WO2012166056A1

WO2012166056A1 - Fragment-based approach for inhibitor design by amide hydrogen/deuterium exchange mass spectrometry

Info

Publication number: WO2012166056A1
Application number: PCT/SG2012/000196
Authority: WO
Inventors: Ganesh Srinivasan ANAND; Srinath KRISHNAMURTHY
Original assignee: National University of Singapore
Current assignee: National University of Singapore
Priority date: 2011-06-02
Filing date: 2012-06-01
Publication date: 2012-12-06
Anticipated expiration: 2013-12-02

Abstract

Methods of screening small molecule fragments having a low binding affinity to a target protein, to identify fragments of interest or to identify binding sites of interest within a target protein, are described in which ligand-peptide binding index (LPBI) values are assigned to test peptides, and composite LPBI values are determined, by which small molecule fragments with composite LPBI value of interest are identified, and the relevant binding sites within the target protein are determined. LPBI values may be computer generated.

Description

FRAGMENT-BASED APPROACH FOR INHIBITOR DESIGN BY AMIDE HYDROGEN/DEUTERIUM EXCHANGE MASS SPECTROMETRY

RELATED APPLICATION

This application claims the benefit of U.S. Provisional Application No. 61/429,549, filed on June 2, 2011. The entire teachings of the above application are incorporated herein by reference.

BACKGROUND OF THE INVENTION

Fragment-based drug discovery (FBDD) is a new field that is based on identifying small chemical fragments and then growing them or combining them to produce molecules for exploration in therapeutic applications. In contrast to conventional structure based drug discovery approaches where up to a million or greater lead compounds are tested for high affinity binding to a particular target, FBDD screens a limited number of molecules which bind with very low affinities (Dissociation constant (KQ)) ranging from high μΜ to mM) to the target protein. 'Hit compounds' are then chemically modified to generate lead compounds with higher target affinity and specificity. The inherent drawback of using low affinity binding fragments is the inadequacy of effective screening methods. The limitations of current assays include poor dynamic range to enable screening of low affinity binders, economy of protein and fragment samples, throughput and turnaround time of analysis.

SUMMARY OF THE INVENTION

The present invention pertains to methods of screening small molecule fragments having a low binding affinity to a target protein, to identify fragments of interest, as well as to methods for identifying a binding site of interest within a target protein. The methods include either a) generating a deuterium exchange kinetic profile of the target protein in an unliganded state and in a ligand bound state in the presence of a natural ligand/substrate or substrate analog, and determining an optimal labeling time (TL) of deuterium exchange by measuring the relative deuteration exchange values for the unliganded state and for the ligand bound state determined in step (a) for peptide regions within the protein which show a maximum difference in relative deuteration exchange values between the unliganded state and the ligand bound state (the "test peptides"); or (b) utilizing a standardized chart of T_L for the target protein. Small molecule fragments having a low binding affinity (e.g., small molecule fragments obtained by screening a library of fragments for binding characteristics to the target protein) are used. The small molecules can be, for example, peptides, organic molecules, and/or any other ligand capable of binding to the target protein. Such molecules can have a binding affinity, for example, of under 10 mM (e.g., 100 μΜ to 1 niM). The small molecule fragments are incubated with the target protein (e.g., in solution in which the concentration of the target protein is 10-50 μΜ), for the period of time T_L, and then a Ligand-Peptide Binding Index (LPBI) value is assigned for each of the test peptides as a ratio of change in number of deuterons (D) exchanged upon fragment binding to the number of heavy atoms (HA) of the small molecule fragments (AD/HA). In a preferred embodiment, the ratio is 2000:1. A composite LPBI value is formed for each of the small molecule fragments by summing the individual LBPI values of step (d) for the test peptides; and at least one small molecule fragment with a composite LPBI value of interest is identified. Such small molecule fragments of interest can be used, for example, as building blocks for a drug candidate. Furthermore, if more than one fragment is identified, they can be assembled together, such as by a combinatorial chemistry approach. In addition, identification of a small molecule fragment of interest allows identification of its binding site in the target protein, and thereby identifies a binding site of interest (e.g., a catalytic active site, an allosteric site) within the target protein.

