Lennarz W.J., Lane M.D. (eds.) Encyclopedia of Biological Chemistry. Four-Volume Set . V. 4

Подождите немного. Документ загружается.

proliferation and cytokine production in response to

their respective ligands expressed on dendritic cells. For

example, mice deﬁcient in the OX40–OX40L inter-

action show impaired T-cell proliferative capabilities

and cytokine production in response to viruses such as

lymphocytic choriomengiitis virus (LCMV) or inﬂuenza

virus, and the ability to generate or sustain memory

T cells is diminished. Additionally, some evidence

indicates that several receptors such as CD30, Fas, or

HVEM may play a role in positive and negative selection

in thymocyte differentiation. Thus, TNFR family

members can contribute to immune tolerance.

On B cells, CD40 is the most prominent costimula-

tory receptor where it interacts with its ligand, CD40L,

on T cells to initiate T-cell-dependent B-cell antibody

responses. Mutations in CD40 or its ligand affect a

number of immune functions including immunoglobulin

class switching; in humans, X-linked hyper-IgM syn-

drome is due to defects in CD40L structure and function

manifesting as an inability to switch immunoglobulin

M to IgG classes.

ORGANOGENESIS AND DEVELOPMENT

A number of TNFRs play roles in the organization of

tissues. The lymphotoxin system mediated by LT

and its ligand LT

, play a critical role in lymph node

development; mice deﬁcient in either receptor or ligand

completely lack lymph nodes and Peyer’s patches.

RANK and its ligand RANKL mediate osteoclast

differentiation and therefore are important for main-

taining normal bone homeostasis. Osteoporosis, a

condition of bone thinning, can be caused by over-

expression of the soluble RANKL decoy receptor,

OPG, which is induced by estrogen and prevents

normal bone formation. RANK-RANKL signaling also

supports the development of mammary gland alveolar

buds important for lactation, and consequently an

absence of RANK induces accelerated apoptosis of

mammary epithelial precursors. EDAR, XEDAR, and

TROY, and the ligand EDA, regulate the development

of ectodermal tissues including hair follicles. In fact,

mice or humans with deﬁciencies in EDAR or EDA

lack primary hair follicles and sweat glands. Finally,

NGFR is important for the development of sensory

neurons; acting alone it induces apoptosis, but upon

association with other nerve growth factor receptors it

mediates the differentiation and survival of neurons.

NGFR-deﬁcient mice develop cutaneous sensori-

neuronal defects.

Viral Targeting of TNFR

Since TNFRs and their ligands play a critical role in

mediating immune defenses against invading pathogens,

and control death and survival fates for cells, it is not

surprising that viruses have evolved mechanisms that

directly target the TNF family in order to subvert

immune effector mechanisms. In fact, viruses when

viewed collectively have targeted virtually every step

of the TNFR signaling pathway, from ligand binding

to activation of apoptosis. For example, infection by

some viruses, including cytomegalovirus (CMV),

herpes simplex virus (HSV), adenovirus, and HIV,

induce the expression of FasL and/or TRAIL on the

host cell resulting in the probable killing of recruited

immune effector cells that express the cognate TNFR

death receptor. In contrast, adenovirus down-regulates

the proapoptotic receptors Fas and TRAILR from the

surface of infected cells, thereby preventing the killing

of the host cell harboring the virus. Another

mechanism of immune evasion employed by pox-

viruses is the expression of soluble, secreted orthologs

of TNFR2 or CD30 that inhibit the interaction of

their respective ligands with their cellular receptors.

In some cases, viruses and TNFR signaling have been

adapted to function cooperatively. For instance,

signaling through the LT

R or TNFR1 can prevent

the replication of human CMV in ﬁbroblasts without

inducing death of the infected cell thereby allowing

continued existence of CMV within the host. This

molecular example of host– virus coexistence might

provide a mechanism for CMV to persist latently in

individuals with a strong host immune system, and re-

emerge to cause disease in individuals who are

immunosuppressed.

