. Analysis of fmole quantities of peptides is now possible with modern mass spectrometric techniques, significantly enhancing the potential for detecting protein‐protein interactions by cross‐linking.

Clinical Relevance

Protein‐protein interactions underlie many critical cellular processes that are important in both the normal functioning of cells and disease states, including prion diseases and Alzheimer's diseases. Chemical cross‐linking has proven invaluable in determining the identity, spatial organization and sites of interaction between the protein components of numerous enzyme and structural protein complexes, further defining their role in physiological processes. Additionally, chemical cross‐linking is used by industries in both the areas of therapeutics and diagnostics to couple proteins with different functions. For example, enzymes catalyzing specific reactions can be tethered to proteins that bind specific targets, facilitating either their detection or destruction.

Copyright information

© Springer-Verlag 2005
Show all