Judy GibbsCorning IncorporatedLife SciencesKennebunk, ME

Table of ContentsIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Colorimetric Assays . . . . . . . . . . . . . . . . . . . . . . 2

Fluorescent Assays . . . . . . . . . . . . . . . . . . . . . . . 7

Luminescent Assays . . . . . . . . . . . . . . . . . . . . . 10

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

IntroductionOver the years, the enzyme immunoassay thatEngvall and Perlmann first described has takenmany different forms. Today there are heteroge-neous, homogeneous, cell-based, colorimetric,fluorescent and luminescent, to name just a few,versions of the original ELISA. They all haveantibody-antigen complexes and enzyme reactionsin common. In this technical bulletin,, the fifth in the series, we will focus on the enzyme linkedimmunosorbent assay and discuss three types ofdetection systems — colorimetric, fluorescent, and luminescent.

All ELISA, regardless of the detection systememployed, require the immobilization of anantigen or antibody to a surface (Corning LifeSciences ELISA Technical Bulletin No. 1). They alsorequire the use of an appropriate enzyme labeland a matching substrate that is suitable for thedetection system being used. Associated with theenzyme-substrate reaction are several require-ments, such as timing and development condi-tions, that need to be optimized to result in aprecise, accurate and reproducible assay.

Selecting the Detection System -Colorimetric, Fluorescent,Luminescent MethodsELISA Technical Bulletin - No. 5

Life Sciences

Colorimetric AssaysColorimetric assays result in a coloredreaction product that absorbs light in thevisible range. The optical density of thereaction product is typically proportionalto the amount of analyte being measured.

Selecting the Appropriate Enzyme LabelThe most common enzymes used aslabels for ELISA are 1) horseradish per-oxidase, 2) calf intestine alkaline phos-phatase, and 3) E. coli ß-D-galactosidase.These enzymes are typically used becausethey each meet most, if not all, of the cri-teria necessary to produce a sensitive,inexpensive, and easily performed assay.

These criteria include:

◗ stability at typical assay temperatures:4°C, 25°C, and 37°C,

◗ greater than six months shelf life when stored at 4°C,

◗ commercially available, ◗ capable of being conjugated to an

antigen or antibody, ◗ inexpensive,◗ easily measurable activity, ◗ high substrate turnover number,◗ unaffected by biological components

of the assay.

By far, the two most popular enzymes are peroxidase and alkaline phosphatase.Each has their advantages and disadvan-tages. Both are quite stable when handledand stored properly, and both can bestored at 4°C for greater than 6 months.Both are also commercially available asfree enzymes and as enzyme conjugates(enzyme labeled antibodies, etc.) and arerelatively inexpensive. However, there are some differences between these twoenzymes that should be considered whenchoosing one for an assay.

Peroxidase is a small molecule (MW~40,000) that can usually be conjugatedto an antibody in a 4:1 ratio. Due to itssmall size, it rarely causes steric hin-drance problems with antibody/antigencomplexes bound on a surface. Peroxidaseis very inexpensive compared to alkalinephosphatase. Several substrates, yielding

either soluble or insoluble reaction prod-ucts, are commercially available for per-oxidase. Since all peroxidase reactionsrequire hydrogen peroxide, purchasingcommercially available substrates is rec-ommended because these preparationscontain stabilized hydrogen peroxidewhich adds to their value and usefulness.

The major disadvantage associated withperoxidase is that it is incompatible withmany preservatives, such as sodium azide,that are used to reduce microbial contam-ination in many biological buffer solutions.Sodium azide, even in low concentrations,inactivates peroxidase activity. Othercompounds or elements that interferewith peroxidase activity are metals foundin water and endogenous peroxidasesfound in biological specimens. Thesedisadvantages can be overcome by usingsterile buffers without preservatives, usingreagent grade type II water, and pretreat-ing specimens suspected of having highperoxidase levels with hydrogen peroxideprior to use in an assay. Typically, non-bound biological components are washedaway prior to the addition of the enzyme,so endogenous peroxidase activity isusually not an issue.

Alkaline phosphatase is approximately dou-ble the size of peroxidase (MW ~86,000).This means that one will typically see alower enzyme to antibody conjugationratio. It also means that the larger molec-ular size of alkaline phosphatase can causesteric hindrance issues due to close packedantigen-antibody complexes. This canresult in lower activity than expected forthe estimated number of bound enzymemolecules (which is sometimes consideredresponsible for the “high dose hook” phe-nomenon). Alkaline phosphatase is slight-ly more expensive than peroxidase, but isconsidered to be more stable. Substratesfor alkaline phosphatase range from solu-ble to insoluble; many can be signalenhanced to increase sensitivity.

The major disadvantage associated withusing alkaline phosphatase is that it isinactivated by chelating agents, acidic pH(< 4.5), or inorganic phosphates. Thismeans that buffers must be specific foralkaline phosphatase, and one cannot usestandard assay phosphate buffered saline

2

solutions as diluents or wash solutionsthat come in contact with the enzymeduring an assay. However, chelators(EDTA) and acidic pH are typically usedas convenient and inexpensive stoppingreagents for alkaline phosphatase reactions.

ß-galactosidase is the least used of the threetop enzymes for ELISA. This enzyme isquite large; its four subunits combinedhave a molecular weight of greater than300,000. Its size is most likely the reasonwhy it is the least popular. For unexplainedreasons, ß-galactosidase also suffers fromantibody-induced inhibition. An advantageof ß-galactosidase is its enhanced reactionrate in the presence of alcohols, whichlends itself as a suitable enzyme for assaysperformed on hydrophobic membranesurfaces (i.e., dot blot applications) thatrequire alcohol to wet out.

For colorimetric assays, either alkalinephosphatase or peroxidase is a suitableenzyme. Both enzymes have a wide rangeof substrates that yield qualitative andquantitative results.

Selecting a Suitable SubstrateFor all enzyme-linked immunoassays, thefinal stage is the addition of the enzymesubstrate. The substrate is chosen for itsquantitative yield of a colored, fluorescentor luminescent reaction product. For col-orimetric assays the rate of color develop-ment is proportional, over a certain range,to the amount of enzyme conjugatepresent.

A suitable substrate must be chosen tomeet the assay requirements of the assaybeing performed. Substrates can produceeither insoluble or soluble colored reactionproducts. Typically, insoluble reactionproducts are desired for membrane-basedassays, such as dot blots. An insolublecolored dot is produced at the site of thereaction. Along with being a visual andsometimes permanent record, the intensityof the colored product can be measuredusing densitometry. However, insolublereaction products are not practical forsolution immunoassays performed inmultiple well assay plates. Substrates thatform soluble reaction products are bettersuited for ELISA.

Both peroxidase and alkaline phosphatasehave substrates that yield soluble coloredreaction products. The decision as towhich substrate is the best for any type of assay depends on the sensitivity desired,the timing requirements, and the detec-tion device to be used. For assays thatneed to be very sensitive (able to detectlow amounts of analyte), the most desir-able substrates produce intensely coloredreaction products at very fast reactionrates. For assays that require a largedynamic range (typical analyte amountsspan a wide range of concentrations),substrates that produce reaction productover a long time (15 to 30 minutes) and result in a broad range of analyte-dependent color intensities are the mostdesirable. For assays with a timed end-point, a chemical inhibitor is added to thereaction after a defined time that stopsfurther color development. This allowsdetection to be performed within a rea-sonable time; for this, a substrate that hasa “slow” reaction rate (15 to 30 minutesto completion) is optimal. This “slow”reaction rate allows the technician (orautomated equipment) to start the reac-tion and stop the reaction at a reasonablepace. However, when kinetic analysis ofthe enzyme-substrate reaction is used, asubstrate that has a “fast” reaction rate (5 minutes or less) should be used. In thiscase, the substrate is added, and the rateof conversion of substrate to coloredreaction product is immediately meas-ured. The reaction is usually measuredover discreet and short time intervals (i.e., 10 seconds) for 2 to 5 minutes.

