Biotinylation

Explore biotinylation products  Crosslinking Technical Handbook

The interaction between biotin (vitamin H) and avidin is a useful tool in nonradioactive methods of purification, detection, immobilization, labeling, viral vector-targeting and drug targeting systems. The extraordinary affinity of avidin for biotin is one the strongest known non-covalent interactions of a protein and ligand (Ka=10¹⁵M^-1) and allows biotin-containing molecules in a complex mixture to be discretely bound with avidin conjugates. The bond formation between biotin and avidin is very rapid, and once formed, it is unaffected by extremes in pH, temperature, organic solvents and other denaturing agents.

Chemical structure of biotin. Biotin, also known as B-vitamin B7 (formerly vitamin H and coenzyme R) is water soluble. The molecule is comprised of a ureido ring joined with a tetrahydrothiophene ring. A coenzyme for carboxylase enzymes, biotin is required for the synthesis of fatty acids, isoleucine and valine. Biotin is also involved in in gluconeogenesis.

Proteins that are biotin labeled (i.e., biotinylated) are routinely detected or purified with avidin conjugates in many protein research applications, including the enzyme-linked immunosorbent assay (ELISA), western blot analysis, immunohistochemistry (IHC), immunoprecipitation (IP) and other methods of affinity purification, cell surface labeling and flow cytometry/fluorescence-activated cell sorting (FACS).

Besides a strong affinity for avidin, biotin exhibits two characteristics that make the molecule ideal for labeling proteins and macromolecules. First, biotin is comparatively smaller than globular proteins, which minimizes any significant interference in many proteins and allows multiple biotin molecules to be conjugated to a single protein for maximum detection by avidin. Second, as shown in the diagram below, biotin has a valeric acid side chain that is easily derivatized and conjugated to reactive moieties and chemical structures without affecting its avidin-binding function. This feature allows many useful biotinylation reagents to be created.

Biocytin is a derivative of biotin found in serum and urine that has an added lysine group coupled at the ε-amino acid side chain to the valeric acid side chain. As shown below, biocytin is longer than biotin, which makes the molecule useful in making long-chain biotinylation reagents. Biocytin can also be used to make trifunctional crosslinking reagents because of the free carboxylate group and α-amine.

Comparison of biotin and biocytin. Biocytin differs from biotin by the addition of a lysine group attached to the valeric acid side chain.

Biotinylation, also called biotin labeling, is most commonly performed through chemical means, although enzymatic methods are also available. Chemical methods provide greater flexibility in the type of biotinylation needed than enzymatic approaches and can be performed both in vitro and in vivo. Enzymatic methods require the co-expression of bacterial biotin ligase and an exogenously expressed protein of interest that is modified to carry a biotin acceptor peptide, which provides a more uniform biotinylation than chemical methods and can be cell compartment specific. Because of the greater availability of chemical biotinylation reagents and customization, though, this article focuses solely on chemical methods of biotinylation.

All biotinylation reagents have similar features, as diagrammed below, and variations in these features give biotinylation reagents distinct characteristics that are ideal for different types of experiments. Of key distinction is a reactive moiety or group (indicated in red) that crosslinks the biotinylation reagent to either distinct amino acid functional groups or non-distinct domains available on all amino acids; the reactivity of a given reagent is dependent upon the reactive group used. The distance between this reactive moiety and the biotin molecule (indicated in blue) can also be adjusted to increase the availability of biotin for avidin binding, increase the solubility of the reagent, or to make the biotinylation reversible. The structure from the site of amino acid binding, which depends on the type of reactive moiety, to the end of the biotin molecule is called the spacer arm; the distance of these spacer arms varies between biotinylation reagents.

