WO2002082066A1 - Concentration and identification of moderate-to-strong hits in natural products by capillary electrophoresis - Google Patents

Abstract

A method of increasing the concentration of any candidate hit compound (12) in a natural sample that binds to a selected target during capillary electrophoresis, said method comprising the steps of: filling the capillary (10) of a capillary electrophoresis apparatus with a running buffer containing a natural sample; injecting a plug of the selected target (16) into said capillary (10), wherein the concentration of said target (16) is in excess of the concentration of any candidate hit compound (12) in said natural sample; subjecting said target plug to capillary electrophoresis; tracking migration of any said target bound to hit compound (20) at a detection point; collecting said target bound to hit compound detected at said detection point; and isolating said hit compound separate from said target.

Description

TITLE OF THE INVENTION

Concentration and Identification of Moderate-to-Strong Hits in Natural Products by Capillary Electrophoresis

CROSS REFERENCE TO RELATED APPLICATIONS This application claims the benefit under 35 U.S.C. §119 (e) of U.S. Provisional Patent Application No. 60/283,988 filed March 28, 2001, entitled, ON-LINE CONCENTRATION AND IDENTIFICATION OF MODERATE-TO-STRONG HITS IN NATURAL PRODUCTS BY CE-MS, the whole of which is hereby incorporated by reference herein.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR

DEVELOPMENT N/A

BACKGROUND OF THE INVENTION The isolation and characterization of potential drug lead compounds from crude natural extracts (e.g., fermentation broths, plant extracts, microbial extracts) is a complex and time- consuming procedure. This has led to a decreased interest by the pharmaceutical industry in pursuing natural products for new drug compounds . Once an extract containing a potential hit has been identified in a primary screen, the long, arduous task of isolating sufficient hit material for further characterization begins. Typically, this involves scale-up production of more extract (e.g., via fermentation or plant growth), followed by several cumbersome fractionation and purification steps. The isolation process can involve several sequential procedural steps such as liquid-liquid extraction, solid-phase extraction, countercurrent chromatography, and high performance liquid chromatography. With each fractionation step, material losses occur and thus hit compounds in low concentration may be lost . After sufficient hit compound has finally been isolated and purified, it is then typically subjected to structural analysis using a combination of techniques such as mass spectrometry (MS), nuclear magnetic resonance (NMR) , and ultraviolet (UV) spectral analysis. The whole process can take weeks to months. An additional disadvantage for natural products is that previously known, uninteresting compounds are often re-discovered through this process, resulting in a tremendous waste of time and resources. Thus, it would be useful to improve the efficiency of isolation and structural characterization of hit compounds in natural product extracts .

BRIEF SUMMARY OF THE INVENTION The present invention relates to a method of isolating, concentrating, and characterizing moderate-to-strong binding compounds present in complex biological material that bind to a target (e.g., protein, nucleic acid, protein/nucleic acid complexes, other cellular components) of interest using an initial capillary electrophoresis separation and affinity step followed by a detection analysis. The described method is particularly useful for identifying compounds having an equilibrium dissociation constant (K_d) of less than 100 nM (moderate-to-strong hit compounds) . The method results in the isolation and concentration of enough pure compound for dereplication (i.e., obtaining sufficient structural information to determine whether the compound, or "ligand", has previously been discovered and of an interesting structural class) . Exemplary detection methods include, but are not limited to, mass spectroscopy (MS) , nuclear magnetic resonance (NMR) spectroscopy, photodiode array (PDA) detection, light scattering detection, infrared detection, liquid chromatography in association with a subsequent, different detection method (e.g., LC/MS, LC/PDA or LC/NMR) or ultra-violet spectroscopy. BRIEF DESCRIPTION OF THE FIGURES Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof and from the claims, taken in conjunction with the accompanying drawings, in which:

Fig. 1 is a diagram representing movement of a target plug as it progresses through a capillary containing a mixture of potential ligands, or "hit compounds," and inactive mixture components in electrophoresis buffer during a capillary electrophoresis run according to the method of the invention;

Fig. 2 is a description of exemplary steps used to carry out the method of the invention;

Figs. 3A and 3B are histograms showing the results of carrying out the method of the invention to purify and concentrate a strong ligand Methazolamide (MZ) (3A) and a weak ligand Carzenide (CL) (3B) using the target carbonic anhydrase. The results show about a 100-fold increase in the concentration of MZ and about a 2.4-fold increase in the concentration of CL; and Figs. 4A-4C show HPLC chromatographic detection (4A, 4B) and MS/MS analysis (4C) of methotrexate (MTX) purified and concentrated from a crude natural extract according to the method of the invention using the target protein dihydrofolate reductase (DHFR) .

DETAILED DESCRIPTION OF THE INVENTION

The principles of the method of the invention is directed to isolating, concentrating and characterizing moderate-to-strong hit compounds present in complex biological materials and/or natural samples (NS) according to the invention are illustrated diagrammatically in Fig. 1.

