Protein Interaction Analysis - University of Torontolocal.biochemistry.utoronto.ca/stagljar/pdf/The experts talk about... · Protein Interaction Analysis Visit us on the Web at discover.bio-rad.com

6x6

Protein Interaction Analysis

Visit us on the Web at discover.bio-rad.comCall toll free at 1-800-4BIORAD (1-800-424-6723);outside the US, contact your local sales office.

A Powerful EquationThe ProteOn™ XPR36 protein interaction array system, the ultimate in parallel processing.

Bio-Rad introduces the ProteOn XPR36, a surface plasmon resonance optical biosensor. Take the next step from protein discovery to function with label-free protein-protein interaction analysis. XPR™ technologyis a unique 6 x 6 crisscross approach to multiplexing that processes, in parallel, 6 different ligands with 6 different analytes. Perform a complete kinetic analysis, rank a library of antibodies, or test multiple interaction conditions on a single chip, in a single run.

The 6 x 6 interaction array format provides:

� High-throughput parallel sample processing� Full kinetic data� Flexible experimental design

Visit us on the Web at www.bio-rad.com/ad/proteininteraction/

Genome Technology Protein-Protein Interactions 3

www.dualsystems.com

Dualsystems Biotech

Take advantage of the power of yeast protein interaction screening today!

■ Screen full-length integral membrane proteins for interaction partners with our DUALmembrane system

■ Screen soluble proteins or protein fragments with our optimized two-hybrid system

■ Full-length cDNA expression libraries for all screening systems

■ Yeast-based recombinant protein expression

We have the solution:

Looking for novel protein interactions?

All systems available as custom screening services and ready-to-use kits.

Table of contents

Q1: What are the benefits of your preferred assay system(s)? . . . . . . . . . . . . . 5

Q2: How do you eliminate or reduce false positive results? . . . . . . . . . . . . . . . . . 8

Q3: How do you identify structural patterns of multiple interactions? . . . . . . . 9

Q4: What techniques do you use to reduce non-specific interactions? . . . . . . . 10

Q5: What do you do to validate suspected protein-protein interactions? . . . . . . . . . . 11

Q6: Which computational methods do you use to analyze interactions? . . . . . . . 12

Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14

Letter from the Editor . . . . . . . . . . . . . . . . . . . . . . .4

Index of Experts . . . . . . . . . . . . . . . . . . . . . . . . . . .4

4 Protein-Protein Interactions Genome Technology4 Protein-Protein Interactions Genome Technology

For this latest installment ofGenome Technology's technicalreference guide series, we've assembleda team of authorities on protein-proteininteractions. In this guide, thesecontributors share their strategies fordetecting, characterizing, and analyzingproteins on the move.

Every process in a cell is affected by interactionsbetween proteins, which inform everything from theshape of an organelle to the function of a ribosome.Just as we tailor our own conversations depending onsetting, proteins exhibit many different modes ofinteraction. Long-term interactions result in proteincomplexes, while briefer protein liaisons may lead to arange of possible chemical modifications. Proteininteractions are also behind the phenomenon of signaltransduction, by which a message can be relayedacross the cellular landscape.

Because they are so integral to physiologicalfunction, protein-protein interactions are germane tomany lines of research, both basic and clinical. Proteininteraction data may yield important information about

the molecular basis of disease, and many researchersare making moves to use such information in devisingnew therapeutics.

Parsing out the ways in which proteins meet is apretty good starting point for really understanding themolecular conversations that fuel a biological process.That's not to say it's easy. The study of proteininteractions has historically involved expertise inbiology, biochemistry, and biophysics. So although it'sbeyond the scope of this guide to present all of thepossible tactics used in protein studies today, we didattempt to formulate questions that could yieldvaluable advice on detecting and analyzinginteractions, regardless of your discipline.

Keep this guide nearby if you are interested inpicking up a few tricks to spot and monitor proteininteractions in your own system of interest. The expertsbelow cover everything from the relative merits ofdifferent assays to keeping non-specific interactions to aminimum. Also, be sure to check out the resourceguide, which contains reading and resourcesrecommended by our contributors.

— Jennifer Crebs

Letter from the editor

Edward Evans

University of Oxford

Ulf Landegren

Uppsala University

Moritz Rossner

Max Planck Institute of Experimental Medicine

Stephen Michnick

University of Montreal

Igor Stagljar

University of Toronto

Index of experts All of the expert contributors listed below are based inacademic institutions and have no ties to vendorsfeatured in this guide.