The following are the most commonlyused substrates for peroxidase and alka-line phosphatase:

Peroxidase: The three most common sub-strates that produce an insoluble productare TMB (3,3',5,5' tetramethylbenzidine),DAB (3,3',4,4' diaminobenzidine), and4CN (4-chloro-1-naphthol). The mostcommon substrates that produce solublereaction products are TMB (dual functionsubstrate), ABTS (2,2'-azino-di [3-ethyl-benzthiazoline] sulfonate), and OPD (o-phenylenediamine). TMB is a highlysensitive substrate. Due to its rapid reac-

3

tion rate, it is ideally suited for on-linekinetic analysis. It produces a blue colormeasurable at a wavelength of 650 nm.TMB can also be used in endpoint assaysby stopping the reaction with 1M phos-phoric acid. A yellow reaction product isformed upon acidification that is measur-able at 450 nm. ABTS is considered anall-purpose substrate. Although it is lesssensitive than either TMB or OPD, it hasthe widest working range of any substratecurrently available for peroxidase or alka-line phosphatase. The reaction productfor ABTS is a blue-green compoundmeasurable at 405 to 410 nm. Its reactionrate is suitable for endpoint assays and iseasily stopped with 1% SDS (sodiumdodecyl sulfate), which does not changethe color or the absorbance of the reac-tion product. OPD was once the mostpopular substrate for peroxidase. It isslightly less sensitive than TMB. Itsreaction product is yellow and can beread at 490 nm.

Alkaline phosphatase: The most commonsubstrate that produces an insoluble reac-tion product is BCIP/NBT (5-bromo-4-chloro-3-indolyl-phosphate/nitrobluetetrazolium). It is recognized as the mosteffective substrate for immunoblots dueto its stability and resistance to fadingwhen exposed to light. The most widelyused substrate that produces a solublereaction product is p-NPP (p-nitro-phenylphosphate). It produces an intenseyellow color measurable at 405 to 410 nm.An advantage of this substrate is that itcan be allowed to develop for extendedperiods to obtain a corresponding increasein sensitivity. Normally p-NPP has a slowreaction rate which should be allowed 30 to 60 minutes to reach optimal colordevelopment before being stopped with1N NaOH. It is not recommended forkinetic analysis.

Reaction RequirementsMany factors affect the measurement ofenzymatic activity. Some of the mostobvious are:

◗ temperature,◗ pH,◗ ionic strength,

◗ buffer composition,◗ substrate depletion,◗ build-up of product inhibitors,◗ increasing back-reaction as product

concentration increases,◗ denaturation of the enzyme, and in

some cases,◗ exposure to light.

The ones that are of most concern forELISA today are reaction time, tempera-ture and exposure to light. The factors,such as pH and substrate depletion, havebeen addressed, and commercially avail-able reagents have been optimized forcomposition and concentration in orderto control these parameters.

Timing the ReactionIn order to have an endpoint assay thatprovides reliable and consistent results, it is important that the timing of thereaction in each and every well, in eachand every plate, and in each and every setof plates be controlled as precisely as pos-sible. Since enzyme-substrate reactionsare kinetic, timing from the start to theend of the reaction can and will affect thefinal concentration of product developed.To ensure precise timing, we follow thisscheme for every assay that we perform —regardless if only a few wells, an entireplate, or ten plates are involved in the assay:

1. Set timer to the desired and pre-determined substrate incubation time.

2. Start timer with the addition of sub-strate to the first well or set of wells.

3. Use a rhythmic pipetting pattern to add substrate to all the wells.

4. When the timer signals the end of theincubation period, stop the reactionusing the same pipetting pattern andrate that was used to add the substrate.

This scheme assures that all the wells seeactive substrate for the same amount oftime and adds consistency to the assayresults. We use a rhythmic pipetting pat-tern with a 12 channel pipettor for dis-pensing substrate and/or stop solutionfrom row A through row H in a 96 wellplate that takes approximately 30 secondsper plate to complete.

4

The incubation time associated with thesubstrate step in an assay must be prede-termined so that the color formed for thelowest analyte concentration is signifi-cantly higher than the background andthe color formed for the highest analyteconcentration is less than the reader cut-off value. (This value usually ranges from2.0 to 4.0 optical density (OD) unitsdepending on the reader used.) A goodrule of thumb is to choose a high levelOD of approximately 1.0. The followingis a method that can be used to determineoptimal incubation times for the substratestep:

1. Coat the plate with the optimalantigen or antibody dilution.

2. After rinsing away non-bound reagent,block the surface.

3. Add the standard or sample containingthe highest concentration of analyte tobe detected.

4. Incubate as appropriate; wash awaynon-bound analyte.

5. Add the enzyme conjugate. Incubate as appropriate. Wash.

6. Add the substrate solution. 7. Monitor color development. 8. Stop the reaction when the OD is

approximately 1.0. 9. Record the time required to reach

an OD of 1.0. This is the optimalsubstrate incubation time.

Development ConditionsAs mentioned earlier in this bulletin,temperature and light can affect theenzyme-substrate reaction. These twoassay parameters can be the cause of“edge effect”; where OD’s in edge wellsare higher or lower than center wells. Allenzyme reactions are temperature depend-ent. This means that temperature duringthe enzyme-substrate reaction step mustbe kept constant in all the wells. Typically,this reaction step can be performed atroom temperature. Some assays requirethis step to be performed at elevatedtemperatures to increase enzyme activity.Regardless of the assay temperature, thetemperature of the substrate prior toaddition to the plate should be equivalent

to the intended incubation temperature.Room temperature incubations are themost common and are the easiest to con-trol. All plates being processed at roomtemperature should be placed in a locationsuch that none of the plates is situatednear a source of heat or cold. For assaysat elevated temperatures, controlling edgeeffect is more difficult. Since polystyreneis a poor conductor of heat, incubationsperformed in an incubator usually resultin edge effect problems – the outer wellsreach the desired temperature up to 30minutes before the center wells do. Thisresults in higher OD’s in the edge wellsdue to increased enzyme activity in thesewells. In order to maintain constant tem-peratures in all the wells, elevated tem-perature incubations should be performedon a heat block that allows all wells toreach the critical temperature simultane-ously. The easiest way to avoid tempera-ture related edge effect problems is toincubate at room temperature – whichmay mean simply increasing the incuba-tion time to reach the OD acquired at the higher temperature incubation.

Although many substrate systems arebelieved to be stable in the presence oflight, we firmly recommend that the sub-strate incubation step be performed in thedark to avoid edge effects caused by varia-tions in incident light from well to well.Typically, outer wells receive a higherdegree of incident light than center wells,which results in lower or higher thanaverage OD’s in these wells. Plates thatare incubated in a drawer or cabinet thathas a light leak can also be adverselyaffected by incident light (wells closest to the light leak exhibit higher or lowerOD’s). Performing incubations in thedark is the easiest method of assuring thatincident light is not affecting the assay’soutcome. Kinetic assays are usually not aproblem since the incubation is performedwithin the reading chamber of the detec-tion instrument which has to be light-proof in order to function properly.

