A new lab at New York Univ. Medical Center is likely the first of many to offer open-access screening across four enhanced RNA libraries.

Flourescence Microscopy Image

A fluorescence microscopy image of a typical 384-well sample using a Cellomics Arrayscan VTI automated high-throughput microscope.(Image: Chi Yun, RNAi Core Facility).

These days, bioscience resembles a melting glacier. The inexorable force and weight of a wealth of new data coupled with a long-term thaw in restrictions—not to mention a leap in computing power—have conspired to burst the floodgates. Icebergs of knowledge seem to be calving off everywhere in the form of small companies offering services unthinkable just 10 years ago. Take Illumina, a company that entered the spotlight for offering low-cost human genome sequencing. Now they say $1,000 sequences are available and that in less than 10 years most newborns will have their genetic sequences decoded.

But so far these breakthroughs are just window dressing for what is taking place in the background. In order to truly understand the complex mechanics of life and disease, scientists must deal with the scale of DNA, with its billions of base pairs.

One of the best ways to unlock these secrets is to observe and experiment with gene silencing. First observed in plants in 1990, the technique needed another decade of R&D before entering its own after the discovery that double-stranded RNA causes gene silencing. The human genome was sequenced in 2000, and by 2002, major companies such as Merck had set up RNA interference (RNAi) laboratories for basic research. The National Institutes of Health was also setting up dedicated labs, including the Chemical Genomics Center, Rockville, Md., in 2004.

In 2004, RNAi reached academia with the founding of Harvard Univ. Medical Center’s Drosophila RNAi Screening Center (DRSC), which for the first time allowed academics the chance to use gene interference to determine cell functions.

Meanwhile, bioscience picked up even more speed. In the last few years, the value of cross-platform (or cross-species) RNAi screening has emerged, as has the interest in functionalized siRNA libraries of typically useful animal models, such as mice, C. elegans, Drosophila flies, and humans.

At the end of 2008, New York Univ. (NYU) Medical Center responded with the RNAi Core Facility, a new laboratory that offers relatively affordable RNAi screens and access to all four major gene libraries. The brainchild of Ramanuj Dasgupta, assistant professor in the Depts. of Pharmacology and Cancer Institute (Research) who studied gene function at DRSC, is a first for RNAi screening. Open to all academics, the access extends to the screening results themselves, which are eventually available to the public.

PerkinElmer Janus MDT Automated Workstation

The PerkinElmer Janus MDT automated workstation is equipped with an expanded platform to reformat 96- and 384-well plates, as well as a modular dispensing arm. (Image: Chi Yun, RNAi Core Facility).

What is RNAi interference?

Messenger RNA was first discovered in 1960, but another 30 years was needed before actual gene silencing was first observed in plants. In short, RNA interference occurs when small pieces of RNA silence the activity of specific genes.

Genes use specialized enzymes to transcribe (or copy) a strand of a gene’s DNA. The transcription forms messenger RNA (mRNA). The mRNA are moved from the nucleus to another area of the cell where it is translated by ribosomes into a specific protein.

The actual process of gene silencing has multiple steps. First, RNA molecules bind to proteins that slice them into small fragments. The fragments then bind easily to other proteins collectively known as RNA-induced silencing complexes (RISCs). This process removes one of the RNA’s two strands, leaving a chain that is able to connect to naturally occurring microRNA segments. The RISC proteins cut down the mRNA to fragments that cannot be translated.

This means the gene that produced (or copied) the mRNA segment is suppressed, or silenced. Cellular processes ranging from disease protection to growth and death are regulated by RNAi activity.

This natural process is harnessed at the RNAi Core Facility; its convenience has emerged as a favored way to unlock the complex processes of gene networks.