FURTHER READING

Benedict, C. A., Banks, T. A., and Ware, C. F. (2003). Death and

survival: Viral regulation of TNF signaling pathways. Curr. Opin.

Immunol. 15, 59–65.

282 TUMOR NECROSIS FACTOR RECEPTORS

Bodmer, J., Schneider, P., and Tschopp, J. (2002). The molecular

architecture of the TNF superfamily. Trends Biochem. Sci. 27,

19–26.

Chung, J. Y., Park, Y. C., Ye, H., and Wu, H. (2002). All TRAFs

are not created equal: Common and distinct molecular mech-

anisms of TRAF-mediated signal transduction. J. Cell Sci. 115,

679–688.

Dejardin, E., Droin, N. M., Dehase, M., Haas, E., Cao, Y., Makris, C.,

Li, Z. W., Karin, M., Ware, C. F., and Green, D. R. (2002). The

lymphotoxin-

receptor induces different patterns of gene

expression via two NF

B pathways. Immunity 17, 1–11.

Locksley, R. M., Killeen, N., and Lenardo, M. J. (2001). The TNF and

TNF receptor superfamilies: Integrating mammalian biology. Cell

104, 487– 501.

Orlinick, J. R., and Chao, M. V. (1998). TNF-related ligands and their

receptors. Cell. Signal. 10, 543–551.

BIOGRAPHY

Karen G. Potter is an NIH Postdoctoral Fellow in the lab of Dr. Carl F.

Ware at the La Jolla Institute for Allergy and Immunology. She received

her Ph.D. in Cell and Molecular Biology from Duke University in July

2002. Her research interests include the regulation and immunomo-

dulatory functions of the TNFR HVEM.

Carl F. Ware is Head of Molecular Immunology at the La Jolla Institute

for Allergy and Immunology, and a Professor of Biology at UCSD. His

research program investigates the regulation of immunity by cytokines

and viruses. He is an advisor for the National Institutes of Health and is

an editor of scientiﬁc journals. Dr. Ware received his Ph.D. in

Molecular Biology and Biochemistry from the University of California

Irvine, and was an NIH Postdoctoral Fellow at the Harvard Medical

School.

TUMOR NECROSIS FACTOR RECEPTORS 283

Two-Dimensional Gel

Electrophoresis

Gerhard Schmid, Denis Hochstrasser and Jean-Charles Sanchez

Biomedical Proteomics Research Group, Geneva University Hospital, Geneva, Switzerland

Two-dimensional polyacrylamide gel electrophoresis (2D

PAGE) is a high-resolution protein separation technique

exploiting two independent physico-chemical properties

(charge and size) of the protein components to be separated.

During the ﬁrst dimension, under the inﬂuence of an electrical

ﬁeld, charged proteins migrate in a pH gradient until each of

them reaches a distinct pH value, which corresponds to its

isoelectric point (pI). At its pI, a protein has zero net charge.

This separation technique is called isoelectric focusing (IEF). In

the second dimension, the protein mixture previously separ-

ated by IEF is sorted orthogonally by electrophoresis in the

presence of sodium dodecyl sulfate (SDS). Migration of SDS-

coated proteins in a sieving polyacrylamide gel results in a

separation according to their relative molecular mass (M

Nowadays, 2D PAGE is a powerful and widely used tool for

the analysis of complex soluble protein mixtures extracted

from cells, tissues, or other biological samples.

Some Milestones in the

Development of 2D PAGE

HOW IT ALL BEGAN

In 1975, three papers (Klose, O’Farrell, Scheele)

introduced two-dimensional polyacrylamide gel electro-

phoresis (2D PAGE) using denaturing isoelectric focus-

ing (IEF) and SDS-PAGE. At that time, IEF was carried

out in cylindrical rod IEF gels cast in glass capillary tubes

(1–1.5 mm inner diameter). The gels contained syn-

thetic carrier ampholytes, which formed a continuous

pH gradient under the inﬂuence of an electric current.

After IEF the gel rods had to be removed from their tubes

for second dimension separation by SDS-PAGE.