John James

University of Oxford

Surface plasmon resonance (SPR) technology was firstcommercially implemented for analyzing protein-protein interactions by Biacore. Biacore experiments, inour case utilizing a Biacore 3000, involve passing asoluble protein or analyte (e.g. a ligand) over animmobilized binding candidate (e.g. a receptor) insidea very small flow-cell, controlled by clever microfluidics.Because SPR, induced by the phenomenon of totalinternal reflection at the base of the flow cell, issensitive to refractive index changes within the flowcell, changes in the SPR signal can be used to monitorbinding-induced changes in protein accumulationwithin the flow cell in real time.

Historically, there had been many ways to detectprotein-protein interactions, such as radio-immunoassay and ELISA. These approaches did notallow binding to be followed in real time and so wereconfined to the analysis of high-affinity interactionsthat could tolerate a wash step, such as antibody orhormone interactions. Such methods were thusunsuited for the analysis of very weak interactions,such as those expected to occur at the cell surface(tfi<1s), as the interactions would "fall apart" in thecourse of washing.

The Biacore was a major advance because itallowed, for the first time, protein interactions to beobserved without a wash step. Although notexplicitly designed to study very weak interactions,PA van der Merwe and colleagues showed that,using the Biacore, it was possible to observe suchinteractions by comparing the amount of binding inthe test flow cell (with test protein immobilized)versus a control flow-cell (in which a negative controlprotein is immobilized). A second advantage is thatthe Biacore collects data rapidly enough to do kinetic

analyses of very fast interactions. The third keyadvantage is that relatively small amounts of proteinare required for analysis (40-100 ml is all that'srequired to get a minimum dataset). This isimportant because, in order to see very weakinteractions, the analyte needs to be at very highconcentrations, which might not be achievable iflarge volumes of protein were required. The BIAcorenow represents the state of the art for studying weakinteractions.

— Edward Evans

Bioluminescence Resonance Energy Transfer (BRET)relies on the same physical principles as its fluorescentcounterpart, FRET, where a donor fluorophore in closeenough proximity, typically <10nm, to an appropriateacceptor fluorophore can transfer energy through anon-radiative process. BRET uses Renilla luciferase asthe donor, a protein that oxidizes its substrate(coelenterazine) and concomitantly emits light in the400-500nm wavelength range, which is in contrast toFRET where the donor is excited by an external lightsource. By genetically linking luciferase and anacceptor, invariably a green fluorescent protein (GFP)variant to proteins of interest, it is possible to identifyprotein interactions and potentially assignstoichiometries to them.

The major benefit of this approach is thatprotein-protein interactions can be monitored in situ,where the donor- and acceptor-tagged proteins areexpressed as a "BRET pair" in an appropriate cell line,often HEK-293T. This is most lucidly seen formembrane-confined proteins, where there is a realpaucity of techniques that can probe proteininteractions without extracting the target from its


What are the benefits of your preferred assaysystem(s)?

6 Protein-Protein Interactions Genome Technology

Question

native environment, which is likely to causesignificant artifacts for highly hydrophobic proteins.BRET also offers several advantages over FRET.Principally, luminescent detection of light emissiongives an exceptional signal/noise ratio for data,which allows protein-protein interactions to bedetected at more physiological expression levels. Theabsence of an external light source to excite donorfluorophore also obviates problems ofphotobleaching, unwanted excitation of the acceptorfluorophore, and other optical effects.

— John James

We are using a method we have developed ourselvesand published recently to investigate interactionsbetween proteins, directly in genetically unmodifiedcells and tissues (Söderberg et al., 2006). This methoddoes not depend on overexpression of fusion proteins,but observes natural proteins detected in fixed tissuesvia antibodies.

