WO2002042749A2

WO2002042749A2 - Method and apparatus for measuring apoptosis and growth kinetics

Info

Publication number: WO2002042749A2
Application number: PCT/US2001/046195
Authority: WO
Inventors: Vladimir D. Kravtsov
Original assignee: Vanderbilt University
Priority date: 2000-11-21
Filing date: 2001-11-08
Publication date: 2002-05-30
Also published as: AU2002225874A1; WO2002042749A9; WO2002042749A3

Abstract

A computer-based method of performing optical density measurements on a plurality of cell population samples.In the preferred embodiment of the invention, a calibration factor is first established to eliminate the effect of the media used in the wells. Optical density measurements are then made every five minutes on the cell population samples in each of the wells. For measurement of apoptosis, every five minutes it is also necessary to compute for each cell population the slope of the optical density measurement versus time and correct for any effects attributable to the cell populations in the control sample. The values of interest include maximum corrected slope value, the time from initiation of the assay to observance of the maximum optical density value, the time from initiation of the assay to the beginning of apoptosis and the time from the beginning of apoptosis to observance of maximum optical density value.

Description

Method and Apparatus for Measuring Apoptosis and Growth Kinetics

BACKGROUND OF THE INVENTION

This relates to optical measurements and, more particularly, to computer-based methods of performing a large number of such measurements in applications such as the measurements of apoptosis and growth kinetics in cell populations. It is particularly useful in spectrophotometric measurements of optical density in microtitration plates. The pharmaceutical industry has a constant need to test the efficacy of new drug candidates. Often times such testing is done by observing the effect of the new drug candidate on a large number of different cell populations. One observable effect is a change in optical density in the cell population which can be observed by illuminating the cell population with a suitable beam of electromagnetic radiation. Spectrophotometers are conventionally used for this task.

Because of the number of new drug candidates and the number of variables in the populations against which they are tested, it is desirable to test more than one candidate and one population at a time. A convenient way to do this is to establish different test populations in solutions in the wells of a microtitration plate. For example, one standard microtitration plate has 96 wells arranged in an 8 xl2 array. Plates with more wells, e.g., a 384 well plate, are also available. Typically, a small number of wells are used for calibration purposes. At least one well should be a control that contains a cell population sample in appropriate media but nothing of the variables being tested. Another well or wells might contain the media but no cell population at all. These wells are referred to as blanks. Other wells might be empty. The remaining wells contain cell population samples with different combinations of the variables being tested and/or different concentrations. Preferably, the volume of solution in each well is the same. Alternatively, if different volumes are used, the volumes are known and corrections can be made to account for the effects of such different volumes. The use of such plates in optical density measurements, however, results in staggering amounts of data. Optical density measurements typically are performed at five minute intervals for periods of 24 hours or even several days. Thus, every 24 hours will yield 288 measurements per well or over 27,500 measurements for all 96 wells in a standard microtitration plate. Typically, further calculations are required to be made to produce useful data from these measurements. SUMMARY OF THE INVENTION

5 I have developed a computer-based method of performing optical density measurements on a plurality of cell population samples. Preferably, the cell population samples are in media in wells in a conventional microtitration plate such as a 96-well plate. hi accordance with a preferred embodiment of the invention, a calibration factor is first established to eliminate the effect of the media used in the wells. Optical density

10 measurements are then made every five minutes on the cell population samples in each of the wells. For the measurement of the extent of apoptosis, every five minutes it is also necessary to compute for each cell population the slope of the optical_.density measurements versus time and correct for any effects attributable to the cell populations in the control sample. For each cell population sample, the values of interest include the maximum

15 corrected slope value, the time from initiation of the assay to observance of the maximum optical density value, the time from initiation of the assay to the beginning of apoptosis and the time from the beginning of apoptosis to observance of maximum optical density value.

BRIEF DESCRIPTION OF DRAWINGS

20 These and other objects, elements and features of the invention will be more readily apparent from the following detailed description of the invention in which:

Fig. 1 is a schematic illustration of apparatus used in the practice of the invention; Figs. 2 and 3 are a flow chart of a first embodiment of the invention; Fig. 4 is an illustration of the output from the embodiment of Figs. 2 and 3; 25 Fig. 5 is a flow chart illustrating further details of one part of the embodiment of

Figs. 2 and 3;

Figs. 6 and 7 are a flow chart of a second embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION

30 Illustrative apparatus for practicing the invention is shown in Fig. ^ . Such apparatus comprises a spectrophotometer 10, a microtitration plate 20 and a computer 30. The computer includes a keyboard 40, a display 50 and a pointing device 60, such as a mouse. Aside from a computer program that is the subject of the present invention and is loaded in computer 30, the apparatus of Fig. 1 maybe conventional. For example, spectrophotometer

35 10 may be a SPECTRA max ™ 340 spectrophotometer or a Bio-Tek Power Wave X spectrophotometer. Both devices are designed to operate with standard microtitration plates. An on-board processor in the spectrophotometer calculates and reports optical density values to the computer.

