Microtransponders, the miniature RFID electronic chips, as platforms for cell growth in cytotoxicity assays

Abstract

Background:

An electronic radio frequency (RF) microchip, the microtransponder (MTP), has been developed as a platform for assays in the fields of genomics and proteomics. Upon activation by light, each MTP provides a unique RF identification (ID) signal that matches a chip to the specific biological material attached to it. The MTP is powered by a photocell and has an antenna that transmits the signal. The aim of the present study was to explore utility of MTPs as a platform for cell growth in cytotoxicity assays.

Methods:

The MCF-7, MCF-116, A549, or T-24 cells growing on MTPs placed in petri dishes or slide chambers were cultured untreated or exposed to antitumor drugs topotecan, mitoxantrone, or onconase for up to 4 days. Their attachment to- and growth on- MTPs was assessed by fluorescence microscopy and laser scanning cytometry (LSC) and compared with growth on the dish surface in the MTP neighborhood. The MTPs were fixed in ethanol, stained with propidium iodide (PI), and interrogated in flow in the instrument capable to rapidly (up to 10³ MTPs/s) identify their ID signal and measure fluorescence.

Results:

The cells plated on MTPs exhibited similar attachment properties to those plated in culture dishes. When measured by LSC, they had similar mitotic activity, growth rate, and cell cycle distributions as the cells adhering to the culture dish in the neighborhood of MTPs. The fluorescence intensity of MTPs provided information about the cell number per MTP, which made it possible to assess cell growth rate and monitor the cytostatic/cytotoxic effects of the tested drugs.

Conclusions:

The MTP-based system holds promise for the multiplexed cell assays in which numerous different cell lines can be screened for their growth rate or sensitivity while exposed to particular agents in the same vessel. Other advantages of the system are the rapidity of the screening and a very large number of ID codes. Because many cell lines/types can be assayed in a single dish, the system also offers cost savings on tissue culture reagents. © 2006 International Society for Analytical Cytology

Cell viability assays are commonplace in various disciplines including cell biology, biotechnology, pharmacology, and medicine. Owing to the differences in methodology and the data output, these assays can be subdivided into several types. The oldest type is based on the use of radioisotopes, either on incorporation of radio-labeled thymidine (1) or on release of ⁵¹Cr from ⁵¹Cr pre-labeled target cells (2). The hazard associated with storage and disposal of radioisotopes makes these methods unattractive. Another method consists of rapid assays, often semi-automatic or robotic, utilizing multiwell plates as a platform for cell growth. Most often, the cell growth rate in each well is assayed colorimetrically using one of the tetrazolium salts, such as 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide (MTT), as a marker of metabolic cell activity (3–8). Reduction of these water-soluble salts by metabolically active cells results in precipitation of colored formazans which, after solubilization, are analyzed by plate readers that measure light absorption of the dye. The advantages of these assays are convenience, low cost, and rapidity. However, metabolic activity is not always correlated with cell proliferation, as it is in the case of induction of growth imbalance by many antitumor drugs (9–11). These assays, thus, fail to accurately assess the antiproliferative effects of such drugs.

Several viability assays utilize instruments that rely on measurement of cells in flow such as flow cytometers or Coulter counters (12–17). The advantage of flow cytometry is that it allows one not only to identify and count dead cells, but also to analyze the cell cycle effects of the studied drug as well. While it is relatively simple and convenient to assay by flow cytometry cells that grow in suspension, the cells that grow attached require either prior trypsinization or other means of detachment to make them amenable to flow analysis. The detachment step is cumbersome and may introduce bias. The bias, for example, may be related to the fact that dead (late apoptotic or necrotic) cells become detached and, unless specifically collected and pooled with the trypsinized cells, may escape detection (18). Another bias may result from centrifugation, a step used in this procedure, which may lead to selective loss of apoptotic and necrotic cells due to their increased fragility and different density (18).

Miniature card-like carriers that have unique, identifiable codes, similar to barcodes, have been proposed as platforms for growth of attached cells in cytotoxicity assays (19). Different cell types are first associated with different classes of cards. Next, all of the card classes are mixed and deposited in microtiter plate wells, creating a cellular array in each well. After an assay is performed, images of wells are collected and analyzed. This system offers the possibility of exposure of different cell types (lines) to the tested agents in the same culture vessel, i.e., under identical conditions (19).

