EP2483666A1 - In vitro method of determining changes in a protein environment - Google Patents

Abstract

An in vitro method of determining changes in a protein environment. The method comprises selecting a protein of interest, placing a first sample of the protein in a spectroflurometer such that the protein sample may be irradiated with frequency modulated exciting light and the emitted light is detected; obtaining a reference set of light measurements at a selected frequency by measuring the change in phase angle (Φ) between emitted light and excited light and also simultaneously recording the change in signal modulation (m); applying time domain equations or frequency domain equations of the reference set of light measurements to obtain a reference data set; obtaining at least a second set of light measurements at the selected frequency for the protein; applying a time domain equations or frequency domain equations of the second set of light measurements to obtain a second data set; plotting S vs G for the reference data set and the second data set; and, determining whether the protein environment underwent changes between the reference measurement and second measurement by observing if a change in the position of the generated plot points. The spectroflurometer needs a light source which can cause a protein to fluoresce and a device for detecting and measuring the emitted light from the protein. The light source needs to be modulated in the frequency domain or time domain. Changes between the reference measurement and the second measurement are selected from the group consisting of: a) the same protein sample at a different time; and b) the protein in a different milieu. The changes in the protein environment is a conformational change of a single protein, changes in protein-protein interactions and/or changes in protein-excipient interactions.

Description

In vitro Method of Determining Changes in a Protein Environment

[01] This patent application claims priority pursuant to 35 U.S.C. § 1 19(e) to U. S. Provisional Patent Application Serial No. 61/248,295, filed October 2, 2009, incorporated entirely by reference.

Background

[02] Both time domain and frequency domain methods provide information on the nature of the excited state decay of a fluorophore. Data analysis of a fluorophore with a single lifetime component is straightforward. More sophisticated data analysis is required when the emission is heterogeneous, such as fitting to a specific decay model or using the model-independent Maximum Entropy Method. (J.C. Brochon, Maximum entropy method of data analysis in time- resolved spectroscopy. Methods Enzymol. 240 (1994) 262-31 1 ). Emission heterogeneity could be due to the presence of multiple fluorophores (each giving rise to different exponential decays), excited state processes (such as solvent relaxation or Forster Resonance Energy Transfer (FRET)) or non-exponential decays due to processes such as transient quenching. Models used to fit multiexponential decays are usually based on discrete exponential components or continuous distribution functions. (J.A. Ross, and D.M. Jameson, Time-resolved methods in biophysics. 8. Frequency domain fluorometry: applications to intrinsic protein fluorescence. Photochem. Photobiol. Sci. 7 (2008) 1301 -1312; B. Valeur, Molecular Fluorescence, Wiley-VCH, Weiheim, Germany, 2002; and, J. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, New York, 2006.)

[03] An alternative, model-less approach to fluorescence lifetime determinations was introduced in 1984 by one of the current inventors. (D.M. Jameson, E. Gratton, and R.D. Hall, The Measurement and Analysis of Heterogeneous Emissions by Multifrequency Phase and Modulation Fluorometry, Appl. Spectrosc. Rev. 20 (1984) 55-106.) A few years later, this approach was used to correct phase and modulation lifetime measurements for background fluorescence. (G.D. Reinhart, P. Marzola, D.M. Jameson, and E. Gratton, A method for on-line background subtraction in frequency domain fluorometry. J. Fluoresc. 1 (1991 ) 153-162.) This approach to fluorescence lifetime analysis was then largely dormant until recently when several laboratories resurrected this approach and applied it to microscopy, e.g., for studies on live cells, using Fluorescence Lifetime Imaging Microscopy (FLIM). (A.H. Clayton, Q.S. Hanley, D.J. Arndt-Jovin, V. Subramaniam, and T.M. Jovin, Dynamic fluorescence anisotropy imaging microscopy in the frequency domain (rFLIM). Biophys. J. 83 (2002) 1631-1649; A.H. A. Clayton, Q.S. Hanley, and P.J. Verveer, Graphical representation and multicomponent analysis of single- frequency fluorescence lifetime imaging microscopy data. J. Microsc. 213 (2004) 1-5; G.I. Redford, and R.M. Clegg, Polar plot representation for frequency-domain analysis of fluorescence lifetimes. J. Fluoresc. 15 (2005) 805-815; Q.S. Hanley, and A.H.A. Clayton, AB- plot assisted determination of fluorophore mixtures in a fluorescence lifetime microscope using spectra or quenchers. J. Microsc. 218 (2005) 62-67; A. Esposito, H.C. Gerritsen, T. Oggier, F. Lustenberger, and F.S. Wouters, Innovating lifetime microscopy: a compact and simple tool for life sciences, screening, and diagnostics. J. Biomed. Opt. 1 1 (2006) 34016-34024; M.A. Digman, V.R. Caiolfa, M. Zamai, and E. Gratton, The phasor approach to fluorescence lifetime imaging analysis. Biophys. J. 94 (2008) L14-16; and, A.H.A. Clayton, The polarized AB plot for the frequency-domain analysis and representation of fluorophore rotation and resonance energy homotransfer. J. Microsc. 232 (2008) 306-312.) These FLIM/phasor studies (also termed AB or polar plots) have largely been focused on FRET systems, although recently they have been applied to study and characterize cell autofluorescence.