For each well, or the cell population sample in the well, the spectrophotometer generates a set of measurements of optical density values. Thus, in the case of a 24-hour experiment in which optical density values are measured every five minutes, the spectrophotometer generates a 96 x 288 array of optical density values, d ;_j , where i identifies the well and j identifies the measurement and 0 ≤ i <95 and 0 ≤ j < 287. As will be described below, for each set of such measurements from one well, computer 30 calculates a set of slope values. Illustratively, the calculation of slope values begins after three hours of optical density values have been collected and thereafter a new slope value is calculated every time a new optical density value is measured.

Fig. 2 is a flow chart depicting the first part of this process in more detail. The process begins at step 105 when the user selects "Measuring Apoptosis" from a menu of choices presented to him/her. In response to this selection, the computer presents to the user at step 110 a display of a template for a microtitration plate as well as a set of menu offerings or buttons so that the user can initialize the program. For example, the user can specify the wavelength of illumination of the microtitration plate, its temperature, sampling interval, start time, run time and mode of operation. Alternatively, default settings can be used. Further, the user will identify, preferably by using "point and click" procedures with the template, those wells that are controls, blanks and empties. Advantageously, color coding is then used in the template display to identify these wells for the duration of the measurement.

When all the necessary parameters have been specified, the user can then initiate the measurement process at step 120. The spectrophotometer then reads the optical density values for each well (step 125), and displays these data if desired; and the program calculates a calibration coefficient (steps 130-150). At step 130, the program determines the average, OD contr, of the optical density observed in the control wells; and at step 135, it determines the average, OD blank, of the optical density observed in the blank wells. At step 140, it subtracts OD blank from OD contr to produce ΔOD and at step 145 it calculates the calibration constant k = 0.03/ΔOD. Next, at step 150 it checks whether k is between a minimum of 0.5 and a maximum of 2.0.

If the calibration constant is within limits, it is displayed to the user at step 155 and the recommendation is made that the user proceed with the measurement of apoptosis. If the user elects to do so, the value of k is stored. If the user does not elect to continue, the measurement data is saved at step 160 in a backup file and the user is informed of the availability of the file.

Fig. 3 is a flow chart depicting the measurement of apoptosis in accordance with the invention. After the user elects to continue the measurement (Fig. 2, step 155), the program tests at step 205 if enough data has been collected to begin calculating the slope of the data values versus time. This is done because the user interface varies with this condition as described below. Ordinarily, three hours of data (36 sets of optical density measurements) are accumulated before slope calculations begin.

The user interface is similar to that displayed in step 110 of Fig. 2 and includes a display of the template for the microtitration plate as well as a set of menu offerings or buttons. As shown in Fig. 4, the display of the template includes a time plot of the optical density data being measured in each well, the operating parameters of the measurement, and for each well a numerical value that represents the extent of apoptosis. Other values can also be displayed as described below. The menu offerings or buttons allow the user to change all the parameters of the measurement within certain ranges while the measurement is being run. In addition, the menu offerings or buttons allow the user to pause the measurement, to stop the measurement altogether, to select a specific well for further attention, to display the data being plotted for that well at full scale on the display, and to subject that data to additional analysis. Advantageously, split screen capabilities are also available to allow the display of data from some subset of the 96 wells; and capabilities are also provided to overlap the display of data from different wells selected by the user.

If sufficient time has elapsed to permit calculation of slope values, the display also offers the user menu selections or buttons that permit the user to request the calculation of extent of apoptosis and to select a variety of different displays. As shown in Fig. 4, one such display depicts for each of the 96 wells a plot of the measured optical density values versus time as well as a numerical value that represents the extent of apoptosis. Other options allow for the display of the maximum optical density value measured in each well, either in place of the numerical value for extent of apoptosis or in addition to it. If desired, the plot of measured optical density values can be suppressed.

