U.S. patent application number 09/395661 was filed with the patent office on 2002-06-13 for flourescence polarization assay system and method.
Invention is credited to HOYT, CLIFFORD C..
Application Number | 20020070349 09/395661 |
Document ID | / |
Family ID | 26807165 |
Filed Date | 2002-06-13 |
United States Patent
Application |
20020070349 |
Kind Code |
A1 |
HOYT, CLIFFORD C. |
June 13, 2002 |
FLOURESCENCE POLARIZATION ASSAY SYSTEM AND METHOD
Abstract
An instrument is disclosed for fluorescence assays which is
capable of reading many independent samples at the same time. This
instrument provides enhanced throughput relative to single-sample
instruments, and is well-suited to use in general fluorescence,
time-resolved fluorescence, multi-band fluorescence, fluorescence
resonance energy transfer (FRET), and fluorescence polarization.
This invention is beneficial in applications such as
high-throughput drug screening, and automated clinical testing.
Also disclosed are means and methods for a fluorescence
polarization measurement which is highly sensitive, inherently
self-calibrated, and unaffected by lamp flicker or photobleaching.
This fluorescence polarization invention can be practiced on a
variety of fluorescence instruments, including prior-art equipment
such as microscopes and multi-well plate readers.
Inventors: |
HOYT, CLIFFORD C.; (NEEDHAM,
MA) |
Correspondence
Address: |
MARTIN B PAVANE ESQ
COHEN PONTANI LIEBERMAN & PAVANE
551 FIFTH AVENUE SUITE 1210
NEW YORK
NY
10176
|
Family ID: |
26807165 |
Appl. No.: |
09/395661 |
Filed: |
September 14, 1999 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60109618 |
Nov 24, 1998 |
|
|
|
Current U.S.
Class: |
250/458.1 ;
250/459.1 |
Current CPC
Class: |
G01N 21/6445
20130101 |
Class at
Publication: |
250/458.1 ;
250/459.1 |
International
Class: |
G01T 001/10; G01J
001/58; G21H 003/02 |
Claims
I claim:
1. A fluorescence measurement instrument comprising a plurality of
sample regions for receiving samples; excitation means that produce
a first beam; a diffractive optical beamsplitter element that
splits the first beam into plural secondary beams, said plural
secondary beams simultaneously exciting the plurality of sample
regions to effect fluorescence of samples therein; and detection
means for detecting the fluorescence from the plurality of sample
regions.
2. The fluorescence measurement instrument of claim 1, wherein the
detection means comprise a detector with plural independent pixel
regions.
3. The fluorescence measurement instrument of claim 1, wherein the
plurality of sample regions is a multi-well plate having a
plurality of wells having well walls separated by inter-sample
regions, and wherein the plural secondary beams provide
substantially no illumination of the well walls or the inter-sample
regions.
4. A fluorescence measurement instrument comprising excitation
means for providing light; detection means comprising an objective
and a photodetector; a sample region; and a mirror located between
the sample region and the objective for directing light from the
excitation means to the sample region.
5. A fluorescence polarization measurement instrument comprising at
least one sample region for receiving a sample; excitation means
for producing light that is substantially linearly polarized along
a first axis of polarization at the sample region; detection means
that comprise an objective, a photodetector, and a polarization
analyzer; wherein the photodetector provides a plurality of
spatially distinct pixel regions; the objective directs a beam of
fluorescent light from the sample toward a polarizing beamsplitter;
the polarization analyzer divides the beam of fluorescent light
into two linearly polarized secondary beams, one with polarization
axis oriented substantially parallel to the first axis of
polarization and the other with polarization axis oriented
substantially perpendicular to the first axis of polarization; and
the secondary beams of fluorescent light are directed onto the
spatially distinct pixel regions of the photodetector by the
polarization analyzer.
6. The fluorescence polarization measurement instrument of claim 5,
wherein the polarization analyzer is a planar element of doubly
refractive material.
7. The fluorescence polarization measurement instrument of claim 6,
wherein the polarization analyzer is made of calcite.
8. The fluorescence polarization measurement instrument of claim 6,
wherein the polarization analyzer further comprises a second
birefringent element that equalizes the optical path length between
the two linearly polarized secondary beams.
9. A fluorescence polarization measurement instrument comprising at
least one sample region for receiving a sample; excitation means
for producing light that is directed at the sample region to effect
fluorescent emission of the sample and that is substantially
linearly polarized along a first axis of polarization at the sample
region; detection means that comprise an objective, a plurality of
independent detector regions, and a polarization analyzer; wherein
the plurality of independent detector regions comprises one of a
unitary detector with multiple pixel regions and multiple
detectors; the objective collects the fluorescent emission from the
sample region and directs the fluorescent emission in a beam toward
the polarization analyzer; the polarization analyzer divides the
beam of fluorescent emission into two linearly polarized secondary
beams, one with polarization axis oriented substantially parallel
to the first axis of polarization and the other with polarization
axis oriented substantially perpendicular to the first axis of
polarization; the linearly polarized secondary beams are directed
by the analyzer to separate detector regions; and said excitation
means further provide switching means for changing the state of
polarization of the excitation light at the sample region during a
single fluorescence polarization measurement from a first
orientation parallel to the first axis of polarization to a second
orientation parallel to a second axis of polarization which is
substantially perpendicular to the first axis of polarization.
10. The fluorescence polarization measurement instrument of claim 9
wherein the switching means are automated and do not require input
to effect the changing of the state of polarization of the
excitation light from the first orientation to the second
orientation.
11. The fluorescence polarization measurement instrument of claim
10, wherein the switching means comprise a liquid crystal cell and
associated drive circuitry.
12. The fluorescence polarization measurement instrument of claim
10 wherein the plural independent detector regions comprise a
plurality of pixel regions on a unitary detector and the
polarization analyzer comprises a planar element of doubly
refractive material.
13. The fluorescence polarization measurement instrument of claim
11 wherein the plural independent detector regions comprise a
plurality of pixel regions on a unitary detector and the
polarization analyzer comprises a planar element of doubly
refractive material.
14. The fluorescence polarization measurement instrument of claim
10 further comprising calculation means where the fluorescence
polarization is calculated from the four readings comprising the
intensity of the two secondary beams under both the first and
second state of polarization of excitation light.
15. The fluorescence polarization measurement instrument of claim
10 which measures a plurality of sample regions simultaneously.
16. The fluorescence polarization measurement instrument of claim
15 where the plurality of sample regions comprise one of a linear
array of points and a two-dimensional array of points.
17. The fluorescence polarization measurement instrument of claim
14 which measures a plurality of sample regions simultaneously.
18. The fluorescence polarization measurement instrument of claim
17 wherein the switching means comprise a liquid crystal cell and
associated drive circuitry, and the plurality of independent
detector regions comprise a plurality of pixel regions on a unitary
detector and the polarization analyzer comprises a planar element
of doubly refractive material.
19. A fluorescence polarization measurement instrument comprising
at least one sample region for receiving a sample; excitation means
for providing plural beams of excitation light that are directed at
the sample region to effect fluorescent emission and that are
substantially linearly polarized along a first axis of polarization
at the sample region; detection means that comprise an objective,
plural independent detector regions, and a polarization analyzer;
wherein the principal ray of each of the plural beams of excitation
light is substantially parallel to the optical axis of the
objective.
20. The fluorescence polarization measurement instrument of claim
19 where the objective is telecentric at the sample region.
21. A method of measuring fluorescence polarization of a sample,
comprising illuminating the sample to effect fluorescence emission
with a beam of excitation light that is linearly polarized along a
first axis; measuring the intensities of a first component of the
fluorescence emission that is polarized along the first axis and a
second component of the fluorescence emission that is polarized
orthogonal to the first axis while the sample is illuminated with
the beam of excitation light that is linearly polarized along the
first axis; switching the state of polarization of the beam of
excitation light to a polarization state wherein said beam is
linearly polarized along a second axis substantially orthogonal to
the first axis; measuring the intensities of a third component of
fluorescence emission that is polarized along the first axis and a
fourth component that is polarized orthogonal to the first axis
while the sample is illuminated with the beam that is linearly
polarized along the second axis; and calculating the fluorescence
polarization of the sample based on the measurements of the
intensities of the first, second, third and fourth components.
22. The method of measuring fluorescence polarization of claim 21
wherein the measurement of the intensities of the orthogonal
components of fluorescence emission are made simultaneously.
23. The method of measuring fluorescence polarization 21, where the
fluorescence polarization is calculated using a self-calibrating
algorithm that compensates for one of variations in intensity in
the first beam, differing responsivity between measurements of the
two orthogonal components of fluorescence emission, and
photobleaching of the sample.