The methods of the invention offer significant advantages over conventional approaches for FBDD, including small size of protein samples, enhanced throughput, detailed information on conformational dynamics and allostery in solution. It also offers a strategy for combining binding kinetics with binding site analysis-something that is very inefficiently and inadequately addressed by current methods. Fragment-based drug design is a process of linking or expanding multiple fragments to generate high-affinity binders.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 depicts the fragments and natural inhibitors that were tested in pilot studies for binding to the N-terminal fragment domain of Hsp90. In the top box are two high affinity natural inhibitors- radicicol and a geldanamycin derivative (17 A AG), which bind with Dissociation constants of 19 and 33 nM, the structures and molecular weights are from the references listed on the slide. In the bottom box are three fragments that were tested for binding with the N-terminal fragment domain of Hsp90 by amide hydrogen/deuterium exchange- EA1, EA4 and aminopyrimidine (AMP).

Fig. 2 depicts a Butterfly Plot comparison of amide hydrogen/deuterium exchange for the free, unbound and radicicol-bound samples of the N-terminal domain of Hsp90 protein, after 1 min deuterium exchange.

Fig. 3 similarly depicts a Butterfly Plot comparison of amide

hydrogen/deuterium exchange for the free, unbound and radicicol-bound samples of the N-terminal domain of Hsp90 protein, except after 10 min deuterium exchange.

Fig. 4 depicts a modified Butterfly Plot displaying the results for the fragment EA4 following 1 min deuterium exchange. The regions showing decreased exchange upon EA4 binding are a subset of those seen upon radicicol binding.

Fig. 5 depicts a modified Butterfly Plot displaying the results for the fragment EA4 following 10 min deuterium exchange. Magnitude of differences are smaller after 10 min than at 1 min (Fig. 4).

Fig. 6 depicts the structure of the radicicol-bound N-terminal fragment domain of Hsp90, highlighting a particular region that shows differential exchange with certain fragments and natural inhibitors highlighted in red ( residues 126-141). On the right panel is overlaid the isotopic envelope from the mass spectrometer., highlighting how the pattern or 'signature' resembles radicicol ( in dashed boxes).

Fig 7 is a schematic illustration of a computer network in which

embodiments may be deployed.

Fig. 8 is a block diagram of a computer node in the network of Fig 7.

DETAILED DESCRIPTION OF THE INVENTION

This invention pertains to a novel application of amide hydrogen/deuterium exchange mass spectrometry in screening low binding affinity small molecule fragments that bind target proteins, for use, for example, in fragment-based drug discovery (e.g., for generating building blocks of a drug candidate) In addition to identifying small molecule fragments of interest, this invention allows mapping of binding sites of interest within target proteins,-including catalytic active sites and allosteric sites, through identification of small molecule fragments that bind to the target protein. Such small molecule fragments can serve as building blocks for compounds of interest for effective drug design, such as through combinatorial chemistry approaches.

FRAGMENT BASED DRUG DISCOVERY

This invention describes a novel application of amide deuterium exchange mass spectrometry as a screening tool for Fragment Based Drug Discovery (FBDD). The present invention provides a method for screening low affinity fragments that bind to a target protein, combining ligand binding with structural and

conformational dynamics information as an output. This invention gives a new faster and powerful method for screening low affinity fragments that bind to a target protein with binding, structural and functional information as an output.