Returning to Fig. 3, the main steps of the program throughout the many hours of the measurement basically include: generating a display, monitoring the user interface for any changes, making the changes requested and processing the measurement data. More specifically, in the case where there is not enough data to calculate the slope values, the program displays the well template and menu selections or buttons at step 210, monitors the display for selection of a menu offering at step 215, identifies any selection at step 220 and updates the display and any related internal controls. As the optical density measurements become available every five minutes from the spectrophotometer, the program reads these measurements at step 230, stores them in its memory and updates the display at step 235. The program continues in this fashion until sufficient data has been collected to permit calculation of slope values.

At this point, the operation of the program branches to the right hand side of Fig. 3. The operation of the display and the processing of received optical density values is similar to that of steps 210 through 235 and the corresponding steps bear identifying numerals that have been increased by 50. The only difference, as noted above, is that the display has additional capabilities including the ability to select the calculation of the extent of apoptosis (EA).

If the user has selected this feature, the program calculates the extent of apoptosis at step 305 and updates the display at step 310. Regardless of whether this calculation is selected, the program tests if the measurement has been completed (i.e., if the run time has elapsed) at step 315 and either continues on or exits.

Further details relating to step 305 are set forth in Fig. 5. The calculation begins at step 405 by determining the slope over a series of consecutive measurements of optical density versus time. Illustratively, 36 consecutive measurements (3 hours of data) are used in each calculation. Preferably, the slope is determined by making a best fit approximation of a straight line to the 36 consecutive measurements of optical density. As will be appreciated, the straight line has the form: optical density = A+ Bt where B is the slope. Any negative slope value is set equal to zero.

The slope value calculated in step 405 is then corrected by determining the slope value for the corresponding measurements in the control well(s) for the same time period and subtracting that value from the slope value determined in step 405. Again, the slope is determined at step 410 by making a best fit approximation of a straight line to the measurements of optical density in the control well(s) and any negative slope value is equal to zero. If there is more than one control well, the slope values measured in each of the control wells at the same time period are first averaged at step 415. And, at step 420, the value of the slope determined at step 415 is subtracted from the value of the slope determined at step 405 to yield a value which is referred to as Net Sample Slope (NSS). The extent of apoptosis (EA) is then determined at step 425 by multiplying NSS by the coefficient k that was determined in step 145 of Fig. 2. The value of interest for each sample well is the maximum value of the net sample slope (NSS max) or the maximum value of the extent of apoptosis (EA max). Whichever value is used, the value generated at step 420 or 425 is compared at step 430 with the largest value previously stored and the greater of the two values is retained at step 435 for future use. This value is also used to update the display in step 310 of Fig. 3.

Other values are also of interest to the researcher. These include the time (Tm) from the initiation of the assay to observance of the maximum value of the optical density (OD max), the time (Ti) from the initiation of the assay to the beginning of apoptosis and the time (Td) from the beginning of apoptosis to observance of maximum OD value of the sample. Advantageously, these times are available for readout at any point after OD max achieved. Since Tm depends on the time at which OD max was detected, the time of OD max is also stored at step 435 and Tm is determined from this time value at step 440. Typically, Tm equals the time value stored plus a small correction factor on the order of 0.25 hours which accounts for the time during which the cells were exposed to an apoptosis inducer before initiation of the assay.

The beginning of apoptosis is determined by the time at which the first EAmax is registered. This time is designated Ti. Td is simply the difference between Tm and Ti and is calculated at step 450.

Each of the steps 405-450 is performed for each measurement of optical density in each of the wells other than the control sample(s). As the data becomes available it is used to update the display as indicated in step 310. When the measurement is completed, the final screen freezes.

The apparatus of Fig. 1 may also be used to measure growth kinetics. As in the case of the measurement of apoptosis, the spectrophotometer generates a set of measurements of optical density values for each well, or the cell population sample in the well. Thus, in the case of a 24-hour experiment in which optical density values are measured every five minutes, the spectrophotometer generates a 96 x 288 array of optical density values, d _y , where i identifies the well and j identifies the measurement and 0 < i <95 and 0 ≤ j < 287. Proper measurement of grown kinetics requires calibration of the instrument to the particular cell population being measured because the measured optical density varies with the type of cell, hi particular, the amount of light absorbed (or transmitted) by a cell is known to be a function of properties of a cell such as its density and the size of its nucleus. The amount of light absorbed (or transmitted) by a cell population is also a function of its concentration in the medium in which it is distributed. Accordingly, the first step in the measurement of growth kinetics is to calibrate the instrument by measuring the optical density of the cell population of interest at a distribution of known concentrations that span the range of concentrations that are expected to be observed in the course of the measurement of growth kinetics.

