September 15, 2005 (Vol. 25, No. 16)
Enhancing Compound Screening and Data Analysis
The Society for Biomolecular Screenings annual conference, held earlier this week in Geneva, focused on advances in high throughput screening technologies with an emphasis on cell-based assays. These are becoming more important in early drug discovery, and many companies presented data on their unique approaches to further enhance compound screening and data analysis.
EuroScreen (www.euroscreen.com), for example, is focused on developing technologies for assay development and screening of G-protein coupled receptors (GPCR), which play a major role in cell signaling. Functional and cellular testing of these receptors is increasing, with an emphasis on large-scale screening. We presented an innovative use of validated procedures, said Paolo Meoni, Ph.D., director of scientific marketing.
In order to increase flexibility in cellular testing, the company has developed a process to freeze and arrest cell growth of various GPCR-expressing cell lines. Although cellular testing is much closer to physiology than biochemical tests, it claims higher variability due to, among other things, the state of the cell and the number of cells.
So what we have been trying to do is disconnect the cell culture from the testing, explained Dr. Meoni. Cells are cultured in large batches, frozen, and irradiated with gamma rays to arrest their growth. This controls variability between different testing sessions, and provides enough cells for small-scale testing without having to transfer the cell line, which is more expensive.
The idea is to have cells from a given batch that you can use at different times. If you do one experiment every week, you can have different aliquots of cells from the same batch, so it controls the variability between experiments, added Dr. Meoni. Data has shown that this process does not impede the pharmacology of the results.
There are three main applications the company is currently focusing on: profiling, lead optimization, and bulk production. An initial panel of approximately 20 receptors, available as frozen cells, was launched recently.
It will be in some ways, a stand-alone product for a specific number of receptors, and it will be a service for specific applications, like large-scale screening. Well also provide a service of upscaling and freezing, and we already have some batches of ready-to-use cells, summarized Dr. Meoni.
Next Generation of Patch Clamping
Molecular Devices (www. moldev.com) set several goals last year to enhance its automated patch clamping equipment: increase throughput and consistency of data and reduce costs to less than one dollar per well.
Over the past three years, there has been an effort to bring patch clamping into the mainstream of the drug discovery arena, said Alan Finkel, Ph.D., CTO and senior vp, engineering. In response, a few years ago, the company introduced its IonWorks HT system, which uses a proprietary substrate called PatchPlate PPC at the bottom of each well. A small hole (12 microns) was drilled on the bottom of the well, allowing one cell to latch onto the hole. However, this only measures one cell per well and only if it binds.
In order to increase throughput, the company made a plate with 64 one-micron holes drilled into each well.
This enables a high success rate of cells filling all 64 holes around 95%, even up to 99%. We have a technique called leak subtraction that eliminates from the measurement the uninteresting current that flows through unoccupied holes, explained Dr. Finkel.
We can now apply 384 different compounds to each plate and get four times as many results, and the cost per data point comes down (less than $1.00 per point).
In addition, consistency is increased because of the inherent averaging processorcalled Population Patch Clamp (PPC) technology. This records averaged ionic currents from the cell population expression of a recombinant voltage-gated ion channel.
Called the IonWorks Quattro System (launched in early 2005), it supports the original PatchPlate and PatchPlate PPC substrates. The company is working on long-term enhancements for Quattro, but said it is currently useful for lead optimization, medicinal chemistry, and safety testing.
Software Automates Drug Activity Profiling
Genedata (www.genedata.com) has developed software to address the challenges of high throughput, high-content screening, which include quality control and data analysis.
Its Screener software (version 3.0) combines information from screening data, pharmacological properties, and chemical structure and then pinpoints false negatives, reduces false positives, and provides guidance for lead optimization.
Our software evolves and develops through our collaborations, explained Stephan Heyse, Ph.D., head of screening informatics. We use the software in-house and in a way that can be integrated with other vendors software.
Three modules: one for quality control, one for dose-response curves, and one for interpretation of high throughput data (in terms of compound selection), work together to cover the process from raw screening data to final results, usually compounds of interest.
In order to illustrate Screeners abilities, the company presented a case study of a cell-based assay, using an HT screening platform, EVOScreen, made by Evotec (www.evotec.com).
The main goal was to apply our analysis to high-content screening data, where instead of reading one value or signal from a well, you get images, and from those images you can extract parameters, like cell morphology, organization of proteins and changes in cell organs, explained Dr. Heyse.
Algorithms can detect whether there are any artifacts present and automatically compensate. You can detect some forms of toxicity early and remove compounds from your potential list.
Dr. Heyse says that what makes this software unique is the balance of automation and interactive analysis, enabling researchers to systematically relate screening data to compound and assay information via integrated databases. He adds that the company is now focusing on high content screening and compound profiling.
