J A N U A R Y 1 , 2 0 0 1 / A N A LY T I C A L C H E M I S T R Y 99 AA

How good is electrospray ionization MS(ESI-MS) at detecting weakly boundcomplexes? Richard Griffey and col-leagues at Ibis Therapeutics report aspectrum of a complex held togetherby a single hydrogen bond. They alsodescribe the binding stoichiometry forcomplexes of multiple ligands with RNAand an example where a ligand binds totwo different sites. The key is choosingthe right conditions for the ESI-MS experiment.

The experiments focus on low-affinitycomplexes formed with a 27-nucleotideRNA model of the 16S rRNA A-site(16S). As a first step, the researchers ad-just the “harshness” of ESI-MS condi-tions by looking for the spectrum ofammonia-adducted ions of 16S. Thisspectrum appears by either loweringthe capillary–skimmer potential or thedesolvation capillary temperature.

Binding of the ligand 2-deoxystrept-amine (2-DOS) was the basis of severalexperiments. In one, dissociation con-

stants for ligand-to-complex ratios rang-ing from 1:1 to 4:1 were determined.However, the chemistry is even morecomplex, because 2-DOS binds to twodistinct sites on 16S. To investigate thestability of these two forms, the re-searchers turned to colli-sionally activated dissocia-tion and MS/MS. Byvarying the relative disso-ciation energy, the com-plex was dissociated intothe 1:1 complex and thefree 16S and 2-DOS ions.

MS also revealed multi-ple different ligands con-currently binding 16S.One particularly complexexperiment found themono- and bis-ligandcomplexes of 2-DOS and16S, plus a complex with2-DOS and a diaminotria-zole simultaneously boundto 16S. The triazole is be-

lieved to be hydrogen bound to theRNA. Finally, this tour de force studyreports the relative gas-phase activationenergies for the dissociation of a seriesof ligand–16S complexes. (J. Am. Chem.Soc. 22000000,, 122, 9933–9938)

nn ee ww ss

16S [M-5H+]5-

16S + DT

16S + 2-DOS

16S + 2-DOS + DT

16S + (2-DOS)5

1720 1740 1760 1780 m/z

Low-affinity complexes galore. ESI-FT ion cyclotron reso-nance mass spectrum showing the concurrent binding of2-DOS and 3,5-diaminotriazole to 16S.

LLooww--aaffffiinniittyy ccoommpplleexxeess

ANALYTICAL CURRENTS

Here is a handy trick. David Padowitz and Ben-

jamin Messmore of Amherst College investi-

gate exchange between long-chain molecules

in solution and those arranged as a surface

monolayer. However, they use scanning tun-

neling microscopy (STM), which is too slow to

follow the dynamics of individual molecules. To

get around this problem, they introduce a trac-

er molecule that allows the exchange to be fol-

lowed by the “slow-moving” STM.

The system under investigation is the C33H68molecule n-tritriacontane, which is adsorbed

onto graphite. These molecules move on and

off the surface in milliseconds—much faster

than the seconds required for an STM image.

Instead of looking at individual molecules,

the researchers tracked the entire surface.

To do that, tracer molecules, which replace a

central methylene group in the carbon chain

with either a sulfur (thioether) or oxygen

(ether), are mixed with the alkane or with

each other in a ~1:10 ratio. The mixture co-

crystallizes on the graphite surface, and the

heteroatom is easily viewed.

The researchers find that the sulfur is par-

ticularly easy to track, and they can watch

the entire monolayer turn over in a few tens

of seconds. The rate of exchange is deter-

mined from the data, and the researchers

speculate on the mechanism. They also see

different dynamics at the domain boundaries.

(J. Phys. Chem. B 22000000,, 43, 9943–9946)

Follow the bright dots. Two images trackingthe thioethers on the surface. It takes 2 sec-onds to collect an STM image with a 2-sec-ond delay between scans.

DDyynnaammiiccss wwiitthh aa ttrraacceerr

11 00 AA A N A LY T I C A L C H E M I S T R Y / J A N U A R Y 1 , 2 0 0 1

nn ee ww ss

AA rraaiinnbbooww ooff mmoolleeccuullaarr bbeeaaccoonnss

With the introduction of “wavelength-shifting” molecular beacons, the futurefor these fluorescent probes looks espe-cially bright, not to mention colorful.Sanjay Tyagi, Salvatore Marras, andFred Kramer at the Public Health Re-search Institute in New York developedthe new probes, which emit light at var-ious colors yet are excited by a single,

monochromatic light source.Like traditional molecular beacons,

the wavelength-shifting version is a sin-gle-stranded, hairpin-shaped oligonu-cleotide probe that does not emit a sig-nal when the hairpin is closed—that is,when the probe’s fluorophore andquencher are close together. The signalbecomes detectable when the hairpin is

open, and, thus, the quencher and fluo-rophore are apart.

However, there are two fluorophoresin the new molecular beacons—a “har-vester” that absorbs energy from the ex-citation light and an “emitter” that re-leases energy as a fluorescent signal. Theemitter fluorophore is joined to the 5´-end of the probe by a short spacer se-quence, the length of which can be adjusted to optimize the fluorescenceres onance energy transfer between thetwo moieties.