For this purpose a microtitration plate is prepared in which the wells of the plate contain known concentrations of the cell population of interest distributed over the concentration range of interest. Illustratively, the plate is prepared so that seven different concentrations of cells are used in the different wells and four wells each are used with the same concentration of cells. Another four wells are blanks that contain media but no cell population. Preferably, all the cells contain the same volume of solution. By way of example, the seven different cell concentrations maybe 1.0, 1.5, 2.0, 3.0, 4.0, 6.0 and 8.0 units per mL of solution where the magnitude of the units is selected appropriately by the user to include the expected range of observed values.

Fig. 6 is a flow chart depicting the calibration process in detail. The process begins at step 605 when the user selects "Growth Kinetics" from a menu of choices presented to him/her. In response to this selection, the computer presents to the user at step 610 a display of a template for a microtitration plate as well as a set of menu offerings or buttons so that the user can initialize the program. For example, the user can specify the wavelength of illumination of the microtitration plate, its temperature, sampling interval, start time, run time and mode of operation. Alternatively, default settings can be used. Further, the user will identify, preferably by using "point and click" procedures with the template, those wells that are blanks and empties. Advantageously, color coding is then used in the template display to identify these wells for the duration of the measurement.

When all the necessary parameters have been specified, the user can then initiate the calibration measurement process at step 620. The spectrophotometer then reads the optical density values for each well (step 625), and displays these data if desired; and the program calculates a calibration coefficient (steps 630-660). At step 630, for each set of wells having the same concentration of cells the program determines the average, OD aver, of the optical density observed in each of the wells in the set; and at step 635, it determines the average, OD blank, of the optical density observed in the blank wells. At step 640, it subtracts OD blank from each of the values of OD aver determined in step 630 to produce a like number of values of ΔOD and at step 645 it fits these values to a straight line having the form Y=A+BX where the Y values are the values of ΔOD determined in step 640 and the X values are the corresponding values of known cell concentration. At step 650, it calculates the correlation coefficient, R. The line that is calculated in step 645 is displayed to the user at step 655 along with the correlation coefficient. If the coefficient is less than 0.9 a warning message is also presented with advice on how to improve the correlation coefficient. If the user elects to proceed, the process continues as shown in Fig. 7. If the user does not elect to continue, the measurement data is saved at step 660 in a backup file and the user is informed of the availability of the file.

Fig. 7 is a flow chart depicting the measurement of growth kinetics in accordance with the invention. The process begins at step 705. As in the case of the calibration step, the computer presents to the user a display of a template for a microtitration plate as well a set of menu offerings or buttons so that the user can initialize the program. For example, the user can specify the wavelength of illumination of the microtitration plate, its temperature, sampling interval, start time, run time and mode of operation. Alternatively, default settings can be used. Further, the user will identify, preferably by using "point and click" procedures with the template, those wells that are controls, blanks and empties. The display is substantially the same as that of Fig. 4 but, of course, does not provide any display of apoptosis. Advantageously, color coding is then used in the template display to identify these wells for the duration of the measurement.

When all the necessary parameters have been specified, the user can then initiate the measurement process. On default, the measurement of growth kinetics will be performed only once at the end of the experiment, which typically is 48 hours after the experiment begins. However, in some cases, it is desirable to measure growth kinetics throughout the experiment and even at frequencies such as once every five minutes. To accommodate these options, the system measures the optical density at least once every five minutes in each well of the microtitration plate that contains a cell population sample. Measurements are also made on a suitable number of controls such as wells that contain media but not cell population samples (i.e., blanks).

The main steps of the program throughout the many hours of the measurement basically include: generating a display, monitoring the user interface for any changes, making the changes requested and processing the measurement data. More specifically, the program displays the well template and menu selections or buttons at ster 710, monitors the display for selection of a menu offering at step 715, identifies any selection at step 720 and updates the display and any related internal controls at step 725. As the optical density measurements become available every five minutes from the spectrophotometer, the program reads these measurements at step 730, stores them in its memory and updates the display at step 735. The program continues in this fashion until a growth rate calculation is required. This may not happen until the end of the experiment. However, one advantage of the invention is that optical density values are read quite frequently and data is therefore available for computing growth rate pretty much on demand. Advantageously, the user interface includes appropriate displays to allow the user to program how frequently he or she wants the growth rate to be calculated.

At the end of the experiment or at any other time the user has selected, the growth rate is calculated at step 740 and the results are displayed at step 745. Regardless of whether a growth rate calculation is performed, after updating the optical density measurements, the program tests at step 750 if the experiment is completed and either continues on or exits.