If temperature and light exposure is con-trolled, the probability that the assay willsuffer from edge effect problems is greatlyreduced.

5

Mixing the ChromophoreAfter the addition of stop solution and/orprior to reading the OD’s, it is importantto adequately mix the well contents. Thiswill assure both complete cessation of thereaction (endpoint assay only) and evendispersion of the colored reaction prod-uct. Well-to-well precision can be dra-matically improved by adding this step to the protocol.

The physical effect that shaking the plateprior to reading has on the resulting opticaldensity is depicted in Figure 1. Microplatecolorimeters only read the immediatecenter of the well. This means that anunshaken plate will have wells that resem-ble Well A in the diagram. Colored reac-tion product will be concentrated at thesurface of the well where the enzyme islocated. Dissolution of the reaction prod-uct is slow if not mechanically stirred ormixed. The reader will not be measuringthe true optical density within Well A.With good mixing, the colored reactionproduct is evenly dispersed through thesolution in the well as seen in Well B. Thereader can then detect the true opticaldensity of the colored reaction productresulting from the enzyme-substratereaction occurring at the surface.

The data in Table 1 clearly shows theincreased precision that is possible bysimply shaking the plate prior to reading.Out of 26 plates that were shaken beforereading the OD’s, all of the plates pass ourcertification criteria for CV’s less than3.0% and high and low wells less than8% from the mean OD. Over half of theplates that were not shaken do not passour certification criteria. The majorproblem with these plates is low wellssuspected of being caused by a lower con-centration of colored reaction product inthe center of the wells (see Figure 1) asopposed to the edge of the wells. (Note: ourreaders only read the center of the well).

There are several ways to accomplish the

mixing step. Microplate shakers are avail-able that hold 2 to 4 plates at a time sothat plates can be shaken as a group priorto being loaded into a reader. Many micro-plate readers have built-in shaking mech-anisms that allow plates to be shaken justprior to being read. Both of these methods(built-in versus remote shakers), whenused properly, can have very positiveeffects on assay precision.

Detection MethodsFor the purpose of this bulletin, we will focus on the detection methods andequipment used for soluble reactionproducts that absorb light in the visiblewavelength range (> 405nm). There is awide choice of microplate readers avail-able today that range from simple, singlewavelength colorimeters to sophisticatedmulti-wavelength spectrophotometers.The main function of a microplate readeris to measure light absorbed at particularvisible wavelengths of light by a sub-stance, which in most cases is a coloredsolution. Some of the options availablemake the measurement easier to attainand allow analysis of samples that cannotbe measured in the simplest models.

Some readers are capable of only singlewavelength measurements. These areadequate for standard ELISA that havelittle or no interference from subtractablebackground noise. Most readers have theability to read at dual wavelengths. Thisoption is necessary when backgroundnoise from attached cells or minutescratches on the optical surface interferewith accurate detection of the coloredreaction product. Dual wavelength allowsone to read the plate at a wavelength spe-cific for the reaction product and at a sec-ond wavelength that is out of the range ofabsorbance for the reaction product. Theinstrument subtracts the second wave-length from the first, effectively sub-tracting out the background noise that is usually not wavelength specific. The

6

Table 1. Effect of Shaking vs. Non-shakingShake vs. # of Plates # of Plate Low HighNon-shake Run Out-of-Spec CV Well WellShake 26 0 2.0 5.7 6.0Non-shake 23 14 3.1 11.0 6.9

Figure 1. Effect ofShaking vs. Non-shaking

A

B

resulting OD is specific for the reactionproduct. The added cost of dual wave-length capability is minimal and worthpurchasing with any microplate reader.

All readers are capable of endpointanalysis. This is adequate for most assays.However, for increased sensitivity anddynamic range, a reader capable of kineticanalysis is required. Endpoint assaysrequire that the reaction be stopped whena certain predetermined high dose OD is reached. These assays many times lackthe sensitivity to distinguish minorchanges in analyte concentration above acertain limit — typically due to substratedepletion. When kinetic analysis of theenzyme-substrate reaction is employed,this limitation is minimized. Since reac-tion rates are measured in mOD/minuteas soon as the reaction is initiated, minordifferences in analyte concentrations areeasily distinguishable even at the highdose end. Readings are performed before substrate depletion is an issue.

The newest readers available are micro-plate spectrophotometers. These readersare capable of measuring absorbance atseveral wavelengths and can even performcontinuous scans on samples over thewhole range of available wavelengths(including UV in some cases). Thesereaders are probably more sophisticatedthan what is necessary for a standardELISA; however, there are a number of applications (total protein and DNAdeterminations) that can benefit from this feature.

Fluorescent AssaysFluorescent immunoassays (ELFIA) aresimply a variation of colorimetric ELISA.An enzyme converts a substrate to a reac-tion product that fluoresces when excitedby light of a particular wavelength. Therelative fluorescence units (emitted pho-tons of light) that are detected are typi-cally proportional to the amount ofanalyte being measured. In comparison to the colorimetric ELISA, fluorescentimmunoassays are only slightly more sen-sitive. However, they widen the dynamicrange of the assay by allowing very highreadings to be accurately measured asopposed to the 2.0 to 4.0 OD limitimposed on colorimetric assays.

Selecting the Appropriate Enzyme LabelThe three main enzymes that weredescribed in the section for colorimetricassays are the same three that are used forELFIA. However, the order of popularityis different. Alkaline phosphatase is themost widely used enzyme for fluorescentapplications. ß-galactosidase is also usedmore frequently due to its greater theo-retical sensitivity when used with a fluo-rogenic substrate. Peroxidase is rapidlygaining popularity as an enzyme label for fluorescent-based immunoassays.Stabilized substrates for all three of thetop enzymes are commercially available.

Selecting a Suitable SubstrateA fluorogenic substrate is chosen for itsquantitative emission of light followingexcitation. The rate of light emissionshould be proportional to the amount ofenzyme conjugate present. The substrateshould be stable at room temperature andin the presence of normal room lighting.The resulting enzyme-substrate reactionproduct should also have distinctly sepa-rate excitation and emission wavelengths,plus the substrate itself should be non-fluorescent.

The three main enzymes each have oneor two major fluorogenic substrates thatare suited for ELFIA. Alkaline phos-phatase is usually paired up with 4-MUP(4-methylumbelliferyl phosphate), whichis converted to 4-methylumbelliferonewith an excitation wavelength of 360 nmand an emission wavelength of 440 nm.This substrate is dissolved at 0.1 to 0.2mg/ml in 100mM diethanolamine, 1 mMMgCl2, pH 9.6. Commercial liquid prepar-ations of 4-MUP are available; however,our results have indicated that the liquidversion results in a much higher back-ground as compared to freshly preparedsubstrate solutions.

A suitable fluorogenic substrate for ß-galactosidase is MUG (4-methylumbel-liferyl galactoside), which is converted to4-methylumbelliferone. (This is the sameproduct that results from the conversionof 4-MUP, the substrate used with alka-line phosphatase.) There is an increaseduse of 4-MUG to detect reporter geneexpression; a ready-to-use form of this

7

substrate is available for other non-ELISA applications.

Two fluorogenic substrates are currentlyused for ELFIA utilizing peroxidase as the enzyme label. These are HPA(hydroxyphenylacetic acid) and HPPA (3-p-hydroxyphenylproprionic acid). Both require, as expected, the addition of hydrogen peroxide in order to producea fluorescent product. HPPA is the mostwidely used fluorogenic peroxidase sub-strate. Its fluorescent product has anexcitation wavelength of 320 nm and anemission wavelength of 404 nm. Stabilizedfluorescent substrates for peroxidase arenow commercially avaialable.

Methods to Improve SignalFluorometric assays are subject to severalproblems that either non-specificallyreduce or enhance the signal output.Detection of fluorescence is susceptible tochanges in pH, temperature, ion concen-tration, detergent concentration, drying,and the solid matrix, which lead to lightscattering, high background, quenchingand bleaching issues.

Light scattering is a phenomenon causedby the emitted fluorescent light beingbounced around as it encounters mole-cules and/or particles in solution, or thesurface of the microplate. It is importantto use high quality chemicals; however,even the purest of reagents have particlesthat lead to light scattering. Opaque blackplates and strip plates of the highest qual-ity are designed to reduce the scattercaused by light bouncing off the surfaceby absorbing this light. Only light direct-ed toward the detector should be meas-ured. Aside from controlling the purity ofreagents and diluents, and using a qualitymicroplate to minimize the matrix effect,the only other assay technique that canreduce light scatter is preventing bubblesin the well contents.

Background fluorescence and/or autofluores-cence has several sources and is a majorobstacle in the development of 96 wellfluorescent assays. Sources of backgroundinclude:

◗ sample components (hemoglobin,bilirubin, cellular debris, drugs),

◗ diluent components (metal ions), ◗ plate material (type of plastic used), ◗ miscellaneous contamination (dust

particles, fingerprints).

Background can be combated via threeapproaches: assay design, instrumentation,and cleanliness. Several assay design fea-tures can help reduce background. Forhomogeneous assays, sample dilution isthe key. It is important that any interfer-ing substances present in the sample bediluted to a degree that background fromthese substances is minimized. When alow enough dilution cannot be feasiblyachieved, employing kinetic analysis ofthe production of a fluorescent reactionproduct is a good alternative. The inter-fering substances will produce only staticfluorescence, while the specific assayreaction will be kinetic and easily dis-cernible from the background. Time-resolved fluorometry is also a successfuldetection method that reduces non-spe-cific background from being measured.

For heterogeneous assays, backgroundinterference from the sample componentsis rarely an issue since this type of assayemploys a separation step. However, this separation step must be adequate to remove all the interfering substancesprior to the addition of substrate. It isimperative that all reagents used in theassay are of high quality and filteredthrough a 0.45 µm (non-particle shed-ding) membrane prior to use. It is alsoimperative that the microplate used bemolded from non-autofluorescing mat-erial. Opaque black plates and strips aretypically used for fluorescent assays dueto their low background fluorescence.The type of plastic used is critical. Someplasticizers that may leach into solutioncan auto-fluoresce. Most quality fluor-ometers are equipped with features thataid in the reduction or elimination ofmachine-related background. These fea-tures include (i) filters, (ii) adjustable slit-width, and (iii) adjustable gain. Of these,the one that allows the end-user to mosteasily adjust for (subtract) background isgain. If the gain is adjustable, it can beused to achieve the best signal-to-noiseratio.

8

In general, lack of cleanliness is probablythe most common cause of occasionalbackground. Physical items, such asfingerprints and dust, are notorious forincreasing background fluorescence. Thefogging of plate bottoms or instrumentoptics by condensation from temperaturedifferences can also increase non-specificbackground. Typically, high backgroundfrom these physical causes results in poorwell-to-well precision.

Quenching is a problem characterized bya non-specific reduction in signal. Thisphenomenon is caused by the absorptionof the emission by dissolved oxygen.Reagents can be degassed prior to use toalleviate this effect.

Bleaching or fading is also characterizedby a reduction in signal. It is caused by anexcessively long excitation step. Typicallythis is not a problem due to the low powerand short excitation exposure times asso-ciated with today’s fluorometers. However,if substrate incubation steps are performedunder bright light, bleaching can occur. It is recommended that this step be per-formed in the dark. Other sources ofinterference that can distort fluorescentreadings include: (i) temperature varia-tions, (ii) light source stability, and (iii)slit-width. Fluorescence behaves oppositeto most other detection systems. Fluor-escence increases with decreasing temper-ature. It is important to maintain both aconstant intra-assay and inter-assay tem-perature. Variations in temperature fromone assay to another can cause changes in assay sensitivity. We recommend thatassays be performed at room temperatureand include a positive and negative con-trol (as a minimum) on each multiplateper assay.

Opaque Plates to Reduce Crosstalkand BackgroundSince some of the plastics used to manu-facture microplates and 8 well strips areautofluorescent, it is important to choosea plate that is made specifically for fluo-rescent assays. These plates are usuallyopaque black (although white plates canbe used for some fluorescent applicationssuch as TRF). These black plates can

have solid black bottoms or clear bot-toms. Both are designed to reduce well-to-well crosstalk and backgroundfluorescence.

Clear bottom black plates are typicallyused for cell-based fluorescent assays toallow visualization of the cells duringattachment and growth while reducinglateral light transmission or crosstalk dur-ing the detection step. With the advent ofmicroplate fluorometers that are capableof reading from the top or the bottom ofthe well, these plates have allowedincreased versatility associated with assaydesign. Using a clear bottom plate, a dualassay resulting in a colorimetric productfor one analyte and a fluorescent productfor a second analyte can increase produc-tivity and reduce sample requirements.

Why black plates? When an appropriatematerial is selected for the manufactureof these plates, background noise can be significantly reduced and crosstalkvirtually eliminated. The black materialreduces background by being non-autofluorescent and absorbing stray light(reducing the background caused by light scatter).

Background relative fluorescence units(RFU) at excitation 485 nm and emission530 nm for Corning’s assay plates andstripsPlate Type RFsStandard Clear Plate 4Black Clear Bottom Plate 2Solid Black Plate/Strips 1

Although background is relatively loweven for the standard clear plate, it is evi-dent that the black material used for theclear bottom plate and solid plate andstrips reduce background by 50% to75%, respectively.

Percent crosstalk from well to well can be significantly reduced by using highquality black platesPlate Type Percent cross TalkStandard Clear Plate 0.30%Black Clear Bottom Plate 0.05%Solid Black Plate/Strips 0.03%

9

As with background fluorescence, cross-talk reduction is dependent on the mate-rial used to manufacture the plates orstrips. It is important to choose a platethat meets the minimum requirementsassociated with both background andcrosstalk and is consistent from lot-to-lot.

EquipmentThere are currently many different micro-plate fluorometers on the market. Thesedevices vary greatly in performance. Whenchoosing an instrument we advise thatyou ask for the following features:

◗ Light sealed reading compartment. ◗ Adjustable light detector. (The detector

must be positioned as close to the plateas possible to obtain accurate readings.Since plates from different manufacturersvary in dimensions, it is important to havethe ability to adjust for these variations).

◗ Numerically adjustable gain. (Choose areader that has a gain adjustment that isnumerical and not just low, medium, orhigh light intensity options).

◗ Thermal consistency in the readingchamber. (Since fluorescent output isaffected by temperature variations, it isimperative that the reading chamber betemperature controlled such that eachplate has even heating or cooling. Aplate that is warmer on one side versusthe other will exhibit inconsistentfluorescence from well to well).

◗ Internal background controlmechanisms.

◗ Mix and match excitation and emissionwavelengths.

◗ Selectability of number and duration ofread times per well.

◗ Optional: choice of top plus bottomdetection.

◗ Optional: adjusts to 1536, 384, 96, 48,24, 12, and 6 well plate formats.

◗ Features 1 to 5 are necessary if accuracyand consistency are to be realized.

