Nonlinear optical detection of molecules comprising an unnatural amino acid possessing a hyperpolarizability

Abstract

A system for making molecules, and proteins in particular, suitable for detection by a surface-selective nonlinear optical technique. A first use of the invention is for determining a protein's structure in real space and real time. A second use of the invention is to detect a protein or its activity (conformational change). A third use of the invention is for drug screening. A further aspect of the present invention is measuring probe tilt angle orientation in an oriented protein.

Description

INTRODUCTION

Surface-selective techniques such as second-harmonic generation (SHG) have recently been applied to the study of proteins at interfaces and protein conformational change by the use of second-harmonic-active labels^1-3, which are attached to the surface of the protein. Methods for detecting proteins by SHG or sum-frequency generation (SFG) are disclosed wherein the protein is detected by incorporating an SH-active (or SF-active) unnatural amino acid probe into it.

The aim of structure-based drug screening and basic studies of the mechanism of biological molecules requires a tool that can measure structure, and structural change, of biological molecules as they bind to ligands, drugs, etc. Present techniques for determining structural change are mainly confined to NMR (Nuclear Magnetic Resonance) and X-ray crystallography. Neither of these techniques is suitable for measuring structural change in real-time. Moreover, they are time- and labor-intensive and unsuitable for widescale use in drug screening. Furthermore, there are many proteins that are difficult to crystallize (e.g., membrane proteins) and thus whose structures have not been determined.

It is often not convenient to label a protein using standard in vitro methods, as it may require single-site mutagenesis to produce an attachment site for the chemical label, or the labeling itself may introduce a perturbative modification to the structure of the protein. Also, it is often not possible to engineer a single, chemically reactive and orthogonal attachment site for a label into a protein; for example, there may be native and reactive cysteine groups in the case of sulfhydryl-bearing labels. Furthermore, for the purpose of measuring conformational change using a surface-selective nonlinear detection technique such as second-harmonic generation (SHG) or sum-frequency generation (SFG), it is desirable to have the probe (the label) rigidly fixed within the protein frame of reference. Then, movement of the label relative to the protein will be minimized, if not completely eliminated, the signal contrast (i.e., SH signal) between conformations will be maximized, and precise spatial measurements of the probe tilt angle relative to a surface can be used to determine a protein's structure.

SUMMARY

The present invention includes a system for making molecules, and proteins in particular, suitable for detection by a surface-selective nonlinear optical technique. A first use of the invention is for determining a protein's structure in real space and real time. A second use of the invention is to detect a protein or its activity (conformational change). A third use of the invention is for drug screening. A further aspect of the present invention is measuring probe tilt angle orientation in an oriented protein.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1A shows the structure of Aladan. FIG. 1B shows the structure of GB1(¹¹). Residues with Aladan substitutions are highlighted: Leu⁷ and Ala²⁴. Fc and Fab binding surfaces are also shown^9,10. FIG. 1C shows an SHG spectrum of Aladan adsorbed to a mica surface, in PBS.

FIG. 2A shows an SHG spectrum for GB1-Aladan adsorbing to an aldehyde-derivatized glass coverslip, detected by SHG, and plotted as SHG intensity versus time. FIG. 2B shows an SHG spectrum of the adsorbed protein at steady-state, detected by SHG, and plotted as SHG intensity versus wavelength.

FIG. 3 shows a schematic view of the protein A-Fc-GB1 complex binding to the Fc fragment.

FIG. 4A shows an SHG spectrum for GB1-Aladan binding to an IgG-Protein A-glass coverslip. As indicated in FIG. 4B, addition of Fc causes a small decrease in SH signal with the Ala²⁴ Aladan mutant, indicating a change in the orientation of Aladan. FIG. 4C shows an SHG spectrum for addition of Fc to the Leu⁷ Aladan mutant. Signals were detected by SHG and are averaged over 4 seconds. The bars denote the standard error of measurement (SEM).

DETAILED DESCRIPTION

The invention includes a system for making molecules, and proteins in particular, suitable for detection by a surface-selective nonlinear optical technique. A first use of the invention is for determining a protein's structure in real space and real time. A second use of the invention is to detect a protein or its activity (conformational change). A third use of the invention is for drug screening. A further aspect of the present invention is measuring probe tilt angle orientation in an oriented protein.

We introduce a new class of probes—SH-active unnatural amino acids—and demonstrate their use for detecting biomolecules and structural changes by SHG. Aladan is demonstrated to be SH-active by studying it alone on a surface and incorporated into a protein, GB1, rendering the protein detectable. A structural change is observed when an Fc fragment is introduced to GB1, labeled with Aladan at Ala²⁴, and bound to IgG on a surface. No such change is observed if Aladan is substituted at Leu⁷ instead. These results are consistent with a local conformational change of GB1, a change undetectable by fluorescence or X-ray crystallography. SHG with SH-active exogenous labels or unnatural amino acid probes is promising as a structural technique with high angular and temporal resolution. With a narrowly oriented protein population appropriately labeled, the technique could be used to determine conformational change in real time and space, site-by-site.