As illustrated in Fig. 1, in an exemplary experiment, the capillary 10 of a capillary electrophoresis (CE) instrument is filled with running electrophoresis buffer containing a natural sample comprising a potential ligand 12 and inactive mixture components 14 dispersed throughout. The natural sample can be included in the reservoir buffer chambers as well. A plug of target protein sample 16 is injected into capillary 10, the capillary is inserted into the reservoir buffer chambers (not shown) , voltage is applied across the capillary, and the target migrates down the capillary (in a target zone) . As the target migrates, individual target molecules bind to any ligands (potential hit compounds) present in the electrophoresis buffer sample mixture. As electrophoresis continues, the target zone 18 accumulates and concentrates ligands as they bind and, consequently, move with the target. Only a small portion of the target protein binds a ligand at any particular moment in time during migration, as the protein is present in excess of the ligand in the vicinity of the target zone (e.g., 5 μM protein and 1 nM ligand) . A large excess local concentration of the protein target, e.g., about 10 to 50 μM, over the concentration of possible hit compounds will drive the equilibrium of any binding reaction toward the formation of complex and, thus, complete association of the ligand, or "strong hit." In the next moment, the target zone migrates further, and the remaining free, or unbound, target protein is exposed to a new portion of strong hit in the buffer. Consequently, more protein/strong hit complex 20 can form. As a result of these multiple events, the strong hit will be concentrated in the electrophoretic zone containing the target protein. In fact, the strong hit will be spread in this zone between two extreme sub-zones: sub-zone containing the target/hit complex formed in the initial event of target-hit binding and sub-zone of free unbound target, i.e., the portion of the target that never associated with a hit compound or from which a hit compound has dissociated. A weak-binding hit compound, if present in the same sample, will not be concentrated, or will be concentrated much less, with the target protein in its electrophoretic zone as any weak hit/protein complex dissociates very rapidly due to the fast off-rate of the weak hit. Free or unbound target protein then moves away from the dissociated weak hit and interacts with a new portion of the sample, containing additional molecules of the strong hit, present in the running buffer. Significant accumulation will occur for ligands of approximately K_D = 100 nM or higher affinity. Weaker ligands (K_D = 1 - 100 μM) will not accumulate extensively in the target zone. Very weak ligands (K_D > 100 μM) and inactive compounds will not accumulate at all. This is because the off-rates of the weaker affinity ligands are too high to form stable complexes and concentrate in the migrating target zone.

For successful concentration of the strong hit in the migrating target zone, the hit compound's electrophoretic mobility should be different from the mobility of the target protein. (For example, the hit compound may have the same charge as the target so that its mobility is in the same direction as that of the target, but this mobility should then be slower or faster than that of the target.) That way, there is constant exposure/re- exposure of the target zone to "fresh" sample in which there may be molecules of strong hit compounds as yet unbound to the target. If the strong hit migrates in a direction opposite (i.e., has an opposite charge) to that of the target protein, the concentration of the strong hit accumulating in the target zone will be smaller than it would be in the case where the hit and target migrate in the same direction if there is no natural sample in the outlet end buffer reservoir to feed buffer containing additional sample into the capillary. However, if natural sample is in both buffer reservoirs of a CE apparatus during an electrophoresis run, sample and any strong hit can migrate out of the outlet reservoir into the CE capillary toward the target protein, replacing any strong hit that has migrated past the target. This modification would require twice as much running buffer containing sample in each run (in terms of volume) and, therefore, will double the amount of natural sample required.

As illustrated in Fig. 2, an intended target molecule is subjected to electrophoresis as described above and accumulates a potential hit compound in the natural sample as it migrates through the electrophoresis buffer in the capillary (Step 1) . The run is stopped when the target (protein) passes a detector window

(Step 2) . This enables the operator to know where the target zone resides in the capillary. The capillary is then carefully removed from the electrophoresis unit and cut into several small segments (Step 3) depending on the width and position of the target peak(s) for collection, keeping track of the segment containing the target zone with the target and any bound ligand. In one aspect, the position of the target peak(s) may be preliminary determined by performing an exploratory run of the target. In another aspect, other collection methods may be used, for example, by collecting the sample from the end of the capillary (retaining the capillary whole) for an off-line analysis. The segment containing the target and any bound ligands is then flushed, e.g., with HPLC buffer into a fresh tube (Step 4) . This material is then analyzed by, e.g., HPLC-UV to quantitate the amount of ligand present (Step 5) , although other detection methods may be used. The HPLC can be performed with an organic mobile phase to denature the protein and release bound ligands, which are detected by UV absorbance . The amount of ligand is quantified by comparing the HPLC peak area with known standards.

The method of the invention significantly improves the efficiency of isolating small quantities of pure compounds, or ligands, from natural extracts for further analysis. For further characterization, dissociation of the ligand from the target can either occur on-line or off-line. In order to assist complex dissociation and detection of the hit, one can provide a liquid sheath at the capillary electrophoresis-mass spectrometer (CE-MS) interface for example, containing a small amount of organic (e.g., 50% methanol) and weak acid (e.g., 1% acetic acid). Other combinations of organic chemicals and weak acid may be used, for example, TFA, acetonitrile, acetone or ammonium acetate. CE-MS can be applied to a crude NS or a pre-fractionated sample. For example, where the presence of a hit has been previously- detected in a crude sample by any screening assay, that crude sample can be fractionated and the various fractions can be subjected to the method of the present invention.