We propose an alternative approach, namely to use microtransponders (MTPs) as a platform for cell growth and for screening their viability in cytotoxicity assays. This platform has been developed and already used in the fields of genomics and proteomics (20–23; see later). Similar in concept to flow cytometry, the technology offers the possibility to rapidly screen many (hundreds, even thousands) cell lines or cell types. These can be exposed to the tested agents concurrently, in the same culture, i.e., under the same growth and drug exposure conditions. This type of assay provides a method to measure the differences between the cell lines in cell sensitivity to a particular drug. The unique ID of each MTP precludes the possibility of an identification error. This platform also offers the possibility for multiparametric (multivariate) analysis when different constituents of the cell are stained with dyes emitting fluorescence at different wavelengths. An additional advantage of the assay is that concurrent incubation of a multitude of cell lines/types in a single culture offers substantial savings of tissue culture reagents.

Recent emphasis on simultaneous analyses of many types of molecules led to an emergence of so-called “multiplex” assays. The critical requirement in such assays is to encode the identity of groups of molecules. In the microarray assays, the molecules are distinguished by their x,y position in a grid. Other methods to encode have been also described. They include chemical encoding (with halo-aryl tags, secondary amines, peptides, or oligonucleotides), spectrometric encoding (via mass spectrometry tags, colloids, or combinations of fluorescent dyes), graphical encoding (bar-coded aluminum rods, metallic microparticles), and other means [reviewed in (24)]. While multiplex approaches based on particles (beads) offer some advantage over arrays (e.g., faster kinetics and ease of derivatization), they often face challenges due to a limited number of codes, complexities of manufacture, difficulties or time needed to derive the code, or particle size that might be of limited differential for the applications. Therefore, such methods are not always suitable for high-throughput multiplex assays.

The design of the present PharmaSeq instrumentation system is based on two underlying principles: (a) a desire for a large number of codes accomplished through building of an integrated circuit with electronic memory; (b) a rapid readout in a fluidic environment. A unique platform has been developed applicable for multiplex analyses of nucleic acids, proteins, or cells. The heart of the system is an MTP (integrated circuit with electronic memory) and a bench-top MTP-flow reader (MTP-FR). The MTP (Fig. 1) is composed of photocells, ROM, control electronics, and an antenna (21). Potentially, it can store information in 64-bit memory, which allows for more than 10¹⁷ different IDs. The current chips have 42 programmable bits that yield up to 4 × 10¹² ID codes. The photocells, when illuminated, provide power for electronic circuits, modulating the current through the antenna based on the ROM contents. The antenna transmits the ID through a varying magnetic field; the frequency of transmission is in the low-megahertz range. Batches of MTPs are manufactured in silicon foundries on silicon wafers, using the same standard CMOS processes as those used to make memory chips and microprocessors. Postprocessing of wafers involves backgrinding (thinning) and dicing of the wafers into individual MTPs. Both 500 × 500 μm² and 250 × 250 μm² MTP versions have been produced. In the present study, 500 × 500 μm² MTPs were used. The larger size of MTPs allows one to grow greater number of cells on it, thereby increasing the dynamic range of the assay (the differential from minimal to maximal number of cells per chip.

Figure 1. Schematic illustration of an MTP. The MTP is composed of an antenna, photocells, digital and analog circuitry, and read-only memory (ROM). When illuminated with light, it transmits a unique ID via an RF signal. In this scheme, a single-stranded DNA covalently linked to the MTP surface (probe) binds to complementary DNA present in a sample (target) that has been tagged with a fluorochrome (20, 21). Thus, a unique ID is associated with a specific DNA sequence as shown in this Figure. However, a capture protein or cells attached to MTP can be analyzed in a similar fashion. While 500 × 500 × 120 μm³ and 250 × 250 × 100 μm³ MTP versions have been produced, the former (500 × 500 × 120 μm³) were used in the present study. MTPs are tabloid in shape, with the surfaces forming 90° angles, and have relatively sharp edges. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

The MTP surface is silicon dioxide, which is deposited during the electronic manufacturing process as a final passivation layer. This layer was not modified during the course of the study at all. Thus, the surface of the MTP is glass-like, and procedures for cell growth developed for glass surfaces can be used for MTPs.