Determining the right biotinylation reagent to use for a specific application and protein of interest depends on a number of factors that must be carefully considered and optimized, including:

• Solubility—Modifications in biotinylation reagent solubility allow access to target proteins in hydrophobic or hydrophilic environments and influence the solubility of the biotinylated protein.
• Spacer arm length—The availability of biotin for avidin binding is influenced by the length of the spacer arm.
• Cleavability/reversibility—Captured biotinylated proteins can be recovered or purified by cleaving the biotin molecule from the target protein or reversing attenuated biotin–avidin interactions.
• Functional group—Specific reactive moieties bind to any amino acids or to the functional groups of specific amino acids for nonselective or targeted biotinylation, respectively.

(TOP)

The solubility of a biotinylation reagent greatly influences the ability to label target proteins or other macromolecules. Proteins have hydrophobic and hydrophilic regions based on the amino acid side chains and protein conformation, and these regions can promote or restrict biotinylation based on the solubility of the reagent. Furthermore, the hydrophobicity of the target protein microenvironment can either prevent or allow biotinylation based on the solubility of the reagent. For example, surface biotinylation, which is a common method to study the expression or endosomal trafficking of surface molecules, requires that the biotinylation reagent be hydrophilic to prevent it from crossing the hydrophobic cell membrane, thereby restricting biotin labeling to the surface of the cell. Therefore, the selection of the proper biotinylation reagent is dependent upon both the targeted amino acids and the microenvironment of the protein or macromolecule.

Biotinylation reagent solubility is based on the solubility of the reactive moiety, the spacer arm or a combination of both. Some reactive groups are inherently charged and are therefore water-soluble, while uncharged groups require modification (e.g., sulfonation; see section on NHS esters below). A common method to make the spacer arms soluble is to incorporate poly(ethylene glycol), or PEG (see figure below), which is highly soluble and flexible. Spacer arms comprised of PEG can make biotinylation reagents with uncharged reactive groups soluble or those with charged reactive groups more soluble. Additionally, the increased solubility stemming from PEG-containing biotin tags helps prevent biotinylated protein aggregation during long-term storage compared to non-biotinylated proteins.

Poly(ethylene glycol) increases the solubility of biotinylation reagents. A four-ethylene glycol chain (PEG4) was conjugated onto the spacer arm between the reactive moiety (red) and biotin (blue). The reagent shown is Thermo Scientific EZ-Link NHS-PEG4-Biotin.

The ability of avidin to bind to biotin molecules on biotinylated proteins is dependent upon the availability of biotin without steric hindrance from multiple biotins on the same protein. Longer spacer arms can enhance the detection sensitivity of the target protein because more biotin molecules are available for reporter-conjugated avidin binding. The definition of the spacer arm in the context of biotinylation should not be confused with that for the spacer arms used with bifunctional crosslinkers. The spacer arm used for crosslinking is defined as the distance between the two reactive moieties. For biotinylation, the spacer arm is defined as the distance from the end of the conjugated amino acid to the end of the biotin molecule. Thus, the spacer arm length of a biotinylation reagent can be the difference between strong protein detection/purification and weak or no detection/purification because of the availability or unavailability of biotin for binding to avidin conjugates, respectively.

As illustrated in the figure below, spacer arm length is varied in the construction of biotinylation reagents by adding or removing chemical structures between biotin and the reactive moiety. The functional length of the spacer arm depends on both the length of the added chemical structures and the nature of the reactive moiety. Spacer arm length is regulated by hydrocarbons, PEG, or disulfide bonds that allow biotin cleavage.

Examples of variable spacer arm lengths. Chemical groups (black) modify the distance between the reactive moiety (red) and biotin (blue) to regulate the length of the spacer arm. The reagents shown are (A) Thermo Scientific EZ-Link NHS-Biotin, (B) NHS-LC-Biotin and (C) Sulfo-NHS-LC-LC-Biotin.