Once isolated, a number of different detectors can be used to obtain physical and structural properties of the hit compound (dereplication) . Exemplary detectors include, but are not limited to, mass spectroscopy (MS) , nuclear magnetic resonance (NMR) spectroscopy, photodiode array (PDA) detection, light scattering detector, infrared detector, liquid chromatograph (LC) /MS, LC/PDA, LC/NMR or ultra-violet spectroscopy. For example, with today's highly sensitive mass spectrometers, only a few nanograms of hit material may be necessary to get mass data, including the molecular formula. In most cases, mass spectrometry will be used to get mass and/or mass substructure (MSⁿ) on the initial purified material. MS (or MSⁿ) analysis can be performed using direct infusion, High Performance Liquid Chromatography/Mass Spectrometer (HPLC/MS), or a CE/MS interface.

For direct infusion, some protocol must be used to dissociate the hit from the target/hit complex prior to MS analysis. In some cases, the MS spray will be adequate to dissociate the complex. Some denaturant may need to be added to disrupt strong complexes.

HPLC analysis gives additional information of the hit compound such as polarity. The mobile phase utilized in many HPLC protocols will usually be sufficient to dissociate the complex due to the presence of organic and acidic additives. PDA detection is routinely used online with HPLC to obtain a UV/Vis spectrum on the hit compound, which is useful for identifying chromophores . It is possible to interface CE with a MS for direct analysis of the compound after the CE run. Complex disruption can be achieved by using a sheath flow containing denaturing solution at the interface of CE and the MS. The detection limits of modern mass spectrometers are on the order of about 0.1 μM concentration. The use of an electrospray ionization-time-of-flight (ESI-TOF) mass spectrometer provides this type of sensitivity, and also mass accuracy to about 5ppm. Once the hit's molecular weight has been determined, the same experiment can be repeated on an ion trap mass spectrometer (which is very sensitive when used in MS/MS mode) , in order to obtain MS/MS or deeper fragmentation spectra.

In cases where the method has isolated greater than 10 micrograms of hit compound, NMR can be used to obtain additional structural information. Again, the NMR can be interfaced with HPLC if an initial separation step is required prior to NMR analysis. The stability of the protein/hit complex in the capillary can be increased by lowering the capillary temperature during CE. The temperature range can be from about 0°C to about 60 °C, preferably from about 5°C to about 37 °C. By repeating the CE runs at incrementally lower temperatures, more hits of lower affinities can be concentrated in the target protein's electrophoretic zone. For example, under the most stringent CE conditions (for example, temperature of about 37°C) , only a strong hit is detected and identified. Under less stringent CE conditions (for example, temperature of 20°C) , any hit of moderate-to-tight binding strength can be isolated and concentrated by the method.

To change the concentration factor, for example, using longer (e.g., 1 m length) and wider (e.g., 100 μm wide) capillaries can increase the concentration of a protein-ligand complex because more natural sample can be used and thus more hit accumulates in the target zone. A larger volume of NS (e.g., 7.85 μL) can be introduced into the capillary, while the target zone remains the same (e.g., lOnL) . Additionally, in the case of a charged strong hit, the concentration factor will be larger, as more of the strong hit will be migrating into the CE capillary from the buffer reservoirs.

Prior to the performance of the method of the invention, capillary electrophoretic conditions must be optimized to detect the protein and, possibly, a new protein peak representing protein/ligand complex. This can be done by preliminary exploratory runs with and without natural sample. The information from these runs is used to determine when to collect the protein peak(s), e.g., where to cut the capillaries, or when to collect fractions from the capillary.

In the capillary electrophoresis run of the target and the buffer containing the natural sample with potential ligands, the target protein and the target/ligand complex may generate different results since they may potentially have different mobility rates. A preliminary run of the target is useful to determine the migration and mobility profile of the target. Following the method of the invention, the area where there is a shift in the mobility profile of the target indicates the presence of hit compounds where one can collect for further analysis. If a shift is not present, then the area where a target is detected is collected for further analysis.

The present method requires the use of at least one buffer, but may require two: (a) a sample buffer and (b) a running buffer. In most cases these buffers will be different but they may be the same buffer. The "sample buffer" is a pH-balanced solution for preparing samples of the target used to isolate a hit compound. The "running buffer" is a pH-balanced solution used in the capillary electrophoresis apparatus and contains the natural sample. Examples of sample buffers and running buffers include Good's biological buffers (e.g., TES, CAPSO, etc.) or Tris based buffers.

One adjusts the sample and running buffers to each have a

3. suitable pH for preserving the target's functional activity, e . g. , binding activity, and allowing binding of the target to its ligand. As well, the running buffer's pH value should produce suitable capillary electrophoretic profiles of the target alone and/or the target/ligand complex. The running buffer should ideally allow the unbound target to produce a detectable peak when subjected to capillary electrophoresis, within a reasonable time period (e.g., preferably under 10 minutes). Each buffer solution may include appropriate additives, as needed. Suitable buffers are well-known in the art to one of ordinary skill. Various capillary parameters may also be adjusted to allow optimal capillary electrophoresis conditions for a selected target molecule and its ligand. Some capillary dimensions or factors that may be optimized include, but are not limited to: capillary size or diameter; capillary temperature; capillary length; inner coatings for the capillary, if necessary; and any capillary pretreatment, if necessary.