Measuring Reactions Key to Drug Metabolism
Enzymes that catalyze group transfer reactions make up a significant portion of the human proteome. BellBrookLabs (www.bellbrooklabs.com) developed a proprietary high throughput platform to measure these enzymes. We chose this technology because we saw it as a universal way of measuring group transfer reactions, which are important in drug metabolism and include primary drug targets like kinases, explained John Majer, Ph.D., CEO.
Transcreener is a homogeneous assay based on fluorescent polarization. A small tracer molecule is labeled with a fluorescent probe. When bound by an antibody that recognizes the tracer, the molecular weight increases and it rotates.
If it slowly rotates, you get a high polarization value, and if it rotates quickly, you get a low polarization value. So, youre just looking at the change in polarization value based on how quickly the fluorescent molecule spins, said Karen Leman-Leyer, Ph.D., project leader.
The problem with some of the group transfer reactions is you have two products. Most of the current assays focus on the detection of the product that is not common among all the different members of the family. An example is kinases.
In order to detect those individual products, (i.e., phosphorylated peptide or protein) you have to develop a new assay for each substrate used by a subclass of kinases. This can be intensive because there are over 500 kinases. Transcreener detects the product common to the entire family of enzymes, using one set of reagents. It eliminates developing a new assay for each subset of kinases, and can get your drug candidate to high throughput screening mode quickly, stated Dr. Leman-Leyer.
The company is currently developing additional assays, a detection module for UDP (drug metabolism) and a kinase assay that detects ADT, and focusing on other transferase reactionsmethyl transferases, acetyl transferases.
The advantage of our assays is that everything is standardized, there is one set of reagents, the same data output and data interpretation, and comparison of the significance of the data, added Dr. Majer.
SNAP-tag technology consists of a protein tag that labels itself with fluorophores or other substituted groups. Originally licensed from the Swiss Federal Institute of Technology in Lausanne, the protein exists in nature to remove chemical modifications of DNA.
Flexible Fluorescent Protein Labeling
Its a rare type of protein (AGT alkyl-guanine-DNA-alkyl-transferase) because it is not an enzyme; its only able to remove one residue, explained Andreas Brecht, Ph.D., CTO of Covalys Biosciences (www.covalys.com). Its selective and will stay covalently bound.
The company modified this protein so it no longer interacts with DNA, but instead interacts with free guanine derivatives. Applications include: labeling fusion proteins inside living cells and extracts, covalent immobility of proteins to surfaces without prior purification, and detection of fusion proteins in SDS-Page gels.
Their presentation focused on the application of SNAP-tag in living cells. The same fusion protein can be labeled with different fluorophores by exposing cells to different fluorescent substrates.
Our tag will survive fixation protocols, and you can label the protein after fixation, which is important, said Dr. Brecht. Use of two photostable fluorphores and a blocking agent allows performance of complex investigations, like pulse-chase experiments. This allows you to see movement or relocation processes inside the cell. We were able to detect two proteins interacting.
Dr. Brecht said the company is working to use SNAP-tag to purify proteins in one step that will be for functional assays. Its also expanding the wavelength of its dyes to longer wavelengths to reduce auto-fluorescence. A future product will enable researchers to go from labeling to immobilization on protein arrays and microtiter plates.
We also see the potential to diversify the tagto have not just one single SNAP-tag, but two, that react with different substrates and will allow labeling of two different proteins in one cell with the same basic technology.
Monitoring Translocation of Target Proteins
BioImages (www.bioimage.dk) founding technology, Redistribution, tracks proteins in living cells and screens for compounds that affect specific cellular signaling pathways. The primary readout is to monitor the translocation of a target protein tagged with GFP (green fluorescent protein).
Most applications are for secondary screening, lead optimization, target validation, and running assays for the same target in different cell linestypically used with smaller numbers of compounds (tens of thousands) versus high throughput assays.
The company presented a study using a Redistribution assay screen of a small molecule library to profile the PI3K pathway, which is key in inflammation and oncology.
An academic collaboration was established with researchers at MITs Whitehead Institute to examine upstream effectors of this pathway. One compound was shown to exhibit selective anti-profilerative effects in human breast and prostate cell lines. We have some interesting results, but dont expect our full data set until next month, said Len Pagliaro, Ph.D., vp, business development.
New assays include: GPCR assays, a nuclear-hormone receptor assay, a panel for cytotoxicity, along with protein-protein interaction assays in the pipeline.
Our vision is to build a panel of assays that is large enough so you can evaluate signaling space different ways. If you want to focus in one pathway or one small domain with high resolution, you can do thator you can look at a number of different pathways at one point and see where the activity is, and then do other Redistribution assays, explained Dr. Pagliaro.
We’re image-based and translocation-based, with the ability for kinetics, where there are visual changes and you can quantify those differences, he summarized.