24. A method of measuring fluorescence of a sample, comprising
providing an excitation beam of light; reflecting the excitation
beam of light from a mirror onto the sample to effect fluorescence
emission wherein the mirror is interposed between the sample and an
objective which collects the fluorescence emission from the sample;
and directing the fluorescence emission collected by the objective
to a detection means for measurement.
25. The method of measuring fluorescence of claim 24 wherein the
area of the mirror is chosen so as not to significantly occlude the
aperture of the objective.
26. The method of measuring fluorescence of claim 24, further
including the steps of polarizing the excitation beam of light
along a first linear polarization axis, analyzing the state of
polarization of the fluorescence emission and determining the
degree of fluorescence polarization of the sample.
27. The method of measuring fluorescence of claim 24, where all
optical filters through which emission light travels in passing
from sample to detection means are oriented substantially normal to
the optical axis.
28. The method of claim 27, further including the steps of
analyzing the state of polarization of the fluorescence emission
and determining the degree of fluorescence polarization of the
sample.
29. The method of claim 28, wherein the sample exhibits a maximum
emission wavelength that is less than 20 nm from the maximum
excitation wavelength.
30. The method of claim 28, where the sample comprises a Bodipy
probe.
31. The method of claim 28 where the sample comprises an Alexa
probe.
32. The method of claim 28 further including the steps of
separating the emission light into the component which is polarized
along a first linear polarization axis and the component which is
polarized orthogonal to the first linear polarization axis, and
measuring both components simultaneously.
33. The method of claim 32 further including the steps of rotating
the plane of polarization of the excitation beam from the first
polarization axis to a second polarization axis substantially
perpendicular to the first axis, and measuring simultaneously both
the component which is polarized along the first linear
polarization axis and the component which is polarized orthogonal
to the first linear polarization axis while the excitation beam is
polarized along said second axis.
34. The method of claim 32 further including the step of
calculating the fluorescence polarization from the measurements of
the two polarized components of the emission light.
35. The method of claim 33 further including the step of
calculating the fluorescence polarization from the four
measurements comprising the intensity of the two polarized
components of the emission light when said excitation beam is
polarized along said first polarization axis, and when said
excitation beam is polarized along said second polarization
axis.
36. The method of claim 35 where the fluorescence polarization is
calculated using an algorithm that compensates for one of
variations in intensity in the first beam, differing responsivity
between measurements of the two components of emission light, and
photobleaching of the sample.
37. The method of claim 35 wherein the sample exhibits a separation
of 20 nm or less between the wavelength of peak absorption and the
wavelength of peak emission.
38. The method of claim 37 where the sample comprises a Bodipy
probe.
39. The method of claim 37 where the sample comprises an Alexa
probe.
40. A method of fluorescence measurement, comprising providing a
sample that is spatially homogeneous within the region to be
measured; providing a beam of excitation light that is directed
toward the sample; collecting emission light from the sample with
an objective; and recording an image of said sample on a detector
with multiple independent pixel regions arranged in a
two-dimensional array; wherein the size of the pixel regions is
chosen so the image of the intersection of the excitation beam and
the sample produced by the objective covers many pixels in the
detector; the spatial distribution of detector readings for pixels
within the image of the intersection of the excitation beam and the
sample is recorded; and making a determination of measurement
integrity based on the shape and uniformity of the spatial
distribution of detector readings, where abnormal distributions
denote an invalid measurement.
41. The method of claim 40 further including the steps of
polarizing the excitation light along a first linear polarization
axis, analyzing the state of polarization of the emission light and
determining the degree of fluorescence polarization.
42. The fluorescence measurement instrument of claim 1 wherein the
excitation means comprise a laser source.
43. The fluorescence measurement instrument of claim 4 wherein the
excitation means comprise a laser source.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates to equipment and methods for
assaying the amount of optical fluorescence, and the degree of
fluorescence polarization, in samples.
[0003] 2. Background and Description of the Related Art
[0004] Fluorescence Methods and Terminology.
[0005] Fluorescence involves exciting a molecular group with light
of a first wavelength, causing it subsequently to emit light of a
second, longer wavelength. The molecular group is termed a
fluorophore, and the first and second types of light are termed
"excitation" and "emission" light, respectively. Between excitation
and emission, the molecular group is said to be in an excited
state. Depending on the molecular group involved, the time spent in
the excited state can vary widely, from a few nanoseconds to
several microseconds. The duration of the excited state is termed
the fluorescence lifetime. It is common to chemically engineer a
fluorescent marker compound by grafting a fluorophore to a chemical
group that reacts only or primarily with a very specific target
molecule. The resultant fluorescent marker will bind only to very
specific targets, and has fluorescence properties of the
fluorophore.
[0006] Fluorescent markers are used to disclose the presence and/or
location of targets within a sample, which may contain a variety of
other compounds. For reliable detection, the other compounds in the
sample must exhibit a very low degree of fluorescence, or there
must be a way to discriminate between fluorescence emission
resulting from the target compound and that from other compounds in
the sample. Since the mechanism of fluorescence is present to at
least some degree in most compounds, discrimination means are
usually employed. Among the means are discrimination by wavelength
of excitation light, by wavelength of emission light, and by
fluorescence lifetime. Typically, discrimination by excitation
wavelength involves making measurements using excitation light that
has been filtered to contain light of only a selected wavelength
band. Similarly, measurement of emission light through a filter
that admits only a selected wavelength band provides a means to
discriminate by emission wavelength.
[0007] Fluorescence lifetime discrimination is performed by a
variety of methods, the simplest of which is to excite using a
modulated source, and to observe emission using synchronous
detection methods. For example, a target compound with a long
fluorescence lifetime may be detected by exciting the sample with
brief pulses, while measuring emission using a gated detector which
is insensitive for a controlled, brief interval after excitation.
Such a detector will not respond to compounds having a short
fluorescence lifetime, which will have ceased emission by the time
the detector becomes responsive. However, the target species,
having a long fluorescence lifetime, will continue to emit for
considerably longer, and the majority of its emission will be
detected. Alternative approaches can also be used with
single-element detectors, including detection with lock-in
amplifiers, quadrature detection, and other standard signal
analysis techniques. Imaging detection is possible with a gated
intensifier or microchannel plate (MCP), by electronic shuttering,
or by pixel shifting between photosensitive and non-photosensitive
regions of a CCD detector.
[0008] Discrimination by these means is useful for removing
so-called `background` fluorescence arising from optical
components, solvents, culture dishes, and the like, which are
necessary elements in a fluorescence experiment but whose
fluorescence signal is not seen as contributing meaningful
information. That is, these are means for removing unwanted
contaminant signals and for obtaining an enhanced signal-to-noise
ratio in the sample fluorescence measurement.
[0009] It is possible to use discrimination techniques not just to
remove background, but to learn more about the sample itself. For
example, two or more fluorescent markers can be used which have
distinct excitation or emission properties, so a single sample can
be tagged with multiple markers to identify different structures or
entities. The fluorescent signals are then resolved to obtain
information about each marker independently. This technique is
often termed multiprobe fluorescence.
[0010] Discrimination by excitation or emission wavelength may also
be used to learn additional information about a single target
species. For some fluorescent markers, the characteristic
wavelengths of optimal excitation or emission vary with the
chemical properties of the environment, such as the pH, salinity,
concentration of calcium, or the presence of other, very specific
molecules. Observing how fluorescence intensity varies with
excitation wavelength, or measuring the spectrum of emission light,
can provide a measurement of the chemical environment of the
fluorescent marker. These practices are termed fluorescence
excitation spectroscopy and fluorescence emission spectroscopy.
[0011] One case of special significance is fluorescence resonance
energy transfer (FRET), which employs two molecular groups having
carefully related properties. Typically, one group contains a
fluorophore that is characteristically excited at a first
wavelength and emits at a second wavelength. The other group is
characteristically excited at the second wavelength and emits at a
third wavelength. Depending on the presence, concentration, or
molecular conformation of the two groups, the likelihood of
interaction between the two groups is higher or lower. When the
likelihood of interaction is high, energy is transferred from
excited molecules in the first molecular group to the second
molecular group, resulting in emission light at the third
wavelength. Thus, sample emission at the third wavelength is
enhanced and emission at the second wavelength is depressed,
compared to the case when the likelihood of interaction is low.
Typically, a FRET experiment involves excitation at the first
wavelength while monitoring the level of emission at the second and
third wavelength bands. From the ratio of emission at these bands,
the level of interaction is inferred.
[0012] Additional information can be obtained in some cases by
analyzing the fluorescence polarization (FP), which involves
exciting the sample with linearly polarized light and measuring the
degree of linear polarization in the emission light. Excitation
light preferentially excites those molecules having a selected
geometrical orientation relative to the light's polarization
vector. Thus, the population of excited molecules is selectively
oriented, rather than randomly so, at the time of excitation. If
the fluorescence lifetime is comparable to or shorter than the
molecular reorientation time, then the molecules will also be
preferentially oriented when they emit, and the emission light will
be linearly polarized to some degree. By measuring the degree of
polarization, one infers the degree of preferential orientation at
time of emission. It is conventional to refer to the degree of
polarization (DOP) in terms of millipolarization units (MPU),
defined as
DOP=1000*(I.sub..parallel.-I.sub..perp.)/(I.sub..parallel.+I.sub..perp.)
[1]
[0013] where I.sub..parallel. and I.sub..perp. are the intensities
of fluorescence emission polarized in the same sense as the
excitation light and polarized orthogonal to it, respectively.
Related measures are also used to quantify fluorescence
polarization, based on substantially the same information.
[0014] Many factors can affect molecular re-orientation time,
including rotational viscosity, temperature, and whether the
fluorescent molecule is bound to another molecule or not.
Measurement of fluorescence polarization (FP) is often used as a
way to assess whether a molecule is in its bound or free state.
[0015] Equipment Used in Fluorescence Instruments
[0016] It is common to use mercury or xenon arc lamps with optical
filters for fluorescence excitation. These are very useful for
providing ultraviolet (UV) light, which some fluorophores require.
When only visible light is required for excitation, a less
expensive tungsten lamp can be used instead. Most often, lamp
sources are used when fine definition of the illumination region is
not required, or when a large region is to be illuminated. Laser
sources are sometimes employed to illuminate small, well-defined
regions, because of their higher specific radiance and more readily
controlled beam properties. For example, lasers are often used as
excitation sources in confocal equipment, and to create very high
flux densities in e.g. single-molecule detection experiments. They
are limited in that they emit a restricted, often discrete set of
wavelengths in contrast to lamps, which generally produce a
continuous spectrum that can be filtered to provide any desired
band within a range.
[0017] Detection of fluorescence in multiple-sample assay plates is
typically performed by optical scanning means, by sequential
measurements with single-element instruments that are mechanically
stepped across the individual sample regions, or by use of array
detectors with flood-light sources,. Examples of optical scanning
means include Kain, in U.S. Pat. Nos. 5,672,880 and 5,719,391,
where a galvo-driven scanning mirror directs excitation light
toward, and emission light from, one of a plurality of samples.
This allows a single lamp and a single detector to be used, and
minimizes the number of moving parts required.
[0018] In systems such as the Analyst from LJL Biosystems
(Sunnyvale, Calif.), a single-element reader is mechanically
stepped across the various samples in a sample plate. The Analyst
arrangement enables use of a specialized optical design where the
excitation and emission optics define a sample volume in an
approximately confocal arrangement, with different optical axes
angles. However, readout time and instrument cost are increased
because of the need for mechanical stepping.
[0019] These approaches use a single excitation source and a single
detector when measuring total fluorescence levels. Typical detector
types include avalanche photodiodes (APD), photomultiplier tubes
(PMT), or other devices capable of operating at low light levels.
This is important, as high sensitivity is vital to fluorescence
assay instruments. And since only a single detector is required,
the cost or complexity of that element can be increased somewhat
without great consequence.
[0020] In contrast, other systems use an array detector such as a
silicon charge-coupled device (CCD) or photodiode array to measure
fluorescence from many samples at once. This parallelism increases
the instrument throughput, and because modem CCD detectors have
low-noise readout circuitry, the detector does not impose a
significant penalty in terms of reduced sensitivity. The Arthur
system from E.G.&G. Wallac (Gaithersburg, Md.) utilizes a flood
source to illuminate many samples at once, and a CCD sensor to
measure the fluorescence emission.
[0021] Fluorescence Polarization Considerations
[0022] For FP analysis, additional components are employed. It is
necessary to polarize the excitation source, which is readily
achieved with conventional polarization optical elements for the
visible and UV, such as dichroic sheet polarizer, polarizing
beamsplitter cubes, and crystal polarizers such as Glan-Taylor or
Rochon prisms.
[0023] The polarization state of the emission light is analyzed
using one of several approaches. In one class of FP instruments,
emission light is polarized using a linear polarizer that is
rotated to two orthogonal settings while a detector is read, to
measure the components I.sub..parallel. and I.sub..perp.. Typically
a sheet dichroic type of polarizer is used, but use of any linear
polarizer would result in a similar overall function. This type of
instrument, referred to in the present application as a
sequential-measurement FP reader, has a number of drawbacks.
Because it measures the two states in time-sequence rather than
simultaneously, its accuracy is degraded by fluctuations in the
lamp or laser source. One can employ a reference detector to
monitor the source fluctuations, and numerically compensate for
variations by e.g. division, but this approach is not entirely
successful. Further, intrinsic changes in the sample during the
process of measurement, such as the effects of photobleaching,
cannot be corrected. Finally, mechanical moving parts are normally
used to select alternating polarization states, introducing
reliability concerns.
[0024] Another class of FP instruments uses a polarizing
beamsplitter (PBS). This separates the fluorescent emissions into
two distinct beams according to their polarization state, and these
beams are directed onto separate detectors to measure the
components I.sub..parallel. and I.sub..perp.. This type of
instrument is termed a simultaneous-measuremen- t FP reader in the
present application. It measures both states simultaneously, and so
does not suffer the problems of the sequential-measurement FP
reader just described. However, it has other limitations. A PBS is
a pair of right-triangle prisms with optical coatings on the
hypotenuse, at which face they are cemented or joined to form a
cube. This means that the two detectors must be physically distinct
parts, rather than being two segments of a multi-element detector,
because the images formed by the two beams are not coplanar. The
need for two detectors, two sets of readout circuitry, and
sometimes two lenses, means increased cost and complexity.
[0025] Both of the FP instrument designs described above read a
single sample at a time, and no known commercial FP instrument can
read a plurality of samples at once. This is a weakness for
applications such as clinical testing and high-throughput drug
screening, since single-sample systems inevitably have lower sample
throughput. While one can conceive of an instrument based upon an
array of detectors, polarizers, and sources, the construction of an
instrument with competitive price and performance to existing
single-sample instruments has not been achieved. This is a
significant limitation of the prior-art designs. A system described
in pending U.S. application "Fluorescence Imaging System" by Hoyt
and Miller could be used for detection of multiple samples within
an assay plate, and can select alternating polarization states
using liquid crystal elements means rather than mechanical means.
However, it is a sequential-measurement FP reader with all the
inherent problems of this approach, as explained above.
[0026] Another inherent limitation of all prior-art systems is
their need for calibration. There is generally some polarization
dependence in the transmission of lenses, in the reflection from
mirrors, and the like. So, the amount of light reaching the
detector(s) is altered by these elements, which are normally
present in a practical system. For a simultaneous-reading PBS-type
instrument, this is compounded by the fact that the two detectors
employed generally have somewhat different quantum efficiencies,
and are measured by different electronic circuits. Due to the
differences in the optical elements when transmitting the two types
of polarized light, and possible detector differences, the system
has different responsivity for measurement of I.sub..parallel. and
of I.sub..perp.. Since there is no way to assess the relative
proportions of I.sub..parallel. and I.sub..perp., this voids the
measurement of FP unless the system can be calibrated.
[0027] The factors producing disparate responsivity between
I.sub..parallel. and I.sub..perp. are not necessarily constant in
time, nor are they the same at all wavelengths. Consequently,
calibration must be undertaken separately for each wavelength band
of fluorescent emission, and must be repeated at intervals to
accommodate aging in components and circuitry. These problems are
most severe when two detectors are used, as in a simultaneous
measuring system.
[0028] In summary, the aforementioned art includes fluorescence
measurement instrumentation for optically scanning a plurality of
samples on a plate, or mechanically stepping a single-point system
to a plurality of samples, or for flood-illuminating a plate and
imaging a plurality of samples using a CCD detector. It provides
methods for measuring the degree of fluorescence polarization
through simultaneous-measurement or time-sequential measurement of
orthogonally polarized emission components, but all suffer from
significant limitations. All require calibration to obtain
high-accuracy readings. Those which employ simultaneous-measurement
of orthogonally polarized fluxes have higher cost and parts count,
while those employing sequentialmeasurements often require moving
parts; they are sensitive to fluctuations in the lamp or laser used
to excite fluorescence; and their accuracy can be compromised by
the inevitable photobleaching of the sample itself. Nor does any
prior art system enable measuring the fluorescence polarization of
many samples at once. Thus, no prior art system provides a
self-calibrated measurement of fluorescence polarization, with high
accuracy and the capability to measure many samples at once, for
reading multi-well plates, microscope slide samples, and similar
applications.
SUMMARY OF THE INVENTION
[0029] It is an object of the present invention to describe a means
for measuring fluorescence which can read many samples at once, in
parallel, for high throughput screening. A further object is to
enable reading plates, pipettes, tubes, microscope slides, and a
wide variety of formats with little or no change to the instrument
hardware. Another object is to provide for measurement of
multi-band fluorescence, time-resolved fluorescence, fluorescence
emission spectroscopy, and fluorescence excitation spectroscopy in
a single instrument. A further object is to provide enhanced
sensitivity measurements, to enable use of smaller sample sizes and
lower concentrations. Yet another object is to provide a method and
means for fluorescence polarization measurements which are
inherently accurate with no need for calibration, and which do not
suffer degraded accuracy despite fluctuations in the excitation
source. A final object is to provide means and methods for
improving the accuracy and sensitivity of fluorescence polarization
measurements made with existing fluorescence instrumentation.
[0030] Other objects and features of the present invention will
become apparent from the following detailed description considered
in conjunction with the accompanying drawings. It is to be
understood, however, that the drawings are designed solely for
purposes of illustration and not as a definition of the limits of
the invention, for which reference should be made to the appended
claims.
[0031] The various features of novelty which characterize the
invention are pointed out with particularity in the claims annexed
to and forming a part of the disclosure. For a better understanding
of the invention, its operating advantages, and specific objects
attained by its use, reference should be had to the drawing and
descriptive matter in which there are illustrated and described
preferred embodiments of the invention.
[0032] The present invention provides for a fluorescence
measurement instrument comprising excitation means, a plurality of
sample regions, and detection means; where the excitation means
produce a first beam and a diffractive optical beamsplitter element
that splits the first beam into plural secondary beams; the plural
secondary beams excite the plurality of sample regions
simultaneously to effect fluorescence; and the detection means
detect the fluorescence from the plurality of sample regions.
[0033] The present invention further provides for a fluorescence
measurement instrument comprising excitation means, a sample
region, and detection means; where the excitation means and the
detection means comprise an objective and a photodetector; where
light from the laser source is directed toward the sample region by
a mirror located between the sample region and the objective.
[0034] The present invention additionally provides for a
fluorescence polarization measurement instrument comprising
excitation means, at least one sample region, and detection means;
where the excitation means produce light that is substantially
linearly polarized along a first axis of polarization at the sample
region; where the detection means comprise an objective, a
photodetector, and a polarization analyzer; where the photodetector
provides a plurality of spatially distinct pixel regions; where the
objective directs a beam of fluorescent light from the sample
toward the polarizing beamsplitter; where the polarization analyzer
divides the beam of fluorescent light into two linearly polarized
secondary beams, one with polarization axis oriented substantially
parallel to the first axis of polarization and the other with
polarization axis oriented substantially perpendicular to the first
axis of polarization; and where the secondary beams of fluorescent
light are directed onto the spatially distinct pixel regions of the
photodetector by the polarization analyzer.
[0035] Another embodiment of the present invention is a
fluorescence polarization measurement instrument comprising
excitation means, at least one sample region, and a detection
means; where the excitation means produce light that is directed at
the sample region to effect fluorescent emission and that is
substantially linearly polarized along a first axis of polarization
at the sample region; where the detection means comprise an
objective, a plurality of independent detector regions, and a
polarization analyzer; where the plurality of independent detector
regions comprises one of a unitary detector with multiple pixel
regions and multiple detectors; where the objective collects the
fluorescent emission from the sample region and directs the
fluorescent emission in a beam toward the polarization analyzer;
where the polarization analyzer divides the beam of fluorescent
emission into two linearly polarized secondary beams, one with
polarization axis oriented substantially parallel to the first axis
of polarization and the other with polarization axis oriented
substantially perpendicular to the first axis of polarization;
where the linearly polarized secondary beams are directed by the
analyzer to separate detector regions; where said excitation means
further provide switching means for changing the state of
polarization of the excitation light at the sample region during a
single fluorescence polarization measurement from a first
orientation parallel to the first axis of polarization to a second
orientation parallel to a second axis of polarization which is
substantially perpendicular to the first axis of polarization.
[0036] Finally, the present invention provides for a method of
measuring fluorescence polarization, consisting of illuminating a
sample to effect fluorescence emission with a beam of excitation
light that is linearly polarized along a first axis measuring the
intensities of a first component of the fluorescence emission that
is polarized along the first axis and a second component of the
fluorescence emission that is polarized orthogonal to the first
axis while the sample is illuminated with the beam of excitation
light that is linearly polarized along the first axis; switching
the state of polarization of the beam of excitation light to be
linearly polarized along a second axis substantially orthogonal to
the first axis; measuring the intensities of a third component of
fluorescence emission that is polarized along the first axis and a
fourth component that is polarized orthogonal to the first axis
while the sample is illuminated with the beam that is linearly
polarized along the second axis; calculating the fluorescence
polarization based on the measurements of the intensities of the
first, second, third and fourth components.
BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In the drawings, wherein like reference numerals denote
similar elements throughout the several views:
[0038] FIG. 1 shows a side-view of the illumination and collection
optics for a fluorescence instrument according to the present
invention;
[0039] FIG. 2 shows how the optics and sample may be arranged to
reject specular reflections and thus improve the detection limit in
the fluorescence measurement;
[0040] FIG. 3 shows use of a diffractive optical element to produce
a selected number of beams with approximately equal intensity,
separated in angle of propagation, from a single input beam,
[0041] FIG. 4a shows in side-view a fluorescence instrument capable
of reading simultaneously several samples arranged in a line, as
viewed from along the axis defined by the samples;
[0042] FIG. 4b shows in side-view a fluorescence instrument capable
of reading simultaneously several samples arranged in a line, as
viewed transversely to the axis defined by the samples;
[0043] FIG. 5 shows in side-view a fluorescence instrument for
performing measurements of time-resolved fluorescence of several
samples, incorporating a modulated beam for excitation and a
detector with a photosensitivity that is time-gated according to
control means;
[0044] FIG. 6 illustrates schematically the timing relationship
between modulation of the excitation source and time-gating of the
detector photosensitivity;
[0045] FIG. 7 shows the use of a polarizer and polarization
modulator in optical series with a diffractive optical element, to
produce a selected number of beams with approximately equal
intensity, with a polarization state that is well-defined, which
may be adjusted for all beams together;
[0046] FIG. 8 shows a means for monitoring the polarization state
of one of the polarization-modulated beams;
[0047] FIG. 9 shows the use of a double-refractive calcite slab to
spatially separate the polarization components of a beam onto
distinct photosensitive regions of a detector;
[0048] FIG. 10a shows a fluorescence instrument taking a first
measurement as part of an assay of fluorescence polarization;
[0049] FIG. 10b shows a fluorescence instrument taking a second
measurement as part of an assay of fluorescence polarization;
[0050] FIG. 11 shows an instrument for measurement of fluorescence
polarization according to the present invention, using a prior-art
optical arrangement for illumination;
[0051] FIG. 12 shows an alternative detector arrangement for
measurement of fluorescence polarization according to the present
invention, utilizing two detectors and a polarizing beamsplitter
cube.
DETAILED DESCRIPTION OF THE PRESENTLY PREFERRED EMBODIMENTS
[0052] In this detailed description of the present inventive
fluorescence instrument certain terms are synonymous in meaning and
interchangeably used. The term waveplate and retarder are both used
to denote an optical retarder element having a selected optical
retardance. Wavelength band and wavelength range are both used to
denote a contiguous range of wavelengths, which typically spans a
few nanometers or more, but may be monochromatic in some cases such
as when discussing laser light or spectral line emission from
lamps. Fluorescence instrument, instrument, fluorescence reader,
and plate reader all refer to an instrument for quantifying the
amount, polarization, or time-evolution of fluorescent light from a
sample.
[0053] FIG. 1 depicts a fluorescence reader 10 in accordance with
the present invention. Light rays 11 reflect from a mirror 12 and
are directed onto a sample 16. Ideally, the light rays form a
relatively compact bundle, so the mirror is small and blocks a
negligible portion of the aperture of lens 13. Rays 17 of
fluorescent emission are imaged by a primary objective 13 to form
an image of the source at detector 15. A barrier filter 14 rejects
light having the wavelength of the excitation light, and transmits
light having selected wavelengths that are characteristic of the
emission light.
[0054] A number of aspects of this arrangement bear discussion.
First, the arrangement is well suited for multi-spectral imaging.
To perform fluorescence excitation or emission spectroscopy, only
the wavelength of excitation light 11 and the barrier filter 14
need to be varied. These are readily achieved by means of filter
wheels, gratings, or tunable filters such as liquid crystal tunable
filter (LCTF) or acousto-optic tunable filter (AOTF) devices. These
components provide high speed tuning (50 ms or less).
[0055] Second, there is no dichroic epi-illumination member, such
as are commonly used in fluorescence microscopes. The mirror 12 is
normally chosen to be reflective at a wide range of wavelengths
including the visible and near-UV range. So, there is no need to
interchange this element in order to adjust the excitation or
emission wavelengths. This eliminates a major barrier to
multi-spectral imaging that is present in many fluorescence
instruments.
[0056] A third benefit is that the excitation light does not pass
through objective 13. This eliminates the possibility of
fluorescence being generated within the objective, or of excitation
light being scattered from lens surfaces back toward the detector.
The instrumental sensitivity is therefore improved, relative to
fluorescence instruments where excitation light passes through the
objective on its way to the sample.
[0057] Another advantage is that by eliminating the dichroic an
improved measurement of fluorescence and of fluorescence
polarization is obtained. It is well-known that a dichroic element
used at non-normal incidence has a different spectral response for
light polarized in the plane of incidence (p-polarized) than for
light polarized orthogonal to the plane of incidence (s-polarized).
As the dichroic is not perfectly transmissive for both states (S-
and P-), there is some loss of emission light by reflection at the
dichroic, which reduces the total signal. And, the differential
spectral response between the S- and P- states means that the
amount of reflective loss varies with the state of polarization.
Simply put, the dichroic partially polarizes the emission light,
which distorts the reading of fluorescence polarization. These
effects are most pronounced when using samples for which the
emission wavelengths are not widely separated from the excitation
wavelengths, as this places the greatest requirements on the
dichroic element.
[0058] Fourth, the objective only needs to operate at the emission
wavelengths. Compared to an epi-illumination objective which must
also operate well at the excitation wavelengths, which are often
located in the ultraviolet region, this objective sees a more
restricted wavelength range and the lens design is simplified as a
result.
[0059] Another benefit of this arrangement is illustrated in FIG.
2. This shows excitation light 21 reflecting from mirror 22 and
passing to sample 24 which has a relatively flat specular surface.
Rays of excitation light 23 reflected from the sample 24 pass back
to mirror 22 and are reflected to a surface 28 which is out of the
field of view of objective 25. This traps the reflected light and
prevents it from contributing to the image formed at detector 27,
further reducing the amount of unwanted excitation light at that
component, and improving sensitivity. This method can be used
effectively with samples which are prepared in the form of slides,
multi-well plates illuminated from below, gel plates, and other
sample formats having relatively flat specular surfaces.
[0060] Excitation light 11 may be supplied by laser, a laser diode,
an LED, or a lamp with a compact arc. If a lamp is used, a bandpass
filter is used to select an excitation wavelength range. If a laser
or LED source is used, a filter may not be necessary, but a filter
or tuning means may be employed if the source is capable of
multi-wavelength operation.
[0061] When constructing instruments for use with multi-well
plates, the source is preferably a laser and the beam is not
focused to a point at or within the sample. Rather, in passing
through the sample, the beam defines a right cylinder with a height
as set by the sample thickness and a diameter set by that of the
beam. It is normally best not to illuminate the sample well walls,
since the optical properties at the walls often are not
representative of the bulk sample material. Not illuminating the
walls reduces or eliminates fluorescence components arising from
the plate material itself, from nonspecific binding, or from other
unwanted sources. This result is readily achieved by choosing a
suitable laser beam diameter and insuring that it is placed to
avoid illuminating the walls.
[0062] In many cases, no further optics are needed to realize this
goal. However, it may be desirable to incorporate a lens to produce
a converging beam, rather than a nominally collimated beam, with a
smaller beam diameter at the sample. It may also be desirable to
use spatial mode filters, collimators, optical fibers, coupling
optics, and all other optical means, either separately or together,
to produce an improved beam profile or to couple various sources
into the fluorescence instrument, as is well known in the art. The
need for these is dictated by the sample type, source properties,
and the like, according to the requirements of the particular
instrument being built.
[0063] The present invention can easily read a large number of
samples at once. FIG. 3 illustrates a multiple-beam source 30
consisting of an incident beam 31, a beam-dividing diffractive
optical element 32, and multiple output beams 33a, 33b, 33c, and
33d. The diffractive element 32 produces a selected number of
distinct beams according to its diffractive properties, and
commercially available devices are available from MEMS Optical Inc.
(Huntsville, Ala.) including the model 1011-488 for producing 16
output beams in a fan pattern, with a separation of approximately 2
degrees between the endmost beams. The output beams are of high
optical quality and have approximately equal energy in each beam.
In this way, distinct illumination spots of approximately equal
power are available for exciting a plurality of samples, using a
single illumination source.
[0064] FIG. 4a shows a side-view of an instrument that incorporates
a multiple-beam source. In this view, the various beams are
separated in the dimension extending into and out of the plane of
the drawing, and excite a plurality of samples. The design is
identical to that shown in FIG. 1, except that multiple beams,
samples, and detector elements are used. The transverse side-view
4b of the same instrument illustrates more clearly the relationship
between the beams, samples, and detector regions. Distinct beams
41a-d reflect from mirror 42 and excite samples 46a-46d. The
samples are arrayed in a line, with an inter-sample spacing equal
to the separation between beams. Mirror 42 has sufficient length to
span the array of samples, but it is small in the orthogonal
dimension, so its area is minimized and it does not greatly occlude
objective 43. Fluorescent emission 47a-d from these samples is
collected by objective 43 and forms images at detector elements
45a-d. Filter 44 rejects stray or scattered excitation light and
defines the wavelength band for the emission signal.
[0065] The instrument 40 has significantly higher throughput than a
single-sample fluorescence instrument, since it simultaneously
reads many samples. This performance increase is achieved with
little added complexity or cost: mirror 42 is extended in one
dimension, a diffractive optical element is added to create
multiple beams, and a pixelated detector is used to provide
independent detector elements. Optical performance is not degraded
relative to a single-sample system, although the lens must
adequately image multiple samples rather than a single sample. This
requirement is not difficult to meet using designs in the existing
optical art. Multiple objectives are not required. While the
excitation beam is divided among N samples, resulting in lower flux
density and lower fluorescence signals, many lamp and laser sources
provide sufficient flux that lowered excitation flux is not a
practical concern; indeed, it is often necessary to include
neutral-density filters in existing systems to avoid sample
overexposure and premature photobleaching.
[0066] The arrangement of excitation and emission optical paths in
the present invention yields an important benefit in
polarization-sensitive measurements. Beams of excitation light 41a
through 41d are nearly parallel to the optical axis of the
objective 43. If they derive from plural light sources, the
direction of propagation for each beam may be adjusted separately;
if they derive from a single light source and a beam division
element, the beams appear to fan out from the virtual image of the
division element as reflected by mirror 42. In either case, the
multiple beams can be made to have propagation vectors which lie
within a few degrees of the optical axis of objective 43. This
near-coincidence of optical axes for excitation and collection in
the present invention is termed coaxial illumination.
[0067] If the propagation of beams 41a-41d are parallel to the
optical axis of objective 43, and objective 43 is telecentric in
the optical path region containing samples 46a-46d and in the image
plane, then the axis of polarization of excitation light will not
be distorted by the imaging system. That is, orthogonal
polarization states in the excitation beams 41a-41d will appear
orthogonal in the image plane at detectors 45a-45d. Similarly,
rotating the plane of polarization by an angle d.theta. in any of
the beams 41a-41d will result in a rotation by the same angle
d.theta. at the detectors 45a-45d. This property enables
quantitative measurement of polarization states. Substantially the
same result is obtained if excitation beams 41a-41c lie within a
few degrees of the optical axis of objective 43. The apparatus of
this invention may be used for time-resolved fluorescence, as
illustrated in FIG. 5. Fluorescence instrument 50 incorporates a
beam 51 having an intensity that is modulated by element 151 in
response to control signal 152. Detector 55 incorporates adjustment
means 58 for altering its photosensitivity in response to control
input 153. These enable time-resolved measurements of fluorescence,
to resolve species with different excitation lifetimes. Objective
53 collects the emission light rays 57 from sample 60 and filer 54
transmits light in a selected wavelength range to the detector
55.
[0068] A simple type of time-resolved fluorescence is diagrammed in
FIG. 6. Timing diagram element 61 indicates the flux level in beam
51, as it is varied between a high flux level 63 and a low flux
level 64. Bursts of excitation light have a time-duration indicated
by 67. In concert with this, the detector photosensitivity 62 is
varied between a highly responsive state 65 and a non-responsive
state 66. The detector is non-responsive to light during the period
of high excitation flux. After a delay indicated by 69, it becomes
photosensitive for a time-duration indicated by 68, then becomes
non-responsive again. After a delay indicated by 70, the cycle is
repeated.
[0069] This discriminates between two populations of molecules, one
of which has a relatively long fluorescence lifetime and one of
which has a short fluorescence lifetime. The latter, after being
excited, quickly decay during the interval indicated as 69. When
the detector becomes highly responsive, this species is no longer
emitting. Thus the detector records only the species with a long
fluorescence lifetime during the interval 68. Such a scheme is
often used to selectively view fluorescent markers which have been
chemically designed to have a long fluorescence lifetime, while
selectively ignoring background fluorescence with a short
fluorescence lifetime.
[0070] More complex versions of time-resolved fluorescence may be
practiced using the arrangement of FIG. 5, such as phase-sensitive
detection and quadrature detection. These may be performed at
various frequencies to effect a dispersion measurement. The present
invention provides a means to perform time-resolved measurements on
many samples at once, by incorporating multiple beams and multiple
detectors as described above. The excitation beam may be modulated
prior to its passage through the diffractive element, so only a
single modulator is required.
[0071] If a laser or laser diode is used, pulsed or Q-switched
operation can be employed to provide a high depth of modulation at
little or no additional component cost. Or, a modulator such as a
Pockels cells or integrated optical modulator can be used with a
continuous (CW) laser. Any of the techniques known in the laser art
for intensity modulation may be employed, and depending on the
desired goal, one may wish to implement an on-off modulation, or
modulation that produces a prescribed intensity pattern such as
sine-wave modulation. For lamps, pulsed operation is usually
preferred because higher peak flux levels are achieved, but
modulation producing a sine-wave or another prescribed pattern
could also be employed.
[0072] The detector responsivity can be altered by a microchannel
plate or intensifier tube, which are well-known in the art as
providing means for gating a detector's photosensitivity in a few
nanoseconds. Alternatively, a detector may be used which has
integral electronic shuttering means to render it non-responsive
when this is desired. Certain types of charge-coupled device (CCD)
detectors provide this feature. Any technique which alters the
photosensitivity of a detector would be suitable, so long as it
provides the desired modulation.
[0073] FIG. 7 illustrates a system 80 for controlled modulation of
the polarization state of one or more beams. A linear polarizer 84
transmits only those components 89a-89d having a specified
polarization state, to a polarization modulator 85 which imparts a
selected polarization rotation or retardation to all beams 83a-83d.
This arrangement eliminates any polarization effects which may be
introduced to the incident beam 81 by diffractive beamsplitter 82,
that might cause the polarization state to differ from one beam to
another. Suitable linear polarizer elements include a sheet
dichroic polarizer such as HN-38S from Polaroid Corp. (Norwood,
Mass.), or a Glan-Taylor polarizer from Karl Lambrecht (Chicago,
Ill.). Polarization rotator elements include passive optical
components such as waveplates, or active electro-optical components
such as liquid crystal cells. Suitable waveplates include NRZ type
film from Nitto Denko (San Jose, Calif.). A waveplate may be
mechanically engaged and disengaged from the beam to effect
switching, or the waveplate may be rotated. It is well-known that
when a half-wave plate is placed in a linearly polarized beam with
its slow axis at an angle .phi. to the polarization axis of the
beam, the polarization axis of the beam is altered by 2.phi. in
passing through the half-wave plate. Liquid crystal cells include
variable retarder cells from Meadowlark Optics, preferably with the
slow axis of the liquid crystal cell oriented at 45 degrees to the
polarization axis of linear polarizer 84. Such a cell transforms
linearly polarized light beams 89a-d to the orthogonal polarization
state, or to left- or right-circular polarization, when it exhibits
retardance values of .lambda./2, .lambda./4, and 3.lambda./4. The
beam is unaltered at a cell retardance value of 0 or .lambda..
[0074] The polarization state of one or more beams may be monitored
using the polarization measuring arrangement 100 of FIG. 8.
Incident light 101 reflects from reflector 103 to produce a beam
171. This beam has polarization components 108a and 108b, which are
conventionally referred to as the S and P states of polarization.
Polarization analyzer 104 transmits a single polarization state of
light 109 in beam 105 to a photodetector 106 which is connected to
readout electronics 107.
[0075] While FIG. 8 indicates that polarization state 109 is the
linearly polarized S state, it could be the P state or any other
linear polarization state; it could also be any circular or
elliptically polarized state. Polarization analyzers 104 for
transmitting circular or elliptical states can be constructed using
methods of the prior art, such as by combining a linear polarizer
with an optical waveplate. The polarizer and waveplate types listed
above are suitable for this purpose. Suitable photodetectors
include photodiodes, photomultiplier tubes, or CCD detectors. These
components are all available from Hamamatsu Corporation (Middlesex,
N.J.).
[0076] Reflector 103 may be a partial reflector such as a
beamsplitter or an uncoated glass window, or it may be the
high-reflectivity mirror used to direct excitation light to the
sample. In the case of a partial reflector, beam 102 contains
significant flux and fluorescence measurements may be made while
the polarization measurement system 100 is operating. In the case
of a highly reflective mirror, there is little or no flux in beam
102. Instead, the polarization measurement system 100 operates in
alternation with actual measurements of fluorescence from samples.
Components 104 and 106 are located in the optical path normally
used for the sample, and directly view the excitation beam.
Polarization measurement system 100 may be mechanically engaged or
disengaged from the beam, or it may be permanently located in the
beam at a location downstream from the position 172 where the
sample is placed.
[0077] The system 100 is often useful in connection with the
controlled polarization modulation system 80 of FIG. 7. Modulation
system 80 may be used to adjust the polarization state while the
result of each adjustment is measured with polarization measuring
system 100. Polarization measuring system 100 acts as a sensor that
provides feedback to the polarization adjustment process.
[0078] Other aspects of the system 100 become important when
controlled adjustment of polarization state is sought. Typically,
the intensity of laser or lamp sources is not perfectly controlled.
Variations in lamp intensity render it difficult to maximize a
given polarization state by seeking a maximum in the signal from
photodetector 106. Rather, it is preferable to select polarization
analyzer 104 to transmit the polarization component 109 which is
orthogonal to the desired polarization state. By seeking a minimum
in the orthogonal signal 109, the desired component is maximized.
Since a value of zero is sought, modest fluctuations in the
incident beam 101 do not hinder the adjustment process.
[0079] If multiple beams are used in the fluorescence system, it is
preferable that the beams have essentially the same polarization
state,. Then, by monitoring the reading of photodetector 106 in any
given beam, the polarization component 109 is determined for all
beams at once.
[0080] In a multiple-beam arrangement, beamsplitter 103 may sample
more than one beam and direct the plural sample beams through
plural polarization analyzers to plural detectors, each of which is
connected to readout electronics. The polarization analyzer 104
need not be the same in each beam. Rather, the various polarization
analyzers 104 may be constructed to select various polarization
states 109. By seeking a minimum (maximum) intensity in a given
photodetector 106, the content of the associated polarization state
109 is minimized (maximized) for all beams. This process may be
repeated for the signals from various photodetectors 106, to
produce a sequence of polarization states that contain a minimum
(maximum) content of various polarization states 109.
[0081] FIG. 9 illustrates a polarization beam separator 110 which
is especially well-suited to the present invention. Light beam 111
encounters double-refractive element 112 having an extraordinary
axis 117 which is inclined at a non-zero acute angle to the beam.
In passing through element 112, beam 111 is separated into beams
113 and 114. Beam 114 contains polarization component 116 and
passes through element 112 without substantial deviation, while
beam 113 contains linear polarization component 115 which is
orthogonal to component 116 and is spatially deflected. The
propagation of light through double-refractive crystal elements is
described in standard optics texts such as Max Born and Emil Wolf,
"Principles of Optics" (Pergamon Press, New York, corrected 6th
Edition, 1980 and 1993).
[0082] The thickness, composition, and extraordinary axis
orientation of element 112 are selected to completely separate
beams 113 and 114. Typically, element 112 is constructed of calcite
because of its high value of n.sub.e/n.sub.o, but quartz, lithium
niobate, ammonium dihydrogen phosphate (ADP), potassium dihydrogen
phosphate (KDP), oriented liquid crystal polymers, or any other
double-refractive material may be used. Calcite elements may be
obtained from Karl Lambrecht (Chicago, Ill.).
[0083] Because the optical path length in the double-refractive
element is different for beams 113 and 114, these beams will form
images at different focal planes if polarization separator 110 is
placed in a convergent imaging system. This path length difference
can be compensated by a second double-refractive element 118 with
an extraordinary axis 119 that is substantially normal to the
propagation axis of beams 113 and 114. This second element 118 does
not significantly displace either beam, but introduces a
compensating path length so beams 113 and 114 will come to a focus
at the same plane. The composition, material, and orientation used
in element 118 can be readily calculated according to the prior art
for birefringent optics, to achieve the function of compensating
the optical path difference introduced by element 112.
[0084] A complete system 140 for fluorescence polarization (FP)
measurement is pictured in FIG. 10a. Excitation light beam 141
passes through a system for controlled modulation of polarization
142, emerges as beam 143 with linear polarization state 144,
reflects at mirror 145 with polarization state 147 and excites
sample 146. Fluorescent emission light ray 148 is collected by
objective 149 and passes through a filter 150 which selectively
transmits fluorescent light in a selected wavelength band and
rejects stray or scattered excitation light. Light ray 148
encounters double-refractive element 151 and is spatially separated
into components 152 and 153 with linear polarization states 156 and
157, respectively, which are sensed at photodetector elements 154a
and 154b oil photodetector array 154. The signal from detector 154a
indicates the intensity of light having the same polarization state
as the excitation beam, while the signal from detector 154b
indicates the intensity of light which has the orthogonal
polarization.
[0085] From the measurement depicted in FIG. 10a, a rough measure
of the fluorescence polarization (FP) may be had. If the signal
levels at detector elements 154a and 154b are termed A and B,
respectively, the degree of polarization (DOP) may be calculated
as:
DOP=1000*(A-B)/(A+B) [2]
[0086] The limitations of this approach are follows. It is
relatively straightforward, using the present invention, to produce
light having a high degree of polarization purity 147 at the
sample. It is within the existing optical art to produce an
objective 149 for the current invention which substantially
preserves linear polarization components in fluorescent light ray
148 without conversion into circular or elliptical polarization
states. So, the optical system 140 excites the sample with a
single, pure linear polarization state and properly resolves the
fluorescent emission into its orthogonal components.
[0087] However, objective 149, filter 150, and double-refractive
element 151 typically exhibit different transmissions when passing
light in polarization state 156 versus state 157. And, the response
of photodetector elements 154a and 154b are not identical. So, the
readings A and B do not accurately indicate the relative proportion
of each polarization component in the sample emission light. If the
responsivity of photodetector elements 154a and 154b differs by 1%,
the measurement of FP will be in error by 10 milli-polarization
units (MPU). Similarly, differential transmission of the two
polarization states of 1% would produce the same error of 10 MPU.
Many applications require accuracy of 1-2 MPU, which exceeds the
capability of present-day detectors and optics when FP is
determined by a single measurement of the type shown in FIG.
10a.
[0088] An improved determination of FP is obtained when the
measurement of 10a is combined with a second measurement indicated
in FIG. 10b, which differs from 10a in that excitation light 143
exhibits polarization states 144' and 147' which are orthogonal to
the states 144 and 147. Fluorescent emission 148' is characteristic
of the sample 146 under illumination by the orthogonal polarization
state. It is similarly resolved into distinct beams 152' and 153'
which are sensed at photodetector elements 154a and 154b. The
signal from detector 154a indicates the intensity of light having
the orthogonal polarization state to the excitation beam, while the
signal from detector 154b indicates the intensity of light with the
same polarization as the excitation beam.
[0089] If the intensity readings obtained at elements 154a and 154b
under the conditions of FIG. 10b are termed C and D, an improved
measure of the fluorescence polarization (FP) is given by:
DOP=1000*[(A-B+D-C)/(A+B+C+D)] [3]
[0090] The method of taking two measurements under conditions of
FIG. 10a and 10b, and reducing the results according to equation
[3], yields a perfect determination of FP when the intensity of
excitation beam 147 and 147' is equal for the two measurements.
This is readily shown. Suppose that the overall collection
efficiency of the optical system, incorporating losses in all
optical elements and responsivity in the photodetector, is
expressed as .alpha. for light having polarization state 156, and
.beta. for light having the orthogonal polarization state 157.
Signals A through D are then:
A=.alpha.I.sub.156=.alpha.I.sub..parallel. [4a]
B=.beta.I.sub.157=.beta.I.sub..perp. [4b]
C=.alpha.I.sub.156=.alpha.I.sub..perp. [4c]
D=.beta.I.sub.157=.beta.I.sub..parallel. [4d]
[0091] where I.sub..parallel. and I.sub..perp. refer to the
fluorescence emission in the same polarization state as the
excitation beam, and the orthogonal state, respectively.
[0092] When equation 3 is evaluated, the result is:
DOP=1000*(.alpha.I.sub..parallel.-.beta.I.sub..perp.+.beta.I.sub..parallel-
.-.alpha.I.sub..perp.)/(.alpha.I.sub..parallel.+.beta.I
.sub..perp.+.beta.I.sub..parallel.+.alpha.I.sub..perp.) [5a]
=1000*[(.alpha.+.beta.)*(I.sub..parallel.-I.sub..perp.)]/[(.alpha.+.beta.)-
*(I.sub..parallel.+I.sub..perp.)] [5b]
=1000*(I.sub..parallel.-I.sub..perp.)/(I.sub..parallel.+I.sub..perp.)
[5c]
[0093] exactly as defined by equation 1. There is no need to
compensate for, or calibrate, the optics or detectors. Readings of
FP obtained this way are inherently self-calibrating.
[0094] This felicitous result is only obtained when the excitation
beams in polarization states 147 and 147' have equal intensity.
This is difficult to achieve in practice, due to fluctuations in
the excitation source beam 141, to polarization-dependent losses in
the polarization modulator 142, mirror 145, and to
polarization-dependent losses in such other components as may be
present in various realizations of this invention. These lead to
variations between the intensity of excitation flux in the setting
shown as 10a and that shown as 10b. Further, the lamp fluctuations
are random, so cannot be calibrated out.
[0095] A better method for calculating FP from the same
measurements A, B, C, and D is the following:
DOP=1000*[A-B+.gamma.(D-C)]/[A+B+.gamma.(D+C)] [6a]
[0096] where
.gamma.=[(A*B)/(C*D)].sup.1/2. [6b]
[0097] Equation 6 yields a perfectly accurate measure of FP even
when the source intensity varies between the measurement of A and
B, and the measurement of C and D. The accuracy is obtained whether
the variation is systematic, as may arise from optical elements
142, 145, or others not pictured; or random, as may arise from lamp
fluctuations.
[0098] Denote the intensity of excitation light during the
measurement of A and B as I.sub.0, and that during the measurement
of C and D as kI.sub.0. Then, we may write equations analogous to
4a-4d that incorporate the difference in intensity:
A=.alpha.I.sub.156=.alpha.I.sub..parallel. [7a]
B=.beta.I.sub.157=.beta.I.sub..perp. [7b]
C=k.alpha.I.sub.156=k.alpha.I.sub.195 [7c]
D=k.beta.I.sub.157=k.beta.I.sub..parallel. [7d]
[0099] The veracity of equation is then apparent by direct
substitution. Substituting first into 6b,
.gamma.=[(A*B)/(C*D)].sup.1/2=[(.alpha.I.sub..parallel.*.beta.I.sub..perp.-
)/(k.alpha.I.sub..perp.*k.beta.I.sub..parallel.)].sup.1/2 [8a]
=[.alpha..beta.I.sub..parallel.I.sub..perp./k.sup.2.alpha..beta.I.sub..par-
allel.I.sub..perp.].sup.1/2=1/k [8b]
[0100] and then into 6a,
DOP=1000*[A-B+(D-C)/k]/[A+B+(D+C)/k] [9a]
=1000*[.alpha.I.sub..parallel.-.beta.I.sub..perp.+(k.beta.I.sub..parallel.-
-k.alpha.I.sub..perp.)/k]/[.alpha.I.sub..parallel.+.beta.I.sub..perp.+(k.b-
eta.I.sub..parallel.+k.alpha.I.sub..perp.)/k] [9b]
=1000*[(.alpha.+.beta.)*(I.sub..parallel.-I.perp.)]/[(.alpha.+.beta.)*(I.s-
ub..parallel.+I.sub..perp.)]=(I.sub..parallel.-I.sub..perp.)/(I.sub..paral-
lel.+I.sub..perp.) [9c]
[0101] which is in agreement with the definition of DOP given in
equation 1.
[0102] The present invention thus provides a means and method for
performing measurements of fluorescence and fluorescence
polarization which are inherently self-calibrating, by using two
measurements of the type indicated in FIG. 10a and 10b, and
calculating the FP using equations 6a and 6b. The results are not
affected by systematic or random variations in the intensity of the
excitation beam, nor by differences in the transmission of the
optical system for orthogonal polarization states of fluorescent
light, nor by different responsivities of the photodetectors 154a
and 154b used to detect the polarized components of fluorescent
emission.
[0103] In practicing the invention, it is important to ensure that
the polarization states of excitation light used in the two
measurements are indeed orthogonal. In one preferred embodiment, a
coaxial illumination system of the type drawn in FIG. 4b is used.
An argon-ion laser operating at 488 nm is the light source, with a
MEMS beamsplitter to produce 16 beams. A linear polarizer of HN-38S
removes any residual polarization that is not polarized with an
E-field in the plane of incidence with mirror 145. A liquid crystal
variable retarder from Cambridge Research & Instrumentation,
Inc. is used as an electrically-selectable zero-wave or half-wave
retarder, with its crystal axis at 45 degrees to the polarization
vector. The mirror is a first-surface aluminized mirror suspended
by fine metal supports at the proper angle, in front of the
objective. A 90 mm f/2.5 Tamron (Tokyo, Japan) macro lens is used
as the objective, at a 1:1 object: image reproduction ratio. The
filter is a longpass interference filter from Chroma Technology
(Brattleboro, Vt.), which transmits light with wavelengths above
515 nm. The photodetector is a Princeton Instruments MicroMax
cooled CCD camera with a Kodak KAF-1400 sensor chip. A 30 mm
plane-parallel slab of calcite with its extraordinary axis at 45
degrees to the surface normal is used as the polarization
separator. Objective focus is adjusted to lie halfway between the
optimum sample focus for the two polarization states. The calcite
slab is oriented to ensure coincidence of the polarization states
used for analyzing the fluorescence emission, and the polarization
states used for excitation. Proper orientation is achieved when the
plane which includes the extraordinary axis and the normal to the
slab, also includes the E field of one of the polarization states
used for excitation. Orientation may be checked by placing a
non-fluorescent, non-depolarizing target at the sample position,
illuminating it using one of the polarized excitation states, and
observing the signal levels at the two detector spots with the
fluorescence wavelength filter removed. When a minimum is attained
in one of the spots, the detector is properly oriented.
[0104] A polarization measuring system is used, consisting of an
HN-38 linear polarizer carefully oriented to select S polarization
only, located in front of a laser power meter Model 520 mm from
Thorlabs (Newton, N.J.). The drive voltage applied to the liquid
crystal was adjusted while the power meter output was noted. The
drive voltages corresponding to maximum and minimum power at the
meter were noted and used for the measurements of A and B, and C
and D, respectively.
[0105] One of the benefits of this preferred embodiment is that the
detector array consists of pixels in a CCD sensor, which can be
reconfigured under software control. The only required hardware
change in order to read a new arrangement of samples, is to the
beam division optics used to create the multiple excitation beams.
Often, only a single MEMS element need be engaged to change sample
formats.
[0106] Another benefit is obtained when the imaging detector has
significantly finer spatial resolution than the minimum amount
required to resolve individual spatial regions of the sample. This
runs counter to the desire for high signal-to-noise, which favors
the use of the minimum number of pixels, thus minimizing read-out
noise. However, modem CCD cameras readily achieve the shot-noise
limit, at which point photon statistics dominate and noise from the
detector and circuitry are secondary. It is useful to consider this
in some detail: a typical read-out noise is 8 electrons, and for a
back-thinned CCD such as the SPH-5 from Apogee Instruments (Tucson,
Ariz.) the quantum efficiency is in excess of 0.8, so the detector
noise is equivalent to 10 photons. Provided that the overall signal
level exceeds 100 photons per pixel, the measurement will be
shot-noise limited.
[0107] When determining the fluorescence or fluorescence
polarization, the group of pixels corresponding to a given sample
region are chosen based on the sample geometry and the illumination
pattern. All pixel readings within the region are then summed to
derive the total sample flux.
[0108] Use of spatial over-sampling, as this is termed, allows for
an assessment of measurement quality. In many cases, the sample
being measured is essentially homogeneous and the readings of
fluorescence should be relatively free of spatial structure except
for that imposed by the intensity profile of the excitation beam.
The beam profile can be made quite smooth by conventional means
such as spatial filtering to the gaussian spatial mode (0,0) as is
well known in the optical art. Once this is done, the image of the
sample spot should be relatively smooth as well. Presence of dark
or light regions within the sample indicate defects in sample
preparation such as particulates or bubbles. These can be tested
for using image processing techniques such as thresholding against
a known profile, and the like. If a sample has an anomalous
intensity pattern, it can be identified as suspect and that
information may be used to e.g. require a confirming test of that
sample element.
[0109] FIG. 11 pictures a system 160 consisting of an
epi-fluorescence microscope of the prior-art type, outfitted with
additional components to implement the present invention.
Specifically, a polarization modulator 162 is provided to
illuminate the sample in sequence with light 163 of two orthogonal
polarization states; and, the detector 174 includes a
double-refractive polarization separator to produce two distinct
images of the sample, according to the polarization state 176 and
177 of the beams 172 and 173 of fluorescence emission light 168.
Although elements such as the dichroic epi-illuminator 165 may have
significantly different transmission and reflection coefficients
for the two polarization components, a perfectly accurate
measurement of FP is obtained using equation 6 to analyze the
intensities A, B, C, and D at each point in the sample. The
orthogonal polarization states are preferably selected as the state
whose E-field lies in the plane of incidence of the excitation beam
with the epi-illuminator, and the state orthogonal to it. Filters
164 defines the wavelength bands used to excite the sample, and
filter 169 defines the wavelength band for emission light passing
to the detector 174. Objective 167 and projection lens 170 serve
their usual function, and one skilled in the art will appreciate
that various implementations are possible in these elements without
deviating from the spirit of the present invention.
[0110] FIG. 12 illustrates an alternative polarization beam
separator and detector arrangement. A polarization beamsplitter
(PBS) 191 of the conventional right-angle prism type is used to
separate the orthogonal components 196 and 197 of fluorescence
emission into two rays 192 and 193, and direct them to two distinct
detectors 194 and 198. This arrangement is used while the sample is
illuminated in light of orthogonal polarization states, to realize
the measurements of A and B, and C and D. From these, DOP is
calculated using equation 6.
[0111] It is explicitly intended that the optical arrangement may
be used for a variety of measurements including simple
fluorescence, time-resolved fluorescence, multi-band fluorescence,
FRET, and all prior-art methods of fluorescence assays. These will
enjoy the benefits of the improved optical system in terms of
enhanced sensitivity, high throughput, and easy multi-wavelength
operation. The techniques taught in this application can be used
individually and in combination, such as a time-resolved FRET
measurement or other joint modes of use. It is explicitly intended
that the teaching of the present invention be used in concert with
practices of the prior art, such as compensation for dark-readings,
background fluorescence, and the like.
[0112] The optical arrangement of the present invention forms the
preferred embodiment for FP measurements, by virtue of the coaxial
illumination, high sensitivity, and ease of performing measurements
on multiple samples at once.
[0113] Improved FP measurements are possible in a variety of
fluorescence instrumentation, based on the novel two-step
measurement and data analysis methods disclosed herein. These
measurements have been described in terms of the index of FP cited
as DOP in equation 1, but measures of interest other than DOP may
readily be calculated from this approach, and data analysis may be
performed which is functionally or algebraically equivalent to that
described herein without deviating from the teachings and spirit of
the present invention. Similarly, approximate measures may be used
if full accuracy is not required.
[0114] While particular means have been disclosed as preferred
embodiments of the functions which support this measurement scheme,
such as beam division, polarization measurement, and polarization
separation, prior-art means may also be used for these purposes.
Finally, while refinements have been taught which are beneficial in
many instances, such as the improvement of equation 3 to yield
equation 6, or the monitoring of beam polarization by a
polarization measurement system, these may be eliminated where no
benefit accrues from their use in a given application, or where
such benefit is not sought.
[0115] Thus, while there have been shown and described and pointed
out fundamental novel features of the invention as applied to
preferred embodiments thereof, it will be understood that various
omissions and substitutions and changes in the form and details of
the devices illustrated, and in their operation, may be made by
those skilled in the art without departing from the spirit of the
invention. For example, it is expressly intended that all
combinations of those elements and/or method steps which perform
substantially the same function in substantially the same way to
achieve the same or substantially the same results are within the
scope of the invention. Substitutions of elements from one
described embodiment to another are also fully intended and
contemplated. It is also to be understood that the drawings are not
necessarily drawn to scale but that they are merely conceptual in
nature. It is the intention, therefore, to be limited only as
indicated by the scope of the claims appended hereto.
[0116] The invention is not limited by the embodiments described
above which are presented as examples only but can be modified in
various ways within the scope of protection defined by the appended
patent claims.
* * * * *