TECHNICAL IMPROVEMENTS

With the introduction of immobilized pH gradients

(IPGs) by Bjellqvist et al. in 1982, problems associated

with carrier ampholyte pH gradients such as gradient

instability, low sample loading capacity, and difﬁculty to

achieve reproducibility could be eliminated. In an IPG, a

set of acidic and basic buffering groups is covalently

incorporated into a polyacrylamide gel at the time it is

cast. Precast IPG gel strips supported by a plastic ﬁlm

backing are commercially available in a variety of

narrow and broad, linear and nonlinear pH ranges.

This represents an important development and allowed

2D PAGE to become the tool of choice for high-

resolution protein separation (Figure 1). Besides

improvements in the 2D technique, critical develop-

ments in other ﬁelds greatly contributed to the more

widespread use of 2D PAGE. In 1987, Matsudaira

showed that classical N-terminal sequencing by Edman

degradation can be applied to picomole quantities of

proteins electro-transferred onto polyvinylidene diﬂuor-

ide (PVDF) membranes. In 1993, several groups

independently showed that mass spectrometry allows

the identiﬁcation of proteins previously separated by 1D

or 2D gel electrophoresis. Improvements on the level of

the mass spectrometers as well as rapid access to genome

sequence databases for various species from across the

world nowadays allow the high-throughput identiﬁ-

cation of hundreds of proteins separated by 2D PAGE.

More powerful, less expensive computers and appro-

priate software made computer-assisted evaluation of

the highly complex 2D patterns possible.

2D PAGE IN THE FUTURE

After more than 25 years, one can say that 2D PAGE

ﬁnally became a routine protein separation method;

however, its further development is a continuous process.

A major interest, especially for the application of 2D

PAGE in the clinical laboratory, lies in the ﬁeld of

miniaturized and automated 2D protein mapping.

Recently, a fully automated 2D electrophoresis system

(a2DE

) has been presented by NextGen Sciences.

A microchip format that is based on 2D PAGE (digital

ProteomeChip

from Protein Forest Inc.) allows

protein separation, staining, and digital imaging in as

little as 20 min, with a detection sensitivity below 1 pg of

protein.

The Current Protocol

The method described by O’Farrell back in 1975 already

contained the magic bullets of 2D PAGE, i.e., urea,

dithiothreitol (DTT), a nonionic or zwitterionic deter-

gent for the ﬁrst and SDS for the second dimension.

The procedure of 2D PAGE using IPG gel strips in the

ﬁrst dimension is largely based on the work of Go

and co-workers.

SAMPLE SOLUBILIZATION

An ideal procedure results in the complete solubili-

zation, disaggregation, denaturation, and reduction of

all the proteins in the sample. However, dealing with

several thousand different proteins with different

physico-chemical properties makes complete solubili-

zation practically impossible. In general, cells or tissues

are disrupted by various techniques such as grinding in a

liquid nitrogen cooled mortar, sonication, shearing-

based methods, or homogenization. Proteins are

then solubilized in a buffer containing 9 M urea, 4%

3-[(3-cholamidopropyl)dimethylammonio]-1-propane-

sulfonate (CHAPS), 50–100 mM DTT. Various minor

modiﬁcations from this basic, IEF compatible solubil-

ization buffer have been described over the years, trying

to improve the solubilization of certain subclasses of

proteins. DNA or RNA molecules, which can interfere

with 2D PAGE, are removed by incubation with

nucleases. Centrifugation is used to separate the

solubilized proteins from nonsolubilized material. Pre-

fractionation of complex samples is a good choice when

only a subset of the proteins in a tissue or a cell is of

interest or when abundant proteins dominate the sample

as in the case of albumin in plasma.