A preferred capillary inner diameter is within the range of about 10—500 microns, preferably within about 25—100 microns. The capillary length will depend on the amount of time needed for obtaining good capillary electrophoretic profiles of the selected target and/or the target/ligand complex. Longer or narrower total capillary lengths can be used to improve resolution. Longer capillaries will increase the total amount of ligand isolated by the method because there is more natural sample present. Typically, a preferred capillary length is within a range of about 0.5 cm to about 1 meter, most preferably within about 0.5 cm to 40 cm. Optionally, the inner wall of the capillary may be coated with a polymer, a polymer blend, or other suitable material. The inner capillary coating may serve to minimize any electrostatic charge on the capillary wall and to diminish adsorption of a selected target, a ligand, or the target/ligand complex to the capillary wall. As well, the coating may also be pre-treated as needed. For instance, it may be pre-treated with a non-specific protein, such as bovine serum albumin (BSA) , to help prevent target adsorption.

CE may also be carried out in capillaries in the form of open grooves or channels in a planar surface such as a fused silica or polymer microfabricated device or microchip. The migration of the tracked target molecule is followed typically by the use of an on-column detector aligned with a small window etched into the capillary. Alternatively, it is possible to scan the entire capillary. Preferred detection methods use UV absorbance, UV or laser-induced fluorescence, and visible light absorbance. Other on-column detection methods may also be used. As well, one may use on-line detection instruments coupled with the capillary electrophoresis apparatus, which use radionuclide, fluorescence polarization, NMR, mass spectrometry, electrochemical detection and other methods . The detection variable for direct detection can be absorbance at 210 or 280 nm for most proteins and 260 nm for nucleic acids. Indirect detection uses laser-induced (or other) emission of mainly visible wavelengths from dye-labeled target molecules which give high sensitivity. Preferred are fluorescently labeled molecules. Non-limiting examples of fluorescent dyes include fluorescein, rhodamine, tetramethylrhodamine, Texas Red and ethidium bromide. It must be kept in mind, however, that these labels can influence the overall charge on the target molecule and may affect its binding capability. Examples of UV sources and lasers include: deuterium, xenon and mercury lamps; argon, Ar/Kr, HeCd, HeNe, XeCl, KrF, nitrogen and solid state lasers. Some target molecules such as carbohydrates and small molecules, may require pre-capillary derivatization to be detected. Prior to carrying out the isolation method on a natural sample, one determines the most appropriate means of detecting the chosen target and/or target/ligand complex during capillary electrophoresis; i.e., one determines a particular wavelength or other parameter at which the target and/or target/ligand complex is optimally detectable.

The capillary electrophoresis process is adjusted to produce the optimal electrophoretic profiles for the unbound target and, possibly, a target/ligand complex. The profiles, when superimposed, may display at least two distinct peaks corresponding to the selected target molecule alone and to the target when bound to the selected target/ligand complex.

Some electrophoretic parameters to be optimized include, but are not limited to: time, voltage, current and temperature. One can adjust the time or duration of the electrophoretic run so that it is possible to estimate when the target will be detected, alone or in complex with its ligand, at a specific detection point in the capillary electrophoresis apparatus. The detection point may constitute at least one window in the capillary, at which is placed a detector. For instance, in the case of a fluorescently labeled target, there may be a fluorescence detector and an ultraviolet or laser light source to cause fluorescence. The capillary electrophoresis procedure may be set to run for up to 2 hours or even longer, as needed. Typically, but not always, ligands of a particular binding strength have certain similar characteristics. "Moderate-to-tight binding" ligands (MTBL) and "weak-binding" ligands have faster off-rates (K_off) and higher dissociation constants (K_D) than "tight-binding" ligands and form target/ligand complexes that will not accumulate in the target zone during electrophoresis. In contrast, tight-binding ligands have lower dissociation constants and slower off-rates, forming stable target/ligand complexes that remain bound to the target and accumulate in the target zone during electrophoresis as they migrate past a detector during capillary electrophoresis. The characteristics of these ligand groupings are outlined in Table 1.

TABLE 1

Natural samples including, but not limited to, any pure, partially pure, or impure sample that contains complex biological material is considered an appropriate sample to be analyzed by the method of the invention. "Complex biological material" is intended to include any mixture of compounds that may contain compounds that are potentially useful in a biological system, e. g. , whether human, other mammalian, or agricultural. For example, large chemical libraries are frequently generated by combinatorial chemistry to enable investigators to screen extremely large numbers of chemical compounds for potential therapeutic or diagnostic purposes. These libraries can be, in essence, modified biological scaffolds and can be screened advantageously by the method of the invention. Particularly suitable as exemplary natural samples are extracts of terrestrial and marine plants, cells from higher animals including humans, eubacteria, actinomycetes and other bacteria, extracts from non-recombinant or recombinant organisms, microbial fermentation broths, both filamentous and non-filamentous fungi, protozoa, algae, archaebacteria, worms, insects, marine organisms, sponges, corals, crustaceans, viruses, phages, tissues, organs, blood, soil, sea water, water from a fresh-water body (e.g., lake or river), humus, detritus, manure, mud, and sewage or partially pure fractions from isolation procedures performed on any of these samples (e.g., HPLC fractions) . The natural sample may be one that is harvested from the environment and/or cultured under suitable environmental conditions (growth medium, temperature, humidity) . Preferably, the harvested sample is simply diluted to the extent necessary to practice the method of the invention. ^• However, if necessary, the sample material can be treated by any combination of standard processes used by those skilled in the field to prepare the sample for analysis. For example, the crude sample may be subjected to a preliminary treatment such as freeze-thawing, homogenization, sonication, heating or microwave extraction to break down cell walls. The sample could be heated at, e . g. , 50°C for 10 minutes to inactivate destructive enzymes. Non-specific proteins may be added to prevent destruction of proteinaceous targets by heat-resistant proteases. Extraction of cells or culture media with various solvents — such as ethyl acetate, dimethylsulfoxide, ethanol, methanol, ether or water — can be carried out, followed by filtration to remove particulate matter and/or high molecular- weight compounds . The natural sample may also be fractionated by centrifugation, sequential extractions, high pressure—liquid chromatography, thin—layer chromatography, counter—current chromatography, and/or other chromatography techniques before use in the method of the invention. Various fractions of a positive sample may be tested to help guide the detection and isolation of active compounds by the method of the invention.

Finally, the sample may be diluted in aqueous or non-aqueous solution, which may contain salts and buffers such as sodium chloride, sodium citrate or Good's biological buffers. For most samples, the dilution step is required and preferably is the only treatment. However, dilution can also be performed as a final procedure after one or more of the preceding steps. A dilution of about 1:10 to about 1:200 (vol. /vol.) of the original complex biological material sample is usually preferred to achieve reproducible results in the method of the invention. Additional dilution factors may be desirable.

Due to the high resolving power of capillary electrophoresis

(CE) , the target sample may be purified, partially purified, or even unpurified ( e. g. , as in a bacterial extract), as long as the target and/or ligand/target complex give(s) a discernible CE peak. Any molecule that is implicated in a disease process is a potential target. Furthermore, the potential target may be any compound useful in diagnosing a specific condition. Additionally, other categories of target molecules can be contemplated. For example, in the agricultural arena, the target could be a molecule representing an essential function of an insect pest. Examples of target molecules that may be used in the method of the invention include: proteins, nucleic acids, carbohydrates, and other compounds . Some examples of therapeutic target molecules are included in Table 2:

TABLE 2

Mol cular Target Associated Disease (s)

HIV reverse transcriptase AIDS

HIV protease AIDS Carbonic anhydrase Glaucoma

Tubulin Cancer

Thrombin Blood clots

HMG-CoA reductase High cholesterol

Elastase Emphysema, Rheumatoid arthritis

Cyclooxygenase Inflammation p56, p59 tyrosine kinases Cancer

Topoisomerases Cancer

Dihydrofolate reductase Cancer Other examples of appropriate molecular targets include DNA or RNA (used to search for nucleic acid—binding proteins, transcription factors, etc.) ribosomes, cell membrane proteins, growth factors, cell messengers, telomerases, elastin, virulence factors, antibodies, replicases, other protein kinases, transcription factors, repair enzymes, stress proteins, uncharacterized disease-related genes and/or their RNA and protein products, uncharacterized disease-related regulatory DNA or RNA sequences, lectins, hormones, metabolic enzymes, proteases and toxins. This definition also includes any subcomponent of the listed molecules, such as protein subunits, active peptide domains of therapeutic proteins and active regions of small molecules. The target molecule may be chemically, enzymatically, or recombinantly altered to improve its electrophoretic properties ( e. g. , deglycosylated) or subjected to fluorophore or polyion addition to facilitate its separation and/or detection during CE.

The target should be detectable during capillary electrophoresis, as unbound target and/or as target complexed with a moderate- to tight-binding ligand. For instance, it may be detectable by observation of its ultraviolet (UV) or other light absorbance properties, or its fluorescence properties. One may label the target with a detectable tag, such as a tag of a fluorescent or other dye, a radio-label, a chemical tag or other marker. For example, a fluorescently labeled target may be detected by ultraviolet light absorption detection (typically having a micromolar detection limit) or, more preferably, by laser-induced fluorescence detection (typically having a picomolar to low nanomolar detection limit) . An additional advantage of a fluorescent tag is the selectivity provided, particularly in complex samples that may have many UV-absorbing compounds present. The need for a detectable tag, and the type used, will depend on the nature of the target molecule. Proteins and peptides may be labeled by, e . g. , amino labeling of lysine residues or sulfhydryl labeling of cysteine residues. Nucleic acid species and polynucleotides may be labeled by incorporating a labeled nucleotide in an in vitro synthesis reaction. Methods of labeling various targets are well-known in the art.

If desired, one may confirm prior to practicing the method of the invention that a modified target, e . g. , a fluorescently labeled target, retains its functional activity. That is, one can confirm that the labeled target retains a functionally active site by using any available, well-established functional or binding assay whose result depends on a functionally active target.

The following examples are presented to illustrate the advantages of the present invention and to assist one of ordinary skill in making and using the same. These examples are not intended in any way otherwise to limit the scope of the disclosure.

EXAMPLE I

Accumulation of Ligands in Capillary Electrophoresis by the Target

Protein Carbonic Anhydrase II

As shown in Figs. 3A and 3B, the concentration of ligands was performed using the target protein carbonic anhydrase II, which is a target for glaucoma, a serious eye disease. Two ligands were tested to compare the extent of accumulation of a strong ligand and a weak ligand.

In Fig. 3A, the strong ligand methazolamide (MZ, k_d = 20 nM) was present in the capillary electrophoresis buffer and the inlet reservoir buffer chamber at an initial concentration of 100 nM.

The total amount of MZ in the capillary and reservoir chamber can be calculated as follows: Total MZ = [MZ] (Volume) = [MZ] (lπr²) = 11.9 ng, where [MZ] is the initial concentration of MZ (100 nM = 23.6 ng/mL) ; (Volume) is the volume of the capillary and of the inlet reservoir chamber containing MZ (3.5 μL + 500 μL = 503.5 μL) ; (1) is the capillary length (20 cm) ; (r) is the capillary radius (75 μm) ; and the molecular weight (MW) of MZ is 236.27 Daltons and the MW of carbonic anhydrase II is 29,000.00 Daltons.

The target protein carbonic anhydrase II was at a concentration of 50 μM injected in a total volume of 10 nL. Therefore, the total amount of protein in this injection plug is

14.5 ng. 14.5 ng of protein has the capacity to bind 118 pg of MZ

(assuming a one-to-one binding stoichiometry) , so there is more than enough MZ (11.9 ng) in the capillary and the inlet buffer chamber to saturate the target. After electrophoresis, the capillary was cut into equal 0.80 cm segments that were individually flushed with HPLC buffer. The eluants were then analyzed by High Performance Liquid Chromatography using an ultraviolet detection (HPLC/UV) to quantitate the concentration of MZ . The histogram shows the concentration of MZ in several of the segments. The negative value segments (-4, -3, -2, -1) represent the four capillary segments just before the target zone segment

(the protein has migrated through these segments). The "0" segment is the target zone segment that contains the protein and any bound

M . The positive value segments (+1, +2 , +3, +4) represent the four capillary segments just after the target zone (the protein has not migrated through these segments) . Other segments are not shown. Under the conditions used, MZ migrates in the same direction as the protein and has a higher electrophoretic velocity than the protein, so MZ constantly migrates through the target zone and is replenished from the inlet buffer reservoir. This is the reason MZ was included in the inlet reservoir only. For isolating unknown compounds, the test sample can be placed in both the inlet and outlet reservoirs. The capillary segment "0" containing the protein had an MZ concentration of 2.36 μg/mL or 10.0 μM, representing over a 100- fold increase in concentration relative to the starting concentration of 100 nM in the electrophoresis buffer. The total mass of MZ in segment "0" is calculated to be about 330 pg. (If this were an unknown compound, this mass would probably not be enough to perform dereplication, so multiple, identical runs could be run and the samples pooled and concentrated for analysis.) All the "negative" segments contain approximately 100 nM MZ because the MZ is constantly replenished from the inlet reservoir buffer during the run. All the segments after the protein plug were devoid of MZ because MZ itself migrated towards the outlet, moving away from target due to its high velocity and thus never contacting the target during the experiment. This method greatly simplifies the process of concentrating and purifying binding ligands from complex mixtures such as natural extracts for subsequent analysis by a variety of techniques.

As shown in Fig. 3B, a similar experiment was performed with the same target but with a weaker ligand, carzenide (CL, k_d = 1 μM) . CL was present at a concentration of 10 μM in the original electrophoresis buffer. All other parameters of the experiment were the same as with the MZ experiment. The result, shown in the histogram, was a concentration factor of only 2.4-fold in the segment containing the target zone, i.e., the concentration of CL was about 24 μM in the "0" segment. This lower concentration factor is to be expected due to the weaker binding strength, and higher off-rate, of the CL ligand.

Based on the foregoing experiments, the ability of the method to purify and concentrate ligands was shown. Preferably, the method of the invention comprises purifying and concentrating higher affinity ligands. The method eliminates the need for extensive fractionation and purification procedures that are time- consuming and cost inefficient. EXAMPLE II

Affinity Extraction and Isolation of Methotrexate from Crude, Inactive Actinomycete Extract Using the Target Dihydrofolate Reductase

In Figure 4, the method was used to extract a high affinity ligand spiked into an inert natural extract (NE) , in this case, a crude, inactive actinomycete extract. Methotrexate (MTX) is a high affinity (k_d = low nM) antitumor agent that binds to the protein dihydrofolate reductase (DHFR) . While the experimental setup was similar as previously described, an inert natural extract (NE) spiked with MTX was used to see if the method could extract MTX from a complex mixture. The NE was present at a concentration of 1 mg/mL in the capillary and both reservoir buffer chambers. MTX was spiked present in the electrophoresis buffer at a concentration of 20 ng/mL.

The experiment was run using 20 μM of bovine DHFR in the injection plug. Fig. 4A shows an HPLC chromatogram (UV_2X4) of the electrophoresis buffer containing the NE spiked with MTX prior to the experiment. As expected, it is a very complicated profile due to the presence of multiple extract components. The MTX peak is not discernable because of its low concentration combined with the complex profile. Fig. 4B shows an HPLC chromatogram run under the same conditions but on the material eluted from the capillary segment containing the protein and any bound ligands . The peak at 11.377 minutes represents MTX. The earlier peaks in the chromatogram are some contaminating components from the NE.

This experiment demonstrates the ability of the method to concentrate and purify ligands present at low concentrations from complex samples such as natural extracts. Mass spectroscopy/mass spectroscopy (MS/MS) analysis was performed (Finnegan LCQ) on the 11.377-minute peak to absolutely confirm the identity of the MTX (MW = 455.1791 [M+H⁺] ) plus molecular fragment (shown in Fig. 4C) . When isolating an unknown ligand, the ability to do MS and MS/MS analysis would be critical at this stage to help identify its structure.

EXAMPLE III The Method Using ^ΛΛIn-Capillary Protein Immobilization"

Exemplary starting conditions include the following: Concentration of original natural sample (NS) is about 25 mg/mL (the amount of NS is 1 mg in 40 μL) . The concentration of an exemplary strong hit present in the NS is at about 50nM or 25ng/mL (M.W. 500Da) . The capillary dimensions for capillary electrophoresis are 50 μm diameter and 60 cm length, leading to a total volume of 1.1 μL.

The CE capillary was filled with running buffer containing NS. An aliquot of the target protein sample was injected into the CE capillary as a narrow plug.

The target protein concentration is 5 μM, in a total sample volume of 3 μL. The NS concentration in the running buffer is about 2% or a 50-fold dilution. The concentration of a strong hit test compound in the running buffer is 1 nM hit. The total volume of the running buffer with the NS is 200 μL (100 μL on each side of the capillary) .

The electrophoresis is performed with controlled electroosmosis. This means that the CE conditions are optimized such that the electroosmotic flow migrates in a direction opposite to that of the electrophoretic migration of the target protein. By manipulating the velocity of the electroosmotic flow (e.g., by eliminating or modifying the polymeric coating of the capillary inner walls and by choosing appropriate ionic strength of the running buffer and/or pH of the running buffer in a manner well- known to an ordinary skilled artisan (see U.S. Patent No. 6,299,747, the whole of which is hereby incorporated by reference) ) , one can achieve very slow migration of the protein through the capillary. This allows more natural sample to flow through the slow-moving target and thus increasing the amount of strong hit from the natural sample coming into contact with the target.

Any strong hit present in the running buffer will be concentrated with the target protein in its zone by two phenomena. The first is when the migrating target zone continuously encounters " fresh" NS-containing buffer and any unbound, strong hit capable of binding to unbound target protein. The second phenomenon is enrichment of the target zone during slow electrophoresis for any strong hit in NS-containing buffer that is pumped from the outlet buffer reservoir by electroosmotic flow. The electroosmotic flow will carry neutral molecules and molecules that have the charge opposite to the protein into the capillary and through the target zone. The molecules of the same charge as the protein but lower mobility than the protein also will be carried with the electroosmotic flow. The small molecules of the same charge as the protein but higher mobility than the protein will be coming from the inlet buffer reservoir and be enriched when reaching the target zone. Example calculations are presented. The amount of the strong hit concentrated in the target zone by the first phenomenon is 1.1 fmol (1 nM in 1.1 μL capillary volume). The electroosmotic flow velocity is 100 nL/min. The amount of the strong hit migrating with electroosmotic flow through the target zone is 1 fmol/μL or 0.1 fmol/min. The amount of the strong hit concentrated in the target zone during one hour of electrophoresis is 6 fmol (0.1 fmol/min x 60 min) . Total amount of the strong hit accumulated in the target zone is about 7.1 fmol. Thus, the concentration of the strong hit in the target zone is 0.71 μM. After "preconcentration" of any strong hit with the target zone during CE, the target zone can be collected for analysis by several means. The capillary may be cut and the target and bound ligands eluted as in the previous examples. Alternatively, the target zone can be pushed from the capillary into a collection chamber using pressure and then analyzed off-line. In another method, the outlet end of the capillary is placed in a CE-MS interface, and the target zone and any hit contained therein are then allowed to migrate out of the CE capillary and enter into a mass spectrometer.

The advantage of this embodiment is that the protein has a very slow migration in the capillary as the sample migrates through it, so one should be able to pass much more sample across the target than would be possible in the embodiment of Example I. This would be advantageous for ligands present at very low concentrations. Uncoated capillaries are usually used because an electroosmotic force (EOF) is required, so the protein must behave in an uncoated capillary environment (no protein adsorption to walls) . Also, the EOF and electrophoretic mobility of the target must be carefully controlled so that they nearly offset each other and this can be technically difficult and require significant optimization.

While the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art in light of the foregoing description. Accordingly, it is intended that the invention shall be directed to all such alternatives, modifications and variations as they fall within the spirit and broad scope of the appended claims.

Claims

CLAIMS What is claimed is:

1. A method of increasing the concentration of any candidate_. hit compound in a natural sample that binds to a selected target during capillary electrophoresis, said method comprising the steps of:

(a) providing a capillary electrophoresis instrument comprising a capillary having a detection point; (b) providing a running buffer containing a natural sample;

(c) providing a plug of a selected target, wherein the concentration of said target is in excess of the concentration of any candidate hit compound in said natural sample;

(d) injecting the plug of the selected target into said capillary;

(e) subjecting the target plug to capillary electrophoresis ;

(f) tracking migration of any said target bound to hit compound at the detection point; (g) collecting said target bound to hit compound detected at the detection point; and

(h) isolating said hit compound separate from said target.

2. The method of claim 1, wherein said steps (g) and (h) comprise the steps of:

(g) stopping the capillary electrophoresis run when said target bound to hit compound passes the detection point;

(h) removing the capillary containing the target and any bound hit compound from the capillary electrophoresis instrument; (i) cutting the capillary into several segments; and

(j) collecting said target with any bound hit compound from the capillary segment containing said target with any bound hit compound; and (k) isolating the hit compound separate from the target.

3. The method of claim 1, wherein said method further comprises the step of first determining a mobility profile of said target.

4. The method of claim 1, wherein said concentration of said target is about 10 to about 50 μM.

5. The method of claim 1, wherein said hit compound has an equilibrium dissociation constant (K_D) of less than 100 nM.

6. The method of claim 1, wherein said method further comprises the step of analyzing said hit compound for physical and structural properties .

7. The method of claim 6, wherein said analysis comprises any method selected from the group consisting of mass spectroscopy; nuclear magnetic resonance spectroscopy; photodiode array detection; light scattering detection; infrared detection; liquid chromatography in association with a subsequent, different detection method; or ultra-violet spectroscopy.

8. The method of claim 6, wherein said analysis comprises using High Performance Liquid Chromatography with ultra-violet detection.

9. The method of claim 1, wherein said capillary electrophoresis is carried out at a temperature within a range of about 0°C to about 60°C.

10. The method of claim 1, wherein said capillary electrophoresis is carried out at a temperature within a range of about 5^°C to about 37 °C.

11. The method of claim 1, wherein said capillary is 1 meter in length or less and 100 μm in diameter or less.

12. A method of increasing the concentration of any candidate hit compound in a natural sample that binds to a selected target during capillary electrophoresis, said method comprising the steps of:

(a) providing a capillary electrophoresis instrument comprising a capillary having a detection point and an inlet and outlet buffer reservoir;

(b) providing a running buffer containing a natural sample in said inlet reservoir;

(c) providing a plug of a selected target, wherein the concentration of said target is in excess of the concentration of any candidate hit compound in said natural sample;

(d) injecting the plug of the selected target into said capillary;

(e) subjecting the target plug to capillary electrophoresis;

(f) providing an electroosmotic flow in a direction opposite to that of the target, wherein said electroosmotic flow will move any buffer containing natural sample from said outlet reservoir in the direction opposite to the target; (g) tracking migration of any said target bound to hit compound at the detection point;

(h) collecting said target bound to hit compound detected at said detection point; and

(i) isolating said hit compound separate from target.

13. The method of claim 12, wherein said steps (h) and (i) comprises the steps of: (h) stopping said capillary electrophoresis run when said target bound to hit compound passes the detection point;

(i) removing the capillary containing the target and any bound hit compound from the capillary electrophoresis instrument; (j) cutting the capillary into several segments being careful to retain the segment containing the target and any bound hit compound; and

(k) collecting the target bound to hit compound from the capillary segment containing said target with any bound hit compound;

(1) isolating said hit compound separate from target.

14. The method of claim 12, wherein said method further comprises the step of first determining a mobility profile of said target.

15. The method of claim 12, wherein said buffer containing the natural sample is placed in both said inlet reservoir and said outlet reservoir.

16. The method of claim 12, wherein said concentration of said target is about 10 to about 50 μM.

17. The method of claim 12, wherein said hit compound has an equilibrium dissociation constant (K_D) of less than 100 nM.

18. The method of claim 12, wherein said method further comprises the step of analyzing said hit compound for physical and structural properties.

19. The method of claim 18, wherein said analysis comprises any method selected from the group consisting of mass spectroscopy; nuclear magnetic resonance spectroscopy; photodiode array detection; light scattering detection; infrared detection; liquid chromatography in association with a subsequent, different detection method; or ultra-violet spectroscopy.

20. The method of claim 18, wherein said analysis comprises using High Performance Liquid Chromatography with ultra violet detection.

21. The method of claim 12, wherein said capillary electrophoresis is carried out at a temperature within a range of about 0°C to about 60°C.

22. The method of claim 12, wherein said capillary electrophoresis is carried out at a temperature within a range of about 5°C to about 37 °C.

23. The method of claim 12, wherein said capillary is 1 meter in length or less and 100 μm in diameter or less.