BRIEF DESCRIPTION OF THE FIGURES

[04] Figure 1 shows a phasor plot of L-tryptophan at various pH values.

[05] Figure 2 shows the quenching of NATA and lysozyme using the quencher acrylamide.

[06] Figure 3 shows a phasor plot, at 70 MHz and 280 nm excitation, of NATA (circles) and lysozyme (open triangles) as a function of increasing temperature.

[07] Figure 4 A shows a phasor plot for dynamin 2 alone (circle), bound with GDP- (closed triangle), and GTPyS -bound (open triangle).

[08] Figure 4 B. Monomeric HSA (closed circle) plotted with furosemide bound HSA (open triangle) and D-thyroxine bound HSA (closed triangle). Dimeric HSA (open circle), which can be found up to 10% in lyophilized HSA, has a unique decay/phasor point compared to monomeric HSA.

[09] Figure 4 C. Phasor plot showing point movements due to protein-protein interaction.

[010] Figure 5 shows phasor plot of iron release from hTF N-lobe.

[011] Figure 6 shows phasor point trajectory of lysozyme denaturation as a function of increasing GuHCI (solid line) and urea (dashed line) concentration.

[012] Figure 7 shows a phasor plot of denaturation of GFP-SNAP25-BFP from OmM (gray star) to 2M (gray circle) with GHCI (Black line). The universal circle (gray line) is illustrated for reference. Number illustrate the GHCI concentration in mM at the respective point.

[013] Figure 8 shows a phasor plot of LHn/A and LHn/A-Dynorphin at 50MHz and the affects of addition of GuHCI to 50mM.

[014] Figure 9 show LHn/A, Dyn-LHn/A With Cryo-Protectant 5% PEG.

DESCRIPTION

[015] In Phasor Plot analysis the raw fluorescence lifetime data, either from the frequency domain or the time domain, can be reprocessed using a vectoral approach to produce a graphical 'linear' rather than 'exponential' representation. A semicircle is generated in the plot using the assumption that a single lifetime is in play Points generated within that semicircle occur as a result of multiple lifetimes due to the presence of multiple fluorophores or interactions between fluorophores or between fluorophores and their environment.. Any change in a specific protein's environment, say due to protein unfolding (denaturation), subunit oligomerization or subunit dissociation, fragmentation due to proteolysis or more subtle allosteric responses or isomeric relationships of protein molecules can be picked up during this type of analysis. In essence subtle movements in the conformation of the protein or changes in its interactions with other molecules induce changes in the immediate environment of the protein's fluorophores which can change the lifetime of their emission. The fluorescence measured can be from the intrinsic fluorescence of a protein, e.g., from the amino acids Tryptophan or Tyrosine, or from extrinsic fluorescence by labeling a protein with a fluorophore, for example amine or sulfhydryl reactive probes.

[016] The present application discloses an in vitro solution phase method that tracks conformational changes in a protein and/or changes to the protein milieu.

[017] In one embodiment, stability of proteins of interest over time periods may be observed.

[018] In other embodiments, changes in buffers, additives, pH conditions, or temperature changes may be observed.

[019] In other embodiments, mapping kinetic changes in proteins due to enzymatic turnover or oligomerization may be observed.

[020] In another embodiment, screening batch to batch protein preparations to rapidly validate the exclusion of potential protein refolding problems may be observed.

[021] The term "protein milieu" means the setting or surrounding of a protein, and includes, for example, pH, temperature, ionic strength, excipient, formulation, etc.

[022] The term "therapeutic protein" refers to a protein that are often extracted from animal cells or engineered in the laboratory for pharmaceutical use. The large number of therapeutic proteins are recombinant human proteins manufactured using non-human mammalian cell lines that are engineered to express certain human genetic sequences to produce specific proteins. Therapeutic proteins are used to relieve patients' suffering from many conditions, including various cancers (treated by monoclonal antibodies and interferons), heart attacks, strokes, cystic fibrosis and Gaucher's disease (treated by Enzymes and blood factors), diabetes (treated by insulin), anemeia (treated by erythropoietins), hemophilia (treated by blood clotting factors), and botulinum toxin for a variety of disorders.

[023] The term "industrial protein" refers to a protein from plant, animal or microbial origin that can be produced and/or processed on an industrial scale. Industrial proteins are increasingly used in food products because of their functional properties - both techno-functional properties (such as gelling and emulsifying behavior and taste) and bio-functional properties (such as nutritional value and physiological activity). The functionality is determined by (bio)chemical characteristics and is affected by the kind of protein, by the isolation procedure and by processing. Furthermore, the functionality may deliberately be altered by modification procedures. EXAMPLES

[024] Typically an ISS Chronos spectroflurometer (ISS Inc., Champaign IL) equipped with time resolved capabilities is used to acquire data (either frequency or time domain acquisition modes can be utilized). Sample in the μΜ-ηΜ concentration range is used in an 800μΙ quartz cuvette with a 1 cm optical path length. Frequency modulated exciting light from a 280nm LED at 280nm is passed through a 280-20 band pass filter (Melles Griot, Voisins Le Bretonneaux ) to the sample chamber. 300 nm LEDs as the light source may also be used. (B. Barbieri, E. Terpetschnig, and D.M. Jameson, Frequency-domain fluorescence spectroscopy using 280-nm and 300-nm light-emitting diodes: measurement of proteins and protein-related fluorophores. Anal. Biochem. 344 (2005) 298-300.) Emitted light is collected via a 320nm WG cut-on filter (Melles Griot, Voisins Le Bretonneaux ) and processed with ISS's Vinci software. The excitation was modulated at 50MHz and used to generate discrete and average lifetimes based on phase and modulation data supplied by the instrument. The phase and modulation data provides the parameters S and G as defined for the frequency domain by Equations (1 ) and (2) below, where φ is the phase angle and m is the modulation.