Luminescent AssaysLuminescent immunoassays, like fluores-cent immunoassays, are variations of thestandard ELISA. An enzyme converts asubstrate to a reaction product that emits

photons of light instead of developing avisible color. Luminescence is describedas the emission of light from a substanceas it returns from an electronically excitedstate to ground state. The various forms ofluminescence (bioluminescence, chemilu-minescence, photoluminescence) differ inthe way the excited state is reached. Forexample, photoluminescence is simplyfluorescence; the excitation is initiated by light at a particular wavelength. Bio-luminescence is characterized by the useof a bioluminescent compound, such asluciferin and firefly luciferase. Chemi-luminescence is light produced by achemical reaction.

Both bioluminescence and chemilumines-cence are widely used for immunoassaysand will be discussed in this bulletin as“luminescence”. Luminescent assays, inparticular enhanced luminescent assays,are very sensitive and have a wide dynam-ic range. It is believed that luminescenceis the most sensitive detection methodcurrently in use due to the ability ofsignal multiplication and amplification.Luminescent reactions are measured in relative light units (RLU) that aretypically proportionate to the amount of analyte present in a sample.

Selecting the Appropriate Enzyme LabelAs with the previous two detection sys-tems described in this bulletin, the topthree enzymes used for luminescence are peroxidase, alkaline phosphatase, andß-D-galactosidase. However, the mostwidely used enzyme for luminescentimmunoassays is peroxidase. Peroxidasecan be used with either bio- or chemilu-minescent systems and is easily enhancedto allow prolonged detection of intenselight (glow luminescence) which makes itcompatible with all size microplate assayformats.

Selecting a Suitable SubstrateA luminescent substrate should be chosenfor its:

◗ low background luminescence in theground state,

◗ ability to produce intense light in itsactive state,

10

Corning assay plates are availablein a variety of sizes and materials.

◗ ability to produce stable light emissionover a prolonged (minutes) period oftime, and

◗ commercial availability (quality andconsistency).

The substrate should be stable at roomtemperature during the duration of theassay.

The three most-used enzymes have oneto several suitable substrates.

Peroxidase has the most extensive list ofsuitable substrates, which include (i) lumi-nol, (ii) polyphenols and acridine esters,and (iii) luciferin. The reaction of peroxi-dase with luciferin is considered biolumi-nescence. In this reaction, peroxidasereplaces the in vivo enzyme, luciferase.The other substrates are chemilumines-cent compounds. Polyphenols are actuallya class of substrates that include pyrogal-lol, purpurogallin, gallic acid, and umbel-liferone. All polyphenols are known fortheir excellent signal to noise ratio andextremely rapid light decay. Polyphenoland acridine ester substrates can only beused in conjunction with luminescentdetectors equipped to handle “flash” reac-tions. The most popular substrate usedfor immunoassays is luminol. Luminol iscommercially available in a stabilizedform. It is the most suited for clinicaldiagnostic tests due to its properties whenused in an enhanced luminescence system.Commercially available luminol is provid-ed with an enhancer (phenols, naphthols,aromatic amines, or benzothiazoles) thatacts as an enzyme protector and allowsthe reaction to proceed for many minuteswithout substantial decay in light output.Typically, light emission stabilizes in lessthan 2 minutes, and sustained emissionlasts for approximately 20 minutes or more.

Enhanced luminescence is characterized bythe following desirable features: intenselight emission, prolonged light emission,low background, no preincubation step,and substrate that can be added severalminutes prior to detection. As long ascommercial preparations of luminol areused, control of the reaction pH is not aconcern. However, if the substrate is a“home-brew” preparation, pH must bestabilized at about 8.5 to allow both per-

oxidase activity (optimal at pH 5.5) andlight emission (optimal at pH 12.0) tooccur. If the pH varies much above orbelow 8.5, either the enzymatic activity or the luminescent detection will be neg-atively affected. As mentioned, luminol-based chemiluminescence is well suitedfor microplate-based immunoassays; inaddition, this system is also recommendedfor DNA probe assays.

Alkaline phosphatase and galactosidaseeach have one preferred substrate.AMPPD (3-(2'-spiroadamantane)-4-methyl-4-(3'-phosphoryloxyphenyl-1, 2-dioxetane, disodium salt) is the substratemost commonly used with alkaline phos-phatase. A similar substrate, AMPGD (3-(2'-spiroadamantane)-4-methoxy-4-(3'-ß-D-galactopyranosyloxyphenyl-1,2-dioxetane), is routinely used with ß-galactosidase. Both substrates arecompatible with commercially availableenhancers. AMPPD is typically sold as akit that includes additional buffers andreagents matched to the substrate foroptimal performance.

Glow versus Flash Detection MethodsThere are two distinct methods of detect-ing luminescence — flash and glow.

Flash luminescence is transient in natureand reaches maximal light intensity with-in seconds or milliseconds. Due to thespeed at which the reaction occurs, it isnecessary to start the reaction while thereactants are in front of the photomulti-plier tube or other light detection device.Starting the reaction consists of addingsubstrate and complementary reagents orbuffers and subsequent mixing of all theassay components. Of paramount impor-tance is that there be a constant timeinterval between the addition of thestarting reagents and the time that themeasurement takes place. For a microplate-based assay, this requirement is met bycoordinating this step within the readingchamber. Reagent addition and lightmeasurement takes place in a stepwisemanner one well at a time.

An alternate detection method is glowluminescence, which is a steady-statekinetic approach to signal generation.

11

Glow luminescence is actually a largenumber of transient signals that occur insequence and result in a constant signal.Unlike colorimetric or fluorescent reac-tions, the light produced is not accumu-lated as color or fluorescence can be, sothe light emitted must be intense and theenzyme reaction prolonged in order toobtain sufficient signal. The positiveaspects of glow luminescence is that

◗ the reaction can be started outside ofthe detection instrument, thuseliminating the need for internalinjection and mixing within the reader,

◗ the procedure is simple,◗ the results are sensitive, ◗ excellent enhanced substrate systems

are commercially available, ◗ the procedure is highly suited for

microplate-based assay formats.

This type of luminescent reaction can bemeasured using a luminometer, capturedon photographic film, or recorded viaimage analysis. Our experience with lumi-nescence involves glow reactions. An areathat we found to be crucial to obtainingreproducible results is the enzyme-substrate reaction time. Although it isreported that glow reactions are expectedto be stable for at least 20 minutes, wefound our results more consistent fromwell-to-well and from plate-to-plate whenour incubation time was short. We rec-ommend allowing a 2 minute stabilizationperiod after substrate addition and thenimmediately reading the plates. Bothsignal strength and precision can beimproved by optimizing the enzyme-substrate reaction incubation time. Note: luminescent reactions do notrequire a stopping step.

Opaque Plates to Reduce Crosstalkand BackgroundChoosing the best plate for luminescenceis crucial to developing a reliable assay. It is important to choose a plate that isspecifically designed for luminescence.These plates are usually opaque white. Aswith the opaque black plates discussed

earlier, these white plates can have solidwhite bottoms or clear bottoms. Bothversions are designed to reduce crosstalkand background luminescence due to thespecially formulated composition of thewhite resin.

White clear bottom plates are typicallyused for cell-based luminescent assays,such as a luminescent cell proliferationassay. These plates allow visualization ofthe cells during attachment and growthand prevent lateral light transfer fromwell to well during the detection step.Clear bottom plates are also useful fordual-assays that result in a colorimetricproduct for one analyte and a lumines-cent product for a second analyte. Anexample of the versatility of a white clearbottom plate is the ability to stain cells toobserve and assess structural changes andto coordinate these observations withmeasurements of cell proliferation vialuminescence — all in the same plate.