Second-harmonic generation (SHG) is highly sensitive to the net, average orientation of SH-active molecules on surfaces, and has recently emerged as a technique for detecting biomolecules and their conformational changes. As most biomolecules are not intrinsically SH-active, they must be labeled with probes to render them detectable. To date, exogenous probes have been used to do this, but second-harmonic-active unnatural amino acids offer important advantages for the long-range goal of precisely and directly determining structural changes in real time, and may be used for both buried and surface sites. Results of the first known SH-active unnatural amino acid, Aladan, are presented here. Aladan is found to be SH-active by detecting it at an interface, both alone and incorporated into the B1 domain of protein G (GB1), a globular immunoglobulin-binding protein, at both buried and exposed sites. The tilt angle of Aladan alone adsorbed on a mica surface is determined by polarization experiments, and its nonlinear polarizability α⁽²⁾ is found to be ca. 10⁻³⁰ esu. Aladan GB1 mutants are detectable by SHG, either when coupled covalently to a derivatized glass surface or bound to IgG immobilized via protein A. Addition of an Fc domain to this GB1 complex causes a small but defined change in the SH signal when Aladan is incorporated at site Ala²⁴, but not at Leu⁷, consistent with a local conformational change of GB1. This structural change is not apparent in either X-ray crystallography and fluorescence studies, demonstrating that SHG can detect subtle orientational changes, including protein-protein interactions in which no significant rearrangements occur.

Second-harmonic generation (SHG) is well known in the physical sciences for studying molecules on surfaces and is especially useful for measuring orientation. More recently, SHG has emerged as a sensitive technique to detect and study the conformational changes of biomolecules using SH-active probes^1,2. Labeled proteins that are adsorbed or covalently immobilized on surfaces produce an SHG signal, which is due to the average, net orientation of the nonlinear polarizability of the SHG label relative to the surface plane. Specifically, the SH intensity is given as I_SH=G(χ_s⁽²⁾)²I², where I_SH is the second-harmonic intensity, G is a constant that depends on the experimental geometry and wavelength, and I is the intensity of the fundamental beam. The nonlinear susceptibility, χ_s⁽²⁾, carries the details of the SH-active molecules on the surface via the equation:χ_s⁽²⁾=N_s<α⁽²⁾>  (1)where N_s is the surface density of the molecules, the brackets denote an orientational average, and α⁽²⁾ is their nonlinear polarizability, a quantum-mechanical property that determines the probability of producing a second-harmonic photon from two, incident photons of the fundamental beam. Measurements of χ_s⁽²⁾ provide information about the orientation of a molecule on the surface. For example, when α⁽²⁾ is dominated by a single element ζζζ⁽²⁾ along the molecular axis ζ and the azimuthal distribution of the molecules are random in the plane of the surface, the only elements of χ_s⁽²⁾ that do not vanish are:χ_s⊥⊥⊥⁽²⁾=N_s<cos³ θ>α_ζζζ⁽²⁾  (2A)χ_s⊥∥∥⁽²⁾=χ_s∥⊥∥⁽²⁾=χ_s∥∥⊥⁽²⁾=1/2N_s<cos θ sin² θ>α_ζζζ⁽²⁾  (2B)where θ is the polar angle between ζ and the surface normal, and the subindices ⊥ and ∥ refer to the directions perpendicular and parallel to the surface, respectively³.

The SH light is coherent and directional, so collection and isolation of the SH beam is simplified, and because the fundamental and the second-harmonic are well separated spectrally, cross-talk, which can plague fluorescence measurements, is non-existent with SHG. Photodegradation of the probe occurs relatively slowly via two-photon-induced absorption, allowing measurements over relatively long timescales. The trade-off with SHG is signal strength—it is orders of magnitude weaker than fluorescence. However, only SH-active molecules immobilized on the surface contribute second harmonic light since randomly diffusing molecules near the surface produce no signal; their orientational average, from Equation 1, is zero. Therefore, SHG is intrinsically equipped to discriminate between surface-bound and free molecules.

The SH signal reports on the orientational average of the probes, and thus changes due to conformational change. In previous work, ligand-induced conformational changes were detected by monitoring the SH intensity with calmodulin and adenylate kinase adsorbed non-specifically to surfaces^1,2. SFG and SHG have also been applied to study protein adsorption phenomena at various interfaces with protein alone or co-adsorbed with an SH-active probe^4-8. The current study uses GB1, a well-studied IgG-binding streptococcal protein that has been useful in a variety of structural and biophysical studies. GB1 binds to both the Fab and Fc domains of IgG, and structures of these complexes have been determined^9,10, in addition to that of the uncomplexed protein^11,12. We have previously incorporated the synthetic amino acid Aladan at multiple sites of GB1 for studies of protein solvation^13,14 Like other donor-acceptor fluorophores, Aladan undergoes a large increase in dipole moment upon excitation, leading to significant solvatochromic fluorescence shifts.

Unnatural amino acids (UAA's) offer a means of labeling proteins at both buried and exposed sites, and as innate components of the protein they should report on structural changes with more sensitivity and fidelity than probes attached via sidechain-reactive linkers¹³. We report here that the synthetic amino acid Aladan shows good activity in SHG measurements, making it the first known SH-active UAA. We estimate the free Aladan hyerpolarizability to be ˜10⁻³⁰ esu and measure the average orientation of Aladan adsorbed on a mica surface as 48°. Incorporated into GB1, we detect the protein when it is covalently coupled to a surface or bound via an IgG-protein A complex. Addition of Fc to the latter system causes a small change in the SH signal when Aladan is substituted at the Ala²⁴ residue, but not at Leu⁷, consistent with a local change in GB1 conformation.