— Ulf Landegren

Protein-fragment complementation assays (PCAs) havebeen variously called "split-protein" or, in the case offluorescent protein PCAs, renamed biomolecularfluorescence complementation (BiFC), thoughdescribed method is based on original work of Ghoshet al., 2000. We use the acronym PCA to describe allassays based on protein engineered fragmentcomplementation assays for detecting protein-proteininteractions because it is general and followshistorically from the literature. PSAs:

1) allow for direct detection of protein-proteininteractions in vivo and in vitro in any cell type

2) allow detection of protein-protein

interactions in appropriate subcellularcompartments or organelles

3) allow detection of interactions that arespecifically induced in response todevelopmental, nutritional, environmental,or hormone-induced signals

4) allow monitoring of kinetic and equilibriumaspects of protein assembly in cells

5) allow screening for novel protein-proteininteractions in any cell type

6) is not a single assay, but a series of assays; thus,an assay can be chosen because it works in aspecific cell type that is appropriate for studyingthat class of interaction

7) are inexpensive, requiring no specializedreagents beyond those necessary for aparticular assay and off-the-shelf materialsand technology

8) can be automated, allowing high-throughput screening

9) are designed at the level of the molecularstructure of the enzymes used; because ofthis, there is additional flexibility indesigning the probe fragments to controlthe sensitivity and stringencies of the assays

10) can be based on enzymes whose activity canbe determined by multiple assay strategies,including by survival-selection or productionof a fluorescent product

11) that are based on fluorescent proteins orproteins that bind to fluorescent ligands(e.g. DHFR) can be used to determinelocations of protein complexesunambiguously.

— Stephen Michnick

What are the benefits of your preferred assaysystem(s)?


Our preferred system is the Split-TEV system, whereinteraction partners are fused to inactive fragments ofthe TEV protease. We can monitor interactions withTEV protease specific reporters. The benefits of thesystem are:

1) Flexibility: We can use fluorescent orluminescent reporter proteins that either canbe directly activated by proteolytic cleavageor that are indirectly activated as reportergenes activated by a protease-dependenttranscription factor. Practically, any otherreporter gene may be used if preferred. Wehave already also used resistance-conferringreporter to monitor survival as readout.

2) Sensitivity: By using rapamycin-inducedinteractions of FRB and FKBP at themembrane of living cells, we can detectsignificant activation with 0.001 nMrapamycin added.

3) Full-length proteins: We can monitorinteractions also of full-length membraneproteins in living cells, but the TEV-tags haveto be placed properly.

4) Applicability to living cells: The interaction ismeasured within living cells. So far, we havesuccessfully tested a variety of heterologouscell lines (e.g. COS1, Hela, PC12, NIH3T3,CHO) and primary cells (neurons, astrocytes,ES cells).

— Moritz Rossner

In my lab, we primarily use yeast-based geneticscreening assays to identify and characterize protein-protein interactions (PPIs). These assays include the well-known Yeast Two-Hybrid (YTH) system, originally

invented by Stan Fields in 1989, and the split-ubiquitinMembrane Yeast Two-Hybrid assay (or MYTH)developed in our lab in 1998.

Yeast-based assays have several advantageswhen compared to conventional biochemical assays:since all interactions are detected within growing,intact cells, the likelihood of introducing artifacts dueto the destruction of cellular compartments by celllysis and harsh extraction conditions are diminished.Furthermore, both direct and indirect PPIs can bedetected. In addition, as no washing or purificationsteps are involved, there is greater opportunity todetect weak or transient interactions. Yeast-basedassays also offer flexibility, as demonstrated by thegreat number of variations on the original YTHsystem, which allow, for example, the identificationof protein complexes (so-called yeast three-hybridsystems), modification-dependent interactions (e.g.phosphorylation or acetylation dependent), protein-RNA interactions, protein-DNA interactions, smallmolecule-protein interactions or the identification ofcompounds which alter a PPI.

The major goal of our research is to identifyproteins associated with numerous yeast and humanintegral membrane proteins using the MYTH assay ona systematic scale to provide comprehensivemembrane PPI maps. Due to their pivotal role in manycellular processes, their direct link to human diseasesand their extracellular accessibility to drugs, theidentification of proteins associated with integralmembrane proteins is desirable. However, due to theircomplex chemical properties, membrane proteins aredifficult to purify for in vitro assays and ill-suited foridentification of their interacting partners using thecurrent in vivo assays. Because (continued on p.13)


Question


SPR as we implement it, using a Biacore 3000, is notideally suited to screening large numbers ofcandidate ligands for a protein, because only threecandidates can be tested simultaneously. OtherBiacore machines recently released can be used forthis application. However, SPR is ideal for testingproposed interactions because false positive resultsare extremely rare, provided precautions are taken toexclude non-specific interactions. The purity of theinjected analyte and the immobilized material is, ofcourse, also important. Finally, care must be takenthat the method of immobilization used for thebinding partners will not also capture the analyteindependently of its protein-protein interactions. Togive a trivial example, which we have seen all toooften with new users, if a mouse Fc-fusion constructis immobilized via an antibody against all Fc isotypes,you cannot then test for antibody binding, becauseit too will be captured by the immobilized secondaryantibody.

— Edward Evans

Because a BRET experiment involves the expressionof potential interacting proteins as a BRET pair, it isnot well suited to a high-throughput approach toanalyzing protein-protein interactions. As long ascare is taken to minimize and account for the effectsof non-specific interactions, false positive resultsshould be rare.

— John James

At present our method is not used in high-throughput, and we can use appropriate controls toconfirm that we do not have nonspecific signals.

— Ulf Landegren

1) PCAs are designed at the level of the molecularstructure of the enzymes used; because of this,there is additional flexibility in designing theprobe fragments to control the sensitivity andstringencies of the assays.

2) Non-interacting proteins. A PCA response shouldnot be observed if non-interacting proteins areused as PCA partners; nor should overexpressionof a non-interacting protein alone compete forthe known interaction.

3) Partner protein interface mutations. A point ordeletion mutation of a partner that is known todisrupt an interaction should also prevent a PCAresponse.

4) Competition. A PCA response should bediminished by the simultaneous overexpressionof one or other of the interacting proteins that isnot fused to a complementary PCA fragment.

5) Fragment swapping. An observed interactionbetween two proteins should occur regardless ofwhich interacting proteins is attached to whichPCA fragment.

6) Induction or inhibition of interactions in responseto developmental, nutritional, environmental, orhormone-induced signals linked to the interactioncan validate a novel interaction as beingbiologically relevant to a specific cellular process.


Since we always measure with a kineticallyuncoupled reporter (min range for the direct and hrange for the indirect reporters), we first reduce thetime between transfection and analysis. Anotheroption with our system is to reduce, in transienttranfections, the amount of

How do you eliminate or reduce false positiveresults?

(continued on p.13)

This is what SPR is most suited to doing, i.e. allowingyou to precisely characterize the binding propertiesof a particular protein interaction. With proteins ofknown structure, a series of point mutations canthen be made in one of the partners in order to mapthe binding surface using Biacore-based bindingassays. The obvious caveat here is that each mutantmust fold correctly in order to be informative. Manyresearchers make alanine mutants on the surface ofa protein to look for the binding site, the so-called"alanine scanning" approach. However, thisamounts only to removal ofthe side chain, andtherefore identifies onlythose residues contributingfree energy to binding.Since not all residuescontribute binding energy, itfollows that this will give anincomplete map of the binding surface.

We therefore recommend an initial screen ofsurface residues identified from the known structureusing what we call "drastic" mutations. This involveschanging the charge and increasing the size of aresidue. In practice it usually means mutatingeverything to arginine except histidine, lysine, andarginine itself, which are all changed to glutamicacid. Such changes are likely to completely disruptbinding if the residue is involved in the bindinginterface in any way and should give a cleardelineation of the interacting surface.

If a mistake is made and a structurally importantresidue is mutated, this will completely destroyexpression of the protein. This is very importantbecause if the protein expresses well it means that a

surface residue has been targeted and it eliminatesthe chance of false negatives, i.e. the falseconclusion that a residue is important for bindingwhen it is in fact required for folding. Alaninemutations can later be used to determine the sourceof binding energy.

— Edward Evans

Although structural information is invaluable inmaking sure the BRET assay is implemented correctly,it does not rely on explicit knowledge of protein

structure to analyze potentialinteractions. Proteins areexpressed as a pair, thereforeonly allowing one potentialinteraction to be investigatedat one time. Where structuressuggest a pairwiseinteraction, however, it is

simply a case of expressing the two targets as donor-and acceptor-tagged pairs in a suitable cell line andperforming the BRET assay.

— John James

Our methods, termed P-LISA for proximity ligation insitu assay, gives staining patterns in microscopicspecimens. We have shown that the method permitsdetection of sets of three interacting proteins.

— Ulf Landegren

Variation of polypeptide linker length betweeninteracting proteins and reporter protein fragmentscan be used to evaluate whether an interaction islikely direct or indirect.



How do you identifystructural patterns ofmultiple reactions?

