Integrated Fluorescence Analysis System

This invention relates to fluorescence imaging and, more particularly, to selective fluorescence imaging using fluorescence spectral properties preselected by flow cytometer analysis.

Fluorochromes are frequently bound to molecular cell components to yield quantitative information about the cell components. Several different fluorochromes may be present in a given cell sample to yield a number of emission spectra. Further, the emission spectrum from a given fluorochrome may be altered as a result of changes in the physiological state of the cells.

Fluorescence is generally detected by one or more detectors that are responsive to specific wavelengths of interest. Cytometry, either flow cytometry or image cytometry, may use this fluorescence as some measure of individual cells or cellular constituents. Cytometry can differentiate between effects that affect all cells equally and effects that cause subpopulations of cells within the population to change. See generally Clinical Cytometry, M. Andreeff ed., New York Academy of Sciences, New York, N. Y. (1986), incorporated herein by reference. It will be appreciated that an increasing number of fluorochromes with different sensitivities to changes in cell physiology and labeled monoclonal antibodies that recognize specific antigens impose a requirement for an increasing number of spectral channels for fluorescence analysis.

U.S. Pat. No. 4,905,169, issued Feb. 27, 1990, to Buican et al., incorporated herein by reference, teaches cytometry apparatus, and particularly flow cytometry apparatus, for simultaneously resolving a plurality of spectral properties from a total fluorescence spectrum. A Fourier-Transform (FT) spectrometer encodes spectral information in a waveform (interferogram) on which it then performs a numerical, rather than optical, spectral analysis. When installed in a flow cytometer, the FT spectrometer analyzes the total fluorescence emitted by single particles as they cross an excitation laser beam and resolves this emission into a set of separate fluorescences, each described by a characteristic emission spectrum which has been selected by the operator. Thus, a set of numbers (spectral parameters) is produced for each particle, describing the spectral properties of the individual particles in the sample. Taken separately, each spectral parameter represents an enhancement of a specific fluorescence, since it is obtained by eliminating all other emission spectra from the total fluorescence.

Image cytometry devices are also used for detecting fluorochromes in biological specimens and may be implemented as a laser-scanning confocal microscope. See generally Handbook of Biological Confocal Microscopy, J. Pawley ed., Plenum Press, New York, N.Y. (1989, 1990), incorporated herein by reference. However, fluorescence data is obtained on a pixel-by-pixel basis over the biological sample and it is difficult to quantify specific cell subpopulations in the sample. This is normally done by physically sorting a selected subpopulation by flow sorting, recovering the cells, and placing them on a slide for morphological analyses. Accordingly, the present invention incorporates Fourier-Transform technology into the laser-scanning confocal microscope to enable image enhancement for a selected superposition of fluorochrome spectra to effect a virtual sorting of a subpopulation previously identified.

It is an object of the present invention to generate concurrent multiple images from a laser-scanning confocal microscope, each enhanced for a particular emission spectrum.

It is another object of the present invention to generate an image from a biological specimen enhanced for an emission spectrum selected from flow cytometer data.

One other object of the present invention is to enable a virtual sorting of cells in a biological sample without physical sorting.

Additional objects, advantages and novel features of the invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

To achieve the foregoing and other objects, and in accordance with the purposes of the present invention, as embodied and broadly described herein, the apparatus of this invention may comprise an integrated fluorescence analysis system. A flow cytometer is provided with birefringent optics for simultaneously measuring a first plurality of spectral wavelengths present in a first fluorescence spectrum from a first sample. The fluorescence spectrum is resolved by a first processor means into first numbers representing the intensity of spectral components of the first plurality of spectral wavelengths. An imaging cytometer is also provided with birefringent optics for simultaneously measuring a second plurality of spectral wavelengths present in a second fluorescence spectrum from a second sample. A second processor means then resolves the second fluorescence spectrum into second numbers representing the intensity of spectral components of the second plurality of spectral wavelengths. Communication means connects the first processor means with the second processor means so that the first or second numbers can be input to the second or first processor means, respectively, for enhancing spectral components of the second or first samples, respectively.