AMIDE HYDROGEN/DEUTERIUM EXCHANGE MASS SPECTROMETRY (AMIDE HXMS)

Amide HXMS is a powerful technique that allows measurement of deuterium exchange at backbone amide hydrogens in proteins (reviewed by Hoofnagle, A.N., Resing, K.A. and Ahn, N.G. (2003) Protein analysis by hydrogen exchange mass spectrometry. Annu. Rev. Biophys. Biomol. Struct. 32, 1-25.). Monitoring the exchange provides important insights into protein structure and dynamics in solution. The method involves dilution of protein samples of interest in deuterated buffer, which is the aqueous buffer prepared in Deuterium oxide (D₂0). The extent of exchange varies across the molecule and is dependent both on solvent accessibility and dynamics of the protein backbone at that region. This pH- dependent process is slowed down several orders of magnitude at pH 2.5. After allowing deuterium exchange to occur at the pH and temperature of interest, pH and temperature of the samples are rapidly adjusted to pH~2.5 and 0 °C to 'quench' the deuterium exchange reaction. Samples are then subjected to complete proteolytic digestion with the protease pepsin and analyzed by mass spectrometry. The extent of deuterium exchange is reflected in the alteration of distribution of the isotopic envelopes of the individual fragment peptides. Average deuterons exchanged in each fragment peptide are calculated from the mass spectral envelope by publically available and commercial centroid determination software.

METHODOLOGY

In the methods of the invention, a group of small molecule fragments having a low binding affinity to a target protein, are investigated. The term, "small molecule fragment," refers to a small (e.g., less than or equal to 300 kDa in size, such as less than or equal to 300 kDa; less than or equal to 250 kDa; less than or equal to 200 kDa; less than or equal to 150 kDa; and less than or equal to 100 kDa) molecule having a KD ranging from low μΜ (e.g., 5 μΜ; 10 μΜ, 50 μΜ; 100 μΜ) to low mM (e.g., 10 mM; 5 mM; 1 mM; or other) . A "low binding affinity," as referred to herein, refers to a binding affinity that is less than or equal to 10 mM, preferably less than or equal to 5 mM, even more preferably equal to or less than 1 mM (e.g., from 100 μΜ to 1 mM, inclusive). The small molecules can be, for example, peptides, organic molecules, short nucleic acid sequences (e.g., DNA sequences, RNA sequences, etc.), and/or any other ligand capable of binding to the targeted protein. The "target protein" can be any protein of interest; for example, representative target proteins include, for example, biological enzymes or membrane proteins such as receptors (e.g., receptor proteins that are anchored in membranes).

In one embodiment, a group of small molecule fragments of interest to be used in the methods of the invention can be identified through conventional fragment screens. For example, conventional fragment screens are initiated with a relatively large population of fragments (e.g., 2000 fragments or more) in their library. The fragments are screened using preliminary screening methods specifically suited to the target protein of interest to identify a panel of fragments that bind to that protein, such as calorimetry-based binding assays. Through these low resolution HTS screening methods, a group of small molecule fragments of interest is selected for the methods of the invention using Amide

Hydrogen/Deuterium exchange mass spectrometry (Amide HXMS). The group of small molecule fragments of interest can be, for example, 1000 or fewer fragments, or 500 or fewer fragments; in a preferred embodiment, the group is 200 or fewer, such as 100 or fewer, fragments.

In one embodiment, a first step of the invention is to obtain deuterium exchange kinetic profiles of the target protein in the unliganded state and in a ligand bound state (i.e., in the presence of a natural ligand/substrate or substrate analog) over time. In representative examples, the time series can range from 10 or fewer seconds, to 3 hours. In certain representative embodiments, a typical starting time series is 30 seconds, 1 minute, 2 minutes, 5 minutes, and 10 minutes.

This step identifies the timepoint that is the most optimal for fragment screening. The timepoint of deuterium exchange where the greatest magnitude changes upon binding of the natural ligand/substrate or substrate analog is selected as the labeling time (TL) for ligand fragment screening. This optimal labeling time is determined by measuring the relative deuteration exchange values for the unliganded state and the native ligand bound state quantitated at each timepoint. Alternatively, the labeling time (TL) can be obtained from a standardized chart of TL values for the target protein, that can be generated using routine methods such as those described above. In preferred embodiments, the hydrogen/deuterium exchange reactions are carried out in triplicate, or more. For a difference in deuterium exchange to be considered significant, guidelines are employed: a difference in exchange between one state and another should be greater than the sum of the standard deviations in deuterium exchange for the two states.