Equation (1 ) S = m s §

Equation (2) G = m cos φ

[025] the time domain is utilized wherein the decay of the fluorescence is recorded with time after excitation of the fluorescence with a brief pulse of light and the S and G functions are calculated from the time domain by Equations (3) and (4) below:

Equation (3)

Equation (4)

where ω can be chosen as the repetition frequency of the pulsed excitation source or another value which depends on the kinetics of the excited state process under investigation and l(t) is the observed fluorescence intensity at time t.

[026] The bandpass filter FF01-280/20-25 or FF01 -295/15-25 (Semrock, Rochester, NY) was used where appropriate with the excitation light and the emission collected through longpass filters (WG315 or UK330) or a 357/50 nm bandpass filter. Polarizers were set at magic angles to eliminate polarization effects. (G.D. Reinhart, P. Marzola, D.M. Jameson, and E. Gratton, A method for on-line background subtraction in frequency domain fluorometry. J. Fluoresc. 1 (1991 ) 153-162.) Reference lifetime standard of NATA at pH 7.5 (2.70 ns at 25°C and 2.95 at 20°C) was used at excitation wavelengths of 280 nm or 300 nm. (A.H. Clayton, Q.S. Hanley, D.J. Arndt-Jovin, V. Subramaniam, and T.M. Jovin, Dynamic fluorescence anisotropy imaging microscopy in the frequency domain (rFLIM). Biophys. J. 83 (2002) 1631-1649.) For those experiments in which temperature was varied, glycogen (0.00 ns) was used as a reference. Absorbance values at the exciting wavelengths were kept below 0.05 to avoid inner filter affects.

[027] For measurements involving acrylamide quenching, small aliquots of an 8 M stock solution were added to the sample cuvette to obtain the desired acrylamide concentration. For Kl quenching, NaCI was present to maintain the ionic strength between each measurement. The stock solution of Kl (3 M) contained a small amount (10^"4 M) of Na₂S₂0₃ to prevent the formation of l₃ ^", which absorbs in the excitation region of tryptophan fluorescence.

[028] Raw data are plotted to generate a phasor plot using a routine written in Matlab software.

[029] Example 1 . Tryptophan lifetime/phasor as a function of pH.

[030] The sensitivity of tryptophan fluorescence to a number of factors (pH, temperature, polarity of solvent/environment) makes it ideal to monitor alterations in the protein matrix. Such sensitivity has also made interpretation of fluorescence changes in response to specific biological events difficult at best, highlighting the complexity of this fluorescent probe. We examined the excited state properties of L-tryptophan under changing conditions to illustrate the application of the phasor method for analyzing the heterogeneous lifetime decay of proteins. The phasor plot for L-tryptophan as a function of pH (at 25 MHz) is shown in Figure 1 . As the pH is increased from 6.0 to 1 1 .0, the zwitterion form of tryptophan is converted to the anion form; each of these forms has its own unique lifetime and quantum yield values. (J.A. Ross, and D.M. Jameson, Time-resolved methods in biophysics. 8. Frequency domain fluorometry: applications to intrinsic protein fluorescence. Photochem. Photobiol. Sci. 7 (2008) 1301 -1312; A. White, Effect of pH on fluorescence of tryosine, tryptophan and related compounds. Biochem. J. 71 (1959) 217-220; W.B. De Lauder, and P. Wahl, pH dependence of the fluorescence decay of tryptophan. Biochemistry 9 (1970) 2750-2754' and D.M. Jameson, and G. Weber, Resolution of the pH-dependent heterogeneous fluorescence decay of tryptophan by phase and modulation measurements. The Journal of Physical Chemistry 85 (1981 ) 953-958.)

[031] Figure 1 shows a phasor plot of L-tryptophan at various pH values. The black points represent pH 6.0 (circle), pH 9.5 (open triangle), and pH 1 1 .0 (square) data at 25 MHz. The lifetime data were collected with 280 nm excitation at 25°C. At pH 9.5 there should be a -1 : 1 anion to zwitterion molecular ratio (based on the pK_a values of tryptophan), however, one notes that the phasor point corresponding to pH 9.5 is not midway between the high and low pH points. The exact distance of the phasor point along the line joining the starting and ending points on the universal circle depends not only on the relative concentration but also the quantum yields of the species in question (in the case of L-tryptophan, the ratio of the lifetimes and quantum yields of the anion to zwitterion forms is ~3). (D.M. Jameson, and G. Weber, Resolution of the pH-dependent heterogeneous fluorescence decay of tryptophan by phase and modulation measurements. The Journal of Physical Chemistry 85 (1981 ) 953-958.) The frequencies utilized will also weight the fractional contributions of the components differently, i.e., the lower frequency phasor points will weight the longer lifetime component while higher frequencies will favor the shorter component. (R.D. Spencer, and G. Weber, Measurement of subnanosecond fluorescence lifetimes with a cross-correlation phase fluorometer. Ann. N. Y. Acad. Sci. 158 (1969) 361-376.)

