WO1992020815A1 - 'stable end-point' microculture tetrazolium assay (se-mta) - Google Patents

Abstract

The subject invention relates to a method for the measurement of mammalian cell growth or inhibition by a newly designed 'stable end-point' microculture tetrazolium assay (SE-MTA) which permits analysis by formazan colorimetry and by image analysis microdensitometry.

Description

"STABLE END-POINT" MICROCULTURE TETRAZOLIUM ASSAY (SE-MTA)

BACKGROUND OF THE INVENTION TECHNICAL FIELD

The subject invention relates to a method for the measurement of mammalian cell growth or inhibition by a newly designed "stable end-point" microculture tetrazolium assay (SE-MTA) which permits analysis by formazan colorimetry and by image analysis microdensitometry.

BACKGROUND INFORMATION

The discovery and development of anticancer drugs has been an important function of many, large research facilities. Since 1955, several hundred thousand chemical compounds have been screened for anti-cancer activity using in vivo transplantable urine leukemia models and human solid tumor xenografts. Such drug screening programs have been effective in identifying and/or developing the majority of agents employed for the clinical management of cancer patients. While a significant number of agents have been identified, therapeutic efficacy has been limited, for the most part, to leukemias, lympho as, and some types of solid tumors. In the late 1970s and early 1980s, the potential utility of "disease-oriented" jn vitro drug screening with humor tumor specimens was evaluated. Although the human tumor colony formation assay identified interesting compounds for development, use of fresh surgical specimens was found inadequate for large-scale drug screening. However, the fact that hundreds of well- characterized human cancer cell lines were available from repository sources and : iical institutions worldwide prompted a study of the feasibility of using panels of human tumor cell lines for "disease- oriented" drug screening. The fact that 1) in vitro colorimetric assays could be effectively utilized to measure growth and drug sensitivity in cell lines derived from a broad cross-section of human solid tumors, 2) individual cell lines exhibit stable and reproducible drug sensitivity profiles over successive passages and time, and 3) analysis of such assays could be performed in an efficient manner with plate colorimeters resulted in the replacement of the in vivo modes of anti-cancer drug screening with in vitro based methods of anti-cancer drug screening. The n vitro screen is capable of evaluating >10,000 compounds per year in each of 60 human tumor cell lines. While initially q Microculture Tetrazolium Assay (MTA) was employed for the large-scale primary screen, this assay was later replaced by an alternate colorimetric procedure, the SRB assay, due to a "time-unstable" end-point in the original MTA procedure. The SRB assay appears to be better suited to the requirements for large-scale screening.

Compounds exhibiting activity in the primary n vitro screen generally require some form of follow-up testing to confirm and extend initial findings. In the context of such "secondary" testing of active compounds, it is reasonable to suggest that the newly-devised "stable end-point" MTA may play a significant role. While the primary screen is designed to identify compounds with tumor- type specificity or other forms of differential activity, secondary testing is conceptually required 1) to confirm the extent of selectivity in a broader subpanel of cell line tumor types, 2) to define "effective" in vitro concentration and "maximum drug effect," and 3) to select cell line(s) appropriate for further detailed pharmacologic evaluations.

In particular, it is important that such secondary assays are capable of detecting the full range of drug effect and discriminating between cytostatic and cytotoxic drug activity. Moreover, it is important that secondary assay methodologies are not susceptible to the artifactual loss of drug-induced cell detachment.

The "stable end-point" MTA described herein addresses each of the elements required for secondary n vitro testing of potential anti-cancer drugs. Furthermore, the "stable end-point" MTA represents a methodology which should be directly applicable in other areas of experimental biology and medicine for which MTT-based assays have been decribed in the scientific literature.

At the present time, plate colorimeters are employed for the automated analysis of growth/growth inhibition of mammalian cells cultivated according to a variety of colorimetric bioassay methodologies (see, e.g.. Alley, et al., Cancer Research 48:589- 601 (1988)) . In addition, image analyεers have been employed to measure a variety of cultured cell parameters, for example, colony formation by human tumor cells in soft agar cultures (Alley, et al., Br. J. Cancer 52:205-14 (1985)) and hepatocyte- mediated activation or inactivation of experimental chemother.apeutic agents (Alley, et al.. Cancer Res. 44:549-56 (1984) and Appel, et al., Cancer Chemother. Pharmaceol. 17:47-52 (1986)).