Measurements may be made using any suitable experimental set-ups and techniques. An exemplary set-up has been described previously^1,2. Briefly, a Ti:S oscillator (Coherent; Mira 900) pumped by a solid-state DPSS laser (Spectra-Physics; 5W Millenia) was tuned to 780 nm (0.8 W) and focused into a prism (BK-7) that was optically coupled to the coverslips or mica surfaces for total internal reflection (TIR). The incident angle of the fundamental at the surface was 73° (critical angle=61°). A monochromator and color filter were used to remove the fundamental and select the SH signal and its intensity was measured by a PMT. A well for buffer was defined by an adhesive gasket (Grace Biolabs) and filled with phosphate buffered saline (PBS). Human IgG (normal serum) and Fc (IgG-Fc) were obtained from Bethyl Laboratories. Mica (Grade IV) was obtained from SPI Corporation and was freshly cleaved for the experiments. Protein A- and aldehyde-derivatized glass coverslips were obtained from Xenopore Corporation. Aladan and Aladan GB1 mutants (synthesized as described;^13,14 were introduced to the wells at 1 μM; IgG was added at 3 μM; and IgG-Fc was added at 6 μM. GB1-Aladan was immobilized on the aldehyde-derivatized surface by exposing the protein to the surface for several minutes followed by washing with buffer. For the experiments with IgG, the protein was incubated with the protein A coverslips for two hours followed by washing with PBS. All experiments except the measurements to determine χ⁽²⁾ were carried out with the fundamental p-polarized and the second-harmonic beam unpolarized.

We chose Aladan (FIG. 1A) for this SHG study because of its absorption wavelength, large Stokes shift and difference dipole moment¹³. The peak absorbance of Aladan in buffer is 390 nm¹³ and the Δμ of its chromophore is large, variously estimated between 5 D¹⁵ and 20 D¹⁶. Aladan was dissolved in methanol and exposed to a freshly cleaved mica surface and allowed to adsorb in PBS. An SH background of 3500 cps was measured, due to the mica surface and to water polarized in its vicinity (χ⁽²⁾ and χ⁽³⁾ contributions, respectively^17,18). After several hours, the signal with Aladan reached a plateau of ˜10⁶ cps as shown in the SHG spectrum of 1C. The quadratic dependence of the SH signal on the fundamental power confirmed that Aladan is indeed SH-active. Two-photon-induced fluorescence, which appeared at ca. 550 nm, is well separated from the second harmonic light. Washing Aladan out of the well reverted the signal to the background level, demonstrating that the molecule adsorbs reversibly. The average orientation was determined by polarization measurements assuming a close packed monolayer of probe and a narrow orientational distribution. The SH intensity was measured as a function of input and output polarization to determine the nonlinear susceptibility elements χ_zzz⁽²⁾, χ_zxx⁽²⁾, and χ_xzx^(2)3,19. Ratios of the elements were used to calculate an average orientation of the probe on the mica surface of 48°. Power losses by the prism and the optics after the sample were taken into account in the calculation. The probe's nonlinear polarizability is estimated to be α⁽²⁾˜10⁻³⁰ esu.

Next, Aladan mutants of the immunoglobulin-binding protein GB1 were used to show that an SH-active UAA can be used to detect protein on a surface. These Aladan-substituted proteins have previously been shown to fold and function similarly to wild-type GB1¹³. Aladan was incorporated at two positions as shown in FIG. 1B: Ala²⁴, a surface residue at the N-terminus of the helix and in proximity to the Fc binding surface, and Leu⁷, a buried residue in the first β strand. An X-ray structure of the Fab-GB1 complex previously showed that Fab fragment binds edge-on to this first β strand⁹. The Ala²⁴ mutant was introduced into the well in contact with an aldehyde-derivatized coverslip. The SH signal immediately rose above the background as the GB1 lysines reacted with the aldehyde-derivatized surface (FIG. 2), reaching a plateau at about 30,000 cps. Wild-type GB1 produced no increase in the SH background upon addition (data not shown). The production of second-harmonic light by GB1-Aladan demonstrates that the protein self-orients to some degree upon binding to the surface, possibly due to the arrangement of lysines on the protein surface. The wavelength dependence of the signal shows the characteristic lineshape of the second-harmonic (FIG. 2B). The peak of Aladan absorption is ca. 390 nm, so the signal is resonantly enhanced. The SH signal after washing decreases slightly to 25,000 cps, indicating that some of the protein was reversibly bound to the surface.

We next wanted to test whether SHG could detect any conformational change upon addition of Fc to GB1. X-ray structures of the Fc-GB1 and Fab-GB1 complexes show no global conformational changes compared to the uncomplexed GB1, though the modest resolution (3.5 Å) of the Fc-GB1 complex would make subtle changes difficult to detect. For this experiment, we used glass coverslips with protein A covalently coupled to their surface. These coverslips were incubated with IgG, which binds to protein A via its Fc domain as shown in FIG. 3. After washing, GB1 with Aladan substituted at Ala²⁴ was added for binding to the exposed IgG Fab domain, causing an increase in SH signal of ˜600 cps as shown in FIG. 4A. In a control experiment, addition of Ala²⁴ mutant to protein A coverslips without IgG caused no increase in SH signal, indicating that GB1 binds specifically to IgG (data not shown). The much lower SH signal of GB1 bound to IgG-Protein A compared with to the aldehyde-derivatized surface is due to the reduced number of surface sites available for binding given the much larger size of the protein A-IgG complex than GB1 alone (>200 kDa for the complex vs. 6 kDa for GB1). Excess, unbound GB1 was then removed from the well by washing with PBS, but the SH signal remained constant. Addition of Fc causes a small and reproducible decrease in the SH signal (140±60 cps; 4 second average), which is detectable within the experimental noise, as shown in FIG. 4B. The SH signal of GB1-Aladan is remarkably stable over time on the protein-derivatized surfaces (no measurable change in signal over 10's of seconds) without the presence of an oxygen scavenger, indicating minimal photobleaching of the probe. As a control, Fc added to IgG-Protein A without GB1 produces no change in the SH background (data not shown). In another control, Aladan was incorporated at Leu⁷, a buried site away from the Fc binding region. No change in SH signal was observed upon adding Fc to this mutant bound to IgG-Protein A (FIG. 4C).