[032] Example 2 - Quenching studies.

[033] Quenching of a protein's intrinsic tryptophan fluorescence, with molecules such as , N0₃ ^", Cs^", acrylamide and molecular oxygen (S.S. Lehrer, Solute perturbation of protein fluorescence. The quenching of the tryptophyl fluorescence of model compounds and of lysozyme by iodide ion is commonly used to probe the dynamics of protein matrices. (Biochemistry 10 (1971 ) 3254-3263; M.R. Eftink, and C.A. Ghiron, Dynamics of a protein matrix revealed by fluorescence quenching. Proc. Natl. Acad. Sci. USA 72 (1975) 3290-3294; and J.R. Lakowicz, and G. Weber, Quenching of fluorescence by oxygen. A probe for structural fluctuations in macromolecules. Biochemistry 12 (1973) 4161-4170.) Chemical quenchers may provide information on the exposure of a tryptophan residue to the solvent as a consequence of collision of the excited fluorophore with the quencher molecule (dynamic quenching) or, in some cases, by formation of a ground state dark complex (static quenching) (B. Valeur, Molecular Fluorescence, Wiley-VCH, Weiheim, Germany, 2002). Both steady-state and time-resolved methodologies can be used to derive quenching information.

[034] In Figure 2, we examined the effects of quenching molecules on the phasor plot of tryptophan containing samples. NATA was chosen as a model system as it has a single exponential decay under our conditions and the quenching with acrylamide is expected to move the phasor towards shorter lifetimes along the universal circle. The protein lysozyme was used as the photophysics of the tryptophans in this protein have been well investigated. (T. Imoto, L.S. Forster, J.A. Rupley, and F. Tanaka, Fluorescence of lysozyme: emissions from tryptophan residues 62 and 108 and energy migration. Proc. Natl. Acad. Sci. USA 69 (1972) 1 151 -1 155; and S.S. Lehrer, and G.D. Fasman, Fluorescence of lysozyme and lysozyme substrate complexes. Separation of tryptophan contributions by fluorescence difference methods. J. Biol. Chem. 242 (1967) 4644-4651 .)

[035] Figure 2 shows the quenching of NATA and lysozyme using the quencher acrylamide. NATA (circles) and lysozyme (open triangles) phasor points at 89 MHz, with 300 nm excitation at 20°C, with the addition of various concentrations of acrylamide (0 - 0.2 M with NATA and 0- 0.5M with lysozyme). The data was recorded at 89 MHz and 20°C using a 300 nm LED as the excitation source; emission was observed through a WG315 longpass filter. As expected, the position of the phasor points is dependent on the quencher concentration. The lifetime of NATA is single exponential at each quencher concentration and follows a clockwise trajectory along the universal circle towards G=1.0; S=0 (i.e. 0 ns). With lysozyme, one notes that the phasor points are all within the universal circle indicating the heterogeneous nature of the lifetime data. Addition of acrylamide results in shorter lifetimes, indicating one or more of the tryptophan residues in lysozyme are sensitive to dynamic quenching, and the subsequent shift in the phasor points in a clockwise direction. Addition of the quencher iodide (data not shown) also shifted the phasor point, following a similar trajectory to shorter lifetimes, albeit to a lesser extent, as with the acrylamide quenching.

[036] Example 3 - Temperature effects.

[037] Temperature affects the phasor points for NATA and lysozyme were examined. As shown in Figure 3, the phasor points corresponding to NATA as a function of temperature remain on the universal circle and move clockwise as the temperature increases, indicating decreased lifetimes. The phasor points for lysozyme also progress clockwise with increasing temperature, however, these points begin to migrate more toward the universal circle with increasing temperature indicating a decrease in the lifetime heterogeneity.

[038] Example 4 - Protein/liqand and protein/protein interactions

[039] The sensitivity of tryptophan fluorescence to its environment makes it an ideal probe to monitor local and global structural movements and to provide insights into protein conformational changes associated with ligand binding. Using two protein systems, we investigated how phasor plots (derived from the intrinsic protein fluorescence) change during interactions with different ligands. Dynamin 2 is a 100 kDa GTPase associated with the pinching off of the plasma membrane during vesicle fission. (S.D. Conner, and S.L. Schmid, Regulated portals of entry into the cell. Nature 422 (2003) 37-44.)

[040] Dynamin 2 comprises 5 domains (an N-terminal GTPase domain, a middle domain, a pleckstrin homology (PH) domain, a GTPase Effector Region (GED) and a proline/arginine rich domain (PRD)) and contains 5 tryptophan residues, 4 of which are in the PH domain, while 1 tryptophan is in the C-terminal PRD domain. In Figure 4A, the phasor points of dynamin alone, GDP bound dynamin and GTPyS (a slowly hydrolyzable GTP analogue) bound dynamin are presented. The data shown were collected with 300 nm excitation at 84 MHz and 20°C. The addition of the guanine nucleotides clearly causes the phasor point to shift, and the position of the phasor point depends on the precise nucleotide utilized. These results indicate that there is a conformational change around one or more tryptophan residue(s) associated with guanine nucleotide binding and moreover the GDP and GTPyS ligands elicit different responses indicating that the conformation of the protein matrix is not identical in the two cases.