In the methods of the invention, peptide regions within the target protein are explored. The peptide regions include, for example, a non-redundant set of peptides from various regions of the target protein, such as set from regions within one or more ligand binding pocket(s). A high resolution x-ray crystal structure of the ligand-bound protein can be used, for example, to identify the ligand binding pocket and the peptides within that region. Alternatively or in addition, a set of

nonoverlapping peptides based on the sequence of the target protein can also be used. Typical regions range from about 5 to 20 amino acids in length, inclusive. The peptide regions that show the maximum difference in relative deuteration values between the unliganded state and for the ligand bound state (referred to herein as "test peptides") are identified. For example, if the peptides are ranked in descending order of shift in relative deuterium exchange, the top non-redundant and

nonoverlapping peptides (e.g., at least 10 or more, such as 20 or more or 30 or more)] can be selected as the 'test peptides' or the 'test peptide set'. The number can be varied, depending on the size of the protein; generally, larger proteins will have larger groups of peptides.

In certain embodiments of the invention, in the next step of the method, the small molecule fragments are incubated with the target protein. They are incubated at a molar excess of fragments to ensure saturation of the protein (A ratio of 2000: 1 for fragment to target protein, such as at concentrations of approximately 10- 50 uM, would facilitate saturation of a fragment ligand that binds with low mM affinity). They are incubated for a time period (duration) equal to TL determined as described above.

It is noted that the methods of the invention facilitate high throughput analysis of all fragments, by quantitating only the peptides for each fragment. In the methods of the invention, a Ligand-Peptide Binding Index (LPBI) value is assigned for each of the 'test peptides'. This value is the ratio of change in number of deuterons exchanged upon fragment binding to the number of heavy atoms (AD/HA). Each ligand fragment thus has assigned a LBPI score for each of the test peptides. For each fragment, the LBPI scores for each test peptide is summed up to yield a composite LBPI value for that small molecule fragment. For example, to facilitate calculation of the composite LBPI score, the LBPI scores can be tabulated for each peptide, and then the LBPI scores of all the ligands can be ranked in descending order.

Small molecule fragments with composite LPBI values of interest can then be identified. For example, all fragment ligand Composite LBPI values can be ranked in order, and the top value or values can be identified as the value(s) for which the small molecule fragment(s) are of interest . A fragment "of interest," as used herein, refers to a fragment that has the desired characteristic(s) (e.g., a particular LPBI value). For example, such fragments may be useful for generating building blocks, e.g., of a drug candidate. For example, the fragments of interest represent the 'lead' molecules targeting a therapeutic protein of interest. If desired, more than one fragment can be selected (identified) as being a small molecule fragment of interest. Multiple (e.g., 2 or more) small molecule fragments of interest can be combined to generate drug candidates. In certain embodiments, a

combinatorial chemistry approach can be used to assemble the small molecule fragments of interest. For example, chemistries of 'lead' molecules affecting distinct sets of peptides on the target can then be combined to yield hybrids with higher binding and improved specificity. If desired, validation of combinatorial chemistry can again be achieved through repeating the steps described above. An iterative approach can verify the utility of such molecules as drug candidates (e.g., for biological or therapeutic drug testing).

In additional embodiments of the invention, the methods described above can be used to identify binding sites of interest within a target protein. A small molecule fragment with a composite LPBI value of interest can be identified, as described above; such a small molecule fragment binds to the target protein at a particular site (its binding site, which can be, for example, a particular peptide region of the target protein); this site (peptide region) is identified as a binding site of interest within the target protein. In certain embodiments, the binding site of interest can be a catalytic active site, or an allosteric site.

