Patent Application
20090155919
Achieving high throughput analysis is a serious problem in the context of drug discovery as thousands of compounds must be screened to determine reactivity. However, some of the same issues are found in other types of clinical analysis such as in a commercial laboratory. A common problem with high throughput screening is that there may be multiple steps required to determine if there has been a reaction between two molecules. This applies to both screenings that determine the identity or presence of one of the molecules, such as is utilized in clinical labs, and to high throughput screening for pharmaceutical development. The reason why multiple steps are used is that most reactions do not cause a color change or other easy to identify physical property, so there are often difficulties in determining whether a reaction has occurred. To solve this problem, a pre-step of “tagging” one of the molecules using a fluorophor or a quencher, or otherwise providing an additional reaction to identify that a reaction has taken place, e.g., using an ELISA reaction, is necessary. Pre-tagging a component can be expensive and time consuming, even if it speeds up the screening reaction. There may be stability problems with the tagged component or enzymes used in the reaction. This pre-step is not a major issue if only a few assays are conducted, but if hundreds or thousands of assays are carried out in a day, the pre-step can require substantial set up time or the use of additional staff. Thus, there is a need for an assay system that is rapid and accurate but does not rely on tagging one of the components.
It has long been known that binding or other chemical reactions between two molecules are normally exothermic or endothermic; that is, they either give off or require the addition of heat in order to occur. If the reaction takes place in a liquid, this change in energy state can change the temperature of the surrounding solution. While this temperature change has been used to a limited degree to determine the course of reactions, the temperature differences can be difficult to measure, particularly if small levels of reactants are used as is often the case in drug screening,. In addition, the methods that have been used to quantify these changes, such as resistance measurements, are not readily adaptable to high throughput screening.
Optical measurements without a pre-tagging or other pre-step would appear to have promise but there are inherent difficulties. Kromoscopy, which utilizes a near infrared (“NIR”) analogy to color perception in the visible region, has shown sensitivity beyond classic spectroscopic measurements and chemical measurements in some circumstances. Kromoscopy relies on the illumination of the object or liquid with broadband radiation (an analog of white light in the visible), and use of a series of spectrally overlapping filters to detect the reflected, emitted or transmitted radiation to determine the object's relative “color.” This approach is discussed in U.S. Pat. No. 5,321,265, the disclosure of which is incorporated herein by reference. This method provides high sensitivity to low concentrations of molecules but is not easily adapted to high throughput screening. However, it has been found that infrared Kromoscopy in aqueous solutions can be sensitive to changes in temperature of the water. While these temperature changes can be easily ignored when making measurements on constituents such as hemoglobin, they are more difficult to deal with for low concentration materials such as glucose because the size of the signal is much smaller relative to the changes arising from changes in temperature. In fact, it appears that very small differences in temperature may need to be corrected for in a Kromoscopic analysis. These small temperature changes are along the same order of magnitude as those for the liquid in reactions and thus it may be possible to use the means of correcting for temperature as a measuring tool. By selection of appropriate filters, one can determine or correct for the effect of temperature on the optical properties of the water.
The present invention measures the modification of the optical properties of water caused by the small temperature changes from the heat of reaction and utilizes these measurements in high throughput screening methods. Normally, a thermal block containing a series of thermally isolated wells is used so that a reaction can take place in each well without affecting the temperature of any other well. Any type of thermally isolated reaction chamber could be used in lieu of a thermal block. Alternatively, the spatial or temporal changes in the optical properties of a well or test tube is measured. A chemical or biological reaction, such as a binding reaction, is allowed to occur in each well or chamber and the optical properties of the all of the wells are monitored, preferably at the same time, using optical, most preferably Kromoscopic, measurements. A measurement of the change in the optical properties caused by a temperature change in a well, preferably monitored using Kromoscopic measurements, can be used to determine if reaction occurred.
