In vitro mammalian cell-based assays constitute an important part of any biomo-lecular screening effort. The main advantage in using whole-cell-based assays is that a cell model may be predictive of the desired physiological response. Data obtained from a cell-based assay not only relates to the target of interest but can on occasion predict such parameters as compound membrane permeability, general cell toxicity, and cell type specificity and selectivity. The disadvantage of using a cell-based assay is that it is hard to determine the exact mechanism of compound activity without a battery of parallel assays. For example, if the primary screen detects the expression of the gene of interest during the course of the assay, a parallel cell-based assay measuring nonspecific inhibition of transcription/translation machinery is required to eliminate nonspecific com pound effects. The major benefit of miniaturizing cell-based assays is the decrease in the number of cells required to complete the screen. However, multiple criteria have to be met in order to accomplish cell-based assay miniaturization.
Miniaturization of the mammalian cell-based assay poses a number of unique challenges as compared to the other assay types utilized in HTS. Frequently, engineered immortalized cell lines are used to monitor the activity of the gene of interest in the cell-based screens. The most common cell-based formats currently used in HTS are the receptor binding assays and the reporter gene assays. Homogenous assays, where all the components are mixed and the signal is read directly without the necessity of washing cells, are preferable for miniaturization, because technology is only now becoming available that is capable of washing 1536 plates. Additional requirements have to be considered to miniaturize cell-based assays successfully, including signal strength and detection, efficient cell dispensing, evaporation control, and the concentration of the organic solvents in the assay.
Signal detection is limited by the available optical reader systems and therefore has to be either fluorescent or luminescent based. When the detector is CCD-camera-based, the signals obtained from 96-well and 1536-well formats are comparable, because signal per unit pixel is measured. However, when a photomultiplier-based detection system is used, the transition from 96-well to higher density plates displays a concurrent decrease in signal. This is partly because only a small portion of the photomultiplier tube is used for the measurement, and also because the system measures total signal and not a signal per unit area. However, photomultipliers tend to be one or two orders of magnitude more sensitive than CCD cameras and consequently can usually detect much lower signals. Therefore cell lines that express high levels of the reporter gene, or where the signal can be amplified, are usually more suitable for miniaturization.
Different reporter gene models have been characterized for use in HTS [53-58]. The firefly luciferase reporter gene is an enzyme of choice due to the high detection sensitivity (Xem = 510 nm) and the amplification provided by the luciferase enzyme. The luminescence readout signal is generated by the light production resulting from the ATP and Mg2+-dependent cleavage of the luciferin substrate. In addition, different systems have been recently described that lead to significant signal amplification and production of steady light ''glow'' rather than the light ''flash'' previously used . Other luciferase reporter genes can also be utilized, including the enzyme cloned from Renilla reniformis, which is ATP and Mg2+-independent for dual glow signal in the same cell (Packard Instruments Co., Meriden, CT). The use of the later luciferase genes is limited by the short half-life. The disadvantage of both methods is that both require cell lysis and therefore are not amendable for kinetics studies.
The P-lactamase reporter gene is also potentially suitable for miniaturization. Several commercial P-lactamase substrates, that are cleaved to produce ei ther fluorescent or chemiluminescent products by the enzyme, generate a reasonable signal and should be adaptable to a miniaturized format (Clontech Laboratories, Inc., Palo Alto, CA; Molecular Probes, Eugene, OR; Tropix-PE Biosystems, Bedford, MA). However, these methods also require cell lysis and are thus limited in utility. Recently a P-lactamase fluorogenic substrate has been described for use in live cell kinetic studies, applying FRET technology . This substrate has been shown to be membrane permeable and displays low cellular toxicity. This substrate should overcome many of the disadvantages of the previously described substrates.
Green fluorescent protein (GFP) and GFP derivatives are alternatives for the noncatalytic fluorescent reporter [61,62]. This system allows noninvasive tagging and monitoring of cellular proteins. The use of a GFP reporter in miniaturization requires a very high level of GFP expression because of the low quantum yield of GFP. This requirement limits the choice of the gene promoter to strong viral promoters such as SV40 or CMV and is usually less suitable for ''native'' mammalian promoters.
The secreted form of human placental alkaline phosphatase (SEAP) has previously been shown to be an efficacious reporter gene [63,64] and should be amenable to assay miniaturization. Since SEAP is secreted by the cell into the culture supernatant, there is no need to prepare cell lysates in order to monitor gene expression. The same cell preparations can be sampled multiple times and used for kinetic studies. The concern in using SEAP in miniaturized format is that it may be difficult to eliminate background due to endogenous alkaline phos-phatase (AP), which is usually abrogated in standard methods by heating the test samples to 65°C (SEAP unlike other APs is heat-stable). However, chemical inhibitors of AP, which are utilized to decrease background, have been recently introduced (Tropix-PE Biosystems, Bedford, MA). Both chemiluminescent and fluorescent substrates are available to monitor SEAP enzymatic activity (Clontech Laboratories, Inc., Palo Alto, CA and Tropix-PE Biosystems, Bedford, MA).
The time that it takes to complete a cell-based assay is also a major consideration when contemplating miniaturization. A well volume must be used that will allow for sufficient cell density to produce the total signal needed for detection, but not so small that cell death is induced due to overcrowding. With miniaturization, the surface-area-to-volume ratio increases approximately 6.5-fold (Table 4). This provides both a benefit and a limitation to the cell-based assays. The advantage of the increased surface-area-to-volume ratio is that it allows better gas exchange between the cell culture media and the surrounding environment. However, this increased surface-area-to-volume ratio also increases the rate of evaporation, and therefore special precautions must be taken to ensure that the volume change during the assay does not affect the results.
The organic solvents that are often used to maintain compound libraries, both discrete medicinal chemistry and combinatorial in origin, cause unique prob lems in a cell-based assay. Most mammalian cells show high sensitivity to the solvents, such as DMSO, commonly used for compound storage and can only be maintained in less than 1% solvent concentration. Thus test compounds must be delivered in a way that minimizes solvent concentration in the final assay, requiring either very precise delivery of nanoquantities of tested compounds or drying of compounds prior to testing.
One of the primary concerns in miniaturization of cell-based assays is the ability to dispense the cells at a high density in a relatively small volume (1-2 |L) while maintaining cell integrity. Piezoelectric or ink-jet liquid dispensers may be suitable for certain cell-based assays, where the pathway under study is not associated with stress response(s). When the activity of the gene of interest is implicated in a stress response, the use of more conventional mechanical liquid handlers, based on either positive or negative displacement, is recommended in order to avoid stressing the cells.
Taking all the above concerns and considerations into account, a cell-based assay has been adapted to the 1536-well plate format using a total volume of 3 |L per well . This system utilizes a human T cell line transfected with the luciferase reporter gene under the control of the promoter of the gene of interest.
Figure 9 Miniaturization of cell-based assay. Representation of screening data of random compounds from eight 96-well compound plates in duplicate consolidated into one 1536-well plate. Inhibitors of the expression of the gene of interest were identified using a T cell line transfected with the luciferase reporter gene under the control of the promoter of the gene of interest.
Figure 9 represents the screening plate, where eight 96-well compound plates in duplicates were consolidated into one 1536-well plate. The data indicates that identification of inhibitors of the gene of interest can be successfully performed in a miniaturized format.
Was this article helpful?