FIRST DIMENSION: IEF

IEF separates proteins according to their isoelectric

point. Proteins diffusing away from their pI immediately

gain charge and migrate back. This focusing effect

allows proteins to be separated on the basis of very small

charge differences. Simpliﬁed handling and improved

performance made the commercially available precast

IPG gel strips (from 7 to 24 cm in length) the method of

choice for most applications. IPG strips with a pH range

from 3 to 10 will display a wide range of proteins,

whereas narrower pH ranges allow to zoom into a

particular pH range. With a set of narrow range,

overlapping pH gradients, a wide pH range can be

covered at increased resolution. There are several ways

to load the sample on the IPG gel strip. Efﬁcient entry of

the solubilized proteins, especially hydrophobic ones,

into the gel is always a critical point. IPG dry strips need

to be rehydrated before use in IEF. Traditionally, samples

are applied on a discrete point via sample cups

(Figure 2A). Cups are placed at either the acidic or

basic pH extreme of the strip. Under these conditions,

most of the proteins in the sample will be charged and

immediately start migrating under the inﬂuence of an

electric current, thereby minimizing sample loss through

precipitation. Alternatively, the sample can be applied

over the whole IPG gel surface during strip rehydration.

This method is of special interest for more dilute samples

since larger volumes can be loaded. For strip rehydra-

tion, a solution containing 8 M urea, 2% CHAPS,

10 mM DTT, 1% carrier ampholytes, traces of Bromo-

phenol Blue (tracking dye) shows excellent results for

most samples. As for the solubilization solution, various

modiﬁcations were described. A general IEF protocol

starts with the sample entry phase at low initial voltage

(200 V). Voltage is then increased in a stepwise manner

up to the desired ﬁnal focusing voltage (up to 8000 V).

Optimal focusing time depends on the nature of the

sample. In general, overfocusing is not deleterious below

a total of 100 kVh. State-of-the-art IEF systems allow

simultaneous migration of up to 12 strips under

temperature-controlled conditions.

SECOND DIMENSION: SDS-PAGE

Prior to the second dimension, the IPG gel strips are

equilibrated in a solution containing SDS. This anionic

detergent binds to the majority of proteins in a constant

mass ratio (1.4 g SDS/g protein), such that the net charge

per mass unit becomes approximately constant. This

allows protein separation in a polyacrylamide gel

according to their relative molecular mass. For an ideal

transfer of the proteins separated in the IPG strip onto

the second dimension gel, the equilibration step should

result in the complete resolubilization of all proteins in

the IPG gel. Again, the enormous chemical diversity

FIGURE 1 High-resolution protein separation of human plasma by

2D PAGE, silver staining.

TWO-DIMENSIONAL GEL ELECTROPHORESIS 285

present within most samples makes it impossible to use

ideal conditions for all its components. Composition of

the equilibration solution and especially duration of the

equilibration steps represent a compromise between

accessing poorly soluble proteins with a tendency to

adsorb to the gel matrix and diluting highly soluble

proteins out of the gel strip. Equilibration in a buffer

containing 6 M urea, 50 mM Tris–HCl buffer (pH 8.4),

2% w/v SDS, 30% w/v glycerol proved to be a good

compromise. For protein reduction, IPG strips are ﬁrst

incubated in equilibration buffer containing 2% w/v

DTT for 12 min. In a second step, proteins are alkylated

by incubation in equilibration buffer containing 2.5%

w/v iodoacetamide (IAA) and traces of bromophenol

blue (tracking dye) for 5 min. In vertical SDS-PAGE, the

equilibrated strip is placed on the surface of the second

dimension gel and embedded in molten agarose

(Figure 2B). Most commonly, the tris-glycine buffer

system described by Laemmli is used. State-of-the-art 2D

gel electrophoresis systems allow the simultaneous

migration of up to 24 large format gels (20 by 24 cm)

under temperature-controlled conditions.

VISUALIZATION

Most of the staining methods used for conventional 1D

SDS-PAGE can be applied to 2D gels. Because of its very

high sensitivity (down to 1 ng/spot), silver staining is

probably most commonly used. However, the use of

glutaraldehyde in the silver-staining procedure should be

avoided for subsequent protein identiﬁcation by mass

spectrometry. Recently, MS-compatible, ﬂuorescent

stains with sensitivities comparable to silver became

available. At the time being, application of this stains in

large-scale and high-throughput 2D PAGE is hardly

affordable. In addition, special imaging systems are

required, whereas a simple desktop scanner can be

sufﬁcient for “visible” stains. This still leaves plenty of

room for other MS-compatible, but less sensitive stains

such as coomassie blue and zinc negative staining.