In more detail, the present invention provides a method of determining whether there is an interaction between a ligand and a target in solution based on a thermal change in said solution. The ligand and the target are allowed to interact in a solution and The solution is optically monitored for changes in temperature that are indicative of an interaction between the ligand and the target. The vessel which is monitored, such as a well or test tube, is kept thermally isolated to ensure that any change in temperature is from the reaction of said ligand and said target. There are several ways to thermally isolate the reaction vessel. It can be isolated by enclosing the majority of the vessel in a thermal block or a temperature control unit can be used. If a thermal control unit is used, there could be heating components, cooling components, or components that provide heating and cooling. Alternatively, monitoring each well or vessel temporally or spatially can indicate if a reaction has occurred without the complete thermal isolation.
A preferred method of carrying out the present invention has the reaction take place in a well in a multi-compartment well plate such as a 96 or 384 compartment well plate. If individual tubes are used, they are preferably in a multi-tube array. The method of the invention can take place in solution, preferably an aqueous solution, or at least one of the target and the ligand may be bound to a solid support such as the reaction vessel or well. Alternatively, a separate solid support could be used.
To carry out a preferred type of Kromoscopic measurement, each of the wells in a multiwell plate is illuminated with broadband infrared radiation and the radiation transmitted, or reflected from the solution is detected at a detector. Alternatively, multiple sources, such as light emitting diodes (“LED”) may be used as illumination sources. The preferred illuminating radiation for aqueous based systems is near infrared radiation having a wavelength of about 900-1500 nm. If other systems are used, the wavelength may be adjusted. The detector may include a plurality of detection units, and each of the detection units detects a specific region of the spectrum, normally by filtering the detection unit to achieve the desired performance. Alternatively, a single detector having multiple filters may be used. The filters may be in the form of a filter wheel or some other device that provides different filtering at different times, or a modified Bayer plate, having a grid or array of different filters may be used. Normally, the detected region of the spectrum for each of the filters or detection units has at least partial overlap with the detected region of the spectrum for another of the filters or detection units. The change in temperature of the solution due to the interaction of the ligand and the target causes a change in the optical properties of the solution that can be distinguished from external temperature changes based on a variety of methods including temporal or spatial information. For example, a transient change in the temperature of the system can cause a change in temperature in a steady gradient pattern or having a spatial gradient across a well, while a change due to reaction is likely to be a sharper temporal spike or a change without the spatial gradient. While it is possible that each of the wells in a multi-compartment well plate can viewed optically with a scanning head that is scanned to measure the temperature of each well separately or the multi-compartment well plate is moved beneath an optical device to measure the temperature of each well separately, a device that measures distribution within the wells, and all the wells simultaneously, is preferred.
The present invention is based on the recognition that the heat of reaction of a binding process can modify the optical properties of the surrounding liquid and that can be used to determine if binding has occurred. Kromoscopy provides an optical platform sufficiently sensitive to measure the small changes that occur based on a change in the optical properties of a liquid such as aqueous solutions from the heat of reaction. In addition, Kromoscopy provides a platform that can be used to measure a number of wells containing reactants at once, thus making it amenable for high throughput screening.
Kromoscopy utilizes broadband illumination of a sample followed by detection with detection units having spectral overlap characteristics. Water is well known to have an absorption spectrum with small shoulder changes in absorbance at 960 nm and 1450 nm. These spectral characteristics can be used with infrared illuminating radiation and known detectors to provide a sensitive assay system. For frequencies up to about 1100 nm, Si detections have high sensitivity and low cost. For wavelengths higher than 1100 nm, and including 1450 nm, InGaAs detectors can be used. However, the InGaAs detectors have about 100 times lower sensitivity than Si detectors and are much higher cost. Therefore, for most uses, Si detectors are preferred.
The advantages of using Si detectors outweigh the disadvantages of not going out to 1450 nm. Si detectors are used in a number of devices including digital cameras and other CCD devices. In digital cameras, an IR blocking filter is used to block the IR wavelength and a Bayer plate imparts color to the photograph. Removal of the Bayer plate and the IR blocking filter and insertion of proper filters can allow digital cameras to be used as detection units. In the alternative, any common Si detector can be used with proper filtering. If two or more detection units are used, each should have an overlap frequency with the other(s) near the 960 nm band of water. Use of these filters allows the ratio of the values obtained to be correlated to a shift in the optical properties caused by a change in the temperature of the liquid.