Two types of information are read by a MTP-FR: (a) the ID transmitted via radio frequency (RF) is read and decoded; (b) fluorescence intensity integrated over the chip is measured. All data are stored for further analysis. The MTP-FR is described in the Materials and Methods section. MTP technology has already been applied to DNA assays (20, 21, 23).

MATERIALS AND METHODS

Cells

Human breast carcinoma MCF-116 and MCF-7, human pulmonary adenocarcinoma A549, and human transitional cell carcinoma T-24 cells were obtained from the American Type Culture Collection (ATCC; Manassas, VA). The cells were maintained in culture in 25-ml Falcon flasks (Becton Dickinson, Franklin Lakes, NJ) in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS), 100 U/ml of penicillin, 100 μg/ml of streptomycin, and 2 mM L-glutamine (all from Gibco/BRL Life Technologies, Grand Island, NY), at 37.5°C, at 5% CO₂. During the exponential phase of growth, the cells were trypsinized, rinsed, and suspended in full medium. Aliquots of these suspensions were then transferred either to petri dishes or to chamber-slide culture vessels, each containing MTPs (see later).

Preparation of MTPs

Two hundred to 300 MTPs were submerged in 3 ml of 70% ethanol in a sterile petri dish (10 × 35 mm²; Becton Dickinson). Ethanol was discarded after 60 min, and the dishes were allowed to completely dry under sterile conditions overnight.

Seeding Cells on MTPs

In early experiments designed to test whether cell growth can be maintained on MTPs, the cell suspensions (10⁴–5 × 10⁴ cells/ml) in fresh medium containing FCS, antibiotics, and L-glutamine, as described earlier, were added to petri dishes containing ethanol-sterilized MTPs. The petri dishes were then transferred into an incubator, and the cultures were harvested after the desired time periods. For the experiments in which cells growing on MTPs were examined by fluorescence microscopy or by laser scanning cytometry (LSC; Refs.23 and24), the ethanol-sterilized MTPs were transferred by forceps from the petri dishes to two-chambered culture vessels containing microscope slides at their base (Lab-Tek; Nalge Nunc International, Naperville, IL). Cell suspensions were added to each chamber, the chambers were transferred into the incubator, and the cells on MTPs were cultured for various time intervals for up to 4 days.

Inhibition of Cell Growth by Drugs

Sterile glass or plastic 1-cm diameter coverslips were placed on the bottom of the 3-cm diameter wells (MULTIWELL™ 6-well plates; Becton Dickinson, Franklin Lakes, CA). Approximately 100 sterilized (as mentioned earlier) MTPs were then placed on the surface of each coverslip and submersed in 3 ml of the respective culture medium. The cells in the parent cultures were trypsinized, counted, suspended in medium, and seeded in a 2-ml volume of medium into the wells containing MTPs on the coverslips. Knowing the total petri dish surface area and the area of the MTP (500 × 500 μm²; 2.5 × 10⁵ μm²), the cell density was adjusted to have ∼20–40 cells per area of a single MTP, and the cells were transferred to an incubator. Such low initial density ensured that the cultures on MTPs did not become confluent throughout the experiment. After the interval equivalent of the particular cell doubling/cell generation time following cell inoculation on the MTPs, different concentrations of the studied drugs (see figure legends) were added to the culture; the control cultures were treated with an equivalent volume of the vehicle. After incubation with the drug, the medium was withdrawn, the transponders/cultures rinsed twice with drug-free medium, new fresh medium added, and the cells/transponders were incubated for an additional 24, 48, and 72 h, and then harvested. The cultures were terminated by rinsing with phosphate-buffered saline (PBS), and the cells/transponders were fixed in 70% ethanol. All four steps were carried out in the same well (petri dish) by carefully adding/removing media and the fixative, without turning over the MTPs. In the pilot experiments, we had observed that transponders remained attached to the bottom of the well or to the coverslip throughout this procedure. Thus, the cells were always on the transponder's surface facing up, exposed to the medium, and then to fixative.