Proteins labeled in vivo by the use of radioactive amino

acids (metabolic labeling) can be visualized by exposing

the dried 2D gel to an X-ray ﬁlm or a storage phosphor

screen for use with a PhosphorImager system. As for 1D

SDS-PAGE, proteins from 2D gels can be electro-

transferred on membranes, which are subsequently

probed with antibodies. This so-called Western blot

technique allows highly sensitive and speciﬁc visualiza-

tion of a protein of interest.

Applications of 2D PAGE

The combination of a high-resolution protein separ-

ation tool (2D PAGE) with a highly sensitive protein

identiﬁcation tool (MS) offers an enormous potential.

In 1994, on the occasion of the ﬁrst Siena 2D

Electrophoresis meeting, the expression “proteome”

was proposed by Marc Wilkins for the description of

the proteins expressed by a genome of a species, an

organ, or a cell at a particular moment under

particular conditions. Since then, the ﬁeld of proteo-

mics, i.e., the analysis of proteomes, is booming. The

global approach taken with proteomics allows the

massively parallel analysis of expressed proteins,

including their isoforms originating from posttransla-

tional modiﬁcations (PTMs). Global gene expression

can also be studied at the level of mRNA. However, it

needs to be emphasized that there is not a good

correlation between mRNA abundance and protein

amount in a cell at a given time and that PTMs

FIGURE 2 (A) Sample application for the ﬁrst dimension of 2D

PAGE – samples are loaded in cups at the cathodic end of the IPG strip.

(B) Strip transfer to the second dimension of 2D PAGE – an

equilibrated IPG strip is placed on the top of a large format

polyacrylamide gel.

286 TWO-DIMENSIONAL GEL ELECTROPHORESIS

cannot be predicted from gene sequences. In general,

applications of 2D PAGE can be divided into

descriptive protein mapping without a need for

protein quantiﬁcation and differential analysis, where

protein quantiﬁcation is essential.

PROTEIN MAPPING AND 2D PAGE

ROTEIN DATABASES

The classical approach, 2D PAGE coupled with MS,

allows the establishment of 2D reference maps. One aims

at identifying as many proteins as possible. A typical

workﬂow is depicted in Figure 3. With the number of

completely sequenced genomes increasing daily, protein

identiﬁcation became possible for a growing number of

species. Over the last years, many 2D reference maps

have been established using different tissues, body ﬂuids,

or cell lines as sample (for an overview see http://ch.

expasy.org/ch2d/2d-index.html). Due to the enormous

complexity of cells from higher eukaryotes, characteriz-

ation of their complete proteome in a single step is still

impossible. Isolation of subcellular fractions such as

organelles, macromolecular structures, and multiprotein

complexes represents a possibility to reduce the com-

plexity of the entire cell.

DIFFERENTIAL ANALYSIS

Digitized 2D maps can serve more than just descrip-

tive purposes. With specialized software packages,

quantitative image analysis of the complex 2D patterns

can be performed. This allows one to compare protein

expression levels between different conditions for

thousands of proteins in parallel. In general, one

aims at identifying only the proteins showing differ-

ential expression in the samples under comparison.

Since the early 1990s this approach has been used to

tackle various biological questions. Comparison of

proteomes from tissues or cells representing the

healthy and the diseased state, respectively, allows

the detection of putative molecular markers of the

disease. Such a global approach is particularly useful

in polygenic diseases such as cancer and cardiovas-

cular diseases. Analysis of cell differentiation and

monitoring therapy are other ﬁelds of application of

differential analysis.

FIGURE 3 Typical proteomic workﬂow using 2D PAGE as protein separation tool.

TWO-DIMENSIONAL GEL ELECTROPHORESIS 287

Limitations and Perspectives

of 2D PAGE

In spite of all beneﬁts, 2D PAGE is not without

limitations. On the one hand, there are intrinsic

limitations of 2D PAGE and, on the other hand,

there are general limitations encountered when dealing

with complex protein mixtures.