Previous applications of image analysis systems for in vitro drug sensitivity testing have been limited to measuring size, shape, volume and numbers of colonies which survive various drug exposures (see, e.g.. Alley et al., Br. J. Cancer 52:205-21 (1985)). While one report by Baker et al (Cancer Res. (1986)) describves the use of image analysis microdensitometry to measure growth and growth inhibition of cultured human tumor material, it is based upon the use of a non-vital dye which has extremely high background due to non-specific deposition of stain upon the culture surface. On the other hand, the vital dye staining procedure employed in the "stable-end point" MTA has virtually no stain depostion upon culture surfaces which lack viable cells. The only "background" associated with the "stable end-point" MTA procedure is a "soluble" formazan stain which is effectively neutralized from deposition by prota ine sulfate (P/S) rinses.. Hence, the only "background" detected with densitometric anlalysis of P/S-rinsed cultures are minor blemishes in the plastic of individual wells. The fact that such aberrations usually cover small surface areas and have intrinsically minimal optical density levels results in extremely low integrated optical density (IOD) values (approximately the product of surface area times average optical density of the blemish) .

Image analysis microdensitometry to date has been employed primarily by industry to measure densities of fabrics, fibers and other materials. Applications of such a general methodology in the areas of biology and experimental therapeutics apparently have been limited to three key factors: 1) the lack of specialized software to perform and translate IOD (or other optical density measurements) on cultured cells into relevant growth or growth inhibition indices, 2) documentation of scientifically acceptable methods to apply such procedures, and 3) reasons to employ densitometric analysis in place of the more accessible and widely- used (but far less sensitive) methodologies (i.e., those based upon spectrophotometry or colorimetry. The present invention represents a novel attempt to address each of these factors and to provide a foundation for the further development and refinement of the method, a method which will become extremely applicable to experimental therapeutics.

All U.S. patents and publications referred to herein are hereby incorporated by reference.

SUMMARY OF THE INVENTION The present invention relates to a method of measuring cellular growth or growth inhibition using a "stable end-point" microculture tetrazolium assay (SE-MTA) in combination with either formazan colorimetry or image analysis microdensitometry. The novel method overcomes many of the disadvantages of known methods utilized for the measurement of cell growth especially in fluid culture systems. Such a method is especially useful in evaluating the "effective" drug concentration and "maximum drug effect" of potential anti-cancer compounds. In addition to a fluid-based culture technique, the present invention relates an alternative agar-based, culture format which may be used to assay materials which may contain sediment. The agar-based SE-MTA permits cells to grow attached or non-attached under an agar "plug" but prevents 1) direct contact between cells and sediment as well as 2) loss of cells during mechanical manipulations prior to analysis. All of the newly-designed SE- MTA procedures employ a specialized culture rinsing step which blocks "background" formazan production and permits storage of plates for up to 2 weeks prior to analysis by colorimetry or by microdensitometry. The new methodology appears to be well-suited for the evaluation of any natural product or synthetic agent known or suspected of perturbing cell attachment and/or growth due to physical interactions.

Furthermore, by aiding in the identification of anti-cancer drugs, the present invention will benefit those individuals who are not responding, or who may not respond in the future, to currently available therapeutic drugs.

In particular, the present invention relates to a method of screening a drug for its effect on cell growth comprising the steps of: a) suspending cells in a fluid culture medium; b) adding the resulting product of step a) to a container means; c) adding said drug to said container means; d) rinsing said cells with a solution of protamine sulfate; and e) measuring cell mass in said container means whereby the effect of said drug on growth of said cells is determined.

The effect of the drug on cell growth is determined by use of densitometric means or colorimetric means. The present invention also includes a method of screening a drug for its effect on cell growth comprising the steps of: a) suspending cells in an agar-containing culture medium; b) adding the resulting product of step a) to a container means; c) adding said drug to said container means; d) rinsing said cells with a solution of protamine sulfate; and e) measuring cell mass in said container means whereby the effect of said drug on growth of said cells is determined.

Again, the effect of the drug is determined by use of densitometric means or colorimetric means. The P/S buffer permits cultures to be stored in the refrigerator for at least two weeks prior to analysis. Thus, one may choose and schedules an appropriate type of analysis at convenient intervals. Without P/S buffer rinses, analyses of the MTA would have to occur within 30 minutes of terminating cultures.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 represents the sequence of steps in each type of microculture tetrazolium assay (MTA) procedure. Protocol I shows the standard MTA procedure (i.e. , the original protocol has a time unstable endpoint) . Protocol II shows the procedure for the "stable end-point" MTA performed by formazan colorimetry. Protocol III shows the "stable end- point"- TA performed by image analysis microdensitometry.