(TOP)

The strength of the avidin–biotin interaction and its resistance to dissociation influences the ability to elute biotinylated proteins from an avidin-immobilized support or remove avidin-conjugated reporter probes from a biotinylated sample. Harsh, denaturing conditions (8M guanidine•HCl, pH 1.5 or boiling in SDS-sample loading buffer) are required to efficiently dissociate avidin-biotin complexes, which damages the support and denatures the proteins so that they do not maintain any biological activity. To overcome this limitation, various advancements have been developed over the years.  One approach is to use modified versions of biotin such as cleavable biotin, iminobiotin and desthiobiotin. Another approach is to modify the avidin/streptavidin resin so that it exhibits lower affinity towards biotin.

Cleavable biotinylation reagents allow the purification of biotinylated proteins after capture or the removal of avidin-conjugated protein from a biotinylated sample. As shown below, cleavable biotinylation reagents are designed with disulfide bonds in the spacer arm. Under reducing conditions (50 mM dithiothreitol, 10 mM 2-mercaptoethanol or 1% sodium borohydride), the disulfide bonds are cleaved, releasing the biotin tag and any avidin conjugate bound to it.

Cleavable biotinylation reagents. These reagents are designed with a disulfide bond that is cleaved under reducing conditions (dotted line) to release the biotin tag (blue) from the tagged protein (green; not to scale).

Pyridyldisulfide biotinylation reagents, such as Thermo Scientific EZ-Link HPDP-Biotin, are activated for labeling sulfhydryl groups via disulfide bond formation. Therefore, disulfide bond reduction not only releases the once-labeled protein from avidin, it also exactly reverses the labeling reaction.

An alternative to using cleavable biotinylation reagents to release purified proteins is labeling them with modified versions of biotin. Iminobiotin is a cyclic guanidino analog of biotin is elutable from avidin because of its considerably weaker binding affinity for avidin (Ka=10⁸M^-1 vs. Ka=10¹⁵M^-1, respectively) that is pH-dependent. Iminobiotin-tagged proteins bind to avidin conjugates at pH 9, but the avidin–iminobiotin complexes dissociate at pH 4 to allow the captured protein to be purified without denaturation. Although the purified proteins are still biotinylated, this method releases biologically functional proteins from avidin conjugates.

Chemical structure of iminobiotin agarose. Each porous iminobiotin agarose bead is 45 to 165 µm in diameter and contains trillions of iminobiotin groups.

Desthiobiotin is a single-ring, sulfur-free analog of biotin that binds to streptavidin with nearly equal specificity but less affinity than biotin (Ka=10¹¹M^-1 vs. Ka=10¹⁵M^-1, respectively). Consequently, desthiobiotinylated bait proteins and their interacting partners can be eluted readily and specifically from streptavidin affinity resin using mild conditions based on competitive displacement with free biotin. For pull-down assay experiments with biological samples, this soft-release characteristic of desthiobiotin also helps to minimize co-purification of endogenous biotinylated molecules, which remain bound to streptavidin upon elution of the target protein complexes with free biotin. The modified avidin–biotin affinity system also eliminates the need to use harsh elution conditions that might disassociate complexes and/or damage the target protein or cell. Desthiobiotin-based techniques are ideal when using native or recombinant proteins that are not expressed with a fusion tag and when isolating captured proteins under native conditions, such as targeting intact cells or cell surface proteins.

The soft-release characteristic of desthiobiotin above reagents minimize the isolation of naturally biotinylated molecules that can interfere with results and also eliminates the use of harsh elution conditions which can disassociate complexes and/or damage the target protein or cell. Various soft-release options have been utilized to develop protocols for a variety of applications (e.g., active site probes, RNA-protein pull-down).

(TOP)

The goal of biotinylation is to label a protein of interest in such a way that the normal biological function of the protein is not significantly interrupted. Even though biotin is small, biotinylation can interfere with normal protein function if the biotinylation reagent is conjugated to amino acids that regulate protein activity such as binding to substrates. Many different reactive moieties are available to reduce potential interference by targeting different specific amino acid functional groups, including:

• Primary amines (-NH2)—This group is located at the N-terminus of each polypeptide chain and in the side chain of lysine (Lys, K) residues.
• Sulfhydryls (-SH)—This group is located in the side chain of cysteine (Cys, C). Often, as part of a protein's secondary or tertiary structure, cysteines are joined together via disulfide bonds (–S–S–) between their side chains.
• Carboxyls (-COOH)—This group is located at the C-terminus of each polypeptide chain and in the side chains of aspartic acid (Asp, D) and glutamic acid (Glu, E).
• Carbonyls (-CHO)—This aldehyde group can be created by oxidizing carbohydrate groups in glycoproteins.

Nonselective biotinylation reagents are also available to label proteins or macromolecules with no available primary amines, sulfhydryls, carboxyls or carbonyls.

As shown in the antibody diagram below, proteins commonly contain sites for multiple reactive moieties. Therefore, the selection of the right reactive moiety is critical for sufficient labeling that does not interfere with protein function.

Protein functional group targets located on a representative protein. This illustration depicts the generalized structure of an immunoglobulin (IgG) protein.  Heavy and light chains are held together by a combination of non-covalent interactions and covalent interchain disulfide bonds, forming a bilaterally symmetric structure. The V regions of H and L chains comprise the antigen-binding sites of the immunoglobulin (Ig) molecules. Each Ig monomer contains two antigen-binding sites and is said to be bivalent. The hinge region is the area of the H chains between the first and second C region domains and is held together by disulfide bonds. This flexible hinge (found in IgG, IgA and IgD, but not IgM or IgE) region allows the distance between the two antigen-binding sites to vary. Also shown are several functional groups that are selectable targets for practical bioconjugation.

The reactive groups listed in the section below are commonly used for all crosslinking applications and are described in detail in a separate Crosslinking section of the Protein Methods Library. Therefore, only a brief discussion of the mechanism of each reactive moiety is discussed in this section, with more emphasis being on these reactive groups in the context of biotinylation reactions.

Amines are the most commonly targeted functional groups for biotinylation because of the abundance of lysine side chain ε-amines and N-terminal α-amines. N-hydroxysuccinimide (NHS) esters readily form stable bonds with primary amines, and the reactive group is easily incorporated and stabilized into a variety of useful, ready-to-use biotinylation reagents. NHS-esters do not carry a charge and must first be dissolved in an organic solvent and then diluted into the aqueous reaction mixture. These reagents can cross the hydrophobic cell membrane to conjugate to and therefore biotinylate proteins restricted to intracellular compartments or in hydrophobic microenvironments.

NHS-esters can be modified to be water-soluble by sulfonating the N-hydroxysuccinimide ring to form sulfo-NHS esters, which do not require organic solvents to be used in aqueous reactions. Sulfo-NHS-ester biotinylation reagents carry a charge and therefore cannot cross intact cell membranes; surface biotinylation experiments rely on this characteristic to limit biotinylation to extracellular proteins only.

Tetrafluorophenyl (TFP) esters comprise a commonly used amine-reactive group that has similar reactivity with primary amines but is more hydrophobic than NHS. TFP biotinylation reagents can function at a slightly more alkaline pH than NHS esters and show greater stability against hydrolysis in aqueous solutions.

(TOP)

Sulfhydryl groups, which are found in exposed cysteine residues, are the second most-common targets for biotinylation. Because sulfhydryls are usually less prevalent in proteins, they are often targeted for biotinylation when primary amines are located in protein active sites; this usually results in more limited labeling than when biotinylating primary amines, but the biological function of the protein can be preserved.

Sulfhydryl-reactive biotinylation reagents require free sulfhydryl groups. Disulfides will not react with these reagents until the sulfhydryl groups are freed under reducing conditions. Alternatively, lysines can be modified with free sulfhydryl groups using thiolation reagents such as Traut's Reagent, SATP, SAT(PEG)4 or SATA. Because these reagents target free sulfhydryls, biotinylation reactions must be performed in buffers that lack reducing agents. EDTA can also be included in the buffer formulation to chelate trace metals that promote disulfide bond formation.