Some of the advantages of the methods described herein include highly improved throughput, robustness, conformational dynamics information in solution, and minimalized workflow timeframe. The methods provide means for identifying small molecule fragments of interest that interact with a target protein, thereby facilitating creation of drug candidates or other relevant molecules such as through use of the fragments of interest as building blocks for candidate molecules; the methods also provide binding site analysis and identification of relevant binding sites of interest within a target protein. All measurements take into account the inherent dynamics of the target proteins. Importantly, the methods allow for narrowing down the possible interaction sites on the target protein. In contrast to other methods (e.g., surface plasmon resonance, fluorescence-based binding, calorimetry, or enzyme-based assays), the present methods allow defining of the binding site, such as by looking at a structure of a target and identifying the regions that show decreased or increased exchange. Furthermore, the methods provide clear mapping details to examine induced conformational change and allostery. The methods work in solution, require minimal amounts of protein and ligands, and do not require immobilization (e.g., on a chip), allowing easy adaptation to investigate membrane proteins.

EXEMPLIFICATION: UTILITY OF AMIDE HYDROGEN/DEUTERIUM

EXCHANGE MASS SPECTRONOMETRY FOR FRAGMENT-BASED DRUG DESIGN

The N-terminal domain of Hsp90 from yeast was explored using the methods described herein. The N-terminal domain binds to the high affinity natural inhibitor radicicol, as determined by X-ray crystallography (Huth et al. (2007), Chem. Biol. Drug Des. 70:1-12; Roe et al. (1999), J. Med. Chem. 42:260-266;

Protein Data Bank (PDB ID 1BGQ). This test protein that used to establish the utility of amide hydrogen/deuterium exchange mass spectrometry for fragment- based drug design. Results are shown in the figures. The mass spectrometry data in the figures have been collected on a Synapt High Definition Mass spectrometry from Waters Corporation, Milford, MA and the results are displayed using Waters Commercial software, DynamX (v 1.0) and HDX Browser (development pre-release software whose features were incorporated into the commercial DynamX software).

Fig. 1 depicts the fragments and natural inhibitors that were tested in pilot studies for binding to the N-terminal fragment domain of Hsp90. In the top box are two high affinity natural inhibitors- radicicol and a geldanamycin derivative (17 AAG), which bind with Dissociation constants of 19 and 33 nM, the structures and molecular weights are from the references listed on the slide. In the bottom box are three fragments that were tested for binding with the N-terminal fragment domain of Hsp90 by amide hydrogen/deuterium exchange- EA1, EA4 and aminopyrimidine (AMP). (Murray et al. (2010), J. Med. Chem. 53: 5942-5955; Brough et al. (2009), J. Med. Chem. 52: 4794-4809). It should be noted that the fragments are much smaller in molecular weight than high affinity inhibitor: the smaller the dissociation constant (KD), tighter the affinity of the compound for the target protein. Thus EA4 binds Hsp90 ~ 26000X weaker than radicicol.

Fig. 2 depicts a comparison of amide hydrogen/deuterium exchange for the free, unbound and radicicol-bound samples of the N-terminal domain of Hsp90 protein. The plot is a Butterfly Plot and generated by proprietary commercial software- DynamX ( v. 1.0)( Waters Corporation, Milford, MA™). In this variation of the Butterfly plot, all the pepsin digest fragments of the target protein are listed from the N- to C-terminus ( X-axis). The distribution of peptides is not to scale, as it is a function of pepsin's specificity of cleavage at certain sites on the target protein. On the Y-axis are represented a ratio of the deuteriums exchanged relative to the total number of deuterium exchangeable amides available. The gray plot is the exchange ( after 1 min deuterium exchange) for free unliganded Hsp90 and the black plot is that with radicicol bound. 'Bubbles' indicate regions where the radicicol- bound Hsp90 shows lower deuterium exchange compared to the free protein.