“We recommend an initial screenof surface residues from theknown structure.” — Edward Evans


The key to avoiding non-specific interactions is toensure that high-quality soluble protein is used asthe analyte. It is therefore important to have a verygood expression system so that you have enoughprotein to purify properly. We favor mammalianexpression systems that fold proteins well and purifyour proteins via an affinity tag followed by gel-filtration (at least). The analyte should be as close tothe native form as possible and be entirely free ofaggregates. Where possible, we use protein straightafter size exclusion gel-filtration on an HPLC systemfor binding analysis using the Biacore, as even verysmall amounts of aggregate may affect measuredbinding properties and these can build up even afterovernight storage at 4°C or following a freeze-thawcycle. Of course, the analyte protein could benaturally prone to non-specific interactions, so it isalso important to have a good control immobilizedfor comparison.

— Edward Evans

Non-specific interactions are an inevitableconsequence of protein diffusion, which is greatlyenhanced when proteins are confined to thecrowded environment of the plasma membrane.Confirming that the expression levels of fluorophore-tagged molecules are within a physiological range,e.g. by FACS, can minimize unwanted interactions aswell as by ensuring that the protein is correctlylocalized within the cell, usually by microscopy. Wehave found that with our BRET experiments, allmembrane proteins give significant but non-specificenergy transfer at the cell surface. In order toaccount for this, proteins of known stoichiometry areused to distinguish these random interactions from

specific protein-protein association. By varying therelative amount of acceptor- and donor-taggedproteins it is further possible to gain quantitativeinformation about the interaction, whereindependence from these changes in theacceptor/donor ratio is the hallmark of randominteractions. Expression of proteins that are knownnot to interact as a BRET pair can also help to identifywhat the profile of this type of interaction wouldlook like in comparison to the unknown interaction.

— John James

Controls are important to set a reliable background.The background is usually a bit higher forinteractions of two membrane proteins compared toa membrane and one soluble protein.

— Moritz Rossner

Non-specific interactions are less of a problem inyeast-based systems than in biochemical assays, asthere are no purification steps involved. However,non-specific interactions can occasionally occur, forexample if two proteins that are normally located inseparate cellular compartments are co-expressed inthe yeast cell. These proteins may interact, althoughunder normal circumstances such an interactionwould never be observed since the two proteinswould not co-localize in the same compartment.

Non-specific interactions can be controlledeither by increasing the stringency of selection (forexample by using the triple reporter strains), or bylowering the expression levels of both a bait and preyprotein, or, finally, by carrying out additionalconfirmation assays after the screen.

— Igor Stagljar

What techniques do youuse to reduce non-specificinteractions?

The Biacore is itself the ideal validation method — itallows detailed characterization of individualinteractions. When high-throughput instruments aremore widely used (e.g. Biacore FlexChip),interactions identified by screening one analyteagainst hundreds of candidates would then beindividually validated on a more conventionalinstrument such as the Biacore 3000 (or the updatedT100).

— Edward Evans

Where BRET analysis shows an interaction betweenproteins and there is structural information availablethat suggests a potential binding interface,mutagenesis provides a means to directly testwhether this interface is responsible for the observedbinding. By using mutations that should have adrastic effect on the interaction, a decrease in BRETvalues compared to the wild type is strong evidencethat this region is important. Where detailedstructural information in not available, chimericproteins can be used to narrow down the region ofthe molecule that is responsible for the observedinteraction. Performing the BRET analysis onhomologous proteins can also shed light on whethera suspected interaction is likely to have beenconserved through evolution and hence may have afunctional consequence.

— John James

Induction or inhibition of interactions as measuredby PCA in response to a pathway-specific stimulusacts as a first-pass validation that a novel interactionis biologically relevant to a specific cellular process.


So far, we have been using mainly well characterizedpairs, but I am sure that for novel interactions thesequential application of at least two non-complementary methods is essential.

This means that you should not verify a FRETresult with a BRET assay and not a Split-TEV resultswith a Spli-Luci assay. If possible, a Split-Enzymeassay should be complemented with a biochemicalapproach (e.g. CoIP) if sensitivity is not limiting.

— Moritz Rossner

The common approach is to verify a particular PPI byco-immunoprecipitating the two proteins fromhuman cell extracts. Furthermore, we are using amodified Bioluminescence Resonance EnergyTransfer (BRET) assay developed in the Bouvier labthat monitors in real time the interactions betweenan integral membrane bait protein and its interactorin living human cells.

Alternatively, co-localization of the two putativeinteraction partners can be used to infer if the twoproteins might interact in their native environment.

If a putative interaction can be confirmed by anyof the above-mentioned methods, the next step isfunctional assays. Such assays include loss-of-function screens (e.g. knockout mice or RNAi-mediated gene suppression in cell lines ororganisms), gain-of-function (overexpression in celllines or transgenic mice) studies, or in vitro enzymaticassays.

— Igor Stagljar


What do you do to validatesuspected protein-proteininteractions?


Question

Initial analysis of sensorgrams (the traces producedduring SPR) can be done using the Biaevaluationsoftware provided with Biacore systems. This allowscomparison with controls, subtraction ofbackground signal, and other simple manipulationsof the data for comparison between experiments.This software can also be used to analyze carefullycollected kinetic data for individual interactions. Forrapid estimation of affinities, equilibrium bindingexperiments can be analyzed in any mathematicalsoftware with a curve fitting function. Followingsubtraction of background response (from thecontrol flow cell), binding response is plotted againstthe concentration of analyte injected and fitted to ahyperbolic curve as dictated by the binding model (inmost cases, for monomeric proteins we use a simple1:1 Langmuir model). Analysis of changes in affinitywith temperature by curve fitting to the non-linearvan't Hoff equation can also be used to obtainthermodynamic constants for an interaction.

— Edward Evans

The detected light emission from the donor andacceptor molecules must all be corrected forbackground signals and especially for the presenceof donor emission in the acceptor channel, whichcan be very significant. These values can then beused to derive the ratio between acceptor and donoremission, which corresponds to the BRET value.When attempting to derive stoichiometric data froma BRET analysis, any dependence of the BRET valueson the acceptor/donor ratio must be fitted to aprescribed hyperbolic equation using a non-linearfitting algorithm that is found in most graphingsoftware. Analysis of structural information about

the protein of interest can be invaluable indetermining the most appropriate terminus of theprotein to fuse the BRET fluorophores to, as well asgiving an approximate hydrodynamic diameter of theprotein that may affect the maximal level of energytransfer observed.

— John James

Broadly-based Bayesian inference tools to linkinteractions to gene regulatory and biochemicalnetworks.


Interaction databases such as DIP, BIND, or GRID areimportant tools to quickly check whether a particularinteraction has been identified before. They are alsohelpful when you have to collect data on putativenew interactors.

On a more complex level, support vectormachines (SVMs) have witnessed increasedapplication in the past years, especially whenprocessing interaction data from high-throughputscreens. SVMs can be used to attach a confidencelevel to a particular interaction based on informationthat includes the Gene Ontology annotations ofbiological process, molecular function, proteinlocalization, transcriptional regulation, andessentiality of the genes encoding the proteins. Allthese parameters can help one to judge whether theidentified PPIs are real or artifactual.

Lastly, we use the Osprey and the NAViGaTORsoftware packages developed in the Tyers andJurisica lab for visualizing and analyzing PPI networksin 2D and 3D format.

— Igor Stagljar

Which computational methods do you use to analyze interactions?

DNA, either of the reporter or one interactionpartner. Stable expression of at least one partner isanother option and increased the stimulus-dependency of a GPCR b-arrestin interaction assayseveral-fold in our hands.

— Moritz Rossner

Q2: How do you reduce falsepositives? (continued from p.8)

In yeast-based assays, false positive interactions arebest reduced by using careful controls. In the so-called"bait-dependency test," putative interactors isolatedin a screen are re-assayed against the "bait" proteinused in the screen and a set of carefully chosencontrols. Only those "prey" proteins which interactwith the original bait but not with any of the controlsis considered a true interactor and is pursued forfurther analyses. If this assay is performed with care,most false positives can be excluded early on.

Other strategies include checking the identity ofisolated putative interactors against a referencedatabase of commonly identified false positives (forexample, the Golemis lab keeps a valuable list offalse positives identified in YTH screens) or usingcomputer-based algorithms to filter out falsepositives which spuriously interact with many baits.

One of the most effective strategiesimplemented in the past years has been thegeneration of YTH strains harboring multiplereporter genes. Compared to first generation strains,these "triple reporter strains" greatly reduce thenumber of false positives in a typical MYTH screenbecause activation of all three reporters is required.