Figure 2 represents the stability of formazan colorimetry (A) and densitometry (B) in the "stable end-point" MTA. Replicate plates of a four-day growth assay of HT-29 cells were stored under refrigeration following P/S buffer rinses for 0 (α) , 2 (+) , 7 (Δ) , and 14 (X) days prior to analysis. f Figure 3 represents typical "stable-endpoint" MTA growth evaluation of human tumor cells analyzed by formazan microdensitometry. Depicted are IOD values (+) for replicate wells (n=3/density) inoculated on day 0 with 1.25 - 640,000 cells (serial dilutions of 2) . Depicted also are the assay detection limited (x) and regression line (-) for values falling within the "linear response" range for a given culture duration. Figure 4 represents typical concentration- effect plots for the "stable-endpoint" microculture tetrazolium assay measured by formazan colorimetry (A) and by microdensitometry (B) . Depicted are the mean (Q) + ISD for vehicle control cultures (n = 6, 100%), drug-treated cultures (n = 3/each of 6 concentrations) and the mean (+) for day 1 proliferation control cultures (Time-zero for drug addition) . Depicted also are phar acologic indices (GI-50, mLC, LC-50 and MDE) and the respective detection limits (x) for each analysis methodology. Figure 5 shows the components of the agar- based SE-MTA culture methodology of the present invention.

Figure 6 shows typical examples of sedimentation for crude natural products applied to wells of flat-bottom microculture plates containing fluid medium (row E-H) and soft-agar "plugs" (rows A-D) : Plant extract N3335 (A) , fungal cell extract F90542 (B) , and fungal cell extract F103687 (C) . Figure 7 shows drug sensitivity profiles for selected human tumor cell lines treated with cytochalasin E (NSC 175151) for 7 days in fluid (left) and under-agar (right) culture formats. Assays were analyzed by formazan colorimetry. Data represent the mean +/- 1 standard deviation with n=6 for vehicle control cultures and n=3 for cultures containing each drug concentration. Depicted are the mean (D) ± ISD for vehicle control cultures (n = 6, 100%), drug-treated cultures (n = 3/each of 6 concentrations) and the mean (+) for.day 1 proliferation control cultures (Time-zero for drug addition) . Depicted also are pharmacologic indices (GI-50, mLC, LC-50 and MDE) and the respective detection limits (x) for each analysis methodology. Figure 8 represents a comparison of pharmacologic indices associated with fluid culture and under-agar culture assays of cytochalasin E in selected human tumor cell lines. (Plots based on the assays presented in Figure 7.) Figure 9 shows a comparison of pharmacologic indices associated with fluid culture and under agar-culture assays of cisplatin in selected human tumor cell lines.

Figure 10 represent typical growth profiles for a human tumor cell line (HT-29) cultivated in fluid (top) and in under-agar (bottom) culture formats as measured by densitometric image analysis.

Figure 11 shows a comparison of drug- sensitivity profiles and assay detection limits associated with formazan colorimetry (top) and densitometric image analysis (bottom) . A single, fluid-culture plate and a single, soft-agar culture plate containing A-549 cells were exposed to actinόmycin D continuously for 6 days prior to performing each mode of analysis.

Figure 12 shows a typical concentration- effect plot for the "stable end-point" microculture tetrazolium assay. Depicted are the mean (D) +/- 1 SD for vehicle control cultures (n=6, 100%), drug- treated cultures (n=3/each of 8 concentrations) and the mean (+) values for the day 1 proliferation control cultures (time-zero for drug addition) . Depicted also are the assay detection limit (X) and multiple pharmacologic indices (GI-50, LC-50, and MDE) .

DETAILED DESCRIPTION OF THE INVENTION The method of the present invention utilizes a specific vital dye, MTT, in the microculture tetrazolium assay (MTA) , which has been modified to achieve a "stable end-point". This "stable end- point" MTA can be analyzed either by a conventional formazan colorimetry procedure or by a newer image analysis microdensitometry method which offers several technical advantages. While imaging procedures have been employed in a variety of biotechnological applications, it is the unique combination of multiple individual culture, rinse, storage and analysis steps described herein which may permit one to overcome the limited detection range and several other technical obstacles associated with techniques currently utilized for the in vitro evaluation of potential anti-cancer compounds.