Systems of this type are sensitive to 0.001° C. Using the type of small wells in a common 96 well plate (about 0.2-0.4 ml), the heat of reaction from a small number of molecules can cause more than the requisite heat increase to be measured, assuming the well is sufficiently thermally isolated.
A test system to show the effectiveness of the present invention can be constructed using two wells of a 96 well plate and a pair of digital cameras. One well (the reaction well) includes a solution of the target compound and the other (the control well) does not have the target compound. While a solution reaction is preferred in most cases, the well may be coated with one reactant. Methods of coating the wells of a 96 well plate are well known, as are preparing a control well. If the target compound is a protein and standard plastic 96 well plates are used, the sample solution is normally placed in the reaction well followed by a coating solution of a material like bovine serum albumin to minimize nonspecific reaction. The control well may have the same initial solution as the reaction well except it is lacking the target molecule and then the control well is also coated with the coating solution.
The 96 well plate 100 is placed in a thermal block or water bath 120 to keep the temperature constant. The thermal block or water bath 120 may have a control unit (not shown) having one or more heating or cooling units to keep the temperature constant. The reaction well 101 and the control well 102 of 96 well plate 100 are located so that each may be visualized by the same two or more cameras 150. Two cameras, 150a and 150b, are illustrated, but more may be used. Cameras 150a and 150b are standard digital cameras having the Bayer plates and IR filters removed and they are optically directed to view the wells. Preferably, cameras 150a and 150b are located such that the optical path is equivalent in length; that is, they are optically congruent. Each camera (150a and 150b) has an associated filter (160a and 160b, respectively) that limits the wavelength range that can reach the internal silicon (Si) detector (not shown). Each of the two or more filters 160 are centered about different wavelength but have a partial overlap in wavelengths they allow through with at least one other 160. For example, one camera (150a) could have an RT-830 filter 160a (Hoya) while the other could have a RM-90 filter 160b (Hoya). These filters have an overlap near the 960 nm band of water.
If the target or the ligand is pre-bound to the well, the other is added in an aqueous solution in a pre-determined amount. If neither the ligand nor target is pre-bound, they are both added to the well in aqueous solution so that the total volume is a known amount. The target and ligand, if they form a binding pair, react and the energy of binding heats (or cools) the aqueous solution in the well. Even though the temperature difference is small, the change in temperature causes a change in the optical properties of the aqueous solution. This change in optical properties modifies the absorbance of the solution and the Si detectors measure different values, changing the ratio of absorbance from one detector unit to the other because of the difference in the filters. If more than two detector units are utilized, even more accurate measurements can be made.
Using cameras 150 with the associated filters 160 as the detection units, the infrared light from each well can be identified spatially by proper assignment of pixels. Thus, all 96 wells can be viewed simultaneously. The control well can be “seen” at the same time as the reaction well because the camera detector unit shows spatial difference. The Si detector gives a value for each pixel in the viewing field and using software, one can assign certain pixels to the location of each well. Although this system has been described for only a reaction well and a control well, many more wells could be viewed simultaneously. Since a 5 megapixel camera has 5000 pixels, assigning 25 to each well of a 96 well plate would still leave over half unassigned. The reaction and measurements can take place in seconds or less and the next plate can be moved in position using convention plate moving machinery standard in the field of high throughput screening. The measurement cycle restarts, leading to the ability to screen many samples in a short time. Alternatively, each well or a subgroup of the wells could be viewed and the plate 100 is moved so that each well or subgroup is viewed in series. While this has certain disadvantages in speed and the fact that one is not viewing the wells simultaneously, it allows more pixels to be assigned to each well so it might provide better temperature discrimination.
The foregoing examples are merely illustrative of the invention and are specifically deemed not limiting. The present invention is described in the following claims.