INTRINSIC LIMITATIONS OF 2D PAGE

Hydrophobic Proteins

Very hydrophobic proteins such as integral membrane

proteins and nuclear proteins were shown to be under-

represented in 2D gels. They have a tendency to stick to

IPG gel matrix, and re-solubilization for the second

dimension becomes a problem. Incorporation of

thiourea, sulfobetaine surfactants, and organic solvents

in the solubilization buffer showed some improvements.

However, very hydrophobic proteins will probably

remain a major limitation of IEF and therefore also of

2D PAGE.

Basic Proteins

Horizontal streaking on the 2D gel is usually observed in

the region with pH . 7. This phenomenon is particularly

prominent when basic narrow range pH gradients are

used. The streaking is due to the disappearance of the

reducing agent from the basic part of the strip.

Oxidation of the protein thiol groups results in the

formation of inter- and intra-chain disulﬁde bridges.

Recently, oxidation of the protein thiol groups to mixed

disulﬁdes using hydroxyethyl disulﬁde (DeStreak

) has

been introduced, showing massively improved focusing

for the very basic proteins.

Automation and Miniaturization

For many years, the low degree in automation and

miniaturization of 2D PAGE has hindered this labor-

intense technique from becoming routinely used in

clinical laboratories and in industry, where high

throughput counts and sample amount is limited.

Promising developments toward both automation and

miniaturization can be observed and their availability

seems to be very close.

GENERAL LIMITATIONS

A biological sample usually represents a complex

protein mixture containing thousands of proteins with

different physico-chemical properties and widely vary-

ing expression levels. In human cells the dynamic range

exceeds 6 orders of magnitude and in human plasma it

reaches more than 12 orders of magnitude. In contrast,

2D gels and silver staining offer a maximum of 4 orders

of magnitude. Consequently, application of unfractio-

nated samples allows one to see only the tip of the

iceberg. In order to access the low abundance proteins,

samples need to be prefractionated, using traditional

biochemical separation techniques, for example.

PERSPECTIVES OF 2D PAGE

For many years, 2D PAGE has been the core technology

for protein separation in proteomics research. This

dominant role is now being challenged by various

non-gel-based approaches such as protein chips, serial

liquid chromatography, or capillary electrophoresis.

They can be very demanding in terms of data manage-

ment andbioinformatic resources, which mightbe limited

for most standard academic research institutions. In fact,

these liquid-based techniques should be considered as

complementary approaches to 2D PAGE, depending on

the biological question one wants to answer. If one aims

at identifying as many proteins as possible in a given

sample, 2D PAGE coupled to MS no longer represents the

method of choice. However, if one aims at differential

analysis of protein expression, 2D gels remain very

powerful. This is mainly due to the possibility to look at

PTMs offered by 2D PAGE. PTMs such as phosphoryl-

ations, glycosylations, or proteolytic processing can be

very important for the function of a protein.

FURTHER READING

Chambers, G., Lawrie, L., Cash, P., and Murray, G. I. (2000).

Proteomics: A new approach to the study of disease. J. Pathol. 192,

280–288.

288 TWO-DIMENSIONAL GEL ELECTROPHORESIS

rg, A., Obermaier, C., Boguth, G., Harder, A., Scheibe, B.,

Wildgruber, R., and Weiss, W. (2000). The current state of two-

dimensional electrophoresis with immobilized pH gradients.

Electrophoresis 21, 1037–1053.

Jung, E., Heller, M., Sanchez, J.-C., and Hochstrasser, D. F. (2000).

Proteomics meets cell biology: The establishment of subcellular

proteomes. Electrophoresis 21, 3369–3377.

Rabilloud, T. (ed.) (2000). Proteome Research: Two-Dimensional Gel

Electrophoresis and Identiﬁcation Methods. Springer, Berlin,

Heidelberg.

Rabilloud, T. (2002). Two-dimensional gel electrophoresis in proteo-

mics: Old, old fashioned, but it still climbs up the mountains.