[041] HSA, which contains a single tryptophan residue (Trp 214), has been the subject of numerous fluorescence studies, including time-resolved studies, see for example (S. Kasai, T. Horie, T. Mizuma, and S. Awazu, Fluorescence energy transfer study of the relationship between the lone tryptophan residue and drug binding sites in human serum albumin. Journal of Pharmaceutical Sciences 76 (1987) 387-392; G. Hazan, E. Haas, and I .Z. Steinberg, The fluorescence decay of human serum albumin and its subfractions. Biochim. Biophys. Acta 434 (1976) 144-153; P. Marzola, and E. Gratton, Hydration and protein dynamics: frequency domain fluorescence spectroscopy of proteins in reverse micelles. The Journal of Physical Chemistry 95 (1991 ) 9488-9495; J.R. Lakowicz, and I . Gryczynski, Tryptophan fluorescence intensity and anisotropy decays of human serum albumin resulting from one-photon and two-photon excitation. Biophys. Chem. 45 (1992) 1 -6; D.M. Davis, D. McLoskey, D.J. Birch, P.R. Gellert, R.S. Kittlety, and R.M. Swart, The fluorescence and circular dichroism of proteins in reverse micelles: application to the photophysics of human serum albumin and N-acetyl-L- tryptophanamide. Biophys. Chem. 60 (1996) 63-77; and, M.K. Helms, C.E. Petersen, N.V. Bhagavan, and D.M. Jameson, Time-resolved fluorescence studies on site-directed mutants of human serum albumin. FEBS Lett. 408 (1997) 67-70.)

[042] HSA is commonly targeted for drug uptake studies and often these drug interactions are studied via changes in tryptophan fluorescence. In Figure 4B, we show the phasor point, collected using 300 nm excitation at 84 MHz and 25°C, of the intrinsic fluorescence from monomeric HSA. In the presence of of furosemide or D-Thyroxine, two drugs known to bind to HSA and to quench the tryptophan fluorescence, the phasor point shifts clockwise, indicating a shortening of the average lifetime. (J.R. Voelker, D.M. Jameson, and D.C. Brater, In vitro evidence that urine composition affects the fraction of active furosemide in the nephrotic syndrome. J. Pharmacol. Exp. Ther. 250 (1989) 772-778; and C.E. Petersen, C.E. Ha, D.M. Jameson, and N.V. Bhagavan, Mutations in a specific human serum albumin thyroxine binding site define the structural basis of familial dysalbuminemic hyperthyroxinemia. J. Biol. Chem. 271 (1996) 191 10-191 17.)

[043] These shifts in the phasor point indicate that drug binding induces changes in the protein matrix near the tryptophan residue, and also that the conformational state of the protein differs depending on the nature of the drug, likely due to different binding sites (thyroxine has been shown to have multiple sites) on HSA. We also note that the phasor point corresponding to dimeric HSA, isolated form the monomeric protein using size-exclusion chromatography, differs from the monomeric phasor point, shown in Figure 4B.

[044] One may also expect, from the previous sets of experiments, that protein-protein interactions could also be investigated using intrinsic tryptophan fluorescence and phasor plot analysis. For two interacting proteins with distinct average lifetimes, complex formation may be expected to change one or both protein conformations and result in a phasor point that deviates from the "normal" linear combination typically seen for non-interacting mixtures.

[045] Figure 4C shows a set of data (acquired with 280 nm excitation at 43 MHz and 20°C) for thrombin, anti-thrombin, lysozyme, and mixtures therein. Antithrombin and lysozyme are not predicted to interact and therefore a solution containing the two proteins should produce a phasor point that falls directly on a line between their individual points. This outcome is clearly observed (Figure 4C) for the phasor plot of a 1 : 1 (1 μΜ) mixture of the two proteins. On the other hand, thrombin/antithrombin is known to form a tight complex (D. Beeler, R. Rosenberg, and R. Jordan, Fractionation of low molecular weight heparin species and their interaction with antithrombin. J. Biol. Chem. 254 (1979) 2902-2913) and may be expected to produce a distinct phasor point away from the linear combination.

[046] The phasor point of the thrombin/antithrombin (1 : 1 at 1 μΜ for each protein) indeed shifts inward away from the line connecting the points corresponding to the two pure proteins, indicating a change in intrinsic fluorescence upon protein interaction. This demonstration shows how phasor plots can provide a facile indication of protein interaction.

[047] Figure 4 A shows a phasor plot for dynamin 2 alone (circle), bound with GDP- (closed triangle), and GTPyS-bound (open triangle). Figure 4 B shows monomeric HSA (closed circle) plotted with furosemide bound HSA (open triangle) and D-thyroxine bound HSA (closed triangle). Dimeric HSA (open circle), which can be found up to 10% in lyophilized HSA, has a unique decay/phasor point compared to monomeric HSA. Figure 4 C shows a phasor plot showing point movements due to protein-protein interaction. Solid lines between antithrombin (circle), lysozyme (closed square) and thrombin (closed triangle) are the projected linear movement of phasor point for non-interacting species. Mixtures of antithrombin/lysozyme (both at 1 μΜ, open square) and antithrombin/thrombin (gray triangle 0.5:1 , open triangle 1 : 1 ) are shown on the plot. The dashed line represents the projected phasor movement for increasing concentrations of antithrombin in the presence of thrombin. Cross-hairs seen in the upper part of each figure represent the statistical error for each phasor point under the expanded phasor plot scale.