What's in an RNAi laboratory?
When Ramanuj Dasgupta arrived at NYU Medical Center, he was able to draw on his own experiences in RNA inteference screening to build a laboratory that met several goals, including high throughput, high flexibility, reliable quality, and reasonable cost. The RNAi Core Facility was completed at the end of 2008.
Some of the equipment available includes:
• Cellomics Arrayscan VTI automated high-throughput microscope
• Matrix Technologies Wellmate, a liquid dispensing system for buffers and antibodies with stacker chimneys
• PerkinElmer EnVision, a multi-label luminescence/fluorescence plate reader
• PerkinElmer Janus automated workstation with modular dispensing arm
• PerkinElmer Janus Varispan automated workstation for cherrypicking siRNAs
• PlateStak automated microplate handlers for increased storage capacity
• Laminar flow hoods and centrifuges

The RNA libraries sourced from Ambion include:
• siRNA libraries for the human annotated and mouse druggable genomes
• Human pre-miR and anti-miR miRNA libraries
• Whole genome and subset Drosophila dsRNA libraries
• Bacterial feeding libraries for the C. elegans genome

RNA's value emerges
The role of RNA screening is becoming increasingly important, says Dasgupta, because working with siRNA is experimentally easy and reliable.

Director of the RNAi Core Facility, Dasgupta joined NYU primarily to help boost the role of RNAi in the university’s research. The school recognized the value of being able to attract potentially valuable research as well having the ability to perform advanced R&D on its own.

His past training in both mouse and fly RNA screening helped Dasgupta when it came time to build the laboratory.

“The set up is something unique for NYU in the sense that we are the only ones who offer multi-species libraries under one roof,” says Dasgupta, who recognized the value of cross-platform studies after a major validation occurred this way at the DRSC laboratory. Over time, he says, researchers have realized that whole genome libraries are very good for developing unbiased screens.

“To test for cell function you want cross-species validation. The (RNAi Core) Facility allows, for example, researchers to do a screen for Drosophila or mouse or human genome and then look for functions essentially by cherry-picking” promising samples elsewhere, says Dasgupta.

The ability to test between species has become attractive for researchers hoping to strengthen their conclusions through corroboration and has also spurred collaboration between labs.

“Everyone is interested in looking for cell function. You could also perhaps go to a fly lab to collaborate, and they could build genetics and do prior chemistry,” says Dasgupta, then perform final screens at NYU.

Arrayscan VTI

The Arrayscan VTI uses a Catalyst Express robotic arm with 45-plate capacity to feed plates into the microscope. (Image: Chi Yun, RNAi Core Facility).

RNAi Core Facility
Located in NYU’s Cancer Institute’s Tisch Hospital building and open for just a few months, the RNAi Core Facility was built with the support of the Kimmel Stem Cell Center, the NYU Cancer Institute, and scientists at DRSC.

The RNAi Core Facility hardly resembles an RNAi screening lab one might find at Pfizer, GlaxoSmithKline, or Merck. Yes, it features fluid-handling gear, automated high-resolution fluorescence microscopes, extensive RNA libraries, and robotic sample-handling equipment. But its automation does not approach what is available to the big pharmaceutical firms, who may have 11 or 12 of the same high-throughput microscope that is performing the automated 384-wellplate screens at NYU.

For now, such equipment is far out of the budget for this $1.5 million lab. But that doesn’t hurt the mission, which is decidedly egalitarian. According to the lab’s assistant director, Chi Yun, a cellular biologist turned RNA expert, the RNAi Core Facility represents an early attempt to bring RNA screening capacity to other academics, researchers who, due to budget constraints and lack of access, might otherwise never conduct high-throughput screening on attractive genomic targets.

“When (Dasgupta) was at Harvard, he was able to see how things are set up there,” says Yun, which helped them set up an efficient workflow in a limited space. The lab is small—just about 1,200 ft2—and has just two full-time employees including Yun. But it’s the only open-access lab offering four RNA libraries—human, mouse, Drosophila, and C. elegans. As she describes it, the lab is not “fee for service.” Instead, researchers pay a relatively low rate—often less than $10,000—to gain access to the libraries. After training and optimization of the samples, the researchers are able to use the equipment to conduct their screening. In return, the lab retains the result of their findings, and after two years that data reverts to the public domain.