In another characterization of the present invention, a programmable spectral imaging cytometer includes microscope means for scanning a laser beam over a selected sample image plane to excite a fluorescence spectrum over the image plane. Birefringent optics resolves the fluorescence spectrum into a plurality of pixels, each pixel having a pixel fluorescence with one or more spectral wavelengths. Processor means resolves each pixel fluorescence into numbers representing the intensity of spectral components in the pixel fluorescence. Numbers are stored representing spectral components from a preselected spectrum. A selection means, such as a computer, then selects those pixels having numbers corresponding to the stored numbers to develop a pix pixel image over the image plane corresponding to the preselected spectrum.

In one other characterization of the present invention, a method provides for obtaining from an actual sample a sample representation having preselected spectral components. The actual sample is irradiated with a laser beam to stimulate a fluorescence spectrum from the sample having one or more spectral wavelengths. The fluorescence spectrum is resolved into sample numbers representing the intensity of spectral components in the fluorescence spectrum. The sample numbers are compared with stored numbers representing the preselected spectral components. Portions of the sample are then identified that have a fluorescence spectrum corresponding with the preselected spectral components.

The accompanying drawings, which are incorporated in and form a part of the specification, illustrate the embodiments of the present invention and, together with the description, serve to explain the principles of the invention. In the drawings:

FIG. 1 is a schematic in block diagram form of an integrated fluorescence analysis system according to the present invention particularly showing flow cytometry components.

FIG. 2 is a schematic in block diagram form of an integrated fluorescence analysis system according to the present invention particularly showing imaging cytometry components.

In accordance with the present invention, a sample is irradiated with a laser beam to produce a fluorescence response spectrum having a plurality of wavelengths. Birefringent optics, as described in U.S. Pat. 4,905,169, incorporated by reference, enable a Fourier analysis to resolve the fluorescence spectrum into numbers representing the intensity of each of the spectral wavelengths of the fluorescence. Thus, a set of numbers representing base spectra can be selected from the total spectrum and thereafter used to provide virtual sorting of biological particles against the base spectra.

As shown in FIGS. 1 and 2, an integrated fluorescence analysis system incorporates a flow cytometer 10 and an imaging cytometer 20, each with birefringent optics, and communicating with one another through a communication channel, such as computer 52. Flow cytometer 10 and imaging cytometer 20 cooperate in the following ways: (a) base spectra for resolving total cell/pixel fluorescence are shared; and (b) the spectral data obtained by one instrument are used to program the other instrument. Then, if the analysis of a cell suspension reveals the existence of a spectrally distinct subpopulation, the average spectrum for that population can be transferred from one instrument to the other to enhance the cell/pixel data and identify cells/pixels that have the same spectral properties. In the imaging cytometer, an image can then be formed with only those pixels containing spectral features of interest. Conversely, the average spectrum of a whole cell of interest, or that of a particular morphological detail, can be transferred to the flow cytometer, where it is used to identify and quantitate cell subpopulations with the same spectral properties.

Referring now to FIG. 1, flow cytometer 10 communicates with imaging cytometer 20 through computer 52. Flow cytometer 10 provides for simultaneously measuring a plurality of spectral wavelengths present in fluorescence produced when a laser beam 12 excites sample materials, such as biological cells, contained in focused flow stream 11. The output fluorescence is input by lens 14 to birefringent interferometer 16, as particularly described in the '169 patent. It will be appreciated that the sample materials may be biological particles such as chromosomes or other cellular material. The particles can be unstained, wherein the fluorescence is intrinsic fluorescence or autofluorescence, or stained with one or more fluorochromes.

Interferometer 16 includes birefringent elements and polarizing elements to introduce a time-varying phase difference between spectral components of the output fluorescence. Detector 22, which may be a photomultiplier tube (PMT), produces an electrical output signal characteristic of the sum of the intensities of the spectral components incident on detector 22. Analog-to-digital (ADC) converter 26 digitizes the electrical signals output from detector 22 for input to array processor 28. Clock 24 is phase-locked with interferometer 16 and synchronizes operation of the system elements. Clock 24 strobes ADC 26, processor 28, and storage array 32 at a frequency that is greater than the frequency at which birefringent interferometer 16 is driven.

Array processor 28 receives the output from ADC 26 for processing. Processor 28 includes a multiplicity of processors equal to the number of spectral channels over which the fluorescence is to be measured. Each processor computes and outputs a single number representing the intensity of one spectral channel, or wavelength, of the fluorescence from samples in flow stream 11. Storage array 32 then stores the computed spectral numbers. Computer 52 enables imaging cytometer 20 to communicate with storage array 32, as hereinafter discussed.