The researchers tested three wave-length-shifting molecular beacons. Eachone used fluorescein as the harvester;the emitter was either 6-carboxyrhoda -mine 6G, tetramethylrhodamine, orTexas red. In all cases, the signal wasdetected predominantly in the emissionrange of the emitter, not the harvester.In addition, the researchers note thatthe new molecular beacons were oftenbrighter than conventional ones, andthey attribute that change to more effi-cient energy absorption. (Nat. Biotech-nol. 22000000,, 18, 1191–1196)

CCyyssttiicc ffiibbrroossiiss ddiiaaggnnoossttiiccaarrrraayyAlthough some mutations that cause cystic

fibrosis can be identified by genetic testing,

the more universal diagnostic technique is

the so-called sweat test, or more correctly,

the pilocarpine iontophoresis test for elec-

trolytes in sweat. Although the sweat test is

convenient and simple to administer, analyz-

ing the samples is not as easy. But Aogán

Lynch, Dermot Diamond, and Matt Leader at

Dublin City University (Ireland) and SENDX

Medical, Inc., hope to change that with a

potentiometric ion-selective microelectrode

array that may replace techniques such as

ANALYTICAL CURRENTS

MMeeaassuurriinngg lliiggaanndd––rreecceeppttoorr ffoorrcceess

Cell–cell adhesion generally dependson specific binding between recep-tors on the surface of one cell andligands on the surface of another.The amounts of force needed tobreak apart these receptor–ligandpairs are typically measured “in bulk”using macroscopic methods or nano -probe techniques at the single mole-cule level. To help develop a morecomplete picture, E. Sackmann andcolleagues at the University of Cali-fornia–Los Angeles, the MPI for Bio-chemistry (Germany), and the Tech-nische Universität München (Germany)describe a middle-ground “mesoscop -ic” method for measuring lig-and–receptor unbinding forces.

In the new method, a “test cell”—in reality, an artificially formed giantvesicle—is immobilized on a substratevia receptor–ligand binding. Then amagnetic bead is attached to the topof the cell, and a vertical force of0.1–2 pN is generated by pulling on the bead. Reflection interferencecontrast microscopy monitors thestructure, and the magnitude and di-rection of the forces are determinedquantitatively.

The approach allows the measure-ment of unbinding forces under bio-logically relevant conditions, the re-searchers say. Because the cell is large,the “pinning center” holding it tothe substrate may be ~1000 Å acrossand may contain ~100 adhesion-pro-

ducing receptor–ligand pairs. Depend -ing on the force applied, the ruptur-ing of single or multiple receptor–li -gand bonds can be studied.

The researchers tested the setupusing the integrin receptor �IIb�3from blood platelets and artificial li -gands and found that the unbindingforces were much smaller than thosemeasured in single molecule studies.The reason for the difference, the re-searchers suggest, is that the newmethod seems to measure leverageforces, which may break the recep-tor–ligand bonds efficiently, whereasother methods measure pure tractionforces. (Langmuir, 22000000,, 16, 8984–8993)

B-field

θFc

θOc

NS

Rc�

h

A “test cell” is held to a substrate by re-ceptor–ligand binding. A magnetic beadattached to the top of the cell allows avertical force to be applied.

J A N U A R Y 1 , 2 0 0 1 / A N A LY T I C A L C H E M I S T R Y 11 11 AA

nn ee ww ss

MMeerrccuurryy mmiiccrrooppuummppss

In the realm where millimeters loomlarge, familiar forces such as gravity andfriction no longer reign. Instead, thekey to success is harnessing the lesser-known forces that have supplanted theirmacroscale oppressors. So when MarcPorter, Jing Ni, Chuan-Jian Zhong, andShelley Coldiron at Iowa State Univer -sity and the Ames Laboratory–U.S. De-partment of Energy wanted to make aminiature pump, they abandoned me-chanical methods and turned to surfacetension.

Others have used surface tension topump small volumes of fluids, a conceptknown as the Marangoni effect. In theearlier work, a surface tension gradientwas generated between two electrodesthat were positioned on the walls of achannel. For example, water-solublemolecules might be transformed from asurface-inactive state to a surface-activestate using electrochemical methods.Controlling this transformation wouldcontrol the surface pressure in the chan-nel, which would drive the fluid.

The Iowa researchers took a com-pletely different approach, developingan electrochemically actuated mercurypump, which is described in the currentissue of Analytical Chemistry (pp 103–110). Porter says the pump was inspiredby the mercury beating heart, a com-mon laboratory demonstration in whichchanges in the accumulated charge onthe surface of a drop of mercury alterthe surface tension. These changes insurface tension, in turn, relax or contractthe curvature of the mercury drop. “This

flame photometry and anion exchange chro-

matography.

The array used is produced commercially

and consists of solid-state microelectrodes

screen-printed over a salt-doped hydrogel

layer. To give a more comprehensive analysis,

the array simultaneously measures levels of

Na+ and Cl–, which are the two most reliable in-

dicators; K+, which is often used to confirm the

diagnosis; and other parameters, such as Ca2+,

pH, pO2, and conductivity. The sample turn-

around time was ~1 min, compared with sev-

eral hours using current analysis techniques.