Why white plates? When these plates are made from a truly opaque, non-luminescent material, crosstalk andbackground can be almost eliminated. A properly formulated white material also increases assay sensitivity by reflect-ing emitted light into the detector. Note:luminescence is not cumulative, so it isimportant that each discreet and transientlight photon reach the detector if sensi-tivity is to be realized. Sensitivity is com-promised if light is allowed to escapefrom the plate undetected or is absorbedby the material chosen to make the plate.An opaque and highly reflective surfaceare the key ingredients for a quality platedesigned for luminescence.

Background in relative light units (RLU)for Corning’s white assay platesPlate Type RLUWhite Solid Plate/Strips 17White Clear Bottom Plates 17

These data indicate that neither the solidor clear bottom plates or strips signifi-cantly contribute to background lumines-cent readings.

12

Assay background in relative light units(RLU) for solid white plates — differentmaterialsPlate Type RLUCorning Plate 68Competitor #1 72Competitor #2 327Competitor #3 2534

As mentioned earlier, the choice for thewhite material is crucial to reducing back-ground from the plastic itself and fromlight transfer from an adjacent well. TheRLU values in wells adjacent to the blankwells reported above were approximately100,000.

Percent crosstalk from well to well can besignificantly reduced using a solid whiteplate as opposed to a clear bottom plate.However, one loses the advantage of visu-ally assessing cell attachment, growth,and structure for cell-based luminescentassays on a solid bottom plate.

Plate type Percent crosstalkWhite Solid Strips 0.01%White Solid Plates 0.02%White Clear Bottom Plates 0.50%

Note: a standard clear plate could not beevaluated for crosstalk due to the limita-tions of the luminometer, which cannotdetect the presence of a clear plate andthus will not initiate reading. Our lumi-nometer has injection ports such that theinstrument can be used for flash lumines-cence. As a safeguard (to avoid reagentinjection into an empty reading chamber),it will not function unless it senses thepresence of a plate in the reading chamber.

EquipmentDue to innovations with photomultipliertubes that can be used as photon coun-ters, 96 well plate luminometers are nowreadily accessible and can be relied on foraccurate and consistent detection of lumi-nescent output. When choosing aninstrument, we advise that the followingfeatures be considered:

◗ light sealed reading compartment,◗ safeguard to prevent accidental injection

of luminescent reagents into the read-ing chamber (flash luminometers only),

◗ control mechanisms for temperaturedrift,

◗ low instrument background at ambienttemperatures,

◗ selectability of read time per well, ◗ optional for flash luminescence:

precision auto-injectors, (Recommend a minimum of two injectors with backflush mechanisms to prevent drippingand liquid jet injection to ensureimmediate and intensive mixing ofreagents.)

◗ optional for flash luminescence: anadequate auto-mixing mechanisminitiated at the time of injection.

◗ optional: easy conversion to 1536, 384,96, 48, 24, 12, 6 well plate formats.

A quality luminometer can enhance theaccuracy, sensitivity and consistency ofassay results.

ConclusionRegardless of the detection technique onechooses, several parameters always needto be controlled:

◗ One must always choose an enzymelabel that is suitable for the applicationand detection method being employed.

◗ Following the enzyme selection, a sub-strate that matches both the enzymeand the detection method must bechosen. It must also meet the assay’srequirements for sensitivity, dynamicrange, and reaction speed, plus if pos-sible, be commercially available as astabilized tablet or solution.

◗ Enzyme-substrate reaction require-ments, such as incubation time anddevelopment conditions (temperature,light, etc.) need to be optimized.

◗ The reaction product, be it colori-metric, fluorescent or luminescent,must be adequately mixed prior todetection to ensure accurate and precise readings.

◗ The proper microplate must be chosen —clear for colorimetric, black for fluor-escence and white for luminescence —and verified as a quality product.

13

◗ Finally, the detection device must beselected, keeping in mind that specialoptions may actually be criticalrequirements; such as a temperaturecontrolled reading chamber.

Each detection method has its own set ofspecial requirements, but if the generalassay precautions outlined in this bulletinare followed, a precise, sensitive andreproducible assay can be developed andconsistently performed.

Technical AssistanceFor additional ELISA technical supportand bulletins or product information,please visit the Corning Life Sciencesweb site at www.Corning.com/lifesciencesor call 1-800-492-1110.

References 1. Bjerrum, O.J. and Heegaard, N.H.H.

Handbook of Immunoblotting of Proteins,Volume 1. Boca Raton, Florida: CRCPress, Inc.: 1988.

2. Craig, J.C., et al. Journal of BiologicalStandardization. 17; 1989; 125-135.

3. Engvall, E. and Perlmann, P. Immuno-chemistry. 8:871; 1971.

4. Gardas, A. and Lewartowska, A. Journal ofImmunological Methods. 106; 1988; 251-255.

5. Graves, H.C.B. Journal of ImmunologicalMethods. 111; 1988; 157-166.

6. Harris, D.A. and Bashford, C.L. Spectro-photometry and Spectrofluorimetry: APractical Approach. Washington, D.C.:IRL Press; 1987.

7. Hemmila, I.A. Applications of Fluor-escence in Immunoassays. New York: John Wiley & Sons, Inc.; 1991.

8. Kenny, G.E. and Dunsmoor, C.L. IsraelJournal of Medical Sciences. 23; 1987;732-734.

9. Maggio, E.T. Enzyme Immunoassay. BocaRaton, Florida: CRC Press, Inc.; 1980.

10. Stenberg, M. and Nygren, H. Journal ofImmunological Methods. 113; 1988; 3-15.

11. Van Dyke, K. and Van Dyke, R. Lumin-escence Immunoassay and MolecularApplications. Boston: CRC Press, Inc.;1990.

12. Vogt, R.F., et. al. Journal of ImmunologicalMethods. 101; 1987; 43-50.

13. Worthington Biochemical Corporation.Worthington Enzyme Manual. Freehold,N.J.; 1988.

14

Corning IncorporatedLife Sciences

45 Nagog ParkActon, MA 01720t 1.800.492.1110f 1.978.635.2476

www.corning.com/lifesciences

Corning is a registered trademark of Corning Incorporated, Corning, New York.Corning Incorporated, One Riverfront Plaza, Corning, NY 14831-0001

Revised 7/01

Katherine E. Strathearn Ph.D.and Mark Rothenberg Ph.D.Corning Incorporated,Life SciencesKennebunk, Maine

IntroductionThe past 30 years have seen an astonishing rate of growth inautomation to aid in research and drug development. Frominstruments such as basic microplate readers, liquid handlers,grippers, and incubators, to highly advanced instrumentationsuch as high content imagers, the microplate-automationrelationship has always played a critically important role.

The relationship between microplates and the equipmentused to run the experiment is fundamental to any assay rang-ing from low to high throughput. By not appreciating thisdynamic, resulting data will be less robust, CVs willbe increased and there will be more opportunities for falseresults. For instance, if a researcher wants to perform afluorescence-based assay, selecting a clear microplate wouldresult in not only high background but also crosstalk betweenwells resulting in inferior data. Additionally, if the microplatereader is optimized for a 384 well normal volume microplateand a researcher attempts to analyze a 384 well low volumemicroplate, the microplate dimensions and focal plane will beinaccurate leading to suboptimal results. Factors suchas those listed above and others (e.g., liquid handling) arecritical to understand when optimizing any assay.