Storage Room

The storage room of NYU Medical Center's Tisch Building is where freezers hold the genomic libraries for the four major species used for RNA research: human, mouse, Drosophila, and C. elegans. (Image: Chi Yun, RNAi Core Facility).

A brief guide to RNAi screening
The screening process is a dynamic one that involves transfer of data between the researcher’s home institution or lab and the RNAi Core Facility.

There are two major types of screens, one with a plate reader and one sourced from high-content imaging.

The plate reader-based assay allows the user to detect fluorescence intensity or polarization and luminescence. The quantitative well information achieved can be analyzed in about a month and requires reagents including luciferase and stains. This screen produces small data files.

High-content imaging is a more sophisticated analysis process that combines high-resolution microscopy with software-enabled image analysis. Both quantitative and qualitative data are generated this way, and while the data files produced by such a method are large—on the order of gigabytes—they are easily analyzed by a variety of applications.

“The actual screen time is only one or two months, but the preparation time is the big time requirement. The samples often need to be optimized,” says Yun. Typically, small laboratory researchers use 96-wellplate formats, which feature larger volumes than the 5 µl used by the Cellomics Arrayscan microscope. As a result, prospective screeners are given a single test plate to ensure optimization of the screen. If results confirm the feasibility of the screen, then the screener may start a pilot screen of six 384-well plates (the first three screening plates in duplicate) from the facility's library.

If the study appears feasible, then the researchers must be trained on the laboratory’s equipment. The process for screening is as automated as is possible with a small budget and a tight workspace, but there are certain pieces of equipment that may be new to guest researchers. Yun and research associate Shauna Katz assist screeners.

Prior to completing an online application at the facility’s website, researchers are asked prepare high-quality optimization data in a 384-well format. This can be a challenge for smaller labs accustomed to 96 wells, but 384 is a necessary for maintaining reasonable throughput.

Users are responsible for the cost of assay consumables and reagents.

The total fee may be from $3,000 plus supplies for a whole Drosophila genome screen for an NYU researcher to more than $10,000 for a whole human genome screen for a researcher outside the NYU system.

Still, it’s nowhere near the $30,000 to $40,000 that many researcher think they'll have to pay, says Dasgupta.

“We maintain an open data policy,” says Yun. Two years after the screen is complete, the data held by the RNAi facility is open to the public. This gives the researcher a chance to make sense of the results and publish while at the same time helping to stimulate further studies.

The first of many to come
On the face of it, an RNAi laboratory is not a good business proposition. Apart from the actual screening, the “business” model has certain problems.

First, says Dasgupta, “it’s not a retail operation. There’s no money in screens. The initial investment can never be recovered.”

This has been a limiting factor for RNA laboratories, but the upsides of discovering cellular pathways and functions outweigh the costs. The hope is that eventually this data will result in real therapies. As for paying for these screens, an oblique approach works best in the academic world.

“If the screens result in one or two grants,” says Dasgupta, “that’s enough to recover the investment and possibly more. The only way it is economically viable is through indirect payment of the costs.”

The other problems are logistical. Optimizing samples for a 384-wellplate, an otherwise efficient and environmentally-friendly format, can be difficult. Libraries take up space, and capital costs for new equipment can be high.

At the RNAi Core Facility, two Drosophila fly screens have been completed and four screens are in the pipeline. Dasgupta has also performed a microRNA screen using the human genome library. And he is collaborating with Mt. Sinai Medical Center to do a mouse screen. But few outside researchers have so far made use of the facility.

“It’s been a challenge to advertise a place like this,” says Dasgupta, who remembers the Boston area as being a place where “everybody and their cousin is doing screens.”

Part of the problem, he says, is that people aren’t used to the availability of the services, especially in the New York City area.

“People are sometimes intimidated and wonder if they can optimize for high-throughput, and that it’s much more expensive than it actually is,” says Dasgupta. “The amount of data you generate for that $10-14,000 is a really good investment for the long term because you get lots of hits and if it’s a good screen you have a lot to work with.”

Published in R & D magazine: Vol. 51, No. 1, February, 2009, p.22-26