[048] Example 5 - Kinetics of Protein-Liqand binding/dissociation

[049] Determination of the rate constants associated with protein-ligand binding/dissociation reactions is a major goal in many studies. Kinetics of such reactions often proceed on a rapid timescale (seconds to a few minutes). Steady-state fluorescence methodologies, such as stopped-flow fluorescence, are frequently preferred over lifetime methods because traditional frequency domain lifetime measurements usually require one or two minutes for reasonable precision, although specialized fast scanning methods are available. (O. Bilsel, L. Yang, J.A. Zitzewitz, J.M. Beechem, and C.R. Matthews, Time-Resolved Fluorescence Anisotropy Study of the Refolding Reaction of the a-Subunit of Tryptophan Synthase Reveals Nonmonotonic Behavior of the Rotational Correlation Time. Biochemistry 38 (1999) 4177-4187; B.A. Feddersen, D.W. Piston, and E. Gratton, Digital parallel acquisition in frequency domain fluorimetry. Rev. Sci. Instrum. 60 (1989) 2929-2936.)

[050] The phasor plot method, recorded at a single frequency, is, however, well-suited for rapidly tracking changes in the phase and modulation data. To illustrate this point and validate the phasor method for tracking kinetics of protein-ligand dissociation, the phase and modulation of the intrinsic fluorescence of human serum transferrin (the isolated N-lobe) were recorded over -300 sec at 80 MHz in pH 6.0 buffer and the presence of a chelator. Human serum transferrin (hTF) is a bilobal glycoprotein that serves as the major transporter of iron in humans (A.B. Mason, and S.J. Everse, Iron Transport by Transferrin, in: H. Fuchs, (Ed.), Iron Metabolism and Disease, Research Signpost, Kerala, India, 2008, pp. 83-123.)

[051] Both lobes, termed the N- and C-lobes, coordinate ferric iron via four amino acid ligands and a synergistic carbonate anion. Numerous studies on hTF have shown that binding of iron quenches the intrinsic tryptophan fluorescence, through radiative and non-radiative means. Rate constants for iron removal are thus determined by tracking the enhancement in fluorescence emission over time under endosomal like conditions (pH ~ 5.5, -150-200 mM salt and the presence of a chelator), which takes place within seconds to minutes. (N.G. James, C.L. Berger, S.L. Byrne, V.C. Smith, R.T. MacGillivray, and A.B. Mason, Intrinsic fluorescence reports a global conformational change in the N-lobe of human serum transferrin following iron release. Biochemistry 46 (2007) 10603-1061 1 ; S.S. Lehrer, Fluorescence and absorption studies of the binding of copper and iron to transferrin. J. Biol. Chem. 244 (1969) 3613-3617.)

[052] Figure 5, which shows the phasor plot (taken at 80 MHz and 300 nm excitation) during iron release (black line) and the points for iron-bound (circle) and apo (triangle), clearly demonstrates the predicted linear progression in the phasor point during iron release from the iron-bound point towards the apo phasor point. Sample heterogeneity, which was expected based on the fitting to a discrete exponential model, is seen in the phasor vectors of hTF N-lobe as each point is inside the universal circle. Calculation of the distance at each point over time can provide information regarding excited state changes during iron removal (i.e. one can recover the fractional contribution of each emitting state using standard linear methods. This kinetic run demonstrates how the phasor method can be utilized for tracking excited-state changes during protein-ligand dissociation, which can complement data obtained using other methods. We also point out that the phasor approach was applied in FLIM by Redford and Clegg (G.I . Redford, and R.M. Clegg, Polar plot for frequency-domain analysis of fluorescence lifetimes. Journal of Fluorescence 15 (2005) 805-815) to follow the kinetics of mixing in turbulent flow at the microsecond time scale.

[053] In Figure 5, iron-bound and apo phasor vectors are drawn as circle and triangle, respectively. The change in phasor during iron removal, at pH 6.0 with 4 mM EDTA over 300 seconds, is drawn as a black line. Over the first -90 seconds there is a rapid change in the excited state data followed by a much slower change. Data shown were collected at 80 MHz and 25°C.

[054] Example 6 - Protein Unfolding/Folding Pathways

[055] Protein folding has been intensely investigated, both theoretically and experimentally, for many decades. One of the most popular experimental approaches to protein folding involves examination of the unfolding pathway(s) by means of chemical denaturation. (K.A. Dill, S.B. Ozkan, M.S. Shell, and T.R. Weikl, The protein folding problem. Annu. Rev. Biophys. 37 (2008) 289-316, A.R. Fersht, From the first protein structures to our current knowledge of protein folding: delights and scepticisms. Nat. Rev. Mol. Cell Biol. 9 (2008) 650-654.)

[056] Chaotropic agents, such as GuHCI and urea, are commonly used denaturants. Unfolding experiments using such chemical denaturants have one major underlying assumption: that the overall, thermodynamic unfolding of the protein is independent of denaturing agent although the structural changes associated with the change are dependent. (W. Pfeil, and P.L. Privalov, Thermodynamic investigations of proteins. II. Calorimetric study of lysozyme denaturation by guanidine hydrochloride. Biophys. Chem. 4 (1976) 33-40; G.I. Makhatadze, and P.L. Privalov, Protein interactions with urea and guanidinium chloride. A calorimetric study. J. Mol. Biol. 226 (1992) 491 -505.)

