O C T O B E R 1 , 2 0 0 5 / A N A LY T I C A L C H E M I S T R Y 3 7 9 A

Biological processes, such as develop-ment, are built on complex networks

of chemical reactions. Rustem Ismagilovat the University of Chicago explains hisfascination for studying these networks.“I’m very interested in chemical net-works, especially those that have spatialand temporal components,” hesays. “Certainly, developmentis a fantastic example of such aprocess because dozens, evenhundreds, of different process-es take place exactly at theright place at the right time.”

Ismagilov teamed up withElena Lucchetta, a graduate stu-dent in his lab; Nipam Patel, aprofessor of integrative biologyat the University of California,Berkeley; and other colleagues toinvestigate the mechanism thatcompensates for variation duringthe embryonic development ofDrosophila melanogaster. The investigatorsexposed the fruit fly embryos to an unnat-ural environment in a microfluidic deviceand discovered that the embryos couldcompensate for environmental perturba-tions and still develop normally (Nature2005, 434, 1134–1138). According toShuichi Takayama at the University ofMichigan, the investigators did “a reallygood job of using microfluidics to ask asignificant biological question.”

In their experimental approach, Is-magilov and colleagues created a PDMSmicrofluidic device that could be con-structed around a live embryo in 1 min.The device was built as interlocking topand bottom halves. A piece of double-sided tape was stuck on the bottom halfof the device, and the embryo was placedon the tape. The second half snapped ontop of the first, like two Lego blocks, tocomplete the device.

While in the device, the embryos wereexposed to temperature differences thatthey wouldn’t normally encounter in thewild. The investigators created a temper-ature step across an embryo by introduc-ing two streams of solutions into the de-vice by laminar flow. One stream was

heated and the other was cooled so thatthe half of the embryo along the anteri-or–posterior axis was exposed to awarmer temperature than the other half.

Creating a temperature step bylaminar flow was easier said than done.“Fluids are quite thermally conductive,”explains Ismagilov. “You have to fightthermal diffusion, which is rapid. For ex-ample, if you were using proteins to cre-ate a step in concentration, that [wouldbe] a lot easier because diffusion is muchslower.” Ismagilov also points out that,in contrast to chemicals, heat can be lostto the walls of the device or tubing.

Embryos were exposed to the tem-perature step (17 °C/27 °C or 20 °C/27 °C) for 150 min and then returnedto room temperature. The densities ofnuclei in the two halves of the embryowere monitored to track the rates of de-velopment. The investigators originally

assumed that the temperature perturba-tion would disrupt the embryos’ growthinto larvae. Although the halves thatwere exposed to warmer temperaturesdeveloped faster than their cooler coun-terparts, the embryos ultimately grewinto normal larvae.

Ismagilov and colleaguesconcluded that compensatorynetworks in the embryos coun-teracted the effects of the envi-ronmental perturbations. Byapplying time-specific reversalsof the temperature step, theyfound that the compensatorynetworks were activated within65–100 min after the start ofembryonic development. Thistime period could provide aclue about the nature of thecompensatory networks.

These days, the researchersare analyzing the compensa-

tory networks in greater detail by com-bining their microfluidic approach with agenetic approach in which a particulargene is removed to form mutant embry-os. “If you combine the two [methods],it might be really powerful,” says Luc-chetta. “You [could] see individual mo-lecular interactions but also [observe]how they function in the network inspace and time.”

Experts say that the method used byIsmagilov and colleagues to tackle theanalysis of chemical reaction networks isparticularly noteworthy. “The work repre-sented in this paper is one of the mostimpressive examples of chemists, engi-neers, and biologists getting together touse cutting-edge tools from each of theirsystems,” says David Bilder at the Univer-sity of California, Berkeley. “This [work]is a real harbinger of a future trend.” aa

—Rajendrani Mukhopadhyay

© 2 0 0 5 A M E R I C A N C H E M I C A L S O C I E T Y

(a) (b)

Cool Warm

Embryo

(a) A schematic drawing of a fruit fly embryo developing inside aPDMS microfluidic device. (b) Thermochromic liquid crystals showa temperature step inside a microfluidic device. The green streamis at 21 °C, and the red stream is at 24 °C. (Adapted with permis-sion. Copyright 2005 Nature Publishing Group.)

b i o s p h e re

Analyzing networks in embryos