The following describes the importance of choosing the cor-rect microplate for an assay, as well as optimizing the readerfor the microplate. Additionally, the importance of properliquid handling and how inaccurate dispensing of a samplemay negatively impact the results is presented. The goal is toarm researchers with the tools needed to develop successfulassays by providing a better understanding of the relationshipbetween microplates, assays, and instrumentation.

Materials and Methods� Multitox-Glo was purchased from Promega (Cat. No.G9272), and the protocol was followed as outlined in themanual.

� Fluorescein was purchased from Sigma (Cat. No. F6377)� The following Corning® microplates were used:

- 384 well Black Solid Bottom (BSB) Tissue CultureTreated (TCT) microplate (Cat. No. 3571)

- 384 well White Solid Bottom (WSB) TCT microplate(Cat. No. 3570)

- 384 well BSB Not Treated (NT) Standard microplate(Cat. No. 3573)

- 384 well BSB NT Low Volume (LV) microplate(Cat. No. 3821)

� Plates were screened using an EnVision® multimode reader(Perkin Elmer).

Results and DiscussionThe Value of Choosing the Correct MicroplateIn any assay, selecting the correct microplate is critical forassay success; it is essential to understand that each wellcontains its own microenvironment for any biochemical orcell-based assay. Microplate dimensions (e.g., well geometry,depth, and surface area), surfaces and color all play importantroles in achieving optimal results. Therefore, when designingan assay the following questions should be addressed (Fig. 1):

1. What type of assay will be performed? (i.e., What type ofsurface will be needed?)a. Biochemical:

i. Not treatedii. Nonbindingiii. Medium bindingiv. High binding

Understanding the Relationshipbetween Automation/Instrumentationand Microplates

SnAPPShotsA brief technical reportfrom the CorningApplications Group

b. Cell-Based:i. Tissue culture treated (TCT)ii. Corning® CellBIND® Surface treatediii. Poly D-lysine (PDL) coatediv. Ultra-Low Attachment Surfacev. Biological coated microplates (collagen, laminin,

etc.)2. What type of detection is required?

a. Fluorescencei. Fluorescence intensityii. Time resolved fluorescence-like assays

b. Luminescencec. Absorbanced. Label-free (e.g. Epic® technology)

3. What type of reader is required?a. Top readerb. Bottom readerc. Top/bottom readerd. High content imager

4. Microplate density and type?a. 96 well normal volume or 96 half area microplatesb. 384 well normal volume or low volume microplatesc. 1536 well microplatesd. Polystyrene

e. Polypropylene cyclic olefin copolymer (COC) –storage and chemical resistance

As mentioned above, selecting the correct microplate for theassay is critical for a successful experiment; it is important tounderstand when to use a black, white, or clear microplate(Table 1). Typically for luminescence-based assays, due tothe low energy produced by the biochemical reaction releas-ing light, the recommendation is to use a white coloredmicroplate. The white colorant allows for increased reflectivecapacity for data capture.

It is widely accepted that for fluorescence intensity assays, ablack colored microplate is the optimal choice. Fluorophores,when excited by an energy source, release a large amountof energy in the form of light. The black colorant used in

2

Figure 1. Flow chart designed to select the best microplate for the assay and instrument.

3

the microplates helps to reduce background by absorbingsome of the emission energy as well as preventing cross-talkbetween wells. Background or noise is of critical concern inany fluorescent assay. Utilization of a clear microplate for afluorescence-based assay may lead to photobleaching, photo-chemical destruction of a fluorophore, and high crosstalklevels between wells resulting in a low signal to noise ratio.For any assay, a high signal to noise ratio is preferred toreduce the possibility of false-positives and negatives. A lowsignal to noise ratio may mask subtle differences in the fluo-rescent signal leading to the potential of false negatives.

In most cases, a white microplate for a fluorescence-basedassay may also lead to a poor signal to noise ratio becausethe overall signal (including background) will be amplified.However, this result is dependent on whether the fluorescentdye is a high or low energy emitter. A low energy emitteryields better results in a white microplate compared to ablack microplate. For example, in Time-Resolved-Fluores-cence Resonance Energy Transfer (TR-FRET) assays astable, low energy emitter fluorescent molecule is used toexamine protein-protein interactions. As a result, a whitemicroplate would be preferred due to its ability to amplifythe low signal produced by the protein interactions. Fig. 2depicts another instance when selecting a white microplatewould be optimal to achieve the highest fluorescent signal.

Understanding the impact of surface area, well depth, andwell number is important when choosing a microplate.When performing cell-based assays it is vital to know thesurface area of that given microplate. Surface area is a dis-tiguishing feature between vendors who sell microplates.It is highly recommended that when seeding microplatesfor any cell-based assay that seeding is performed by surfacearea (cells/cm2) rather then cells/well. Not all microplateshave the same surface area despite having the same numberof wells (e.g., normal volume (NV) vs. LV microplates). Fordetailed information on microplate dimensions and compati-bility with various readers please visit http://www.corning.com/lifesciences/us_canada/en/technical_resources/product_guid/eq_cp.aspx. It may be beneficial to consider miniaturizing anassay into a LV microplate so that the same microplate for-mat can be used while reducing reagent consumption, cost,and still minimizing impact on automation.

Lastly, when selecting a microplate for an assay, it is essentialto understand how the microplate surface can impact theresults. As mentioned above, there are different surface typesbased on the assay that is being performed (biochemical or

cell-based). When performing biochemical assays such asan enzymatic assay, it may be best to use Corning’s NBS™

microplate. This microplate is designed to significantlyreduce (<2 ng/cm2) protein and nucleic acid binding thusproviding researchers with maximum accessibility of assaycomponents in solution compared to other microplates.However, due to the nature of the coating, the NBS micro-plate is not recommended for cell-based assays. For bio-chemical reactions where proteins (antibodies, antigens etc.)are required to bind to the surface, the better option is to useof the High Binding (HB) surface microplate. The HBmicroplate enhances binding of medium to large biomole-cules (>10 kD) that are positively charged, with or withouthydrophobic regions. Furthermore, when performing cell-based assays it is necessary to use a microplate that is designedfor cell attachment (e.g. TCT, Corning® CellBIND® Surfaceor PDL). While some cells adhere well under normal con-ditions, the cells may detach during an assay with multiplewash steps. If cells are detaching during an assay one optionwould be to switch to a microplate precoated with PDL orCorning® CellBIND® Surface treated. Alternatively, otherbiological coatings could be used (e.g., collagen and laminin);however, the coatings and concentrations would need to beoptimized for each cell line. Figure 3 provides an example

600005000040000300002000010000

60 k/cm2 50 k/cm2 40 k/cm2 30 k/cm2 20 k/cm2

BSBWSB

RFU

4000300020001000

0

Figure 2. CHO-K1 cells were plated at various seeding densities for 24 hin phenol red- free 2% FBS containing media. Cells were incubated withMultiTox-GLO GF-AFC reagent for 1 h at 37°C. Fluorescent signal wasmeasured with the Perkin Elmer EnVision® instrument λex 400 nm,λem 515 nm). The results demonstrate that a white solid bottom (WSB)microplate yields higher relative fluorescence units (RFU) than a blacksolid bottom (BSB) microplate.

Table 1. Selecting the Best Microplate for the Assay

Detection Type Sensitivity Background Handling Plate

Colorimetric Low Sometimes Very easy ClearRadiometric Very high No Difficult WhiteLuminometric Very high No Easy WhiteFluorometric Medium to High Varies Easy Black*Label Free Very high No Easy Black with optical biosensors*Note that some assays require a white microplate for optimal results.

on how selecting the correct microplate surface is importantfor assay optimizations. For more detailed information onrecommended microplate surfaces please visit http://www.corning.com/lifesciences/us_canada/en/technical_resources/surfaces.aspx.