Cell Harvesting

After 1, 2, 3, and 4 days of culturing, the cultures containing MTPs were rinsed in PBS and fixed in 1% methanol-free formaldehyde in PBS for 15 min and then were transferred into petri dishes containing 70% ethanol, or directly into 70% ethanol (without prefixation in formaldehyde). The cells attached to MTPs were then stained with propidium iodide (PI; 10 μg/ml in PBS; in the absence or presence of 0.1% RNase A; Sigma Chemical, St. Louis, MO) or diamidino-2-phenylindole (DAPI; 1 μg/ml) and sulforhodamine 101 (SRB, 20 μg/ml) in PBS, for 20 min at room temperature. The cells growing on MTPs in chambers on microscope slides were fixed and stained as mentioned earlier and then examined by fluorescence microscopy or LSC (25, 26). The MTPs from the petri dishes were stained with PI and were analyzed on the MTP-FR, using laser excitation at 532-nm wavelength.

MTP-FR

The MTP-FR is shown in Figure 2. The instrument is composed of the bench-top unit, a computer, and fluid containers. The central component of it is a flow-channel into which MTPs are fed by a computer-controlled fluidics system. MTPs immersed in a moving liquid flow past the RF- and fluorescence-detection subsystems. The fluidic subsystem, consisting of valves, reservoirs, and pumps, controls the liquid movement through the reader, including loading and unloading of the MTPs. The major challenges in the development process were to keep the MTPs in suspension (MTPs have a density of 2.2 g/cm³) and to orient the MTP perpendicularly to the laser beams, activating the photocells and exciting the fluorescence. This was accomplished through a software-aided design of the flow cell geometry (FloWorks, SolidWorks Corp., Concord, MA).

Figure 2. PharmaSeq flow reader instrument. Panel A: The MTP-FR includes the bench-top unit, computer, and flasks for liquids used by the flow reader (A). Sample loading device (sipper) is shown in the cavity in the front panel of the instrument. Panel B: See Materials and Methods for details shown in the block diagram. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

MTPs are activated by laser light that is brought to the ID read position using optical fibers. The resulting RF signal encoding the ID is detected by the pickup coil, amplified by a local pre-amp, and digitized.

The chip-fluorescence intensity is measured using a standard configuration of a laser, emission filters and lenses that feed the light to the photomultiplier. In addition, a timing signal is produced when a transponder moves through the fluorescence detection system using a photodiode detecting a transponder blocking the laser beam. The ID is read in the center of the flow channel using two small RF pickup coils. In the present version of the instrumentation (T3), the fluorescence was measured twice for every MTP, once for the top surface of the transponder, and once for the bottom surface.

The reader includes custom electronics and software, and is controlled by a 1 GHz Pentium-based processor with 256 MB of RAM. The operation of the unit is simple: MTPs are loaded by attaching a vial containing the chips to the sipper unit on the front panel of the instrument (Fig. 2, Panel A). Upon instrument activation, the MTPs are fed from the vial into the flow system, where they pass through the flow chamber, read by the system, and unloaded back to the same vial. The user is presented with the results of the assay on the computer screen.

RESULTS

A series of pilot experiments were initially designed to test adherence of the cells to MTP surfaces and to compare cell growth and viability on these surfaces with standard solid-phase support on multiwell culture plates or slide chambers. In these experiments, the preparation conditions of MTPs to be used were varied, including their sterilization, cell seeding onto MTPs, and harvesting of MTPs at different time intervals during culturing. Based on results of these pilot experiments, a protocol (see Materials and Methods) that allows for efficient and reliable growth of mammalian cells on MTPs was then selected.

Figure 3 shows MCF-7 cells growing for 3 days on a MTP (Panel B) as compared with cells that grow attached to surface of the slide chamber in neighborhood of the MTP (Panel A). It is quite evident that cell attachment and their morphology were similar on both surfaces. The presence of numerous mitotic figures, the marker of cell proliferation, also was apparent among the cells growing on either of these surfaces. Interestingly, Figure 4 illustrates the cell cycle distribution of MCF-7 cells growing for 48 h on the surface of microscope slide culture chambers (Panels A and B), or on the surface of the MTPs (Panels C and D), stained with DAPI, whose fluorescence was measured by LSC. Because a larger surface area of the slide chamber was scanned by LSC compared with the surface of a single MTP (2.5 × 10⁵ μm²), much smaller cell populations were measured on the MTPs. Despite the difference in cell number, the pattern of the integrated vs. maximal pixel DAPI fluorescence was similar for both cell populations, the cells growing on the slide (A) and on the MTP (C). DAPI fluorescence allows one to identify clusters of G₁ vs. S vs. G₂M cells, based on differences in integrated DNA content, and of mitotic (M) and early postmitotic cells (pM) based on their increased degree of chromatin condensation, as shown in this figure (see Refs.26 and27). Also, the cell cycle distribution revealed by deconvolution of the DNA content frequency histograms was similar in both cell populations (B, D). Several MTPs (≫30) were analyzed in such a way, after 24, 48, 72, and 96 h of culturing, and the results, in terms of a similarity in the cell cycle distributions of cell populations growing on MTPs vs. slide chamber surfaces, were essentially the same as shown in this figure.