SHG is directly sensitive to molecular orientation on surfaces and—by applying it to biological molecules—can measure subtle structural changes. The present work is motivated by the goal of using the technique to measure conformational change in real time and space as a function of probe location. An unnatural SH-active amino acid, a class of probes we introduce here, is used to detect GB1 and its conformational change upon binding Fc. There are few techniques available to study structural motion of biomolecules in real space. Fluorescence resonance energy transfer (FRET) is useful for detecting conformational changes^20-22, reporting on relative changes in distance and orientation between two site-specific probes, rather than absolute changes in structure. Other techniques, such as NMR and Raman spectroscopy, are useful for studying protein dynamics at a wide range of time scales^23-25. These techniques rely on an interpretation of spectra to deduce structural changes and therefore do not provide direct real-space information, which may lead to uncertainty in model fitting. NMR has also traditionally been difficult to apply to larger biomolecules (e.g., >30 kDa) and membrane proteins, as well as on structural motions on slower, but functionally important, time scales (e.g., μs to ms). Environment-sensitive dyes have also been used to detect conformational changes, notably when a probe moves between polar and nonpolar environments, or whose distance from nearby quenchers changes^20,26-28. SHG complements these techniques, for example, when it is difficult to incorporate multiple probes into a protein, or when the environment of a probe does not change appreciably with conformational change. SHG is relatively insensitive to quenching and to environmental changes, since it relies on a scattering process rather than emission. Most importantly, the technique is sensitive to absolute changes in probe orientation and has the potential to measure structural changes in real space and real time simultaneously. In the present study, we demonstrate that the free amino acid Aladan is SH-active by detecting it adsorbed to a mica surface. If the orientational distribution of Aladan is the same whether it is incorporated into GB1 or adsorbed directly to a surface, the expected difference in SH signal intensity between the two samples would be about two orders of magnitude apart since I_SHαN_s²; this is in rough agreement with the difference we measure. The signal strength in photons/pulse of surface SHG is governed by the following equation:S≅(256π⁴ω/hc³)|N_sα⁽²⁾|²I²AT  (3)I is the intensity of the fundamental beam incident at 45°, A is the beam cross section at the surface, T is the pulse duration, and ω is the second-harmonic frequency²⁹. An regeneratively amplified system, of the type used to study ultrafast processes (e.g., μJ pulses)³⁰, increases the SHG signal by several orders of magnitude. With Aladan's α⁽²⁾ of ca. 10⁻³⁰ esu, an amplified system, and standard noise reduction methods, the time resolution of the technique should be extendible to time scales of 10⁻⁶ s. To improve the time resolution, tighter focusing, higher fundamental peak power, or probes with higher nonlinear polarizability could be used. For example, a signal enhancement of about a factor of 5 could be achieved by tuning the incident angle of the fundamental to the critical angle³¹. Because the background is similar in magnitude to the GB1-Aladan signal and thus cannot be ignored, the relative phases of the various SH contributions must be known to quantitatively determine the absolute orientation of the probes or their angular change (Δ°) upon adding Fc.

Upon adding Fc to the surface-bound Ala²⁴ GB1 mutant, a small signal decrease occurs. Given the experimental noise levels in our set-up, the lower limit of this signal change is about 80 cps, indicating that a change in the average orientation of Aladan in GB1 occurs upon binding Fc. No observable change occurs in the same experiment with Aladan incorporated at Leu⁷ (FIG. 4C), so these experiments are consistent with a local conformational change of GB1 at the N-terminus of the GB1 helix, rather than a reorientation of the IgG-GB1 complex. Alternatively, following Equation 2, the average orientation of the probe at Leu⁷ may be much less sensitive to reorientation than when it is at Ala²⁴. In this case, the observed change in SH signal could indicate a reorientation of the Fab-GB1 complex when binding to Fc rather than a local conformational change. Structures of both wild-type GB1 and the GB1-Fc complex have been reported, with no apparent structural differences outside of minor sidechain movements in the binding pocket formed largely by the helix, and the 3^rd β strand¹⁰. The Aladan substituted at Ala²⁴ would be expected to reside just outside of the binding pocket, which extends to Glu²⁷ on the helix. In contrast, there is no significant change in Ala²⁴ Aladan fluorescence upon Fc binding (B. Cohen, unpublished data). Compared to UAA's, exogenous probes, such as spin labels or fluorophores attached via sidechain-reactive linkers, are expected to reside farther from the binding pocket and have greater ranges of motion²⁸.

In this instance, SHG is able to detect a small signal change, a change not apparent in the protein by fluorescence or crystallography. With a highly oriented population of probes (and protein), the technique could resolve angular changes as small as 1°¹. GB1, bound to Fab on the surface, is the first study by SHG of a protein specifically ordered on a surface (FIG. 3). The GB1-Fab interaction has been defined crystallographically and is likely to produce an ordered array of GB1 on the mica surface in the present study⁹. Combined with the absolute homogeneity of labeling arising from site-specific Aladan incorporation, this could provide a level of sensitivity necessary to sense to a conformational change not previously detected by other techniques. Detection of conformational changes have also been obtained with an integrin protein (α_vβ₃) and amyloid proteins (β-amyloid and α-synuclein) (J. Salafsky, unpublished data). Another major goal of this research is to quantitatively map conformational change of a biomolecule in real time. To achieve this requires a narrow, or at least a Gaussian distribution of the probes (and thus the protein)³², and there are a variety of methods to specifically immobilize proteins to a surface (e.g., epitope tags, His-tags, antibody binding, etc.). Methods for orienting protein in 2-D in ways which preserve their functionality and freedom to move will be as indispensable to the technique as is growing crystals for X-ray studies. An important test for the technique is to measure the angular change that a site-specific probe undergoes upon ligand binding in a well defined system (e.g., maltose binding protein), and compare the result with that obtained by X-ray or NMR experiments.