Representative isotopic envelopes show the differences between radicicol-bound and free protein. Fig. 3 similarly depicts a comparison of amide hydrogen/deuterium exchange for the free, unbound and radicicol-bound samples, except after 10 min deuterium exchange.

Fig. 4 depicts a modified Butterfly Plot displaying the results for the fragment EA4 following 1 min deuterium exchange. The regions showing decreased exchange upon EA4 binding are a subset of those seen upon radicicol binding. Fig. 5 depicts a modified Butterfly Plot displaying the results for the fragment EA4 following 10 min deuterium exchange. Magnitude of differences are smaller after 10 min than at 1 min.

Fig. 6 depicts the structure of the radicicol-bound N-terminal fragment domain of Hsp90, highlighting a particular region that shows differential exchange with certain fragments and natural inhibitors highlighted in red ( residues 126-141). On the right panel is overlaid the isotopic envelope from the mass spectrometer., highlighting how the pattern or 'signature' resembles radicicol ( in dashed boxes). This qualitative assessment shows that the similarity is not seen with the other combinations of fragments and inhibitors. Importantly, this highlights that amide hydrogen/deuterium exchange is highly sensitive to pick up differences in interactions between two weak-binder fragments ( EA1 and EA4). This

representation is by HDX browser™, a feature that is part of commercial proprietary amide hydrogen/deuterium exchange software DynamX™ from Waters Corporation, Milford, MA.

The results of this experiment highlight the funnel approach to fragment- based drug discovery (FBDD), where HDX-MS can be positioned downstream of a high-throughput binding screen for optimum time utilization. Importantly, the methods do not preclude the option of using HDX-MS entirely for screening Fragments without an upstream high throughput binding screen. Iterative improvements in drug design and validation studies can be performed using these methods. Algorithm for screening fragments by amide hydrogen/deuterium exchange mass spectrometry (HDXMS)

For proteins with known high affinity natural ligands/analogs

The first step is to carry out an HDXMS time-series experiment with the target protein and high affinity ligand/analog. Suggested deuterium exchange labeling times range from 10-30 seconds and longer (a typical starting time series is 30 seconds, 1 minute, 2 minutes, 5 minutes and 10 minutes*). From the HDXMS experiment, a single labeling time is selected as a standard labeling time for the entire fragment library. The labeling time selected is the shortest time point that shows 75% or greater deuterium exchange seen after 10 minutes of deuterium exchange for at least 3 pepsin digest fragment peptides** . A set of peptides

(number: n) are chosen which show the maximum changes in deuterium exchange upon ligand binding. This set of peptides should be a non-redundant set from various regions in the ligand binding pocket, if a high resolution X-ray crystal structure of the ligand-bound protein is available. If a structure is not available, then a set of nonoverlapping peptides, on the basis of primary sequence are chosen.

Once the peptide set (n) and labeling time are optimized, the fragment screen is carried out for the entire set of fragments. Only the peptides that are in the peptide set (n) are selected for analysis. The deuterium exchange is quantified for each fragment, and the Binding efficiency factor for each fragment is calculated from the following formula. This will be a finite value and based on the summation of all the peptides in the test set.

Binding efficiency factor for a fragment=

Where

ADj - Deuterons exchanged in the unliganded state- Deuterium ions exchanged in presence of fragment 'i' ADj- Deuterons exchanged in the unliganded state - Deuterium ions exchanged in presence of the high affinity natural ligand or inhibitor.

n- Number of pepsin digest fragment peptides considered in the test set

MW- Molecular weight of the fragment

C- Number of non- Hydrogen atoms in fragment

The binding efficiency factor is also applied for the high affinity natural ligand as

Binding efficiency factor for the natural Ligand/ analog

C x n x 1000

MW

BEF of fragment

Fragment rank factor= BEF of Natural Ligand/ analogue

This value can range from 0 to above 1 for fragments that are more efficient than the high affinity ligand. Fragments will be ranked according to their Fragment rank factor.