Cluster analyses of Na+, K+, and Cl– levels

together and the Na+ levels alone in 21 sam-

ples yielded well-defined clustering of posi-

tive and negative (normal) samples. Unfortu-

nately, the clustering for Cl– alone, which is

generally regarded as the best single marker

for cystic fibrosis, was not as good because

of interference from NO3–, which is typically

found in the reagent to stimulate sweat.

(Analyst 22000000,, 125, 2264–2267)

DDNNAA ddeetteeccttss lleeaadd

Decades of genetic research are payingoff in unexpected ways. In this case,a deoxyribozyme, which is a catalyti-cally active DNA, is the basis of a se-lective Pb2+ biosensor. According toYi Lu and Jing Li of the Universityof Illinois at Urbana–Champaign, thisunique sensor boasts a >80-fold pref-erence for Pb2+ over other commonmetal ions and a quantifiable detectionrange of 10 nM–4 µM. Moreover, thisapproach suggests that a wealth ofnew biosensors can be developed, be-cause researchers can fish through vastlibraries of random DNA or RNA seg-ments for other recognition elements.

The deoxyribozyme in this bio sen -sor cleaves a single RNA linkage in aDNA substrate when activated by Pb2+.To make this work, the 5´-end of thesubstrate, a DNA/RNA chimera, waslabeled with a fluorophore, and the3´-end of the deoxyribozyme waslinked to a quencher. The enzymeand substrate hybridized together,and the fluorescence was quenched;the addition of Pb2+ led to the cleav-age of the substrate and a ~400% in-crease in fluorescence intensity. The

biosensor retained its selectivity to500 nM Pb2+ in the presence of equalamounts of eight other divalent metalions and under simulated physiologi-cal conditions. (J. Am. Chem. Soc.22000000,, 122, 10466–10467)

RESEARCH PROFILES

fluor

esce

nce

inte

nsity

wavelength (nm)

570 600 630 660 690

80000

60000

40000

20000

0

substrate strand 17DScleavage site

enzyme strand 17E

3'–G T A G A G A A G G rAT A T C A C T C A –5'

5'–C A T C T C T T C T A T A G T G A G T – 3'

GA

CC

CG

G GC

C

A AGT

a.

b.III

I

II

I

II

III

+Enz

+Pb2+

(a) Proposed secondary structure of thedeoxyribozyme (17E) and the substrate(17DS); rA is a ribonucleotide adenosine.(b) Fluorescence spectra of (I) the sub-strate alone and the enzyme with substrate(II) before and (III) after adding 500 nM Pb2+.

11 22 AA A N A LY T I C A L C H E M I S T R Y / J A N U A R Y 1 , 2 0 0 1

nn ee ww ss

property is pretty unique to mercurybecause it’s a liquid metal,” says Porter.“Although you can do the same thingwith gallium, you have to work at high-er temperatures, so it is not as practicalor easy to study.”

In thinking about the mercury beat-ing heart, the researchers realized that“if you’re changing the shape, you’redoing physical work,” Porter explains.The question then became how to har-ness that shape change into a mecha-nism that would actuate fluid flow. Afterperforming some calculations to be surethat the idea was sound and trying afew options, the researchers hit on anapproach that worked.

The “proof-of-concept” pump has twoparts: an open-topped main reservoir anda T-shaped capillary insert. The verticalsection of the T is an open- bottomed

capillary that will fit into the reservoir;the horizontal section is the flow channel,which intersects the capillary. After mer-cury and an aqueous electrolyte are addedto the reservoir, the insert is fitted on top.Capillary action then draws mercury intothe insert, resulting in two concentricmercury col umns. At this point, the rela-tive heights of the two columns reflectthe balance between capillary forces andgravity, Porter says.

To activate the pump, a voltage is ap-plied across two electrodes—one in con-tact with the mercury at the bottom ofthe reservoir, and the other in contactwith the electrolyte at the top of thereservoir. The application of the voltagechanges the surface tension in the outermercury column, which leads to a pres-sure change, and the mercury levels inboth columns shift to a new equilibrium

point, Porter explains. By applying awaveform voltage, the pump can be cy-cled between “fill” mode, in which themercury in the inner column distendsdownward, and “pump” mode, in whichthe mercury in the inner column ex-tends upward, forcing liquid throughthe flow channel. The magnitude of thevoltage determines how much the heightof the inner mercury column changes,and the frequency of the waveform de-termines how fast the pump cycles.

Once the researchers got the basicmechanism working, they arranged fourmercury pumps, which can be actuatedindividually, in a series. The result is, ef-fectively, a miniature mercury peristalticpump with four pistons, which drivesthe fluid in one direction, Porter says.

These days, the researchers are tryingto reduce the scale of the mercury pumpfrom millimeters to several hundred mi-crometers. In addition, they are tryingto figure out how to prevent the mer -cury from coming in contact with thesample. “We’ve been using a thin Teflonmembrane that would push the liquid,”Porter explains, “but we don’t have theright material yet. The ones we have usedso far are too stiff.”

Nevertheless, the researchers are en-couraged by their results so far and evenhope to fly the pump on NASA’s micro-gravity plane one day. Porter notes thatthe team will probably first try a simplerexperiment on the microgravity plane toget used to working with fluids underthose conditions. “But the mercury pumpwork was supported by NASA,” he adds,“so we would like to be able to use it inthat environment at some point.”

––EElliizzaabbeetthh ZZuubbrriittsskkyy

Electrochemicalswitching

E

E

Fluidflow

E

Fill mode

E

Pump mode

The “fill” and “pump” modes of the mercury pump. In fill mode, the mercury in the inner columndistends downward. In pump mode, the mercury in the inner column extends upward, displac-ing fluid in the flow channel at the top. Applying a step voltage switches between the modes.

RESEARCH PROFILES

““FFoouurr--ddeeccaayy”” DDNNAA sseeqquueenncciinngg

Four-color DNA sequencing is the back-bone of every genome project, but mostfour-color sequencing schemes are only~90% accurate. So Linda McGown andHui He at Duke University describe anew alternative in the December 15issue of Analytical Chemistry (pp 5865–

5873): a “four-decay” scheme, whichrelies on frequency-domain fluorescencelifetime detection for multiplexed DNAsequencing.

Four-color fluorescence techniques—in which the four fluorescent tags areidentified by their emission spectra—

have increased sequencing throughput4-fold, because they allow all four basereactions to be run simultaneously in asingle lane. But there is inevitably emis-sion overlap that produces errors inreading the fluorescence spectrum and,thus, misidentification of bases.

J A N U A R Y 1 , 2 0 0 1 / A N A LY T I C A L C H E M I S T R Y 11 33 AA

nn ee ww ss

In contrast to color-discrimination se-quencing, McGown turned to time-re-solved fluorescence, which measures theamount of time it takes the excited dyeto return to the ground state and emit aphoton. It produces a distinct signaturethat follows first order kinetics, makingit far easier to interpret the data. “In asense, it’s the difference between a linespectrum and a band spectrum,” saysMcGown. “Would you rather resolvefour lines or four broad spectral peaks?”

Applying the time-resolved methodto DNA sequencing “just seemed like anatural,” McGown says. But finding setsof four dyes that could withstand theconditions of the sequencingreactions was challenging be-cause most dyes currentlyin use have similar lifetimes.“The companies focus onmaking dyes with differentcolors rather than differentlifetimes,” she explains. Mc-Gown finally settled on twosets of dyes—one that is ex-cited at 488 nm and theother at 514 nm.

Once the bands have beenseparated, McGown’s teamuses a continuous laser witha modulated frequency to excite the dyes. The fre -quency of the emitted pho-ton decreases relative to thefrequency of the excitationlaser, with longer fluorescentlifetimes leading to a greatershift. Those readouts haveyielded an accuracy of 96%for reading DNA fragmentsranging from 41 to 220bases.

The team also used Fouriertransform data analysis in an attempt toseparate the frequency spectrum into theindividual components representing allfour dyes. Just as with color detection,there is some spectral overlap becausethe fragments, especially longer ones,don’t always completely separate onthe electropherogram. The researchersachieved 99% accuracy with two dyes

and 98.5% with three, but technical limi-tations prevented them from accom-plishing the technique using all four.“It’s due to data analysis and propaga-tion of error,” says McGown. Overcom-ing this limitation, she adds, “is just amatter of improving the data analysisstrategy.”

The successful application of the life-time-resolved technique with four basepairs wouldn’t just improve accuracy, itshould also allow researchers to examinelarger fragments, McGown says. Thatwould, in turn, increase accuracy andspeed by cutting down on the numberof enzymatic digestion steps.

The one disadvantage of the currentsystem is that the lifetime fluorescentmeasurement attenuates the signal, whichhurts the detection limit. The difference“depends on several factors, but . . .could be roughly an order of magni-tude,” says McGown. The obvious wayto compensate, she explains, is to usemore sample, perhaps amplifying it with

PCR. “But that can run into problemswith sample overloading, so we’re work-ing on improving the detection limits.”

There are still other challenges ahead.“We need a better four-dye system,” saysMcGown. At 488 nm, the researcherscould get four dyes that had good, in-tense signals and resolvable lifetimes.“The only problem was that one dye ismuch larger than the others, so we hadmobility problems [on the electrophero-gram],” she explains. “You get [onebase] migrating at a different rate.”

Even so, the bottom line is that thesystem could decrease errors in genomesequencing, especially if the researchers

can make improvements in the instru-mentation. “[The new method] has alot of potential to be scaled down intoa simpler, dedicated detection instru-ment,” she explains. “I’d like to im-prove its accessibility . . . to make itsimpler to incorporate into routinemethodology.”

––JJiimm KKlliinngg

Fluo

resc

ence

Inte

nsity

(Arb

itrar

y U

nits

) 1.0

0.5

0.0

1.0

0.5

0.0

1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600

2600 2700 2800 2900 3000 3100 3200 3300 3400 3500 3600 3700 3800 3900 4000

Migration Time (s)

160 180 200 220 240 260 280 300

20 40 60 80 100 120 140

Life

time(

ns)

Total Intensity T (Cy3) C (TMR)

Lifetime-resolved intensity peaks of C-terminated fragments labeled with TMR (blue) and T-terminated fragments labeled with Cy3 (red). Total intensity is shown in black.