[057] The molecular mechanism for protein unfolding by GuHCI and urea are still unclear, even though mechanisms have been proposed as early as 1936. (A.E. Mirsky, and L. Pauling, On the Structure of Native, Denatured, and Coagulated Proteins. Proc. Natl. Acad. Sci. USA 22 (1936) 439-447.) Models describing the direct interaction of the denaturant with the protein and disruption of the hydrophobic interactions via disruption of the hydrogen-bonding network are the commonly accepted mechanisms. (R.B. Simpson, and W. Kauzmann, The Kinetics of Protein Denaturation. I. The Behavior of the Optical Rotation of Ovalbumin in Urea Solutionsl . J. Am. Chem. Soc. 75 (1953) 5139-5152; M. Roseman, and W.P. Jencks, Interactions of urea and other polar compounds in water. J. Am. Chem. Soc. 97 (1975) 631-640.) Recently, Almarza et al. proposed a more detailed model for urea induced protein unfolding in which urea molecules interact with protonated histidines followed by hydrogen bond formation with polar residues leading to hydrophobic collapse of the protein. (J. Almarza, L. Rincon, A. Bahsas, and F. Brito, Molecular mechanism for the denaturation of proteins by urea. Biochemistry 48 (2009) 7608-7613.)

[058] As the detailed molecular mechanisms for protein unfolding due to GuHCI and urea are likely different, it would stand to reason that the intrinsic fluorescence properties will be different during denaturation depending on which chaotropic agent is present. Tryptophan fluorescence is sensitive to changes in the local environment and if the structural changes leading to unfolding of the protein are intrinsically different, then the environment around the tryptophan residues could differ depending on the chaotropic agent, resulting in distinct fluorescence properties. The phasor plot of lysozyme with increasing concentrations of GuHCI (0 - 6 M) and urea (0 - 8 M) is shown in Figure 6. Initial additions of either denaturant cause similar changes, yet, by 2 M (squares) the trajectories of the phasor points diverge, with GuHCI shifting the phasor points in a counter-clockwise direction. The final points between GuHCI and urea are dramatically different indicating unique unfolded forms. Similar deviations between unfolding with GuHCI and urea were observed with monomeric HSA (data not shown). It is clear that the phasor plot method has the potential for illustrating different unfolding pathways, as both proteins tested showed significant deviations in phasor point trajectories depending on the denaturant.

[059] Figure 6 shows phasor point trajectory of lysozyme denaturation as a function of increasing GuHCI (solid line) and urea (dashed line) concentration. Concentrations of 0 M (closed circle), 1 M (open circle), 2 M (open triangle) and 6/8 M (closed triangle) GuHCI and urea, respectively, are highlighted. Data was excited with 280 nm and data plotted were at 70 MHz and 25°C.

[060] Example 7 - Conformation of Clostridial toxin substrate in denaturant.

[061] The mapping of a Clostridial toxin substrate using the Phasor Plot method was accomplished as follows. The protein substrate used is GFP-SNAP25-BFP an approximately 80kDa protein. As part of the characterization of this molecule it was subjected to a 'denaturation titration' experiment using 8M Guanidinium chloride. The unfolding process of the protein is studied using the Phasor Plot mapping as shown in Figure 7. Phasor plot of denaturation of GFP-SNAP25-BFP from OmM (gray star) to 2M (gray circle) with GuHCI (Black line). The universal circle (gray line) is illustrated for reference. Number illustrate the GuHCI concentration in mM at the respective point.

[062] Example 8 - Conformation/dissociation of Botulinum Neurotoxin in denaturant.

[063] Using the same instrumental set up as with Example 7, the Phasor plot method was used to study protein conformational or aggregation state changes induced by the addition of non denaturing amounts of GuHCI. Both samples indicated a relatively large shift in the Phasor plot induced by GuHCI which is indicative of the presence of aggregated or conformationally related forms of the protein being present. In both these samples no cryo-protectant (typically PEG400) was being used. Figure 8: Phasor plot of induced changes to Dyn-LHn/A and LHn/A upon the addition of GuHCI to 50mM. The addition of GuHCI to 50mM is enough to reverse a previously aggregated sample but not sufficiently concentrated to denature the protein. The changes in the data points indicated by color points represent conformational or associative changes in the target proteins.

[064] Example 9 - Conformation/dissociation of Botulinum Neurotoxin in denaturant with PEG added.

[065] The instrumental conditions reagents and general experimental conditions are very similar to Examples 7 and 8. The question posed in this particular example was to study the effect on protein conformation and aggregational changes to a protein sample containing PEG400 as a cryoprotectant. The aggregational state of the protein is perturbed using low concentrations (50mM) of GuHCI. In this case an almost identical phasor plot indicates that no conformational/aggregation changes have taken place to a PEG stabilized sample. [066] Figure 9 shows a phasor plot of protein prep Dyn-LHn/A with added cryoprotectant 5% PEG. After the addition of GuHCI to a final concentration of 50mM there is statistically no change in the Phasor given by the blue and red points. These data indicate that sample aggregation has not taken place under these conditions.