Proteomics 2, 3–10.

Swiss-2DPAGE on the Expasy Molecular Biology Server – http://ch.

expasy.org/ch2d/.

Wilkins, M. R., Williams, K. L., Appel, R. D., and Hochstrasser, D. F.

(eds.) (1997). Proteome Research: New Frontiers in Functional

Genomics. Springer, Berlin, Heidelberg.

BIOGRAPHY

Gerhard Schmid is a Ph.D. student at the Biomedical Proteomics

Research Group (BPRG) of the Clinical Pathology Department,

Geneva University Hospital.

DenisHochstrasseris the Director of the Clinical Pathology Department.

At the academic level, he is a full Professor both to the Department of

Pathology of the Medicine faculty and to the School of Pharmacy of

the Science faculty. His innovations in the methodology of 2D gel

electrophoresis have contributed decisively to the technique’s becom-

ing one of the main protein separation methods used in proteomics.

Jean-Charles Sanchez is the Head of the BPRG since 1995. Working in

the ﬁeld of proteomics since 1989, he contributed to the development

of 2D gel electrophoresis and its application in biomedical research. He

obtained Ph.D. in Biochemistry at the University of Buckingham (UK)

in the ﬁeld of proteomics and diabetes.

TWO-DIMENSIONAL GEL ELECTROPHORESIS 289

Two-Hybrid Protein–Protein

Interactions

Ilya Serebriiskii and Erica A. Golemis

Fox Chase Cancer Center, Philadelphia, Pennsylvania, USA

The two-hybrid system is an artiﬁcially constructed genetic

system intended to facilitate the detection and assessment of

protein–protein interactions. In the two-hybrid system a host

organism, typically yeast or bacteria, is so engineered as to

contain three components. These are a ﬁrst protein fused to a

DNA-binding domain of known speciﬁcity (hybrid 1); a

second protein fused to a transcriptional activation domain

(hybrid 2), that can interact with the ﬁrst protein, constituting

a functional, albeit composite, transcription factor; and one or

more reporter genes transcribed based on the binding of the

composite transcription factor. Many permutations of the two-

hybrid paradigm have been developed, and two-hybrid systems

have become a mainstay of proteomic investigations.

The Yeast Two-Hybrid System

The purpose of the two-hybrid system is to provide a

genetic system that is capable of scoring the degree of

physical interaction between two proteins of interest.

To this end, a simple organism such as yeast

(S. cerevisiae) or bacteria (E. coli) is engineered such

that the interaction of two proteins generates a

scorable signal, that can be used either to gauge the

interaction afﬁnity of two deﬁned proteins, or to

identify interacting partners for one deﬁned protein

by screening an expression library.

The ﬁrst working example of a two-hybrid system

was described in 1989, by Fields and Song. A schematic

showing the system paradigm is shown in Figure 1.

There are three minimal components for a functioning

two-hybrid system. The ﬁrst component is an expression

construct, which synthesizes a ﬁrst protein of interest as

a fusion with a DNA-binding domain (DBD) that binds

a short DNA sequence of deﬁned speciﬁcity (hybrid 1:

frequently termed the “bait”). The second component is

an expression construct that is used either to express a

second deﬁned protein, or random clones from a cDNA

or genomic library, as a fusion with a transcriptional

activation domain (AD) (hybrid 2: frequently termed the

“prey”). The third component is a reporter cassette,

which consists of the DNA-binding site for the ﬁrst

hybrid protein in the context of a minimal promoter,

upstream of the coding sequence for an easily scored

reporter gene.

For the yeast two-hybrid system to work, there are

two preconditions. First, the bait must not be indepen-

dently functional as a transcriptional activator, such that

yeast containing only the bait and the reporter produce

no reporter signal. Second, the bait and the prey must be

capable of localizing to the cell nucleus; hence, proteins

containing signal sequences that efﬁciently target

them to the plasma membrane, or other non-nuclear

compartments, are not generally acceptable as baits.