11 44 AA A N A LY T I C A L C H E M I S T R Y / J A N U A R Y 1 , 2 0 0 1

nn ee ww ss

CChhiipp--bbaasseedd mmoossaaiicc iimmmmuunnooaassssaayyss

For immunoassays, the perpetual goal isto increase throughput by reducing theamount of reagent consumed, shorten-ing the incubation time, and performingmore assays at once. In the current issueof Analytical Chemistry (pp 8–12), Em-manuel Delamarche, André Bernard,and Bruno Michel at IBM’s Zurich Research Laboratory (Switzerland) describe a new method fordoing just that. They calltheir approach microfabri -cated mosaic (“micromosa-ic”) immunoassays and re-port that the assays consumeonly nanoliter quantities ofreagents and perform incuba-tions within a few seconds orminutes.

The works draws upon anongoing effort, led by Michel,which is centered around mi-crocontact processing to de-velop microfabricated devices.The researchers abandon mi-crowells, which are tradition-ally used for immunoassays,in favor of a flatpoly(dimethylsiloxane) sub-strate, onto which they de-posit narrow stripes of antigenusing a network of microflu-idic channels fabricated in sili-con. Then, a second networkof microchannels, perpendicu-lar to the first, delivers the so-lutions to be analyzed. (Theresearchers use simple capil-lary forces to induce flow inthe channels.) Binding theanalytes results in a mosaicpattern of tiny squares—simi-lar to the grids of dots seenwith DNA microarrays—which can be analyzed with afluo rescence microscope.

Researchers have long rec-ognized the value of such amicroarray-type format for im-munoassays. However, despitesome recent success in this

area, protein-based assays have generallybeen more difficult than DNA-based as-says to miniaturize and integrate intosmaller, highly sensitive, practical formats,Delamarche says. For example, gridlikepatterns of DNA oligonucleotides can be“built” on a surface using photolithogra-phy, but the same is not possible for pro-teins, because they can be composed of

hundreds of amino acids. (Antibodiesgenerally have ~1400 amino acids.) Inaddition, proteins can easily lose theirthree-dimensional structure when han-dled under “denaturing” conditions.

The key to making the microfabricat-ed device work was the ability to depositthe proteins with a high resolution, saysBernard. To do this, researchers have

used various approaches, includ-ing inkjet printing, drop-on-de-mand techniques, microcontactprinting, and soft lithography,which is what the IBM teamchose. “What is really pleasingto see,” Delamarche says, “isthat soft lithography not onlyhas the potential to help micro-fabrication, but it can handlefragile proteins and pattern themon a surface.”

In preliminary experiments,the sensitivity and reliability ofthe new approach compared wellwith those of traditional immuno -assays, the researchers say. In thesimple case of two-step assays(antigen–antibody pairs), bindingand detection occurred over aconcentration range of <1–1000nM. More sophisticated sand-wich-type assays (such as en-zyme-linked immunosorbent as-says) could also be performed,and in these cases, the signal in-tensity could be scaled with theamount of analyte present.

Another encouraging findingwas that dilute solutions of pro-teins could be used. Proteinsfrom the solution are depositedonto the surface of the substrateas the solution flows inside themicrochannel, Delamarche ex-plains. The challenge is that solu-tions of proteins are typically di-lute, and the volume is typicallysmall—possibly less than a nano-liter, depending on the geometryof the channel. To ensure thatenough proteins flow through

RESEARCH PROFILES

A Immobilization

θFc

θOc

NS

Rc�

h

B Blocking

C Recognition

D Reading mosaic

E Example with one pair of antigen and antibody

100 µm

Sample 1

Sample 2

Sample 3

BSA

Substrate

Channels

µFN

Antigen 1

FlowAntigen 2

Strategy for performing a micromosaic immunoassay. (A) Antigensare deposited in stripes. (B) Blocking with bovine serum albuminto prevent nonspecific binding. (C) Antibodies are allowed to bind.(D) A mosaic binding pattern results. (E) An array of rabbit immuno -globulin Gs (IgGs) interacting with anti-rabbit-IgGs.

J A N U A R Y 1 , 2 0 0 1 / A N A LY T I C A L C H E M I S T R Y 11 55 AA

nn ee ww ss

the microchannel to entirely coat theexposed region of the substrate, the re-searchers place a spongelike flow-pro-moting pad at the far end of every mi-crochannel. As the dry pad soaks upsolution, it helps draw fluid through the channel.

The short coating and binding timesseen for this device are a result of thesmall channel dimensions, which pre-vent mass transport limitations. “Thispoint is really important, and most peo-ple don’t realize it,” Delamarche says.People doing conventional assays in mi-

crowells typically wait 30 min while theproteins slowly diffuse in all directions,including toward the surface, but insidethe device’s 10-µm-high microchannels,the diffusion path of the protein fromthe solution to the surface is consider-ably smaller. Most of the proteins willdiffuse to the surface of the substratewithin seconds and adsorb. “By per-forming both the patterned deliveryof antigens and antibodies to a surfaceusing microchannels, we make an im-portant economy of solutions, of pro-teins, and time,” Delamarche adds.