The present invention discloses the use of unnatural amino acids which possess a hyperpolarizability for detecting proteins using a nonlinear technique such as second-harmonic generation. Henceforth, these specific unnatural amino acids will be referred to as SHAA's (‘Second-Harmonic-Amino-Acid’). One advantage of using unnatural amino acids (AA's) as probes of protein is that detection can be carried out in vivo—that is, in live cells. For example, the invention could be used to detect conformational change or protein activation in live cells in response to binding of ligands or drugs. The invention could be used for drug screening for compounds that induce or block conformational change in a protein, in the cellular milieu, or which bind, for example, to specific conformations of the protein. By using an oriented protein population, relative to a surface, a highly precise map of structure or conformational change in real space and real time could be built, which would be of use to basic research concerning questions such as protein folding, structure, function and dynamics. The invention optionally may include screening a plurality of compounds (e.g., candidate ligands, drugs, and/or modulators) by sequentially and/or simultaneously assessing their effects on conformation and/or other measurable parameters, for example, using a sample holder such as a microplate having a plurality of sample sites supporting a corresponding plurality of samples or sample systems.

In the case of in vivo detection, the cell itself orients the protein of interest, as, for instance, with membrane receptors and ion channels. This net orientation is a crucial requirement for detection by the nonlinear optical schemes employed in the present invention—such as second-harmonic generation or sum-frequency generation.

There is a large and growing body of literature which describes a number of methods for incorporating unnatural amino acids—including fluorescent ones—into a variety of proteins in both prokaryotes and eukaryotes. For example, the papers of Schultz and coworkers^33,34, Dougherty and coworkers^35,36 and references therein describe the field. U.S. Pat. No. 7,045,337 also describes in detail how unnatural amino acids can be incorporated in vivo. Furthermore, there has been a report in the literature of an unnatural amino acid that is a good candidate for possessing a hyperpolarizability (Aladan) based on its large fluorescence Stokes shift³⁷. The SHAA can be incorporated into proteins in E. coli, a prokaryote, a mammalian cells, S. cerevisiae, etc. Multiple SHAA's can be incorporated into a single protein by techniques known to those skilled in the art.

Whether a given unnatural amino acid possesses a hyperpolarizability can be tested in a number of ways known to those skilled in the art. For example, it can be predicted to occur computationally, tested by incubating the molecule for propensity to generate second-harmonic light when it is contacted with an interface, measured using hyperrayleigh scattering, measured by an EFISH experiment, and so on. However, the final test to determine whether an unnatural amino acid possesses a hyperpolarizability is the presence of second-harmonic emission from the probe within the protein itself.

One aspect of the present invention is the detection of a protein containing a SHAA by second-harmonic generation or sum-frequency generation: for example, detection in vitro at an interface, in vivo in cell membranes, or in vivo in the interior or the cells. With a well defined protein orientation, a protein's structure can be determined in real time and real space (e.g., conformational change detection) by measuring the tilt angle or absolute tilt angle of an unnatural amino acid probe, or a series of such probes, placed in different mutants of the protein. The probes can be incorporated at any site within the protein or at its termini, in any domain, etc. A third aspect of the invention includes a second-harmonic-active moiety or dye molecule that is chemically equipped to react covalently with an unnatural AA; for example, if the unnatural AA incorporated into a protein is Para-acetyl-phenylalanine (pAcF), the second-harmonic-active dye would have appropriate chemistry on it for bonding covalently to pAcF. A fourth aspect of the invention is the incorporation of a SHAA in addition to a second unnatural AA, the second unnatural AA (which will in general not be second-harmonic-active) allows chemically orthogonal covalent coupling of the protein in an oriented manner to a surface derivatized with appropriate chemistry for reaction with the second AA. With a highly oriented protein sample that is SH-active (using the two unnatural AA's), both the baseline SHG signal and the contrast (change in signal with conformational change) are larger.

Furthermore, a highly oriented protein population would enable the determination of the actual conformational change the SHAA undergoes upon conformational change, by determining the tilt angle of one or more probes at one or more sites within the protein as a function of time. The three-dimensional structure of a protein could be determined by making one or more mutants of a protein each containing a SHAA probe placed in a different location (i.e., the probe orientation relative to the surface in each mutant, and therefore the side-chain orientation, can be determined for the probe in each mutant and a model of the overall three dimensional protein structure could be built using this information). Information from steric hindrance methods, statistical methods, molecular dynamics, Ramachandran plots, or energy minimization methods known to those skilled in the art could be used to further aid in determining the structure given the measured probe tilt angles. A time-resolved measurement of the tilt angle of a probe produces a motion picture of a conformational change of a protein as it occurs in real time. Because of SHG's (and SFG's) virtually instantaneous response and sensitivity, spatial orientation of a particular probe (e.g., tilt angle or absolute tilt angle relative to a surface) could be measured in real time at almost any time scale of interest. Although the ideal probes could be unnatural amino acids, the present invention would also work with exogeneous dye probes that are well known in the state of the art (e.g., PyMPO-succinimidyl ester that couples to surface amines of a protein).

The use of an array of proteins on a surface would facilitate, for example, the rapid determination of conformational change in real time for basic research or drug screening, or the three-dimensional structure of a protein that is oriented on the surface, for example, by using a series of mutants arrayed on the surface, which each contain a SHAA probe in a different site of the protein. The absolute tilt angle of each probe can be determined using methods known to those skilled in the art. For example, the papers of Goh and Kemnitz and the papers that cite them describe how absolute tilt angles are measured^38,39. The tilt angles of the probes in the different mutants, determined from their positions in the various sites in the mutant proteins, could thus be used to build a map of the protein's three-dimensional structure using techniques known to those skilled in the art, for example, by using computer modeling and energy minimization methods.