Figure 3. UV light photomicrograph of MCF-7 cells growing on MTPs. Human breast carcinoma MCF-7 cells were grown on MTPs (B) or in their neighborhood on surface of slide chambers (A) for 4 days. The cells were stained with DAPI and SRB 101 and examined by fluorescence microscopy (Nikon Microphot FXA, 40× objective) using incident UV light illumination. Note well-attached and spread cells (including mitotic cells marked by arrows), on the surface of the MTP; bright linear structures in the background are elements of the photocells on the MTP. Because MTPs are not transparent, the photograph was made using incident UV light illumination of the cells stained with DAPI and SRB 101 fluorochromes to concurrently visualize cellular DNA and protein, respectively. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley. com.]

Figure 4. Cell cycle analysis of MCF-7 cells growing on surface of microscope slide chambers (Panels A and B) and on MTPs (Panels C and D) by LSC. MCF-7 cells were seeded on a microscope slide culture chamber containing MTPs on its bottom, and were maintained in the culture for 48 h. The cells were fixed with 70% ethanol and stained with DAPI. Blue fluorescence of the cells located on the bottom surface of the microscope slide chamber (A, B) and on a single MTP (C, D) was measured by LSC. The bivariate distributions (scatterplots) (A, C) represent the integrated DAPI fluorescence (reflecting cellular DNA content) vs. maximal pixel of DAPI fluorescence (reflecting degree of nuclear chromatin condensation); the cellular DNA content frequency histograms (B, D) show the cell cycle distributions.

Results of fluorescence measurement by microscope-based microfluorimetry of individual MTPs on which T-24 cells were cultured for 24, 48, 72, and 96 h and then stained with PI are shown in Figure 5. The observed exponential increase in fluorescence intensity integrated over individual MTPs during 24–96 h of incubation is consistent with their exponential growth phase during the culturing and the expected doubling time ∼12 h (28). It thus represents the growth curve of these cells.

Figure 5. Fluorescence intensity of T-24 cell populations growing on individual MTPs for up to 94 h. The cells were seeded on MTPs, cultured for 24, 48, 72, and 96 h in petri dishes, and harvested. After fixation, the MTPs were stained with PI. The integrated fluorescence intensity of individual MTPs was then measured by a microfluorometer attached to the microscope.

With the present configuration of the MTP-FR, fluorescence was being measured on only one side of the MTPs as they pass through the interrogation channel. Because of the symmetry of the MTPs and their random orientation, there was an equal probability that this was either (a) the “top culture surface” (i.e. with adherent cells), or (b) the bottom surface, essentially devoid of cells. In most cases, identification of the top surface was straightforward because it had strong fluorescence signal compared with rather minimal fluorescence of the bottom surface. The fluorescence intensity threshold, thus, was established to identify MTPs that were measured at their top surface. However, to assure that nearly all MTPs in the analyzed sample (containing numerous MTPs) had their top surface measured, the sample had to be run several times through the interrogation channel: at both ends of the channel they become suspended in small containers from which they re-entered the interrogation channel at random spatial (“top” vs. “bottom”) orientation. This prolonged the time of analysis of larger samples. It should be noted that a version of the MTP-FR is being designed to concurrently interrogate measure on both sides of MTPs, in a single run.

MTPs' speed in the flow channel varied from 300 to 1,500 mm/s, the faster speed being obtained in the newer version of the instrument (T4). Individual MTPs were often measured/read in 50 back and forth passes. The typical CVs ranged from 15 to 25% for the same MTP, although in the most recent fluorescence measurement, the CVs for MTPs exhibiting strong fluorescence were below 10% in many instances.