Importance of Instrumentation SettingsAn essential part when setting up an experiment and validatingall the necessary components (reagents, microplate compati-bility, instrument settings, etc.), is to confirm the instrumentsettings as they relate to the microplate. Settings to considerinclude, but are not limited to, understanding the variousattributes of the instrument (bottom reader vs. top reader,etc.), Z-height, and microplate dimensions (e.g., depth ofwell, well position, length and width of microplate).

Z-height is defined as the height between the well bottomand focal plane. Well bottom elevation and the volume ofliquid will impact the optimal Z-height for an experiment(Fig. 4A). Therefore, for each experiment a researcher wouldwant to re-optimize the Z-height if there is a change in eitherthe microplate type (well height and/or well geometry) orliquid volume. Often minor changes in the Z-height can resultin a decrease in signal and a potential increase in the overallmicroplate CV (coefficient of variance, a statistical measure-ment that analyzes the distribution of data points around themean) (Fig. 4B). In the example shown in Fig. 4B, the opti-mal Z-height is 4 mm. By either increasing or decreasing theheight, the signal in this example expressed as relative fluor-escence units (RFU), and the CV are greatly impacted.

Confirming and validating the correct microplate dimensionsin an instrument is important to achieving optimal results.For example, analyzing a 384 well LV microplate under thesettings for a 384 well NV microplate may result in less thenoptimal results due to higher then expected microplate CVs(Fig. 5A). Figure 5B shows the differences in well geometrydimensions between the Normal and low volume 384 wellmicroplates. Proper alignment and optimization is criticalwhen changing microplate formats. As depicted in the figure,there is a significant difference between the well depth of the

NV microplate (11.43 mm) and the well depth of the LVmicroplate (9.39 mm). This difference in well depth willgreatly impact the optimal Z-height.

Importance of Proper Liquid Handling TechniquesIn addition to optimizing Z-height, accurate dispensing ofliquids into a microplate is important for obtaining highquality data (e.g., excellent CVs). If each well in a microplatehad a different volume as a result of poor liquid handling, theoptimal Z-height for each well would be different. As a result,the data would be inconsistent from well to well leading tohigh CVs and the possibility of false-positives and negatives.

4

Figure 4. (A) Z-height isthe distance betweenthe lens and the sensingvolume and determinesthe focal plane. (B) Per-forming an assay withsuboptimal Z-height canresult in a decrease insignal and increase in CV.

A B

Figure 3. Selecting the correct microplate surface is important for allassays. In the example depicted above two different microplate surfaceswere used for an enzymatic assay, NBS™ and Not treated. The data demon-strate that the NBS microplate greatly enhances enzymatic activity com-pared to the not treated surface. The increase in activity with the NBSmicroplates is an effect of the lack of protein and nucleic acid binding tothe microplate thereby providing researchers with maximum accessibilityof the components in solution.

50 pg/µL Enzyme

0.0E+00

5.0E+08

1.0E+09

1.5E+09

0 20 40 60 80Incubation Time (min.)

RFU

NBS™Microplate

Nontreated

5

Fig. 5C demonstrates that the same XY coordinates used for alarger well microplate (e.g., 384 NV) may not work for asmaller well microplate (e.g., 384 LV) due to the smaller welldiameter. Additionally, for label-free assays accurate liquidhandling parameters are essential for optimal results. Takentogether, microplate readers, handlers and liquid dispensersshould be optimized for each new microplate type, and it ishighly recommended that the instrument be re-optimizedprior to setting up any new assay.

Summary� Each microplate needs to be carefully selected for the assay.

� Instrumentation settings for each assay need to be optimizedto the microplate being used.

� Proper liquid handling is important for achieving optimal CVs.

� Collectively, each of these areas needs to be evaluatedbefore a successful assay can be attempted.

Figure 5. (A) Reading a LV microplate calibrated to a NVmicroplate results in a more dispersed signal across the microplate and a higher CV value. Asdemonstrated above a LV microplate calibrated to LV settings (red triangles) results in a uniform signal across the microplate. However, the samemicro-plate calibrated to NV settings (green squares) leads to a high degree of dispersion across the microplate. This effect is due to the differences in the welldiameter and depth between the twomicroplates. (B) Microplate dimensions for a 384 NV and LV microplate. (C) Dispensing liquid with the proper XYcoordinates is more important in a LV microplate compared to a NVmicroplate due to the smaller area of the well. Poor liquid handling may lead to a scat-tering of data points across a microplate as seen in (A).

A

B

C

NV Plate: OK Not OK

OK Not OK

Scenario 1 Scenario 2

LV Plate:

Reading with NVmicroplate,calibrated with NVCV = 2.17%

Reading with LV microplate,calibrated with NVCV = 11.03%Reading with LV microplate,calibrated with LVCV = 2.52%

RFU

50000

40000

30000

20000

100000 48 96 144 192 240 288 336 384

Well Number

For a listing of trademarks, visit us at www.corning.com/lifesciences/trademarks.All other trademarks in this document are the property of their respective owners.Corning Incorporated, One Riverfront Plaza, Corning, NY 14831-0001 ©

2011

,201

2Corning

Inco

rporated

Printed

inU.S.A.10

/12

POD

CLS-

AN-182

REV3

For additional product or technical information, please visit www.corning.com/lifesciencesor call 1.800.492.1110. Outside the United States, please call 978.442.2200 or contact yourlocal Corning sales office listed below.

Corning IncorporatedLife Sciences836 North St.Building 300, Suite 3401Tewksbury,MA 01876t 800.492.1110t 978.442.2200f 978.442.2476

www.corning.com/lifesciences

WorldwideSupport Offices

AS IA /PAC I F I CAustralia/New Zealandt 0402-794-347Chinat 86 21 2215 2888f 86 21 6215 2988Indiat 91 124 4604000f 91 124 4604099

Japant 81 3-3586 1996f 81 3-3586 1291Koreat 82 2-796-9500f 82 2-796-9300Singaporet 65 6733-6511f 65 6861-2913Taiwant 886 2-2716-0338f 886 2-2516-7500

EUROPEFrancet 0800 916 882f 0800 918 636Germanyt 0800 101 1153f 0800 101 2427The Netherlandst 31 20 655 79 28f 31 20 659 76 73United Kingdomt 0800 376 8660f 0800 279 1117

All Other EuropeanCountriest 31 (0) 20 659 60 51f 31 (0) 20 659 76 73

L AT I N AMER I CABrasilt (55-11) 3089-7419f (55-11) 3167-0700Mexicot (52-81) 8158-8400f (52-81) 8313-8589

® cellgro®

The Corning Family of Brands

®®

Beginning-to-end SolutionsAt Corning,we continuously strive towards improving efficiencies and developing new products andtechnologies for life science researchers.We have scientists working in Corning R&D labs across theglobe, doing what you do every day. Our technical experts understand your challenges and yourincreased need for high-quality products.It is this expertise, plus a 160 year legacy of Corning innovation and manufacturing excellence, thatputs us in a unique position to be able to offer a beginning-to-end portfolio of high-quality, reliablelife sciences consumables.

Seed Expand

Store

Assay

Harve

st

CELL CULTURESOLUTIONS

Storage

Screening

Assa

y Preparation Asssay Development

DRUG DISCOV-ERY SOLUTIONS