In particular, the method of the present invention represents a new approach for the measurement of mammalian cell growth/growth inhibition which is based upon modification and improvement of the original MTA technique and the development of a specific image analysis microdensitometry procedure for MTT formazan. While colorimetric tetrazolium assays are useful for many bioassay applications, the present process will substantially enhance the ability of laboratory investigators to perform detailed, multi-log analyses of cultured mammalian cells exposed to experimental therapeutic agents. In addition, the method is anticipated to serve as a model for the development of other image analysis microdensitometry methodologies. The method of the present invention is comprised of several steps. These steps are as follows: 1) cell inoculation, 2) drug application, 3) MTT incubation, 4) P/S buffer rinses, and 5) formazan colorimetry or image analysis microdensitometry. The last two steps represent a newly-designed methodology which provide a "stable end-point" and vastly improve detection capability of the original MTA methodology.

In particular, the method of the present invention comprises the following steps: a) The cells are first harvested from standard culture flasks or another container means and conventional sterile cell culture techniques are performed. b) The cells are suspended in culture medium at a cell density (or densities) which gives rises to optimal growth for the particular cell cell line. Cell densities are selected on the basis of growth evalautions as shown, for example, in Figures 3 and 10. c) Aliquots of the resulting product of step (b) are dispensed to, for example, 96-well flat bottom microculture plates using standard precision microliter pipetors. d) Culture plates are equilibrated at 37 °C, for example, in a cell culture incubator overnight to allow the cells to recover from harvest. e) As an alternate step to those listed in steps (b) and (c) which are performed at room temperature (25°C) , cells may be suspended in culture medium containing, for example, about 0.3% agarose and dispensed, in aliquots, to a microculture plate while maintaining all solutions at 37°C in order to maintain the fluid culture matrix. (This method is especially suitable for the testing of. insoluble drugs.) After overnight incubation to allow cells to settle/attach to culture well bottoms, plates are refrigerated at 4°C, for example, for approximately 15 minutes in order to cause soft-agar "plug" formation (solid culture matrix) above the cell layer (shown in Figure 5) . f) A fluid culture rediu (grcvth assay) or a fluid culture medium containing a drug and/or drug vehicle (drug sensitivity assay) is applied to appropriate culture wells. Drug solutions are prepared by standard serial dilutions and dispensed by precision microliter pipetors. g) The plates are incubated under standard cell culture conditions for a total of, for example, approximately 4 and/or 7 days depending upon assay duration(s) selected. h) The individual culture wells are then rinsed twice approximately 24 hours intervals with a specialized rinse solution (for example, 2.5% protamine sulfate in normal saline, m/v) and stored in the refrigerator until the type of analysis (i.e., colorimetry or densitometry) selected. i) The optimal mode of analysis is identified based upon visual comparison of maximum degree of growth inhibition in drug-treated cultures with that in day 1 proliferation control cultures. (In particular, if growth in drug-treated wells is less than the cell mass of the day 1 control, than densitometry may be used. If the cell mass is greater than that in the day 1 control plate, colorimetry may be used. However, since the "end- point" is stable for two weeks, one has the opportunity to make a thoughtful decision and to choose the method that will yield the most accurate information under the circumstances. Of course, both methods may also be used if desired.) j) One then, of course, performs colorimetry or microdensitometry in order to measure growth or growth inhibition in each culture well per plate, k) If colorimetry is selected, culture plates are dried overnight, for example, in a vented hood and DMSO, for example, is added to individual wells to extract MTT formazan for colorimetric analysis. If densitometry is selected, plates are analyzed directly without drying or extraction (as desribed in further detail below) .

1) After the colorimetric or densitometric procedure is carried out, the growth and drug sensitivity data must be analyzed by standard mathematical calculations.

One may then generate a graphic display of concentration-effect plots with reference to proliferation controls and assay detection limits. Graphs also contain demarcations of the pharmacologic indices (GI-50, mLC, LC-50, MDE) , as shown in Example III below, to provide a visual aid for the "interpretation" of drug sensitivity profiles. Potentially useful anti-cancer compounds may then be identified based upon observing an appropriate level of activity (based upon GI-50, mLC) or especially detectable appropriate LC-50 index and high degr^ee of cytocidal activity ("maximum drug effect") . One then compares the activity of a series of potential anti-cancer compound based upon relative values of selected pharmacologic indices (GI-50. mLC, LC-50 and/or MDE) . The "stable end-point" procedure of the present invention permits the investigator (1) to select an appropriate mode of analysis (i.e., colorimetry or densitometry) according to experimental needs as well as (2) to schedule such evaluations at convenient times over the course of days rather to be forced to perform analyses within 30 minutes of culture termination.