While hydrophobic NHS esters can be modified to make them water-soluble, sulfhydryl-reactive biotinylation reagents can only become soluble in water by the addition of a hydrophilic spacer arm.

Maleimide groups are highly reactive towards sulfhydryl groups at acidic to neutral pH. BMCC-Biotin (1-biotinamido-4-[4'-(maleimidomethyl)cyclohexane-carboxamido]butane) is a biotinylation reagent that has a maleimide moiety and a cyclohexane ring for greater stability during conjugation and increased spacer arm length. Haloacetyls like iodoacetyl groups also are highly reactive towards sulfhydryl groups but at a higher pH than maleimide (pH 7.5–8.5).

Pyridyl disulfides are distinguished from other sulfhydryl-reactive groups, because the reaction results in the formation of a disulfide bond that can be cleaved to release the biotin spacer arm and purify biotin-free protein. Additionally, the crosslinking reaction releases pyridine-2-thione, which can be detected to monitor the progression of the reaction.

Sulfhydryl-reactive groups. Maleimide, iodoacetyl and pyridyl disulfide groups (all indicated in red) are the most common sulfhydryl-reactive groups used to label proteins with biotin (blue). The reagents shown are (A) Thermo Scientific EZ-Link BMCC-Biotin, (B) Iodoacetyl-LC-Biotin and (C) HPDP-Biotin (the dotted line indicates cleavage site with thiol reductant).

Carboxyl groups are found on the carboxy-terminal ends of proteins and on aspartate and glutamate side chains. Unlike biotinylation reagents that directly react with primary amines and sulfhydryls, though, those that target carboxyl groups require a zero-length crosslinker such as EDC (a carbodiimide) to conjugate to primary amines on the biotinylation reagents. So while the amines on carboxyl-reactive biotinylation reagents are not reactive per se, they are the site of conjugation to target proteins. Besides amines, biotinylation reagents with hydrazide moieties can also be used with EDC to react with carboxyl groups.

Conjugation via EDC occurs at pH 4.5–5.5 and requires buffers devoid of primary amines (e.g., Tris, glycine) and carboxyls (e.g., acetate, citrate). One of the few buffers that meets these criteria is 2-[morpholino]ethanesulfonic acid (MES) buffer. The use of EDC in biotinylation reactions can cause polymerization if the target protein contains both carboxyls and primary amines; thus, optimization of the concentrations of EDC and the biotinylation reagent are required. Additionally, the yield of biotinylation can be considerably increased with sulfo-NHS, which stabilizes intermediate products.

(TOP)

Although carbonyls do not readily exist in proteins, carbohydrate residues on glycoproteins can be modified to aldehydes to be labeled with hydrazide or alkoxyamine derivative biotinylation reagents. Aldehydes on these glycoproteins are generated by the oxidation of carbohydrate sialic acids using sodium periodate. The aldehydes are then reacted specifically with a hydrazide or alkoxyamine at pH 4–6 to form a stable linkage.

Sialic acid residues can also be biotinylated with hydrazide or alkoxyamine derivatives by pretreatment with neuraminidase to generate galactose groups. The galactose and N-acetylgalactosamine residues on whole cells can be selectively biotinylated with these reagents by further treatment with galactose oxidase, which converts the primary hydroxyl groups on these sugars to their corresponding aldehydes.

Mild oxidation of an immunoglobulin with sodium periodate produces reactive aldehydes from the carbohydrate moieties on the Fc portion of the antibody, which then can be alkylated by a hydrazide. This approach is advantageous for use with antibodies because they become biotinylated in a manner that maintains immunological reactivity. This is an ideal method for biotinylating polyclonal antibodies because they are heavily glycosylated. Monoclonal antibodies may be deficient in glycosylation, and therefore success with this method will depend on the extent of glycosylation for a particular antibody.