In iterative screens, this value can be monitored to develop high affinity inhibitors. Improvements in inhibitor design can thus be tracked with every iterative step.

* For most globular proteins, this time series captures deuterium exchange at fast exchanging amides. For some proteins, this time series may need to be extended to longer labeling times.

** Experience has shown 1 minute of deuterium exchange labeling time to be ideal for most globular proteins (data not shown).

For proteins with unknown High affinity ligands

In the case of proteins with unknown high affinity ligands, 10 fragments are chosen at random and HDXMS of these fragments with the target protein is carried out with a labeling time of 1 minute. A peptide set is chosen as the regions that show the maximum difference in deuterium exchange for all the fragments chosen. If none of the chosen fragments show any difference then another set of fragments is chosen. This initial test is done to determine which peptides are to be selected in the peptide list (n).

The Binding efficiency factor equation is modified to the following equation

Binding efficiency f actor= ,_n MW

c

Where

AD; - Deuterium ions exchanged in the unliganded state- Deuterium ions exchanged in presence of fragment 'i'

n- Number of peptides considered in the test set

MW- Molecular weight of the fragment 'i'

C- Number of non- Hydrogen atoms in fragment

Fragments are ranked according to their absolute BEF to determine which fragments can be considered for molecule linking and expansion in iterative FBDD.

Embodiments may employ a computer-based machine or digital processor configured to implement the forgoing calculations. For non-limiting example, Figs 7 and 8 show a computer network in which computer embodiments may be deployed.

Figure 7 illustrates a computer network or similar digital processing environment in which the present invention may be implemented.

Client computer(s)/devices 50 and server computer(s) 60 provide processing, storage, and input/output devices executing application programs and the like.

Client computer(s)/devices 50 can also be linked through communications network 70 to other computing devices, including other client devices/processes 50 and server computer(s) 60. Communications network 70 can be part of a remote access network, a global network (e.g., the Internet), a worldwide collection of computers, Local area or Wide area networks, and gateways that currently use respective protocols (TCP/IP, Bluetooth, etc.) to communicate with one another. Other electronic device/computer network architectures are suitable.

Fig.8 is a diagram of the internal structure of a computer (e.g., client processor/device 50 or server computers 60) in the computer system of Fig. Na. Each computer 50, 60 contains system bus 79, where a bus is a set of hardware lines used for data transfer among the components of a computer or processing system. Bus 79 is essentially a shared conduit that connects different elements of a computer system (e.g., digital processor, disk storage, memory, input/output ports, network ports, etc.) that enables the transfer of information between the elements. Attached to system bus 79 is I/O device interface 82 for connecting various input and output devices (e.g., keyboard, mouse, displays, printers, speakers, etc.) to the computer 50, 60. Network interface 86 allows the computer to connect to various other devices attached to a network (e.g., network 70 of Fig. Na). Memory 90 provides volatile storage for computer software instructions 92 and data 94 used to implement an embodiment of the present invention (e.g., formula for calculating Binding efficiency factor for a fragment, formula for calculating Fragment rank factor, routine or module for ranking or ordering of fragments and displaying indications of resulting identified small molecule fragments of interest, and supporting code detailed above). Disk storage 95 provides non-volatile storage for computer software instructions 92 and data 94 used to implement an embodiment of the present invention. Central processor unit 84 is also attached to system bus 79 and provides for the execution of computer instructions. CPU 84 may also be referred to as a digital processor.

In one embodiment, the processor routines 92 and data 94 are a computer program product (generally referenced 92), including a computer readable medium (e.g., a removable storage medium such as one or more DVD-ROM's, CD-ROM's, diskettes, tapes, etc.) that provides at least a portion of the software instructions for the invention system. Computer program product 92 can be installed by any suitable software installation procedure, as is well known in the art. In another embodiment, at least a portion of the software instructions may also be downloaded over a cable, communication and/or wireless connection. In other embodiments, the invention programs are a computer program propagated signal product 107 embodied on a propagated signal on a propagation medium (e.g., a radio wave, an infrared wave, a laser wave, a sound wave, or an electrical wave propagated over a global network such as the Internet, or other network(s)). Such carrier medium or signals provide at least a portion of the software instructions for the present invention routines/program 92.