Second harmonic generation detection of a protein containing a SHAA is well known to those skilled in the art. The protein can be immobilized or adsorbed to a surface in such a way to produce a net orientation of the SHAA itself (its transition moment) and thus generate a source of second-harmonic radiation that can be detected in a straightforward manner. Conformational change of a protein containing a SHAA can be detected by adding ligands, drugs, etc. to the medium that is in contact with the surface (i.e., the interface) and monitoring a change in a physical property of the second-harmonic radiation (e.g., its intensity or polarization-dependence).

Important drug targets such as ion channels, integrins, kinases, receptors and GPCRs will all benefit from having a means of introducing a probe with a hyperpolarizability into them, without the need for chemical labeling in vitro. These mutant proteins could then be used for drug screening or basic research studies.

The SHAAs could, of course, also be used as probes of protein location or spatial dynamics within a cell. The present invention, of course, is not limited to proteins. Any biological molecule or entity into which an unnatural amino acid can be incorporated is covered by the scope of the invention.

For enzymes such as kinase or phosphatase protein^40-48, important conformational changes occur in loop regions, such as in the activation loop upon binding ligands or drugs. As these loop regions are typically mobile, as exhibited by the lack of order in these regions in the proteins' X-ray-determined structure, it is expected that they will well tolerate amino acid substitutions and/or labeling in or near these regions. The loop regions are well known to those skilled in the art of determining or using the structures of the proteins. Moreover, catalytic function of a kinase requires a conformational change of the ‘activation loop’. As the active structures of kinase proteins across the kinome (there are about 500 distinct kinase proteins encoded by the human genome) are highly similar, drugs that target the active form of a kinase often are non-selective, reacting ‘off-target’ with other kinases. Loop regions that are well known in the art include catalytic loops, WPD loops, PTP loops, recognition loops and activation loops.

In this disclosure, a new class of probes is introduced—SH-active unnatural amino acids—and their use for detecting biomolecules and structural changes by SHG is demonstrated. Aladan is demonstrated to be SH-active by studying it alone on a surface and incorporated into a protein, GB1, rendering the protein detectable. A structural change is observed when an Fc fragment is introduced to GB1, labeled with Aladan at Ala²⁴ and bound to IgG on a surface. No such change is observed if Aladan is substituted at Leu⁷ instead. These results are consistent with a local conformational change of GB1, a change undetectable by fluorescence or X-ray crystallography. SHG with SH-active exogenous labels or unnatural amino acid probes is promising as a structural technique with high angular and temporal resolution. With a narrowly oriented protein population appropriately labeled, the technique could be used to determine conformational change in real time and space, site-by-site.

EXAMPLES

The following examples further describe selected aspects and embodiments of the present disclosure. These examples and the various features and aspects thereof are included for illustration and are not intended to define or limit the entire scope of the present disclosure.

Example 1

An unnatural amino acid, Aladan, which possesses a hyperpolarizability, as determined by standard means known to those skilled in the art, is incorporated into maltose binding protein in E. coli, purified and expressed.

A baseline second-harmonic signal is generated using a Ti:Sapphire laser totally internally reflected from a suitable surface such as a polystyrene coverslip, available commercially from Newport Corp. or Coherent Inc. Details of the experimental set-up and recording of the data are described in Salafsky, 2006³. The surface is mounted on and index-matched to a prism through which the fundamental and second-harmonic beams pass. A rubber gasket is placed on the surface to define a well with a volume of about 100 microliters. Aqueous buffer is added to the well. The wavelength of the fundamental is tuned to about 720 nm and the power is maximized. The SH baseline is recorded continuously using a filter to block the fundamental beam, a photomultiplier and single-photon detection.

The maltose binding protein, with the incorporated second-harmonic-active probe, is then added to the buffer. The protein will adsorb to the surface and generate a second-harmonic beam at about 360 nm, which is then recorded as an increase in signal from the buffer-surface interface. Thus, second-harmonic detection of a protein that has an unnatural amino acid with a hyperpolarizability (a SHAA) is accomplished, in this case in vitro.

Example 2

An unnatural amino acid that is second-harmonic active is incorporated into a single site in 10 different adenylate kinase mutants, each with the probe in a different site, according to procedures known to those skilled in the art. The protein mutants are isolated and purified according to standard means. The mutants are optionally screened for both activity and ability to generate second-harmonic radiation at an interface to determine the kinase that is both most native-like (e.g., in Km, Vmax, etc.) and also competent for generating second-harmonic radiation. For generating the second-harmonic radiation, the kinase proteins are adsorbed to a polystyrene surface.

Example 3

A series of mutants of maltose binding protein, which each possess a SHAA probe at a unique site, is created. A his-tag at the C-terminus is used to orient the protein on a Ni-NTA-bearing lipid bilayer according to protocols known to those skilled in the art. The mutant proteins containing the probes are bound to the bilayer specifically via the His-tag to create a defined, oriented population of protein. The absolute tilt angle of the probe in each mutant is measured according to well-established procedures. The tilt angles of the probes measured by SHG are used to create a three-dimensional model of the side-chain orientation and, by modeling, the three-dimensional structure of the entire protein. This SHG determined structure can be then compared with the X-ray crystal structure to heuristically improve the modeling, if necessary. Addition of lactose will induce a conformational change in the protein that can be resolved as a motion picture in real space and real time by measuring the absolute tilt angle of the probe in one or more of the mutants.