[067] Example 10 - Analysis of Clostridial Toxin Activity Using Phasor Plot Analysis

[068] This example illustrates that data collected from any Clostridial toxin activity assay can be evaluated by phasor plot analysis. Examples of Clostridial toxin activity assays include those disclosed in U.S. Patent 7, 183,066; U.S. Patent 7,208,285; U.S. Patent 7,332,567; and U.S. Patent 7,399,607; each of which is hereby incorporated by reference in its entirety. In addition, the activity of any toxin comprising a Clostridial toxin enzymatic domain derived from a light chain can be assessed by phasor plot analysis, including chimeric toxins are retargeted toxins, such as, e.g., those disclosed in U.S. Patent 6,903, 187; U.S. Patent 7, 183, 127; U.S. Patent 7,244,437; U.S. Patent 7,273,722; U.S. Patent 7,419,676; each of which is hereby incorporated by reference in its entirety.

[069] A Clostridial toxin activity assay is performed and a phasor plot analysis is conducted. The instrumental conditions reagents and general experimental conditions are very similar to other Examples listed herein. In this case the phasor plot indicates a conformational- aggregation change indicative of substrate cleavage by the toxin.

[070] Example 1 1 - Analysis of Formulation Stability of Pharmaceutical Composition Stored Frozen Using Phasor Plot Analysis

[071] This example illustrates that the stability of a formulated pharmaceutical composition can be assessed using phasor plot analysis without any disruption to the packaged composition. Packaged vials comprising a dried pharmaceutical composition (e.g., by lypholization or freeze- drying) are assessed by phasor plot analysis using similar instrumental conditions, reagents and general experimental conditions as disclosed in Examples above. The quality and quantity of the composition are initially assessed at the time of initial packaging. The vials are then stored at the desired temperature, e.g., room temperature, -20 °C, or -70 °C. The vials are periodically assessed over time to monitor both quality and quantity of the pharmaceutical composition, e.g., once every month, once every three months, once every six months, once every year. In this case assessment every three months over a period of three years indicated that the pharmaceutical composition remained stable in that the quality and quantity of the composition after three years was substantially the same as the initial assessment made after packaging.

[072] Example 12 - Analysis of Formulation Stability of Pharmaceutical Composition Stored at Room Temperature Using Phasor Plot Analysis.

[073] This example illustrates that the stability of a formulated pharmaceutical composition can be assessed using phasor plot analysis without any disruption to the packaged composition. Packaged vials comprising a liquid pharmaceutical composition are assessed by phasor plot analysis using similar instrumental conditions, reagents and general experimental conditions as disclosed in other Examples herein. The quality and quantity of the composition are initially assessed at the time of initial packaging. The vials are then stored at the desired temperature, e.g., room temperature, -20 °C, or -70 °C. The vials are periodically assessed over time to monitor both quality and quantity of the pharmaceutical composition, e.g., once every month, once every three months, once every six months, once every year. In this case assessment every three months over a period of three years indicated that the pharmaceutical composition remained stable in that the quality and quantity of the composition after three years was substantially the same as the initial assessment made after packaging.

Claims

CLAIMS We claim:

1. An in vitro method of determining changes in a protein environment comprising:

selecting a protein of interest,

placing a first sample of the protein in a spectroflurometer such that the protein sample may be irradiated with frequency modulated exciting light and the emitted light is detected;

obtaining a reference set of light measurements at a selected frequency by measuring the change in phase angle (Φ) between emitted light and excited light and also simultaneously recording the change in signal modulation (m);

applying time domain equations or frequency domain equations of the reference set of light measurements to obtain a reference data set;

obtaining at least a second set of light measurements at the selected frequency for the protein;

applying a time domain equations or frequency domain equations of the second set of light measurements to obtain a second data set;

plotting S vs G for the reference data set and the second data set; and,

determining whether the protein environment underwent changes between the reference measurement and second measurement by observing if a change in the position of the generated plot points;

wherein the spectroflurometer comprises a light source which can cause a protein to fluoresce;

wherein the spectroflurometer comprises a device for detecting and measuring the emitted light from the protein;

wherein the light source is modulated in the frequency domain or time domain;

wherein changes between the reference measurement and the second measurement are selected from the group consisting of: a) the same protein sample at a different time; and b) the protein in a different milieu, and ,

wherein the changes in the protein environment is a conformational change of a single protein, changes in protein-protein interactions and/or changes in protein-excipient interactions.

2. The method of claim 1 , wherein the protein is a therapeutic protein in a formulation stability study.

3. The method of claim 1 , wherein the protein is part of a quaternary protein structure in either the reference measurement or the second measurement.

4. The method of claim 1 , wherein the protein is an industrial protein in a stability study.

5. The method of claim 1 , wherein emitted light is collimated, isolated from the excitation light using appropriate devices such as optical filters or a monochromator and measured using light sensitive detectors.

6. The method of claim 1 , wherein the frequency domain is utilized by applying at the selected modulation frequency of the sets of light measurements use the frequency domain equations S = m sin φ and G = m cos φ to produce the data sets; and,

wherein the light source is frequency modulatable.

7. The method of claim 1 , wherein the time domain is utilized wherein the decay of the fluorescence is recorded with time after excitation of the fluorescence with a pulse of light and t the time domain equations

where ω can be chosen as the repetition frequency of the pulsed excitation source or another value which depends on the kinetics of the excited state process under investigation and l(t) is the observed fluorescence intensity at time t.