The researchers note that no blindedexperiments were performed (although,in some cases, unexpected cross-reactivi-ty—in which antibodies from one speciesrecognized and bound to antibodiesfrom another species—was observed).The reason, Delamarche explains, is that“we were not comfortable developingthis work on unknown ground.” How-ever, now that the preliminary work isfinished, the team will start using real-life samples to better evaluate the diag-nostic potential of the technique.

––SSaannddrraa KKaattzzmmaann

MEETING NEWS

22000000 EEaasstteerrnn AAnnaallyyttiiccaall SSyymmppoossiiuumm—Gerald Keller reports from Atlantic City, NJ.

TTrraaccee aattmmoosspphheerriicc ggaasseess oonn aa sshhooeessttrriinngg

The necessity of cost control in the analyt-ical laboratory is a nagging problem. ButPurnendu “Sandy” Dasgupta of TexasTech University has a solution, at leastfor measuring some trace atmosphericgases: liquid-core waveguide instrumentsthat smash the usual financial barriersand rival the sensitivity of much moreexpensive spectroscopic instruments.

If you want to measure H2S or CO,then a variety of systems, some of themdisposable, are available from sources asconvenient as the local hardware store.But if you want to measure sub-parts-per-billion levels of formaldehyde, am-monia, or less frequently monitoredgasses, Dasgupta’s fiber-optic threadsmay be the way to go. The sensors arebased on readily available scintillating optical fibers, which are constructed of afluoropolymer (Anal. Chem. 11999999,, 71,1400–1407). When a light-excitationsource is applied to the side of the tube,light propagates down its length and canbe quantified with a relatively inexpensivephotodetector.

The idea came from waveguides thatwere filled with a special scintillationcompound and produced for radiationdetection. A radioactive source would

excite the fiber, and theresulting count could betaken a safe distance away.Later, the fibers were filledwith fluorescent dyes.Cheap to produce, witha long shelf-life, the wave-guides made their way tothe hobby- and education-centric Edmund ScientificCo. There they languishedin the catalog, which de-scribed them as having noknown practical use.

Dasgupta purchased oneand was observing the greenglow from its end in his artificially lit lab-oratory. He then walked down the par-tially sunlit hallway and out into directsunlight, all the time watching the greenglow intensify. Puzzled that the intensi-ty, but not the color, changed with theexcitation light, he consulted a colleaguewho noted simply that the dye, fluores-cein, always glows green. This suggestedto Dasgupta that properties of the exci-tation light would not carry over intothe emitted light, eliminating a sourceof detector interference that plagues con-ventional methods. He translated this into

liquid-filled waveguides. Anadvantage of this approachis that the reaction is oftencarried out in the wave-guide. This approach side-steps any loss of lumines-cence and signal strengthon the way to the detector,which may be a significantproblem during fast reac-tions in other instruments.For this reason, fasterthroughput is also possiblewith the new sensors.

By selecting from a va-riety of dyes and prereac-

tions that can take place outside or with-in the waveguide, different analytes suchas ammonia, hydrogen peroxide, andformaldehyde can be detected, and theresearchers are now working on detect-ing hydrogen sulfide and other gases(Anal. Chem. 22000000,, 72, 5338–5347).After examining several excitation sources,Dasgupta’s favorite was the common, in-expensive, light-emitting diodes (LEDs)found at the local electronics supply store,and with the advent of near-UV and violetdiode lasers and LEDs, he suggests thatthe new approach may really take off.

Penny-pinching wave-guides.

11 66 AA A N A LY T I C A L C H E M I S T R Y / J A N U A R Y 1 , 2 0 0 1

nn ee ww ss

DDNNAA ddiiaaggnnoossttiicc bbuussiinneessss ggrroowwiinngg

The molecular and cytogenetic disease-testing procedure market is estimated tobe $66.3 million in the year 2000 andexceed $100 million in 2005, accordingto a new market research report, The

DNA Diagnostic Business, published inOctober 2000 by Business Communica-tions Company, Inc. (www.bccresearch.com). Sequencing and PCR techniquesremain a large and growing part of the

market for genetic disease testing, andnew disease-specific sequencing proce-dures are continually being developed.

In particular, fluorescent in situ hy-bridization (FISH) cytogenetic geneticdisease-testing procedures are growingfast, with an average annual growth rateof 12.3% domestically and 13% interna-tionally. One of the most rapidly expand-ing cytogenetic areas is Her-2/NeuFISH testing for breast cancer. Consid-erable investments also are being madein emerging markets, such as newbornscreening and preimplantation diagno-sis. Among the DNA diagnostic tech-nologies currently experiencing a largegrowth in research and development areMS and biochips.

In addition to providing marketanalyses, the report profiles severalcompanies in the DNA diagnostic business, including some instrumentcom panies, and discusses some recentgene and technology patent disputes.A section on single nucleotide poly -morphism (SNP) research, includingthe SNP Consortium and database, isalso included.

––LLaauurraa RRuutthh

NNeeww AAddvviissoorryy BBooaarrdd aanndd AA--PPaaggee AAddvviissoorryy PPaanneell mmeemmbbeerrss aappppooiinntteedd

Eight new members from government,academia, and industry have been se-lected to serve 3-year terms on Analyt -ical Chemistry’s Editorial AdvisoryBoard. Established in the 1940s, the

board is a vital link between the editorsand the analytical chemistry community.Each January, membership is rotated asnew appointees replace members whoseterms have expired. The chair of the ACS

Division of Analytical Chemistry servesa 1-year term as ex officio representativeof the Division.