The disclosure set forth above may encompass multiple distinct inventions with independent utility. The disclosure relates information regarding specific embodiments, which are included for illustrative purposes, and which are not to be considered in a limiting sense, because numerous variations are possible. The inventive subject matter of the disclosure includes all novel and nonobvious combinations and subcombinations of the various elements, features, functions, and/or properties disclosed herein. The following claims particularly point out certain combinations and subcombinations regarded as novel and nonobvious. Inventions embodied in other combinations and subcombinations of features, functions, elements, and/or properties may be claimed in applications claiming priority from this or a related application. Such claims, whether directed to a different invention or to the same invention, and whether broader, narrower, equal, or different in scope to the original claims, also are regarded as included within the subject matter of the inventions of the present disclosure.

REFERENCES

• (1) Salafsky, J. S. Journal of Chemical Physics 2006, 125, 074701.
• (2) Salafsky, J. S. Physical Chemistry Chemical Physics 2007, 9, 5704.
• (3) Heinz, T. F.; Tom, H. W. K.; Shen, Y. R. Physical Review A 1983, 28, 1883.
• (4) Salafsky, J. S.; Eisenthal, K. B. Journal of Physical Chemistry B 2000, 104, 7752.
• (5) Polizzi, M. A.; Plocinik, R. M.; Simpson, G. J. Journal of the American Chemical Society 2004, 126, 5001.
• (6) Paszti, Z.; Wang, J.; Clarke, M. L.; Chen, Z. Journal of Physical Chemistry B 2004, 108, 7779.
• (7) Clarke, M. L.; Wang, J.; Chen, Z. Journal of Physical Chemistry B 2005, 109, 22027.
• (8) Kriech, M. A.; Conboy, J. C. Using the intrinsic chirality of a molecule as a label-free probe to detect molecular adsorption to a surface by second harmonic generation Applied Spectroscopy, 2005; Vol. 59; pp 746.
• (9) Derrick, J. P.; Wigley, D. B. Nature 1992, 359, 752.
• (10) Sauereriksson, A. E.; Kleywegt, G. J.; Uhl, M.; Jones, T. A. Structure 1995, 3, 265.
• (11) Gronenborn, A. M.; Filpula, D. R.; Essig, N. Z.; Achari, A.; Whitlow, M.; Wingfield, P. T.; Clore, G. M. Science 1991, 253, 657.
• (12) Achari, A.; Hale, S. P.; Howard, A. J.; Clore, G. M.; Gronenborn, A. M.; Hardman, K. D.; Whitlow, M. Biochemistry 1992, 31, 10449.
• (13) Cohen, B. E.; McAnaney, T. B.; Park, E. S.; Jan, Y. N.; Boxer, S. G.; Jan, L. Y. Science 2002, 296, 1700.
• (14) Abbyad, P.; Shi, X. H.; Childs, W.; McAnaney, T. B.; Cohen, B. E.; Boxer, S. G. Journal of Physical Chemistry B 2007, 111, 8269.
• (15) Samanta, A.; Fessenden, R. W. Journal of Physical Chemistry A 2000, 104, 8972.
• (16) Weber, G.; Farris, F. J. Biochemistry 1979, 18, 3075.
• (17) Berkovic, G.; Shen, Y. R.; Marowsky, G.; Steinhoff, R. Journal of the Optical Society of America B-Optical Physics 1989, 6, 205.
• (18) Ong, S. W.; Zhao, X. L.; Eisenthal, K. B. Chemical Physics Letters 1992, 191, 327.
• (19) Heinz, T. F. Second-Order Nonlinear Optical Effects at Surfaces and Interfaces; Elsevier: Amsterdam, 1991.
• (20) Glauner, K. S.; Mannuzzu, L. M.; Gandhi, C. S.; Isacoff, E. Y. Nature 1999, 402, 813.
• (21) Cha, A.; Snyder, G. E.; Selvin, P. R.; Bezanilla, F. Nature 1999, 402, 809.
• (22) Majumdar, D. S.; Smirnova, I.; Kasho, V.; Nir, E.; Kong, X. X.; Weiss, S.; Kaback, H. R. Proceedings of the National Academy of Sciences of the United States of America 2007, 104, 12640.
• (23) Thomas, G. J. J. Annual Review of Biophysics and Biomolecular Structure, 1999; Vol. 28; pp 1.
• (24) Ishima, R.; Torchia, D. A. Protein dynamics from NMR Nature Structural Biology, 2000; Vol. 7; pp 740.
• (25) Weljie, A. M.; Yamniuk, A. P.; Yoshino, H.; Izumi, Y.; Vogel, H. J. Protein conformational changes studied by diffusion NMR spectroscopy: Application to helix-loop-helix calcium binding proteins Protein Science, 2003; Vol. 12.
• (26) Ghanouni, P.; Steenhuis, J. J.; Farrens, D. L.; Kobilka, B. K. Proceedings of the National Academy of Sciences of the United States of America 2001, 98, 5997.
• (27) Weitzman, C.; Consler, T. G.; Kaback, H. R. Protein Science 1995, 4, 2310.
• (28) Cohen, B. E.; Pralle, A.; Yao, X. J.; Swaminath, G.; Gandhi, C. S.; Jan, Y. N.; Kobilka, B. K.; Isacoff, E. Y.; Jan, L. Y. Proceedings of the National Academy of Sciences of the United States of America 2005, 102, 965.
• (29) Chen, C. K.; Heinz, T. F.; Ricard, D.; Shen, Y. R. Physical Review Letters 1981, 46, 1010.
• (30) Zimdars, D.; Eisenthal, K. B. Static and Dynamic Solvation at the Air/Water Interface Journal of Physical Chemistry B, 2001; Vol. 105; pp 3993.
• (31) Felderhof, B. U.; Bratz, A.; Marowsky, G.; Roders, O.; Sieverdes, F. Journal of the Optical Society of America B-Optical Physics 1993, 10, 1824.
• (32) Feller, M. B.; Chen, W.; Shen, Y. R. Physical Review A 1991, 43, 6778.
• (33) Xie, J. M. & Schultz, P. G. Innovation: A chemical toolkit for proteins—an expanded genetic code. Nature Reviews Molecular Cell Biology 7, 775-782 (2006).
• (34) Summerer, D. et al. A genetically encoded fluorescent amino acid. Proceedings of the National Academy of Sciences of the United States of America 103, 9785-9789 (2006).
• (35) Rodriguez, E. A., Lester, H. A. & Dougherty, D. A. In vivo incorporation of multiple unnatural amino acids through nonsense and frameshift suppression. Proceedings of the National Academy of Sciences of the United States of America 103, 8650-8655 (2006).
• (36) England, P. M. Unnatural amino acid mutagenesis: A precise tool for probing protein structure and function. Biochemistry 43, 11623-11629 (2004).
• (37) Cohen, B. E. et al. Probing protein electrostatics with a synthetic fluorescent amino acid. Science 296, 1700-1703 (2002).
• (38) Goh, M. C. et al. Absolute Orientation of Water-Molecules at the Neat Water-Surface. Journal of Physical Chemistry 92, 5074-5075 (1988).
• (39) Kemnitz, K. et al. The Phase of 2nd-Harmonic Light Generated at an Interface and Its Relation to Absolute Molecular-Orientation. Chemical Physics Letters 131, 285-290 (1986).
• (40) Hubbard, S. R., Mohammadi, M. & Schlessinger, J. Autoregulatory mechanisms in protein-tyrosine kinases. Journal of Biological Chemistry 273, 11987-11990 (1998).
• (41) Huse, M. & Kuriyan, J. The conformational plasticity of protein kinases. Cell 109, 275-282 (2002).
• (42) Nagar, B. et al. Crystal structures of the kinase domain of c-Abl in complex with the small molecule inhibitors PD173955 and imatinib (STI-571). Cancer Research 62, 4236-4243 (2002).
• (43) Noble, M. E. M., Endicott, J. A. & Johnson, L. N. Protein kinase inhibitors: Insights into drug design from structure. Science 303, 1800-1805 (2004).
• (44) Seeliger, M. A. et al. High yield bacterial expression of active c-Abl and c-Src tyrosine kinases. Protein Science 14, 3135-3139 (2005).
• (45) Sicheri, F. & Kuriyan, J. Structures of Src-family tyrosine kinases. Current Opinion in Structural Biology 7, 777-785 (1997).
• (46) Sicheri, F., Moarefi, I. & Kuriyan, J. Crystal structure of the Src family tyrosine kinase Hck. Nature 385, 602-609 (1997).
• (47) Xu, W. Q., Doshi, A., Lei, M., Eck, M. J. & Harrison, S. C. Crystal structures of c-Src reveal features of its autoinhibitory mechanism. Molecular Cell 3, 629-638 (1999).
• (48) Xu, W. Q., Harrison, S. C. & Eck, M. J. Three-dimensional structure of the tyrosine kinase c-Src. Nature 385, 595-602 (1997).