However, for proteins that meet these preconditions,

the interaction between the bait and the prey brings the

transcriptional activation domain encoded in the prey to

the promoter sequence bound by the bait, and activates

the transcription of the reporter gene.

A number of groups have built systems that exploit

this two-hybrid paradigm in yeast, with several of

these systems in common use. The most common DBDs

used in these yeast two-hybrid systems are derived

from the yeast transcription factor Gal4p, or from the

bacterial repressor protein LexA. A variety of different

AD sequences have been used, and include the

activation domain from Gal4p; the activation domain

from the viral protein VP16; or “artiﬁcial” activation

sequences encoded from selected fragments of the

E. coli genome.

Two categories of reporter genes have been used. One

category is colorimetric, exempliﬁed by the bacterial

lacZ gene. Activation of a lacZ reporter causes yeast

colonies to turn blue when plated on medium containing

an appropriate substrate, such as X-gal, or can be

assessed quantitatively over a dynamic range in excess of

1000-fold in liquid assays using a spectrophotometer or

luminometer. The second category of reporter provides

an active selection for yeast growth based on the

strength of interaction between two proteins. These

reporters include genes encoding critical components of

yeast metabolic pathways for amino acid synthesis, such

that expression of the gene is required for yeast to grow

on medium lacking the relevant amino acid (e.g., show

an auxotrophic requirement). HIS3, required for growth

on medium lacking histidine, and LEU2, required for

growth on medium lacking leucine, have been most

commonly used. In general, two hybrid systems now

generally incorporate one colorimetric and one auxo-

trophic selection, both responsive to the same bait. The

use of two different reporter genes greatly minimizes the

background of false positive and false negative inter-

actions in the system, and is essential for library

screening applications, where in excess of 10

clones

must commonly be analyzed.

Because of the ease of manipulation provided by a

genetic system, a yeast two-hybrid system provides a

facile means of exploring the interaction of two known

partner proteins. This can be done in a predetermined

manner, by inserting targeted mutations or deletions into

the AD-fused partner, or based on blind selection, for

example, by using PCR mutagenesis or DNA sonication

to develop a randomized library derived from the AD

partner. Each mutant derivative can be readily screened

for interaction phenotype. However, the most important

use of the yeast two-hybrid system throughout the 1990s

has clearly been for the purpose of screening libraries to

identify novel interacting partners for proteins of

interest. In contrast to other means of identifying

interacting proteins, such as biochemical copuriﬁcation,

two-hybrid screening is extremely inexpensive, requiring

only engineered strains of yeast and standard micro-

biological media. It is rapid, providing the potential of

completing a screen from start to ﬁnish in less than a

month. It is effective at yielding returns for a signiﬁcant

percentage of bait proteins (estimated at . 50%).

Finally, following completion of a successful library

screen, plasmids encoding the novel interacting preys

can easily be isolated from yeast, avoiding the need for

protein sequencing or mass spectrophotometric analysis

required by biochemical methodologies. It is not an

exaggeration to note that recent broad adaptation of

two hybrid screening methodologies across the scientiﬁc

community was a major contributing source of elucida-

tion for key cell signaling pathways.

Advanced Applications

Following the validation of the basic two-hybrid system

as a useful and reliable tool for study of protein–protein

interactions, a number of investigators have explored

the potential of the system to study interactions between

proteins and partner proteins under additional selective

constraints. A ﬁrst step in this direction was the

demonstration that it was possible to identify preys

FIGURE 1 Schematic of yeast two-hybrid components, and proﬁle of reporter activation. Details of system components are described in the text.

Three situations are shown: (1) yeast containing only a bait and two reporters, left; (2) yeast containing bait, prey that does not interact with bait, and

reporters, center; and (3) interacting bait and prey, and reporters, right. For each of these three situations, with a typical auxotrophic and colorimetric

reporter, six independent colonies of yeast are dotted on plates with selective medium: expected results (growth/no growth, or color/no color) are

shown as panel inserts below each schematic.

TWO-HYBRID PROTEIN–PROTEIN INTERACTIONS 291