— Igor Stagljar

the MYTH assay works in intact yeast cells and becauseit's the only technology thus far reported to work as ascreening system to find protein interactors ofmembrane proteins, the system has a great perspectivein proteomics research. For example, MYTH providesan opportunity for therapeutic development byidentifying novel drug targets for the diagnosis andtreatment of many human diseases.

— Igor Stagljar

Q1: What are the benefits of your preferred assaysystem(s)? (continued from p.7)

Talk to Tecan

www.tecan.com

Liquid Handling & Robotics ❙ Detection ❙ Sample Management ❙ Components ❙ Services & ConsumablesSwitzerland +41 44 922 89 22 USA +1 919 361 5200 Other +43 62 46 89 33

HS Pro microarray hybridization stations with ABS™The HS 400TM Pro and HS 4800TM Pro are the perfect solutions for superior microarray processing. Fully automated – frompre-hybridization to slide drying – the instruments maximise the sensitivity and reproducibility you expect from yourmicroarray results.Only Tecan’s HS Pro systems have ABS technology (patent pending) to prevent hybridization artefacts by excluding bubblesfrom the process. A variety of low volume hybridization chambers will even enable processing 2 independent arrays on 1slide, on a throughput of up to 48 slides in a run.The HS Pro systems are part of the Tecan Microarray Suite, a range of instruments and solutions from microarray processingand scanning to data analysis. They are easy to use, require less maintenance, and offer the flexibility to process differentmicroarray applications. HS Pro systems are just one example of how Tecan is in touch with what you need most.To find out more about HS Pro or any of our products, please talk to Tecan on the number below.

If you want to increasesensitivity and avoidmicroarray artefacts…

PublicationsGhosh, I; Hamilton, A. D.; Regan, L. Leucine Zipper Assisted Protein Reassembly:Application to the Green Fluorescent Protein J. Am. Chem. Soc., 2000, 122, 5658-5659.

Söderberg O, Gullberg M, Jarvius M, Ridderstråle K,Leuchowius KJ, Jarvius J, et al.Direct observation of individual endogenousprotein complexes in situ by proximity ligation.Nat Methods. 2006 Dec;3(12):995-1000.

Van der Merwe PA, Brown MH, Davis SJ, Barclay AN.Affinity and kinetic analysis of the interactionof the cell adhesion molecules rat CD2 andCD48.EMBO J. 1993 Dec 15;12(13):4945-54.

List of resourcesOur panel of experts referred to a number ofpublications and online tools that may be able tohelp you get a handle on interacting proteins.

BINDhttp://bond.unleashedinformatics.com/Action

Bouvier Lab BRET pagehttp://www.mapageweb.umontreal.ca/bouvier/bret/index.html

DIPhttp://dip.doe-mbi.ucla.edu/

Dualsystems Biotechwww.dualsystems.com

Golemis Lab False Positive Listhttp://www.fccc.edu/research/labs/golemis/InteractionTrapInWork.html

GRIDhttp://www.thebiogrid.org/index.php

NAViGaTORhttp://ophid.utoronto.ca/navigator

Ospreyhttp://biodata.mshri.on.ca/osprey/servlet/Index

Websites

Sunrise Science Products

° S. cerevisiae and S. pombe formulations

° Amino acid dropout mixtures

° Dropout and yeast nitrogen base

° Sugars, supplements and agar

° Custom mixtures at no extra charge

High-Quality Affordable Ready-to-Shipw w w . s u n r i s e s c i e n c e . c o m • i n f o @ s u n r i s e s c i e n c e . c o m

High-Quality Affordable Ready-to-Ship

GE Healthcare

Success is based onknowledge, not chance

Making the right decision at the right time is crucial to yoursuccess, from basic research through drug discovery anddevelopment, to manufacturing and QC.

So wherever you work, the highest quality, information-richinteraction data from a BiacoreTM system enables you tomake each critical decision with confidence – because everyBiacore system is supported by more than 15 years ofknowledge and experience.

Select the perfect solution for your application – from thecompany that continues to set the standard for label-freeinteraction analysis.

www.biacore.com

A system for success

Protein Interaction Analysis - University of Torontolocal.biochemistry.utoronto.ca/stagljar/pdf/The experts talk about... · Protein Interaction Analysis Visit us on the Web at discover.bio-rad.com

Documents