Referring now to FIG. 2, there is shown a block diagram schematic of an imaging cytometer according to the present invention. It will be understood that the optical analyzing elements, i.e., birefringent interferometer 16', lens 18', detector 22', array processor 28', and clock 24' are identical with the corresponding components 16, 18, 22, 28, and 24 discussed in FIG. 1. A scanning laser microscope 38, 40, 42, 44, and 46 provides the optical input to interferometer 16'. A confocal scanning microscope, e.g., a BioRad MRC-500, can scan over a plurality of image planes within sample 36 to develop a three dimensional image.

Laser beam 38 is input through dichroic mirror 42 to scanning mirror 44, which is driven by scan driver 40 to traverse beam 38 through optics 46 over an image plane within specimen 36. Fluorescence excited from specimen 36 is transmitted to interferometer 16' through microscope lens 46, scanning mirror 44, and dichroic mirror 42. The fluorescence data from specimen 36 is provided as pixel information for a given image plane, i.e., scan driver 40 is synchronized through clock 24' so that fluorescence data are obtained from a sequence of discrete locations over the image plane within specimen 36. It will be appreciated that the fluorescence spectrum from each pixel will be a composite spectrum, i.e., one or more wavelengths, from the fluorescent components in the pixel area. It will also be appreciated that the confocal microscope can scan image planes through the entire volume of a sample so that each "pixel" represents a volume, or "voxel.+ The term "pixel" will be used herein to mean a focal region for scanning microscope 45, whether on a surface area or within a sample volume.

The output from array processor 28' is a set of numbers representing the spectral components of each pixel fluorescence. In one embodiment of the present invention, computer 52 communicates with flow cytometer 10 to obtain one or more numbers indicative of a selected spectral component of specimen 36. As the spectral component numbers from specimen 36 are presented to frame buffers 48, computer 52 selects only those pixels having the selected spectral component or components. The image that is generated for display 54 contains only those pixels having the desired spectral components, i.e., only the desired characteristics such as a particular chromosome or cell or undergoing a particular chemical reaction that produces the selected spectral component or components.

Thus, flow cytometer 10 may examine a serial flow of samples to determine the spectral components of a particular sample of interest. These spectral components are communicated to computer 52 and thereafter used to form the image that is displayed from image cytometer 20 as specimen 36 is scanned by laser beam 38. As a sample is scanned, those pixels having the selected spectral components are enhanced in brightness wherein all cells or subcellular structures with the desired spectral properties appear bright, while all other details appear as a dim background. This difference in intensity allows the identification of those cells that belong to the subpopulation identified in flow cytometry and the analysis of their morphology.

This image enhancement is a "virtual sorting" of the sample components having the selected spectral characteristics. It is equivalent in its results to physical sorting followed by microscopic examination, with the major difference that, in the case of virtual sorting, one sorts numerical data rather than actual cells. Apart from not requiring a physical sorting capability on the flow cytometer, nor the collection of sorted cells and the preparation of separate microscopy samples for each subpopulation of interest, virtual sorting allows large numbers of subpopulations to be analyzed simultaneously from a morphological point of view. The number of simultaneously analyzed subpopulations is limited only by the number of channels in the data processing system of the FT imaging cytometer and by the spectral resolution of the FT spectrometers (typically eight or more), and can exceed by far the number of subpopulations that can be simultaneously physically sorted (typically two to four).

In a "reversed" virtual sorting, the spectral properties of morphologically interesting cells are communicated from image cytometer 20 to flow cytometer 10 to program the FT flow cytometer. In this case, the flow cytometer would rapidly provide population-level data on the abundance of the cells or cellular components of interest. The spectral properties can also be used to program an optical trapping system, such as taught by U.S. Pat. No. 4,887,721, issued Dec. 19, 1989, to separate cells identified as having the same spectral properties as the subpopulation of interest. This technique, an "indirect sorting", is useful when the staining properties of a subpopulation of rare cells are only approximately known, but can be accurately and rapidly determined in flow. Then, the imaging cytometer can accurately discriminate between the rare cells of interest and the rest of the sample, and the automated optical trapping system can scan through the sample, identify the cells, and separate them.

The foregoing description of the preferred embodiments of the invention have been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application to thereby enable others skilled in the art to best utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto.