WO2014059281A1

WO2014059281A1 - Fluorescence polarization assay for detecting high molecular weight molecules in biological fluids

Info

Publication number: WO2014059281A1
Application number: PCT/US2013/064550
Authority: WO
Inventors: Donald A. GIACHERIO; Robert D. BROOK; Jeremy D. HOFF; Alan J. Hunt
Original assignee: The Regents Of The University Of Michigan; Hunt, Karen
Priority date: 2012-10-11
Filing date: 2013-10-11
Publication date: 2014-04-17

Abstract

A point of care (POC) device measures the concentration of large molecular weight molecules or particles in a small volume of complex fluid using a fluorescence polarization assay where binding of a tagged probe molecule to a target molecule is detected as an increase in anisotropy of emitted fluorescent light. Autofluorescence is selectively reduced from the signal by time-gating the emission detection to select for emissions from long-lived fluorophores. This minimizes an otherwise intolerable discordance across population due to variable autofluorescence.

Description

FLUORESCENCE POLARIZATION ASSAY FOR DETECTING HIGH MOLECULAR WEIGHT MOLECULES IN BIOLOGICAL FLUIDS

CROSS-REFERENCE TO RELATED APPLICATIONS [0001 ] This application claims the benefit of United States Provisional

Application No. 61/712,416 filed on October 11, 2012, the full disclosure of which is hereby incorporated by reference.

INTRODUCTION

[0002] In fluorescence polarization assays, a probe molecule is tagged with a fluorophore, and binding of the tagged probe molecule to the target molecule is detected as an increase in anisotropy of emitted fluorescent light. The assay requires that the rotational diffusion time of the tagged probe molecule be comparable to the fluorescent lifetime of the fluorophore, which is typically only a few nanoseconds. Due to this consideration, suitable probes for fluorescence polarization assays have been limited to small molecules (i.e. those with rotational diffusion times of only a few nanoseconds or less). Expansion of the technique to allow the use of larger probe molecules would allow greater flexibility in the selection of both probe and target molecules and thus enable detection of molecules that are undetectable with standard fluorescence polarization assays.

[0003] Detection of binding of large probe molecules in complex biological fluids, however, requires qualitative advances in the approach. First, because of the relatively slower diffusion of large probes, the time between excitation and emission from the fluorophore must be substantially longer, requiring selection of appropriate long-lived fluorophores. However, long-lived fluorophores are substantially dimmer, which poses a signal to noise problem in complex biological fluids which typically exhibit substantial autofluorescence.

SUMMARY

[0004] This disadvantage is now overcome by selectively reducing autofluorescence from the signal by time-gating the emission detection to select for emissions from long-lived fluorophores. This minimizes an otherwise intolerable discordance across population due to variable autofluorescence. Furthermore, time- gating eliminates the need for a blank or a "sample only" control, which substantially simplifies the assay, making it more amenable to miniaturization into a simple clinical assay. Finally we eliminate discordance due to aggregation of molecules or particles in the sample by adding in certain surfactants during sample dilution. The tests are suitable where rapid results and/or small samples are desirable (e.g. ApoB, troponin C, slOOb).

[0005] We also describe a point of care (POC) device that measures the concentration of large molecular weight molecules (e.g. protein or protein complexes such as low density lipoprotein, troponin C or slOOP) particles in a small (<10uL) volume of complex fluid (e.g. blood or sputum). This device enables fast and accurate results for urgent or otherwise time-sensitive measurement of macromolecular markers. DRAWINGS

[0006] The drawings described herein are for illustrative purposes only of selected embodiments and not all possible implementations, and are not intended to limit the scope of the present disclosure.

[0007] Figure 1 is a schematic showing how polarized emission and depolarized emission arise from fluorescence by fluorophores linked to small and large molecules.

[0008] Figure 2 is a graph showing expected anisotropy from a fluorophore with 366ns lifetime and an anisotropy of 0.3 in the absence of rotation.

[0009] Figure 3 is a schematic drawing of a disposable cartridge which performs all necessary fluid handling steps involved in sample preparation, including filtration of large particulates (e.g. blood cells), mixing of analyte with probe, and loading mixed samples into a transparent analysis chamber accessible to the sensor optics.

[0010] Figure 4A is a diagram of a "T-type" set up to measure fluorescence anisotropy, where co- and cross-polarized emissions are measured simultaneously by two photodetectors .

[0011 ] Figure 5A is a diagram of an "L-type" set up to measure fluorescence anisotropy, where co- and cross-polarized emissions are measured serially using a motorized emission polarizer to sequentially route each polarity to a single photodetector.

[0012] Figures 4B and 5B are perspective drawings of a system as mocked up for incorporation into a hand held device. Numbered components are as follows: 1) LED light source; 2) collimating lens; 3) excitation filter; 4) polarizer; 5) focusing lens; 6) emission filter; 7) sample cartridge holder; 8) photodetector; 9) mirror; 10) polarizing beam- splitting cube.

[0013] Figure 6 illustrates how time gating removes autofluorescence background from the measurement of fluorescence anisotropy by collecting emission photons only after the autofluorescent signal has diminished.

DESCRIPTION

[0014] In one embodiment, the invention provides a method for detecting the presence and concentration of a target molecule in a fluid. The fluid is for example a biological fluid, and contains components that fluoresce with an autofluorescent signal characterized by an exponential decay time of T_auto. The method involves adding a composition comprising a fluorescent probe molecule to the fluid to make a test sample, where the test sample includes the probe molecule bound to the target molecule. The method then involves exciting fluorescence in the test sample using linearly polarized light from a pulsed excitation source. Fluorescent emission is detected from the excited sample and the anisotropy of the emitted fluorescence is then determined. Advantageously, the autofluorescent signal is removed by time-gating the fluorescent emission detection.

[0015] In the method, the target molecule is a macromolecule preferably having a molecular weight greater than 10 kDa. Further, the fluorescent probe molecule comprises a ligand conjugated to a fluorophore, where the ligand has a specific affinity for the macromolecule to be detected. Finally, the fluorophore conjugated to the ligand is characterized by having a fluorescent signal that has an exponential decay time that is at least five times longer than the exponential decay time T_auto of the autofluorescent signal.

[0016] In various embodiments, the fluorescent signal of the fluorophore has an exponential decay time that is at least 10 times longer, or at least 100 times longer than the half-life of the autofluorescent signal.

[0017] In various embodiments, the anisotropy of the emitted fluorescence is calculated from digitized intensity of fluorescence in a plane parallel to the plane of polarization of the excitation and from the intensity of the digitized florescence of the plane perpendicular to the plane of polarization of the excitation source. The method is conveniently implemented on a handheld device small enough to be used as a point of care (POC) device.

[0018] In another embodiment, a more specific method is provided for measuring the level of low density lipoprotein (LDL) in blood by measuring polarization anisotropy of a blood sample. The method involves combining the blood sample and a composition comprising a fluorescent probe molecule (e.g., a solution containing the fluorescent probe molecule) to make a test sample. The test sample is then excited with linearly polarized light. The method further involves measuring the fluorescent anisotropy of the test sample, by determining the intensities of fluorescent emission in and out of the plane of the polarized light used to excite the test sample. The level or concentration of the low density lipoprotein is then calculated from the measured anisotropy.

[0019] The test sample preferably contains a surfactant, and the probe molecule preferably comprises an antibody, a fragment thereof, or other molecule having specific affinity to ApoB conjugated to a fluorophore. In various embodiments, the surfactant is selected from albumin, non-ionic surfactants, and anionic surfactants.

[0020] As before, autofluorescence of the blood sample is removed by time- gating the fluorescent emission detection, and the fluorophore used in the probe molecule is characterized by a fluorescent half-life at least 5 times longer than the half-life of autofluorescence arising from non-LDL components of the blood sample. In preferred embodiments, the autofluorescent half-life is less than 10 nanoseconds and the fluorescent half-life of the fluorophore is greater than 50 nanoseconds, greater than 100 nanoseconds, or greater than 250 nanoseconds. In non-limiting examples, time-gating involves exciting the test sample with a pulsed excitation source and collecting fluorescence emitted from the sample beginning after a delay of at least ten nanoseconds, after delay of at least 40 nanoseconds or after delay of at least 50 nanoseconds following the pulse. The time delay in beginning collection of the fluorescent signal is at least equal to the response time of the photo detectors used, which is typically on the order of about 150 nanoseconds. To illustrate, in a particular embodiment, the delay from the pulse before collecting fluorescence is about 190 nanoseconds. Taking into account the 150 nanosecond response time delay of the detector, this involves a further delay of 40 nanoseconds before collecting a fluorescent signal from the target molecule. Once the photodetector for the fluorescent signal has been turned on in this way, time-gating involves collecting fluorescent emission for a suitable time period, such as for 100 to 500 nanoseconds.

[0021 ] In various embodiments, the blood sample is selected from whole blood and plasma and has a sample volume conveniently in the range of 1-10 microliters.

Principle of Operation

[0022] As noted, the methods are advantageously carried out on a handheld device for convenience. The methods are also adaptable to be used on standard laboratory fluorescence equipment.

[0023] The method makes use of a fluorescence polarization assay to detect the binding of a fluorescent probe molecule to target molecules in the form of large (>10kDa) particles. The fluorescent probe molecule is made of a ligand having specific affinity for the target molecule. Non-limiting examples of ligands include an antibody raised against the target molecule, including a monoclonal antibody or an antibody fragment. Further examples include aptamers engineered through in vitro selection or through systematic evolution of ligands by exponential enrichment to bind to the target molecule. Suitable aptamers include oligonucleic acid aptamers (RNA, DNA, or XNA) and peptide aptamers. Suitable compounds exhibit an association constant K_eq for a single epitope of the target molecule that is substantially higher, usually by many orders of magnitude, than for non-specific interactions or cross-reactions with other analyte constituents. In various embodiments, the association constant for a compound exhibiting a specific interaction is 10⁶ or higher.

[0024] The general principle on which this assay is based is described in detail elsewhere (Jameson and Sawyer, 1995). Briefly, the molecule with specific affinity is conjugated to a suitable long-lifetime fluorophore, for example by covalent attachment. The fluorescently labeled antibody (or other suitable molecule having specific affinity) is excited with linearly polarized light. The degree to which the resulting emitted fluorescence is polarized is related to the speed at which the fluorescent particle rotates, which is a function of the molecular weight - a larger complex rotates slower and a smaller complex rotates faster. Faster rotation is correlated to loss of anisotropy in the fluorescent signal, as discussed further below and shown diagrammatically in Figure 1. Figure 1 demonstrates how, when a fluorophore is linked to a large molecule, emitted light (i.e. fluorescence) remains polarized, and polarized emission is observed. On the other hand, when the fluorophore is linked to a small molecule, there is faster motion, with the result that the emission is depolarized. To illustrate, Figure 2 shows expected anisotropy from a fluorophore with 366 ns lifetime and an anisotropy of 0.3 in the absence of rotation.

[0025] Fluorescence anisotropy is described by the rotational correlation time Θ, which can be estimated by:

RT

where η = viscosity, M_r = molecular weight of particle, R = ideal gas constant, T = temperature, v = specific volume of particle, and h = hydration coefficient for the particle.

[0026] The degree of polarization of the emitted light can be described by the anisotropy of emission, r:

^parallel ^perpendickir

r =

J parallel + 21 perpendickir

Θ

where I_paraiiei = fluorescent emission intensity polarized parallel to the polarization of the exciting light, I_perpendicuiar = fluorescent emission intensity polarized perpendicular to the exciting light, r₀ = intrinsic anisotropy in the absence of rotation, and τ = fluorophore lifetime.

[0027] As the rotational correlation time is proportional to particle size (i.e. volume or, approximately, molecular weight), the anisotropy of light emitted from fluorescently labeled antibody (or other molecule with specific affinity for the target molecule) is significantly increased upon binding to its macromolecular antigen or target molecule (e.g. LDL, see Figure 2). The fraction of bound antibody f_b is related to the observed anisotropy r_0t,_s:

N_b r_ohs - r_f

f_b =

N_T r_b - r_f

where N_b = concentration number of bound antibody, Νχ = total concentration of antibody, r_f = anisotropy for unbound antibody, and ¾ = anisotropy for antibody bound to its antigen.

[0028] At equilibrium, the total antigen (i.e. target molecule) concentration can be expressed as 1

tigen]_Total + N_T )

K_eq (i - f_b )

where K_eq is the equilibrium constant for antibody binding to its antigen:

[complex] eq

[antibody] [antigen] eq where [complex] represents the concentration of the antigen + antibody complex.

[0029] In the derivation above, "antigen" is the target molecule to be determined and "antibody" stands for the molecule having specific affinity for the target molecule.

Compensating for discordance between individuals by time-gating fluorescent emission

[0030] Complex fluid samples (e.g. blood) may exhibit substantially varying levels of autofluorescence between individuals, leading to discordance in fluorescent measurement results. This effect is particularly relevant when using relatively dim fluorophores (suitable fluorophores tend to be dim, which is partly a necessary consequence of the decreased frequency of fluorophore excitation due to the long lifetime), where the autofluorescent background may be comparable to the target molecules' fluorescent emission.

[0031 ] The autofluorescent signal is removed by time-gating the fluorescent emission, as illustrated in Figure 6. Here, a pulsed excitation source is used. The excitation source is sampled, and each detected excitation pulse triggers a time-delayed pulse generator circuit that generates a gate pulse with a specified delay and pulse width, which is then ANDed with the photon counting detector. The appropriate delay and pulse width are determined by the lifetime of the dye used, and the lifetime of autofluorescent components to be removed. For example, blood autofluorescent components have a lifetime of approximately 3ns, while the fluorophore bis(2,2'-bipyridine)-4,4'- dicarboxybipyridine-ruthenium di(N-succinimidyl ester) bis(hexafluorophosphate) has a lifetime of approximately 350ns. A suitable delay and pulse width for the gate in this case would be approximately 10ns and 500ns, respectively. The appropriate delay and pulse are determined by the necessary signal to noise ratio, and depend upon the fluorescent lifetime and brightness of both the fluorescent label and the autofluorescent constituents. The delay and window are selected to increase or maximize the ratio of probe fluorescence to autofluorescence, where Φ is the

ratio of photons emitted to photons absorbed, t_\ is the timegate delay, t₂ is the timegate window, and τ is the fluorescent lifetime; while at the same time minimizing the loss of polarization by maximizing J e 'θ dt, where Θ is the rotational correlation time of the bound probe In practice, this maximization can be done experimentally or empirically.

Use of surfactants to prevent discordance and improve sensitivity.

[0032] Discordance in assay results between individuals can result from target molecule aggregation in the samples, substantially affecting the sensitivity and reproducibility of the assay. We have now identified surfactants that act to either 1) prevent aggregation or 2) in the case of LDL, delipidate the LDL particles, leaving monodisperse apoB particles, which are an analogous target for this assay. Examples of agents that reduce aggregation include albumin, anionic surfactants, and nonionic surfactants. A non-limiting example of an anionic surfactant is sodium dodecyl sulfate (SDS), and nonionic surfactants include alcohol ethoxylates, alkylphenol ethoxylates, and ethylene oxide/propylene oxide block copolymers. Examples include Pluronic® F108. Examples of agents that delipidate the LDL particles include nonionic surfactants such as Tween®-20 and Triton® XI 00. In both cases, the surfactants are presented at concentrations that are not expected to interfere with the core detection method (e.g. antibody binding or polarized fluorescence emission). Suitable levels of surfactants include 10 ppm and greater, 100 ppm and greater, 0.1% and greater, and 0.2% and greater. Upper limits of suitable ranges for surfactant concentration are 10%, 5%, 2%, and 1%, where all levels are by weight. Suitable values include 0.5 to 1% for Tween and Triton and 0.005 to 0.1% for Pluronic F108 or F127. Typical values are 0.5% for Tween- 20 and/or 0.05% for Pluronic F108. In embodiments, aggregation of antibodies against the target can also degrade sensitivity and reproducibility. To overcome this, the antibodies can be purified (e.g. chromatography column purification), and/or additives can be used that reduce the aggregation without affecting the core detection method. Examples of such additives include nonionic surfactants such as those discussed above. Specific examples include Tween-20 and Pluronic F127. [0033] It is expected that non-specific adsorption of antibodies and target molecules to the sample chamber surface will also degrade sensitivity and reproducibility of the assay. The above surfactants (e.g. Tween-20 and Pluronic F127) also improve assay performance by reducing such non-specific binding. Characteristics of suitable fluorophore

[0034] This assay makes use of a fluorophore with a lifetime comparable to the rotational correlation time of the sensing molecule (e.g. antibody or antibody fragment). An antibody has a rotational correlation time of approximately 125ns, while an antibody fragment has a rotational correlation time of approximately 50ns. Thus, a suitable fluorophore should have a lifetime ranging from approximately 50ns to 1μ8. As used herein, the lifetime is the exponential decay time, or the time needed for the fluorescence to decay to 1/e (about 37%) of the time zero value. Exponential decay time is commonly used to describe fluorescence lifetimes.

[0035] The fluorophore should also exhibit high intrinsic anisotropy. A high quantum yield is also desirable. Examples of suitable fluorophores:

• metal-ligand fluorophore Bis(2,2'-bipyridine)-4,4'-dicarboxybipyridine- ruthenium di(N-succinimidyl ester) bis(hexafluorophosphate)

(available from Sigma- Adrich, item number 96632)

• PURETEVIE 325 from AssayMetrics (assaymetrics.com) The fluorophore is conjugated to the ligand to make the probe molecule using known covalent reactions, wherein a functional group on the fluorophore reacts with a functional group on the ligand to form a suitable bond. For example, a fluorescent moiety containing a succinimidyl ester is conjugated by covalently binding to free amine groups on the probe molecule. Other common amine-reactive labeling chemistries include isothiocyanates, carboxylic esters, and sulfonyl chlorides.

Usage

[0036] The method can be carried out on conventional lab scale fluorescence equipment that contains or is connected to a computer or computing means to calculate the level of target molecule in a sample based on the measured fluorescence anisotropy. The method can also be miniaturized. In practice, a miniaturized system includes a portable analysis device and a small disposable cartridge. The analysis device houses an excitation light source and polarizer, one or two pairs of analyzers and photodetectors, and a microcomputer. The cartridge houses all necessary fluid handling components; for example a sample loading cell, a filter for separating blood plasma, an antibody loading cell, a microfluidic mixer, and a detection chamber.

[0037] Sample is loaded into the cartridge loading cell. Upon placing the cartridge into the analysis device, the analyte fluid is separated by the filter if necessary, mixed with a fluid stream containing a fluorescently tagged antibody, and injected into a detection chamber. Polarized excitation light illuminates the detection chamber, and the intensity of fluorescent emission is detected through an emission polarizer oriented either parallel or perpendicular to the orientation of the polarized excitation light. The emission

I parallel- 1 perpendicular

anisotropy is then calculated as r =— ^ , and the target antigen

Iparallel^~^l perpendicular

concentration is calculated based on the resultant anisotropy.

[0038] Figure 3 is a diagram of one embodiment of the method. A sample containing whole blood is placed in the loading chamber of a cartridge. Loading the chamber into the device releases a valve that allows the sample to flow through a filtration device to separate out red blood cells and draw the remaining plasma through a pathway into a mixing chamber, where it is combined with a buffer solution containing the fluorescent probe molecule (labeled as antibody storage in Figure 3). The resulting test sample is then drawn into the analysis chamber.

[0039] Figures 4A and 5A show the arrangement of functional apparatus needed to carry out the analyses and determinations of the method. Pulsed light from a source such as an LED or laser is polarized and filtered. The resulting polarized illumination excites fluorescence from the test sample. Rotational diffusion in the excited sample changes the polarization of the emitted light, which passes through polarizers to photodiode sensors. Since rotational diffusion of the probe in the test sample slows upon binding of the probe molecule to the target molecule, the differential of light intensity reaching photodiodes through perpendicular polarizers gives an indication of target molecule concentration. In Figure 4A, there are two polarizers and two photodiode detectors for the emitted fluorescence, in a so-called "T-type" configuration. In Figure 5A, there is only one photodiode detector, and the polarizer is flipped back and forth between in-plane and out-of-plane configuration. In both cases, the in-plane and out-of- plane signals are digitized and provided as input to a computer or other calculating means such as an on board chip or cpu. The concentration of the target molecule is then calculated, with the result being passed to a display such as an on board screen or an external computer. The calculated value for the concentration of the target value can be passed on to other devices or further manipulated before display.

[0040] Figures 4B and 5B show a schematic layout for prototype analysis device to be used for clinical validation. The photodetectors and high-gain pre- amplifiers are housed in electrically shielded boxes. The optically isolated photodetector output is read by a microcontroller which computes target molecule (e.g. VLDL-LDL) concentration and reports it through either a USB computer interface or an onboard digital display, in non-limiting fashion. In various embodiments, the detector is implemented either with a single emission photodetector and moveable emission polarizer (as in Figure 5B, the "L-format") or with two photodetector with two accompanying immovable emission polarizers (as in Figure 4B, the "T-format"). Expected dimensions are less than 6 inches across. Note that for clarity, light stops, electronic components and display are not shown. EXAMPLES

Example 1 - Sample protocol

[0041 ] In a sample protocol, a target molecule such as ApoB is measured in a test sample (analyte) by fluorescence polarization. A buffer solution is used, for example phosphate buffered saline, with optional additives to reduce non-specific molecular interactions and surface binding, such as Tween-20 or Pluronic F127

[0042] A probe is prepared by conjugating a long lived fluorophore to a ligand that has specific affinity for the test molecule. An example is anti-human apoB antibody covalently modified with a long-lived fluorophore such as Bis(2,2'-bipyridine)- 4,4'-dicarboxybipyridine-ruthenium di(N-succinimidyl ester) bis(hexafluorophosphate). [0043] The analyte is typically a biological fluid such as blood serum.

[0044] The test sample (analyte) and probe are diluted in sample buffer. A 160-fold dilution of blood serum and 35ug/mL probe (giving an [antibody]/[apoB] ratio of 0.5 to 2 for physiologic [apoB] levels) work well, but the appropriate analyte and probe concentrations will depend on a variety of factors specific to the device/assay implementation (incubation time, antibody affinity, additives, etc.) [0045] The diluted solution is mixed at room temperature by rocking for a standard period of time. We typically incubate 5 minutes, 15 minutes, or overnight, with or without concurrent mixing. Place solution in optical glass cuvette, and insert into fluorescence polarization detector device. [0046] Acquire co-polarized and cross-polarized emission. Polarizations are measured sequentially for several seconds each, and the measurements repeated until raw emission values converge suitably. We typically aim for a convergence in raw emission of 0.25% standard error. The fluorescent detection is time-gated relative to the excitation pulse to reduce background for autofluorescent blood components such as albumin. We typically use a 190ns delay and 1μ8 acquisition window. The delay includes the device response time (i.e. the electronic delay between photons hitting the detector and registering as detected by the electronics). If the response time for the current instrument is approximately 150ns, then the true timegate delay is approximately 190ns - 150ns = 40ns.

Example 2 - determination of apoB in blood serum. Brief description:

[0047] This example describes an implementation of the technique, specifically the measurement of apolipoprotein B (apoB) in human blood serum via fluorescence polarization using long fluorescence lifetime fluorophores conjugated to monoclonal anti-apoB antibodies. Fluorescent emission is time-gated to reduce autofluorescent background and increase measurement sensitivity.

Preparation of antibody probe:

[0048] Monoclonal anti-apoB antibodies are obtained from commercial vendors (e.g. Meridian Life Sciences, AbCam), or a custom antibody is raised using standard techniques. The antibody is transferred to labeling buffer, 25uM sodium tetraborate at pH 8.4, by gel filtration. The fluorescent probe, bis(2,2'-bipyridine)-4,4'~ dicarboxybipyridine-ruthenmm di(N~succimmidyl ester)

(Sigma, Item# 96632), is suspended in dimethyl sulfoxide at lOmg mL. The fluorescent probe solution is added to the antibody solution at 20-fold stoichiometric excess. The resulting solution is light-shielded and gently mixed by rocking for 4 hours at room temperature. The antibody is then transferred to PBS buffer, and unconjugated fluojOphore removed, by running the solution through a gel filtration column. The labeled antibody is further purified by size exclusion chromatography (e.g. on a GE LifeSciences Superdex 200 10/300 SEC column) to remove antibody aggregates generated during the labeling reaction. The monomeric antibody fractions from this purification are concentrated to lmg/mL and stored in PBS buffer with 50% glycerol at - 20°C.

Device parameters:

[0049] The fluorescent detection is time-gated relative to the excitation pulse to reduce background from autofluorescent blood components such as albumin. We typically aim for a delay of approximately 10 times the fluorescent lifetime of blood serum (~3.5ns), or ~40ns, and a 1 microsecond acquisition window. In the current configuration, the device response time (e.g. the electronic delay between photons hitting the detector and registering as detected by the electronics) is approximately 150ns, so to effect a 40ns delay, the device is set to a 190ns delay (i.e. device response time + desired timegate delay).

Measurement protocol:

[0050] The following steps are carried out:

1) Blood serum and fluorescently labeled probe antibody are diluted in

sample buffer (lOmM phosphate buffered saline, pH 7.4, containing 0.005% Pluronic F108 to reduce binding to cuvette surface). Serum is diluted 160-fold, and antibody is diluted to 35ug/mL (giving an

[antibody]/[apoB] ratio of approximately 0.5 to 2 for physiologic

[apoB] levels). Note that the appropriate analyte and probe

concentrations will depend on a variety of factors specific to the

device/assay implementation (incubation time, antibody affinity,

additives, etc.) 2) The solution is gently mixed on a rocker at room temperature for 15 minutes. The solution is transferred to a standard optical glass cuvette, and inserted into fluorescence polarization detector device.

3) Co- and cross- polarized emissions are acquired by a single photon

counting avalanche photodiode. In the L-format configuration,

polarizations are measured sequentially for several seconds each, and the measurements repeated until raw emission values converge

sufficiently for desired measurement certainty. For example, for a coefficient of variation (CV) less than 5% in anisotropy, a CV of less than approximately 0.4% in the co- and cross-polarized emission

intensities is required. The resulting CV is similar to the target CV of existing commercial [apoB] assays

Calculation of [apoB]

[0051 ] The fraction of bound anti-ApoB fb is related to the observed anisotropy r₀t,_s:

where N_b = concentration number of bound anti-ApoB, Νχ = total concentration of anti- ApoB, r_f = anisotropy for unbound anti-ApoB, and ¾ = anisotropy for anti-ApoB bound to LDL.

[0052] At equilibrium, the total ApoB concentration can be expressed as

where K_eq is the equilibrium constant for anti-ApoB binding to ApoB:

where [complex] represents the concentration of the ApoB + anti-ApoB complex. From this, [ApoB] can be calculated. As there is exactly one ApoB per LDL particle, [ApoB] = [LDL].

[0053] In practice, K_eq may not be known, and buffer conditions and analyte constituents may alter both the association constant of anti-apoB and the photophysical characteristics of the fluorophore, so an experimentally determined relationship between anisotropy r and [apoB] is useful. In the current configuration, we find this relationship to be approximately [apoB] = 4000r, in mg/dL.

Claims

CLAIMS What is claimed is:

1. A method for detecting the presence of a target molecule in a fluid, wherein the fluid contains components that fluoresce with an autofluorescent signal characterized by an exponential decay time of T_auto, the method comprising:

a) adding a composition comprising a fluorescent probe molecule to the fluid, to make a test sample comprising the probe molecule bound to the target molecule;

b) exciting fluorescence in the test sample with linearly polarized light from a pulsed excitation source;

c) detecting a fluorescent emission from the excited sample; and

d) determining anisotropy of the emitted fluorescence,

wherein the autofluorescent signal is removed by time-gating the fluorescent emission detection,

wherein the target molecule is a macromolecule having a molecular weight greater than 10 kDa,

wherein the fluorescent probe molecule comprises a ligand conjugated to a fluorophore, wherein the ligand has a specific affinity for the macromolecule, and

wherein the fluorophore is characterized by a fluorescent signal that has an exponential decay time that is at least 5 times longer than the exponential decay time of the autofluorescent signal.