A typical example of microdensitometry data for the growth assay of a human tumor cell line (SN -12 Kl renal cell carcinoma) in fluid culture medium is provided in Figure 3. Data for day 1 illustrates that a linear relationship exists between Log (Integrated Optical Density) and Log (Inoculation Density) over a wide range of values (10' - ljD⁵ cells/well) . Data for day 2 and day 5 illustrate the ease with which one can determine the maximum cell inoculation density for experimental drug evaluations such that growth controls fall with the linear response range for a given culture duration. It is important to stress that microdensitometric measurements are performed on the original culture plates with no manipulations except rinsing with P/S buffer. By contrast, colorimetric analysis generally cannot detect less than 500 cells/well, and culture plates containing high cell densities

(>50,000 cells/well) often require dilution of DMSO- solubilized formazan so that absorbance values fall within the detection limit of plate colorimeters.

The importance of a large detection range for experimental drug evaluations is illustrated in Figure 4. In this example, the same 96-well microculture plate was evaluated by formazan icrodensitmetry and by formazan colorimetry. The concentration-effect profile determined by colorimetric analysis is "blunted" by a^" detection limit of circa 1% whereas the concentration-effect profile measured by microdensitometry spans the entire range of viable cell mass present in these cultures with a detection limit of circa 0.02%. While both methods detect similar GI-50, mLC, and

LC-50 values, only microdensitometry (a) was capable of detecting lethalities > 50% and (b) permitted visual confirmation of nearly total cell kill in culture wells containing 25-250 uM compound following quantitative analysis.

The current invention differs from related technologies in several ways.

Previously published methodologies require solubilization of cell-generated formazan products with one of several organic solvents for colorimetric analysis. The current methodology requires no organic solvent or solubilization step.

Previously published methodologies have relatively high background and require analysis within 30 minutes of culture termination. The new assay endpoint is stable for more than 2 weeks and has an extremely low level of background. Such stability allows analysis to be scheduled "at convenient times throughout the work week rather than to be closely linked with the time of culture termination on a given day.

The detection range for previously published colorimetric analysis is on the order of 1-1.5 logs. Furthermore, such analysis often requires subsequent dilution steps such that absorbance associated with high cell density fall within the detection range of plate color eters. Also, lower cell densities (<500 cells/well) cannot be distinguished from background formazan production by culture medium. On the other hand, the new methodology permits all- relevant cell densities (i.e, 10 - 500,000 cells/well) to be detected densitometrically over a 4-5 log range with no physical manipulation of cultured cell material. Culture materials are irreversibly "consumed" during the process of colormetric analyses (i.e., formazan stain is dissolved and cell material is dispersed prior to analysis) . On the other hand, culture materials are conserved in the new microdensitometry procedure such that assays requiring or benefiting from closer inspection can be examined by microscope following quantitative analysis.

While image analysis microdensitometry has been employed to measure growth and drug sensitivity of cells stained with conventional cytochemical stains (e.g., Baker, et al., (1986)), such approaches have limited detection ranges (<2 logs) due to high backgrounds associated with non¬ specific stain deposition on the culture surface. On the contrary, the "stable end-point" MTA effectively eliminates non-specific formazan deposition which results in extremely low backgrounds levels and multi-log detection ranges for formazan microdensitometry. The agar-based culture option of the newly- designed MTA prevents loss of non-attached cells due to mechanical manipulations and prevents direct contact between cells and insoluble compounds (see Figure 5) . In addition, agar-based assays exhibit less inter-well variability than fluid-based assays, especially for suspension cell lines (e.g., leukemia and small cell lung carcinoma) .

The newly-devised MTA densitometry protocol permits automated analysis of 96-well microculture plates and provides data in standard ASCII files which can be analyzed efficiently by conventional computer software.

Confirmation of quantitative data with the aid of a microscope following formazan microdensitometry (not possible following formazan colorimetry) permits interpretations of the "status" of cell masses surviving drug exposure.

Furthermore, as noted above, it is important to the discovery and development of new anti-cancer drugs that in vitro assays are capable of identifying highly effective agents. Currently available colormetric procedures (based on a variety of stains) share two limitations of high assay background levels and blunted analytical detection ranges. Such standard colormetric procedures are not capable of detecting agents which exert highly effective tumor cell kill and often cannot discriminate between those agents which have mild- moderate cytocidal activity and those which have prominent cytocidal activity.

In contrast, the newly devised "stable end- point" MTA provides a means to minimize assay "background" levels, and combined with the new procedure for image analysis microdensitometry, to detect highly effective in vitro anti-cancer drug activity. In addition, since the method of the present invention "conserves" culture material, an investigator is able to verify results following quantitative analysis (i.e., examine cultures for existence of drug resistance cell populations) , a capability not possible with currently available colorimetric procedures which "consume" culture material in the process of analysis.

In view of the above-discussion, it is readily apparent that these are many advantages to the present invention. For example, several of the advantages are as follows:

1. The newly-devised "stable end-point" microculture tetrazolium assay permits the mode of analysis (i.e., colorimetry or densitometry) to be selected according to experimental needs and scheduled up to two weeks following assay termination.

A distinct advantage of the SE-MTA (which differs from other colorimetric bioassays) is that the investigator has considerable time (i.e., about 2 weeks) to choose which mode of analysis is more appropriate for a given assay. For example, one can visually inspect drug-treated cultures and compare them with day 1 PC cultures (drug add time - 0) to determine whether a given compound exhibits cytostatic versus cytocidal activity, and if the latter, whether the activity is so profound (circa 50% or less than the PC day 1 level) as to require densitometric analysis.

A related, unique advantage of this method is based upon the point that culture materials are "conserved" with densitometric analysis. Thus, assays requiring or benefiting from subsequent visual inspection (e.g., for the existense of drug- resistant cells) can be evaluated following densitometric analysis.

2. While colorimetric analysis of the MTA has a 1-1.5 log detection range, densitometric image analysis exhibits a 4-5 log detection range. The latter mode of analysis effectively spans the large cell density ranges encountered in microculture growth and growth inhibition assays.

3. Densitometric image analysis described above is an automated procedure which does not require steps of plate drying or formazan extraction by organic solvents or dilution or cell culture materials.

4. Incorporation of agar into the cell inoculation step (maintained in fluid state until cells settle/attach to plastic surface) subsequently prevents cell loss due to mechanical manipulations.

5. An agar "plug" situated above the cell layer prevents direct physical contact between cells and insoluble compounds. In addition, such a culture format permits visual detection/evalaution of material in wells which contain sediment.

6. "Interpretation" of lethal drug activity from concentration-effect plots requires not "only a determination of time-zero cell mass but also acknowledgement of the assay detection limit. Densitometric image analysis is required to detect highly lethal drug activity in microculture plates. The present invention can be illustrated by the use of the following non-limiting examples:

EXAMPLE I "STABLE END POINT MICROCULTURE TETRAZOLIUM ASSAY The elements and optimal parameters employed to carry out the method of the present invention are as follows:

Culture Medium: RPMI 1640

10% Defined Fetal Bovine Serum

2 mM L-Glutamine

Culture Format: 96-Well Microculture Plate 100-1,000 cells/200 ul

Cell Inoculation: Day 0 Drug Application: Day 1

P/S Buffer Rinse: Day 1, Day 4, Day 7

Plate Colorimetry or Densitometry: Any day with 2 weeks of P/S buffer rinse.

Graphic Display^": Primary Reference

Day 4 or 7 Control Growth

(100%)

Secondary References

Day 1 PC (Drug addition

T-0) Assay Detection Limit

Pharmacologic Gl-50 (Drug cone yielding 50% Indices^': net growth inhibition on Day 4 or Day 7} mLC (minimum Lethal

Concentration = Day 1 PC intercept)

LC-50 (Drug concentration yielding 50% of Day 1 PC)

Maximum Drug Effect (% of Day 1 PC)

An example of a graphic display and pharmacologic indices is illustrated in Figure 12. EXAMPLE II

METHOD OF PERFORMING THE SE-MTA AND ASSESSING CELLULAR GROWTH USING COLORIMETRY AND DENSITOMETRY

The methods of utilizing cryopreserved cell stocks, performing cell culture, and preparing cells as well as compounds for growth and drug sensitivity assays in microculture plates are described in detail in Alley et al.. Cancer Res. 48:589-601

(1988). In brief, cells are harvested from expotential-phase maintenance cultures (T-75 cm⁵ flasks) , counted by trypan blue exclusion, and dispensed within replicates 96-well culture plates in 100 μl volumes using a repeating pipet or multichannel pipet. Following a 24 hour incubation at 37^βC, 5% CO-, 100% relative humidity, 100 μl of culture medium, culture medium containing drug or culture medium containing drug vehicle is dispensed within appropriate wells (vehicle control group, n=6; each drug treatment group, n=3) . Peripheral wells of each plate (lacking cells) are utilized for drug blank (n=2) and medium/tetrazolium reagent blank (n=6) "background" determinations. Culture plates are then incubated for 1 to 7 days prior to the addition of tetrazolium reagent. In the present invention, steps for tetrazolium incubation and assay termination have been modified to match conditions for obtaining a stable formazan end-point for soft agar colony formation assays (Alley et al., Cancer Res. 51:1247- 56 (1991)). MTT working solution is prepared just prior to plate addition by diluting 5 mg MTT/ml stock solution 2:5 (v/v) with pre-warmed culture medium containing 10% fetal bovine serum (v/v) . MTT working solution (50 ul) is added to each culture well (100 ug/250 ul total medium volume) and incubated at 37°C for 4-5 hours. Following incubation, plates are inspected microscopically for the degree of growth/growth inhibition and subsequently subjected to colorimetric or densitometric analysis depending upon experimental objectives.

For determinations of growth or approximations of effective drug concentration range in a limited number of culture plates, the original MTA procedure for colorimetric analysis is performed within 30 minutes of assay termination. For detailed comparisons of drug sensitivity amongst multiple cell lines, time and concentration-effect parameters and/or detection of low survival levels, the new stable end-point procedure is performed as follows:

Following tetrazolium incubation, 150 ul of culture supernatant is removed and replaced with 150 ul P/S buffer (2.5% protamine sulfate in normal saline, m/v) . Plates are then maintained at 4°C for 16-24 hours, following which the P/S rinse procedure is repeated a second time. Culture plates are then retained under refrigeration (wells fluid-filled) until 30 minutes prior to densitometric analysis or until 24 hours prior to colorimetric analysis.

A specialized semi-automated densitometry program was developed to perform integrated OP (IOD) measurements of culture-generated formazan in microculture plates using an Omnicon Feature Analysis System II (FAS-II, Dynatech Labs,

Chantilly. VA) . Following a 30-minute instrument warm-up period, scanner sensitivity in manual mode was adjusted to 0% by blocking scanner illumination and to 100% by full-field illumination (through air) from the stage lamp at the step 1 setting. Area calibration was performed using an Omnicon test plate 3. Gray level/optical density calibration was performed using a calibrated photographic step tablet 3 (Catalog #1523422, Eastman Kodak Co., Rochester, NY) . For each assay the peak OP and average OP in proliferation control and/or vehicle- treated control cultures were verified to fall within the FAS-II detection range (0.022 - 1.56 diffuse units) . Analysis is performed at 25X magnification with standard instrument settings for densitometry. The program measures IOD for 7 fields/individual well in the microculture plate (a total of 12.1 x 10⁶ urn', equivalent to 39% of the surface area per well) . IOD data for each culture well (growth evaluation: n=3/density; drug evaluation: n=6/vehicle control, n=3/treatment group and n=6/culture blank) was collected, analyzed and plotted using an IBM PC AT computer equipped with Teliόs communication software as well as Lotus Symphony and Freelance software (Lotuc Corporation, Cambridge, VA) .

"Stable end-point" MTA colorimetry was performed as follows: P/S-rinsed culture plates selected for analysis are dried overnight in a vented containment hood following removal of all but 100 ul supernatant. Complete formazan extraction was performed over the course of 4 hours following addition and occasional agitation of 200 ul fresh DMSO/dry culture well. Aliquots (150 ul) of DMSO extract were transferred to empty microculture plate wells and analyzed as previously described (Cancer Research 51:1247-56 (1991)). Microculture plates containing formazan reagent were utilized for determinations of extraction efficiency and stability (as described below) . Growth and growth inhibition for each mode of culture analysis were initially determined as mean ± 1 SD for each control or treatment group. Drug sensitivity data was subsequently normalized to percent of control values for purposes of plotting and comparing data from different modes of analysis. In addition, where appropriate graphs contain secondary reference markings for proliferation control levels and assay detection limit as well as pharmacologic indices (GI-50, mLC, and LC-50) as a means to illustrate the practical consequences which detection limits have upon pharmacologic measurements.

EXAMPLE III EVALUATION OF CYSTOSTATIC VS. CYTOCIDAL DRUG ACTIVITY "STABLE END-POINT" MTA COLORIMETRY

Cytochalasin E is an agent which is known to have cytcstatic activity in cultured cells and to caus live adherent cells to detach from plastic surfaces. Drug sensitivity profiles for human tumor cell lines treated with cytochalasin E are illustrated in Figures 7 and 8. Concentration- effect plots and pharmacologic indices for each cell line in standard fluid culture format suggest that the agent is highly cytotoxic: survival ^"levels

(percent of control) which drop below the day 1 cell mass intercept (+—+) . In contrast, plots, and indices for the under-agar culture format indicate that the agent is simply cytostatic and cells survive even the highest drug concentration. For day 1, cell mass intercept and no significant lethality is detected. These results demonstrate that the fluid culture system is susceptible to the artifact of drug-induced cell detachment (and loss of viable cells from culture during mechanical manipulations of culture required for analysis) , whereas the under-agar culture sensitivity profiles which appropriately reflect cytostatic rather than cytocidal drug activity. In fact, in this example, the fluid culture system exhibits -very prominent lethality: indices which suggest 100-fold plus greater activity than that observed under agar.

Another example of disparate drug sensitivity- profiles observed for fluid versus under agar culture formats is shown in Figure 9. While similar pharmacologic indices for cisplantin were observed for three adherent cell lines (A-549, U-251, and SN^*- 12K1) , lethality indices (mLC and LC-50) for the non-adherent cell line (HL-60) exhibited values associated with the two culture formats which differ by factors of 10-100. Thus, the SE-MTA under-agar cultures effectively identifies agents which exhibit "false-pdsitive" drug activity due to the artifactual loss of drug-induced cell detachment from standard fluid cultures.

EXAMPLE IV MICRODENSITOMETRY OF THE "STABLE END-POINT" MTA Although formazan colorimetry is capable of detecting drug-induced cell detachment and minimally lethal drug concentrations (shown in Example III), colorometric methods are not always capable of measuring greater levels of lethality due to the limited detection range of standard analytical instrumentation. Although plate colorimeters have been designed specifically for the rapid analysis or colorimetric assays in 96-well culture plates, the detection range and limits for such procedures prevent them from detecting prominent cytocidal drug activity. In contrast, the newly devised procedure for image analysis microdensitometry has an extremely large detection range and is highly capable of measuring the large range of cell densities encountered in microculture growth assays and drug sensitivity assays. Typical growth profiles obtained by this analytical method are shown in Figure 10. For fluid as well as under- agar culture formats, microdensitometry is capable of detecting cell number and optimal inoculation densities for cell growth over a 4-5 log range.

A comparison of drug sensitivity profiles and assay detection limits associated with formazan colorimetry and densitometry is shown in Figure 11. While colorimetric analysis exhibits concentration- effect plots which are blunted in the drug-induced lethality range (i.e. , survivals less tha 5%) and cannot detect the LC-50 index for actinomycin D, densitometric analysis is capable of detecting the full range of drug induced lethality (i-.e., survivals of less than 0.01%). Thus, only densitometric analysis is clearly capable of detecting the "effective" n vitro concentration and true "maximum drug effect" of highly active drugs. In addition, the fact that the densitometric procedure "conserves" culture material permits the investigator to verify results following quantitative analysis (i.e., examine cultures for the existence of drug-resistant cell populations) , a capability not possible with colorimetric procedures which "consume" culture material in the process of analysis.

Claims

WHAT IS CLAIMED IS:

1) A method of screening a drug for its effect on cell growth comprising the steps of: a) suspending cells in a fluid culture medium; b) adding the resulting product of step a) to a container means; and c) adding said drug to said container means; d) rinsing said cells with a solution of protamine sulfate; and e) measuring cell mass in said container means whereby the effect of said drug on growth of said cells is determined.

2) The method of claim 1 whereby said effect is determined by use of densitometric means.

3) The method of claim 1 whereby said effect is determined by use of a colorimetric means.

4) A method of screening a drug for its effect on cell growth comprising the steps of: a) suspending cells in an agar-containing culture medium; b) adding the resulting product of step a) to a container means; and c) adding said drug to said container means; d) rinsing said cells with a solution of protamine sulfate, and e) measuring cell mass in said container means whereby the effect of said drug on growth of said cells is determined. I

5) The method of claim 4 whereby said effect is determined by use of densitometric means.

6) The method of claim 4 whereby said effect is determined by use of colorimetric means.