Claims

1. A method comprising: (a) preparing one or more mutants of a protein by incorporating one or more unnatural amino acid(s) that are second harmonic (SH)-active or sum frequency (SF)-active into the protein;(b) immobilizing the protein mutants on a surface so as to produce a net orientation of the one or more unnatural amino acid(s) relative to the surface;(c) measuring a tilt angle of the one or more of the unnatural amino acid(s) relative to the surface in the one or more protein mutants;(d) determining a structure of the protein using the tilt angle measured in (c).

2. The method of claim 1, further comprising exposing the one or more mutants of the protein to a second molecule.

3. The method of claim 2 wherein the second molecule is a ligand, a compound, a small molecule, a drug, a peptide, an inhibitor, or a protein.

4. The method of claim 1 wherein the one or more mutant proteins are arrayed on a surface.

5. The method of claim 1 wherein the structure is comprised of real space coordinates of the protein.

6. The method of claim 5 wherein the coordinates are of a particular atom, molecule, group of atoms, or group of protein molecules.

7. The method of claim 1, wherein the tilt angle is measured in real time.

8. The method of claim 1, wherein determining the structure of the protein comprises determining a conformational change of the protein.

9. The method of claim 8, wherein the conformational change is determined in real time and space.

10. The method of claim 9, wherein the conformational change is determined in real time and space simultaneously.

11. The method of claim 1, wherein determining the structure of the protein comprises determining structural changes for both buried and surface sites.

12. The method of claim 1, wherein the tilt angle is an absolute tilt angle relative to the surface.

13. The method of claim 1, wherein preparing one or more mutants of a protein comprise genetically incorporating a series of unnatural amino acids into different mutants of the protein.

14. The method of claim 1, further comprising determining a three-dimensional structure of the protein.

15. The method of claim 1, further comprising comparing the determined structure with a X-ray crystal structure of the protein.

16. The method of claim 1, wherein the unnatural amino acid is Aladan.

17. The method of claim 1, wherein the at least one of the one or more unnatural amino acid(s) are incorporated into a buried residue of the protein.

18. The method of claim 1, wherein the one or more unnatural amino acid(s) are incorporated into the protein site-specifically.

19. The method of claim 1, further comprising incorporation of a different unnatural amino acid into the protein.

20. The method of claim 19, wherein the different unnatural amino acid is not second-harmonic active.