IIrraa LLeevviinn, acting director, Division ofIntramural Research, and chief of Molec-

BUSINESS

PEOPLE

PPaacckkaarrdd pplluunnggeess ddeeeeppeerr iinnttoo lliiffee sscciieenncceess

Right now, the place to be is in thelife sciences market. The productionof bio chip and microarray technologyhas become a focus for many instru-ment companies, including PackardBioScience.

In an effort to solidify itself in thelife sciences market, Packard acquiredGSLI Life Sciences in October 2000and created the spin-off companyPackard BioChip Technologies. Thenew company is made up of an instru-ments division and a ventures division.The GSLI acquisition falls under theventures division, which plans to com-bine Packard’s piezoelectric printingtechnology with the confocal lasertechnology of GSLI to produce bio -chips and microarrays. Nevertheless,the two technologies will be used in-dependently, according to Tricia Cald-

well, a spokesperson for Packard Bio-Science. “One technology will createthe chip, [and] the other will analyzeit,” she says. Packard also announcedplans to sell its Canberra Industriessubsidiary, which has focused on nu-clear measurement instrumentation,says Jason Alter, Packard’s directorof business development.

Packard is also collaborating withother companies that are breakinginto the life sciences market. The twomost publicized collaborations arewith Motorola and Oxford Glyco-Sciences. In 1999, Packard licensedits high-through put inkjet printingtechnology to Motorola, a producerof three-dimensional bioarrays. Thecollaboration with Oxford, which isknown for its proteomics work andproducing protein biochips, representsa more forward-looking approach byPackard. “No technologies are beingexchanged between the companies inthe deal, but rather, both companieshave agreed to jointly explore the fea-sibility of producing protein biochips,”says Alter.

––WWiillddeerr SSmmiitthh

1 8 A A N A LY T I C A L C H E M I S T R Y / J A N U A R Y 1 , 2 0 0 1

nn ee ww ss

ular Biophysics in theNational Institute ofDiabetes and Diges-tive and Kidney Dis-eases at the NationalInstitutes of Health,received his B.S. de-gree from the Uni-

versity of Virginia and his Ph.D. fromBrown University. Levin’s research inter-ests include vibrational spectroscopic stud-ies involving lipid–protein interactions anddomain structures in biological membranesand the development of IR, Raman, andspectroscopic imaging techniques.

VViiccttoorriiaa MMccGGuuff--ffiinn,, professor atMichigan State Uni-versity, received herB.A. degree fromEastern MichiganUniversity and herPh.D. from Indiana

University. Her research interests includethe theory and practice of capillary LCand CE, the development of novel laser-based spectroscopic detectors, and the useof computer simulation methods in sepa-ration science and applied spectroscopy.

PPeetteerr SScchhooeenn--mmaakkeerrss,, professorat the Universityof Amsterdam (TheNetherlands) andprincipal researchchemist at the ShellResearch and Tech-

nology Centre in Amsterdam, receivedhis Ph.D. from the Technical Universityin Delft (The Netherlands). His researchinterests include separation techniques,systematic optimization strategies, andapplications of artificial intelligence.

BBaarrbbaarraa LLaarrsseenn,,a senior research as-sociate at DuPont,received her Ph.D.from the Universityof Delaware. Larsen’sresearch interests in-

clude using biological MS for determin-ing the structures of small molecules,identifying protein sequences, and de-veloping new high-throughput screen-ing methods.

MMaarrkk WWiigghhttmmaann,,professor at the Uni-versity of North Car-olina–Chapel Hill, re-ceived his B.A. fromErskine College andhis Ph.D. from theUniversity of North

Carolina. His research interests includethe development of microsensors andmicroelectrode-based techniques, andthe investigations of neurotransmitters.

RRiicchhaarrdd SSaacckkss,,professor at the Uni-versity of Michigan,received his B.S. de-gree from the Uni-versity of Illinois andhis Ph.D. from theUniversity of Wiscon-

sin. His research interests include instru-ments and strategies for high-speed GC,columns with tunable and programma-ble selectivity, and TOFMS for rapidcharacterization of organic compounds.

JJoohhnn FFrreennzz,, di-rector of manufactur-ing science and tech-nology at Genentech,received his Ph.D. inchemical engineeringfrom Yale University.His research interests

include both upstream and downstreamproduction of recombinant protein phar-maceuticals, protein chromatography,process monitoring, and data analysis.

BBrruuccee CChhaassee,,senior research fellowat DuPont, receivedhis B.S. from WilliamsCollege and his Ph.D.from Princeton Uni-versity. His research

interests include IR and Raman vibra-tional spectroscopy of polymeric sys-tems, and vibrational micro scopy andimaging using near-field techniques.

AA--PPaaggee AAddvviissoorryy PPaanneellAnalytical Chemistry has also selectednine new members to serve 3-year termson its A-Page Advisory Panel.

DDeerrmmoonntt DDiiaa--mmoonndd,, associate di-rector of the Nation-al Centre for SensorResearch at DublinCity University (Ire-land), received hisPh.D. at Queen’s

University Belfast (Ireland). His researchinterests include the design, characteriza-tion, and synthesis of molecular receptors,and the development of rapid, multi-anayte sensing systems, which use elec-trochemical and optical sensor arrays.

CCaarrooll RRoobbiinnssoonn,,professor and direc-tor of MS at the Oxford Centre forMolecular Sciences,received her Ph.D.from the Universityof Cambridge (U.K.).

Her research interests include applyingMS to structural biology; investigatingthe mechanisms of protein folding, mis-folding, and disease; and assemblingmacromolecular complexes.

KKllaauuss--DDiieetteerrFFrraannzz,, head of Cen-tral Analytical Serv-ices of Merck KGaA(Germany), receivedhis Ph.D. from theUniversity of Frank-furt (Germany). His

research interests include hyphenatedmethods, automation, chemometrics,miniaturization, and information tech-nologies in analytical science.

HHuubbeerrtt GGiirraauulltt,, professor at the

PEOPLE

2 0 A A N A LY T I C A L C H E M I S T R Y / J A N U A R Y 1 , 2 0 0 1

nn ee ww ss

Ecole PolytechniqueFederale de Lausanne(Switzerland), receiv -ed his Ph.D. fromthe University ofSouthampton (Eng-land). His researchinterests include

the study of charge-transfer reactions at soft interfaces and the developmentof micro analytical devices for proteinanalysis.

NNeeiillss HHeeeeggaaaarrdd,,head of research anddevelopment in theautoimmunologydepartment at theStatens Serum In -stitut (Denmark),received his M.D.

and D.Sc. from the University of Copen-hagen (Denmark). His research interestsinclude the development and applicationof microanalytical methods to study li -gand binding to proteins and peptides.

EEddggaarr AArrrriiaaggaa,,assistant professorat the University ofMinnesota, receivedhis Ph.D. from Dal-housie University(Canada). His re-search interests in-

clude organelle analysis based on CE; micrototal analysis systems; and MS forcharacterizing subcellular interactions,drug trafficking, and proteomic strategies.

SS.. MMiicchhaaeellAAnnggeell,, associate pro-fessor at the Univer-sity of South Caroli-na, received his Ph.D.from North CarolinaState University. Hisresearch interests in-

clude developing methods for remoteand in situ chemical analysis using spec-troscopic techniques (e.g., Raman andlaser-induced breakdown spectroscopy)and fiber-optic chemical sensors.

RRoobbeerrtt DDuunnnn,,associate professorat the University ofKansas, received hisPh.D. from the Uni-versity of California–San Diego. His re -search interests in-

clude the study of protein channel dynam-ics, artificial membrane systems, singlemolecule spectroscopy, and the develop-ment of new microscopy methods.

RRaacchheell LLoooo,, research associate atPfizer Global, receiv -ed her B.S. degreefrom the Universityof Wiscon-sin–Madison and herPh.D. from Cornell

University. Her research interests includeapplying MS to biology, gel elec-trophoresis, and antibacterial drug dis-covery.

22000011 AACCSS aawwaarrddssSeveral scientists in the analytical chem-istry community will receive the 2001American Chemical Society awards at the221st National Meeting in San Diego,

CA, in the spring.

KKllaauuss BBiieemmaannnn,,professor at the Mas-sachusetts Instituteof Technology, willreceive the ACSAward in AnalyticalChemistry, spon-

sored by Fisher Scientific. The award isgiven in recognition of outstanding con-tributions to pure or applied analyticalchemistry. Biemann is known for hiswork in four-sector tandem MS, laserdesorption MS, and the first GC/MSand computer techniques.

EErrnnsstt BBaayyeerr,, professor at the Uni-vesität Tübingen (Germany), will receivethe ACS Award in Chroma to graphy,sponsored by Supelco. The award recog-nizes specific achievements in the field of

chromatography.Bayer is known forhis work in capillaryHPLC, CE, capillaryelectrochromatogra-phy, and couplingthese techniques on-line with MS andNMR.

CCssaabbaa HHoorrvváátthh,,professor at Yale Uni-versity, will receivethe ACS Award inSeparations Scienceand Technology,

sponsored by IBC Advanced Technolo-gies and Millipore. The award recognizesoutstanding accomplishments in funda-mental or applied separation science andtechno logy. Horváth is known for his workin capillary electro chromatography, bio -polymer separations, nonlinear chro ma -

to graphy, and high-speed HPLC.

HHeellmmuutt SScchhwwaarrzz,,professor at the Tech-nische UniversitätBerlin (Germany),will receive the FrankH. Field and Joe L.

Franklin Award for Outstanding Achieve-ment in Mass Spectrometry, sponsoredby Bruker Daltonics. Schwarz is knownfor his work in reaction mechanisms atthe molecular level and in the intrinsic

properties of elusivespecies.

WWiilllliiaamm KKlleemm--ppeerreerr,, professor atHarvard, will receivethe E. Bright WilsonAward in Spectro -scopy, sponsored by

Rohm and Haas. The award recognizesfundamental and applied contributionsin all fields of spectroscopy. Klemperer is known for his work in the geometricand electronic structures of van der Waalsmolecules and weakly bound complexesand modeling the kinetic behavior of the

PEOPLE