The temperature, pH of oxidation and the periodate concentration all affect the reaction with hydrazide derivatives of biotin. Also, because glycosylation varies with each protein, optimum conditions must be determined for each glycoprotein. Each glycoprotein preparation has an optimum pH for oxidation and for the hydrazide-mediated biotinylation. Tris or other primary amine-containing buffers are not recommended for use in either the oxidation or biotinylation steps, because these buffers react with aldehydes and thus quench their reaction with hydrazides and alkoxyamines.

Carboxyl- and carbonyl-reactive groups. Reagents with these reactive groups (red) label carboxyl groups with biotin (blue)when used with EDC. Hydrazides and alkoxyamines can also be used to label aldehydes on carbohydrates of glycoproteins after oxidation by periodate. The reagents shown are (A) Thermo Scientific EZ-Link Amine-PEG2-Biotin, (B) Hydrazide-Biotin and (C) Alkoyxamine-PEG4-Biotin.

Nonselective, photoreactivatable biotinylation reagents are available to label target proteins without available amines, sulfhydryls, carboxyls and carbohydrates. Most photoreactive biotinylation reagents are based on aryl azides, which become activated by UV light (>350nm) and initiates an addition reaction to insert into C-H and N-H sites. Subsequent ring expansion drives the reaction towards binding to nucleophiles, such as primary amines. These reagents can be used in a wide variety of buffers, although acidic pH and reducing conditions inactivate the aryl azide.

Usually, photoactivatable reagents are chosen when primary amines and other functional groups are lacking or when the initiation of conjugation must be timed to a particular point in an incubation period (i.e., by exposure to UV light).

Nonselective, photoactivatable groups. The reactive group (red) is activated by UV light and then nonselectively binds to amino acids to label the protein with biotin (blue). The reagent shown is Thermo Scientific EZ-Link TFPA-PEG3-Biotin.

(TOP)

Determining the extent of biotin modification after a biotinylation reaction can aid in optimizing a particular avidin-biotin assay system and ensure reproducibility in the biotinylation process. The most common method to measure the degree of biotinylation of a sample is using 4'-hydroxyazobenzene-2-carboxylic acid (HABA) dye, which noncovalently binds to avidin in the absence of biotin. When bound to avidin, HABA exhibits a wavelength absorbance at 500 nm (A500), which is proportional to the amount of bound HABA. When a biotinylated sample is mixed with the HABA–avidin complex solution, the biotin displaces HABA for binding to avidin because the association constant of the avidin–biotin interaction is much greater than that for HABA–avidin (6 x 10⁶ M^-1). Because the absorbance of HABA is proportional to its binding to avidin, the amount of biotin present in the solution can be calculated based on the reduction in the A500 signal.

More sensitive assays have been developed in recent years that are based on the same principle of HABA displacement but use fluorescent reporters. Fluorescent biotin assays are significantly more sensitive and require less biotinylated sample than traditional HABA methods. Fluorescent detection requires a fluorescent plate reader, while non-fluorescent methods can be performed with a standard spectrophotometer.

(TOP)

1. Barat B, Wu AM (2007) Metabolic biotinylation of recombinant antibody by biotin ligase retained in the endoplasmic reticulum. Biomol Eng 24:283–91.
2. Emerman AB et al. (2010) Compartment-restricted biotinylation reveals novel features of prion protein metabolism in vivo. Mol Biol Cell 21:4325–37.
3. Hermanson GT (2008) Bioconjugate techniques, 2nd Edition. San Diego (CA): Academic Press.
4. Hofmann K et al. (1980) Iminobiotin affinity columns and their application to retrieval of streptavidin. Proc Natl Acad Sci U S A 77:4666–8.
5. Sugawara K et al. (2005) Voltammetric homogeneous binding assay of biotin without a separation step using iminobiotin labeled with an electroactive compound. Anal Sci 21:897–900.
6. Hirsch J et al. (2002) Easily reversible desthiobiotin binding to streptavidin, avidin, and other biotin-binding proteins: uses for protein labeling, detection, and isolation. Analytical Biochemistry 308:343–357.