Figure 6 presents results of actual fluorescence and ID measurement of individual MTPs automatically, by MTP-FR. The MTPs with the attached T-24 cells from 24- and 48-h cultures were stained with PI and fed into the instrument. Based on their RFID, they were identified by the instrument as from 24- or 48-h cultures. The over threefold higher fluorescence of the MTPs from 48-h cultures measured by MTP-FR is consistent with the growth rate of these cells (28) and with their growth curve while cultured on MTPs (Fig. 5). Large SD values, however, indicate significant variability in fluorescence of individual MTPs.

Figure 6. Fluorescence intensity of T24 cells growing on MTPs for 24 or 48 h measured by MTP-FR. The cells were seeded on MTPs and cultured for 24 and 48 h; the RF identity of each of the MTPs, reporting duration of culturing (24 or 48 h), was known. After fixation, the MTPs were stained with PI in bulk (i.e., the 24- and 48-h groups were mixed together). The integrated fluorescence intensity of individual MTPs was then measured by MTP-FR, concurrently with their ID. High SD, as marked in each bar, of the measured MTPs (n = 10) reveals significant variability of fluorescence intensity between individual MTPs.

To reveal whether cell growth on MTPs was suppressed by antitumor drugs and whether the suppression could be monitored by MTP-FR, the T-24 cells were seeded on MTPs and left in cultures either untreated (control) or treated with analog of camptothecin topotecan, the DNA topoisomerase 1 inhibitor (29), or with the cytotoxic ribonuclease onconase (30, 31). A decrease in fluorescence intensity of MTPs from the cultures exposed to topotecan and onconase measured by fluorescence microscopy was 27 and 24%, respectively (Fig. 7A). Here again, a high variability of the fluorescence measurements was observed between individual MTPs.

Figure 7. A. Effect of topotecan and onconase on growth of T-24 cells on MTPs as measured by MTP-FR. T-24 cells were seeded on MTPs at time 0. Twenty-four hours later, the cultures were treated with 50 nM topotecan (Tpt) or 20 μg/ml of onconase (Onc). The control culture (Ctrl) was treated with equivalent volume of DMSO that served to dissolve Tpt. The cultures were harvested after the next 24 h, the MTPs were fixed, stained with PI, and integrated fluorescence intensity of individual MTPs was then measured by MTP-FR. Note the decrease in fluorescence intensity of the cells treated with topotecan or onconase. B. Effect of mitoxantrone on growth of T24 cells on MTPs. T-24 cells were seeded on MTPs at time 0. After 24 h of culturing, the cultures were treated with 20 or 100 nM of mitoxantrone (Mtx 20 or Mtx 100). The control (Ctrl) culture was treated with an equivalent volume of DMSO. The cultures were harvested after the next 24 h, the MTPs were fixed, stained with PI, and the fluorescence intensity of individual MTPs was measured by MTP-FR as described (A). C. Effects of onconase on growth of A549 cells. The cells were seeded in two sets of cultures in petri dishes containing MTPs on their bottom. After 1 h, onconase at 20 μg/ml concentration was added to the culture (dark-gray bars); the control culture was treated with an equivalent volume of PBS that was used as onconase solvent (light-gray bars). The MTPs collected from the cultures 48, 72, and 96 h after the addition of onconase were fixed, stained with PI, and their ID and fluorescence measured in MTP-FR. D. Effects of onconase on growth of MCF-116 cells maintained on MTPs. The assay procedure—as for Panel C.

In another experiment, T-24 cells growing on MTPs were treated with 20 or 100 nM of DNA topoisomerase II inhibitor mitoxantrone (32) for 24 h. Significant (3–4-fold) decrease in fluorescence intensity of individual MTPs, measured by fluorescence microscopy is apparent for cells growing in the presence of mitoxantrone compared to MTPs located in the control culture (Fig. 7B).

The effect of onconase (20 μg/ml) on growth of human pulmonary adenocarcinoma A549 (Fig. 7C) and mammary carcinoma cells (Fig. 7D) maintained on MTPs was also investigated (Fig. 7D). It is quite evident that fluorescence intensity of MTPs from the onconase-treated cultures was markedly decreased compared to its control.

DISCUSSION

The aims of the present study were to explore whether (a) MTPs can provide a platform that may be used for cell growth in cultures, (b) the cells maintained on MTPs can be fixed, stored, and stained with fluorochromes, (c) the intensity of fluorescence of cells attached to MTPs can be measured in flow by MTP-FR, and (d) the rate of cell growth on MTPs and its perturbation by antitumor drugs can be monitored. All these aims were accomplished. The present data demonstrate that cells of several human tumor lines attached to MTPs in cultures and their growth rates were similar to neighboring cells grown on slide chambers or on the plastic surface of the petri dishes. The cells, thus, were able to progress through the cell cycle, undergo mitosis and cytokinesis, and reattach to the MTPs surface after division. Since similar cell densities were seen on MTPs as in their neighborhood on standard culture vessels, one may conclude that cell viability was unaffected by their growth on MTPs.

Following culturing for up to 96 h, the MTPs could be collected from cultures and fixed, with no detectable cell loss. The data also show that the MTPs with the fixed cells on them could be stored, transported, and subsequently stained with DNA fluorochromes such as DAPI or PI. Fluorescence intensity integrated over individual MTPs bearing the attached cells could be measured with their IDs by our MTP-RF.

We have also been able to monitor growth rate of the cells on MTPs by measuring intensity of fluorescence of MTPs maintained in cultures for up to 96 h (Fig. 6). Perturbation of cell growth induced by the antitumor drugs, reflected by the decrease in intensity of fluorescence of individual MTPs maintained in the drug-treated cultures, was detected as well. This would suggest that the system has a potential to be used in assays monitoring cytostatic/cytotoxic drug effects.

The major advantages of the MTP system are the rapidity of analysis of individual MTPs and the practically unlimited number of ID codes. The designs of the fluidics readout system of the MTP-FR and the flow cytometer are analogous, as both characterize objects in flow to achieve high throughput and fluorescence readout. While in flow cytometry the rate of transfer of cells through the flow chamber can be as high as 10⁴ to 10⁶ cells/s, the present MTP-FR supports a transfer rate of up to 10³ MTPs/s. The time needed to read the MTP ID is 300 μs, and the time to read fluorescence about 1–2 ms. The actual sustained readout times for both fluorescence and the ID in the assay are longer at present to allow for more accurate measurements. While individual cells are interrogated by flow cytometry, the integrated values of fluorescence from whole groups of cells growing on individual MTPs are recorded by MTP-FR. This system, thus, allows one to monitor expansion of cells on MTPs and the cytostatic/cytotoxic effects at much faster rates than, e.g., in the assays based on analysis of cell growth in multiwell plates by dye absorption (MTT) or fluorescence measurement. A large number of ID codes makes it possible to assess the response of literally hundreds of individual cell types (lines) exposed to the drug in the same culture by placing MTPs, each bearing a different cell type (line), in a single culture vessel, for the cytotoxicity assay.

Because relatively small volumes of culture media (per MTP) are needed in such an assay and individual MTPs are relatively inexpensive (a few US cents per MTP), the cost of the assay is minimized.

The MTP-FR system offers a possibility to measure the same MTPs repeatedly even after storage and additional treatment. For example, we have measured fluorescence of T-24 cells attached to MTPs stained with PI in the absence of RNase, recorded the data, then incubated the same MTPs with RNase, and measured them again (data not shown). This allowed us to estimate the relative contribution of RNA and DNA content to the total PI fluorescence, as it was previously described in the case of cells analyzed by LSC (33). Since a change in RNA to DNA ratio is a marker of growth imbalance, known to be induced by many antitumor drugs and predictive of their cytotoxic effectiveness (34), the MTP-FR can be used to assess such drug effects. The generation of MTP-FR currently under development has two-laser excitation that expands its analytical capabilities to multiparametric analysis using more than a single fluorochrome.

While the present study demonstrated in principle the feasibility of using the MTP-based system in cytotoxicity assays, a significant technical problem has to be resolved to have it implemented for practical application. The major limitation of the system at present is the need to carry out some steps manually. Following fixation and staining while still in a petri dish or multiwell plate, in preparation for the readout on the MTP-FR, the MTPs were laboriously collected and transferred manually by forceps. Practical application of MTPs for high-throughput multiplex screening of cytotoxic/cytostatic drug effects would require an automatic or semiautomatic method to transport the MTPs, with the cells attached sequentially from the tissue culture environment to the fixative—to the cell staining station—and then to the MTP-FR. This can be achieved by developing a “conveyor-belt” like transport system that, at the appropriate time, would collect individual MTPs from their culture environment, carry them to the rinse/fixative station, move them to the automatic staining chamber and, finally, deliver them to the MTP-FR.