2. A method according to claim 1, wherein the fluorescent signal of the fluorophore has an exponential decay time that is at least 10 times longer than the half- life of the autofluorescent signal.

3. A method according to claim 1, wherein the exponential decay time of the fluorescent signal of the fluorophore is at least 100 times longer than the half-life of the autofluorescent signal.

4. A method according to claim 1, wherein the anisotropy of the emitted fluorescence is calculated from the digitized intensity of fluorescence in a plane parallel to the plane of polarization of the excitation and from the intensity of the digitized fluorescence in a plane perpendicular to the plane of polarization of the excitation source.

5. A method according to claim 1, wherein the target molecule is a protein.

6. A method according to claim 6, wherein the target molecular is selected from the group of ApoB, troponin C, and si 00b.

7. A method according to claim 5, wherein the target molecule is high density lipoprotein or low density lipoprotein.

8. A method according to claim 1, wherein the test sample further comprises a surfactant to prevent aggregation of the target molecule.

9. A method according to claim 8, wherein the surfactant is selected from the group consisting of albumin, nonionic surfactants, and anionic surfactants.

10. A method according to claim 9, wherein the surfactant is sodium dodecylsulfate.

11. A method according to claim 1, wherein the composition added in step a) comprises a nonionic surfactant and an antibody to the target molecule.

12. A method according to claims 1-11, wherein the fluid is a biological fluid.

13. A method according to claim 12, wherein the biological fluid is blood or sputum.

14. A method for measuring the level of low density lipoprotein (LDL) in blood by measuring polarization anisotropy of a blood sample, the method comprising: combining the blood sample and a composition comprising a fluorescent probe molecule to make a test sample; exciting the test sample with linearly polarized light; measuring fluorescent anisotropy of the test sample, by determining the intensities of fluorescent emission in and out of the plane of the polarized light used to excite the test sample; and calculating the level of low density lipoprotein from the measured anisotropy, wherein the test sample comprises a surfactant, and wherein autofluorescence of the blood sample is removed by time gating the fluorescent emission detection, and wherein the probe molecule comprises an antibody to ApoB or fragment thereof conjugated to a fluorophore, wherein the fluorophore is characterized by a fluorescent half-life that is at least 5 times longer than the half-life of autofluorescence arising from non-LDL components of the blood sample.

15. A method according to claim 14, wherein the fluorescent probe molecule comprises an antibody to ApoB conjugated to a fluorophore.

16. A method according to claim 14, wherein the surfactant comprises albumin.

17. A method according to claim 14, wherein the surfactant comprises nonionic surfactant.

18. A method according to claim 14, wherein autofluorescent half-life is less than 10 ns and the fluorescent half-life of the fluorophore is greater than 50 ns.

19. A method according to claim 18, wherein autofluorescent half-life is less than 10 ns and the fluorescent half-life of the fluorophore is greater than 100 ns.

20. A method according to claim 18, wherein autofluorescent half-life is less than 10 ns and the fluorescent half-life of the fluorophore is greater than 250 ns.

21. A method according to claim 14, wherein time gating comprises exciting the test sample with a pulsed excitation source and collecting fluorescence emitted from the sample beginning after a delay of at least 10 ns following the pulse.

22. A method according to claim 21, wherein the delay following the pulse is at least 50 ns.

23. A method according to claim 21, wherein time gating comprises collecting fluorescent emission for 100 to 500 ns.

24. A method according to claim 14, wherein the blood sample has a volume of 1 to 10 μL.

25. A method according to any of claims 14-24, wherein the blood sample is whole blood.

26. A method according to any of claims 14-25, wherein the blood sampl comprises plasma.

27. A method according to any of claims 14-24, wherein the blood sampl comprises platelets.