Further details concerning step 740 are set forth in Fig. 8. The calculation begins at step 810 by eliminating the effect of any optical absorption in the wells that is not due to the cell population. This is done by subtracting the measured optical density value in the blank well(s), OD blank, from that in each well that contains a cell population, OD sample. This produces the value ΔOD = OD sample - OD blank for each well with a cell population sample.

This data is then converted at step 820 into cell population data using the equation Y=A+BX where A and B are the coefficients of the straight line that was fitted at step 645 to the calibration data and the Y value is the corrected value, ΔOD. Thus, the cell population is given by X=(ΔOD-A)/B.

A growth curve is then determined at step 830 by relating the cell population data for each well to the elapsed time for the experiment and displaying the resulting growth curve. A variety of different measures of cell growth may be used. One such measure is simply a plot of the number of cells at different times throughout the course of the experiment. This plot is typically referred to as the growth curve.

Another advantageous measure is a best-fit slope of the growth curve over three or more periods, typically the first half of the experiment, the second half of the experiment and the entire experiment. Such slopes are calculated and displayed at step 840.

Still other useful measures are the Growth Stimulatory Index, GSI, and the Growth Inhibitory Index, GH, where GSI = (Slope sample/Slope control) X 100 and GH = (1 - Slope Sample/Slope control) X 100.

Advantageously, the program calculates GH only if Slope sample < Slope control and calculates GSI only if Slope sample > Slope control. Accordingly, the program tests for this condition at step 850 and performs the appropriate calculation at step 860 or 870.

Claims

What is claimed is:

1. A computer-based method of performing optical density measurements on a plurality of cell population samples, including at least one control sample, in media in wells in a microtitration plate comprising the steps of: for each cell population sample, repeatedly measuring its optical density over a period of time of at least several hours to obtain a set of measurements of optical density values versus time; for each set of measurements, calculating slopes of optical density values versus time for a series of time intervals extending through said period of time to obtain a set of slope values; for each slope value in a set of slopes values other than a set of slope values for a control sample, subtracting from said slope value the corresponding slope value in the same time interval in the set of slope values for a control sample to produce a net slope value.

2. The method of claim 1 further comprising the step of determining the maximum net slope value for each population sample.

3. The method of claim 2 further comprising the step of determining the time from initiation of an assay of the cell population samples to observance of the maximum

OD value for each population sample.

4. The method of claim 3 further comprising the steps of determining the time from initiation of an assay of the cell population samples to observance of the beginning of apoptosis.

5. The method of claim 1 further comprising the step of displaying plots of the measured optical density values versus time for each population sample.

6. A computer program for performing optical density measurements on a plurality of cell population samples, including at least one control sample, in media in wells in a microtitration plate comprising the steps of: for each cell population sample, repeatedly measuring its optical density over a period of time of at least several hours to obtain a set of measurements of optical density values versus time; for each set of measurements, calculating slopes of optical density values versus time for a series of time intervals extending through said period of time to obtain a set of slope values; for each slope value in a set of slopes values other than a set of slope values for a control sample, subtracting from said slope value the corresponding slope value in the same time interval in the set of slope values for a control sample to produce a net slope value.

7. The computer program of claim 6 further comprising the step of determining the maximum net slope value for each population sample.

8. The computer program of claim 7 further comprising the step of determining the time from initiation of an assay of the cell population samples to observance of the maximum OD value for each population sample.

9. The computer program of claim 8 further comprising the steps of determining the time from initiation of an assay of the cell population samples to observance of the beginning of apoptosis.

10. The computer program of claim 6 further comprising the step of displaying plots of the measured optical density values versus time for each population sample

11. A computer-based method of performing optical density measurements on a plurality of cell population samples in media in wells in a microtitration plate comprising the steps of: preparing a plurality of cell population samples of different known concentrations in media in wells in the microtitration plate, said concentrations ranging from the lowest concentration expected to be measured in the course of the optical density measurements to the highest concentration; measuring the optical density of the cell population samples of different known concentrations; providing for cell growth in cell population samples in media in wells in a microtitration plate; measuring the optical density of the cell population samples after a period of cell growth; and determining the amount of cell growth in the cell population samples by relating the measured optical density from the samples that experienced the period of cell growth to the measured optical density from the cell population samples of different known concentrations.

12. The method of claim 11 further comprising the step of displaying plots of the measured optical density values versus time for each population sample.