In alternate embodiments, the propagated signal is an analog carrier wave or digital signal carried on the propagated medium. For example, the propagated signal may be a digitized signal propagated over a global network (e.g., the Internet), a telecommunications network, or other network. In one embodiment, the propagated signal is a signal that is transmitted over the propagation medium over a period of time, such as the instructions for a software application sent in packets over a network over a period of milliseconds, seconds, minutes, or longer. In another embodiment, the computer readable medium of computer program product 92 is a propagation medium that the computer system 50 may receive and read, such as by receiving the propagation medium and identifying a propagated signal embodied in the propagation medium, as described above for computer program propagated signal product.

Claims

CLAIMS is claimed is:

A method of screening small molecule fragments having a low binding affinity to a target protein, to identify small molecule fragments of interest, the method comprising:

a) generating a deuterium exchange kinetic profile of the target protein in an unliganded state and in a ligand bound state in the presence of a natural ligand/substrate or substrate analog;

(b) determining an optimal labeling time (TL) of deuterium exchange by measuring the relative deuteration exchange values for the unliganded state and for the ligand bound state determined in step (a) for peptide regions within the protein which show a maximum difference in relative deuteration exchange values between the unliganded state and the ligand bound state, wherein the peptide regions within the protein which show a maximum difference are identified as test peptides;

c) incubating each of the small molecule fragments having a low binding affinity to the target protein, with the target protein for the period of time TL, determined in the step (b);

d) assigning a Ligand-Peptide Binding Index (LPBI) value for each of the test peptides as a ratio of change in number of deuterons (D) exchanged upon fragment binding to the number of heavy atoms (HA) of the small molecule fragments (AD/HA);

e) forming a composite LPBI value for each of the small molecule fragments by summing the individual LBPI values of step (d) for the test peptides; and

f) identifying at least one small molecule fragment with a composite LPBI value of interest, thereby identifying a small molecule fragment of interest.

2. The method of Claim 1, wherein the small molecule fragments of step c) are obtained by screening a library of fragments for binding characteristics to the target protein.

3. The method of Claim 1 , wherein the low binding affinity of the small molecule fragments is 100 μΜ to 1 mM.

4. The method of Claim 1 , wherein a ratio of step d) is 2000: 1.

The method of Claim 1, wherein the concentration of the target protein is 10- 50 μΜ.

The method of Claim 1 , further comprising the step of selecting more than one fragment identified in step (f) and assembling the fragments together to produce a drug candidate.

The method of Claim 6, wherein the fragments are assembled together via a combinatorial chemistry approach.

The method of Claim 1 , wherein the small molecule fragments are selected from peptides, organic molecules, and any other ligand capable of binding to the target protein.

a) incubating each of the small molecule fragments having a low binding affinity to the target protein, with the target protein for the period of time T_L, wherein T_L is obtained from a standardized chart of T_L values for the target protein;

b) assigning a Ligand-Peptide Binding Index (LPBI) value for peptide regions within the protein which show a maximum difference in the relative deuteration exchange values between the unliganded state and the ligand bound state, wherein the peptide regions within the protein which show a maximum difference are identified as test peptides, as a ratio of change in number of deuterons (D) exchanged upon fragment binding to the number of heavy atoms (HA) of the small molecule fragments (AD/HA);

c) forming a composite LPBI value for each of the fragments by summing the individual LBPI values of step (b) for all of the test peptides; and

d) identifying at least one small molecule fragment with a composite LPBI value of interest, thereby identifying a small molecule fragment of interest.

A method of identifying a binding site of interest within a target protein, the method comprising:

a) generating deuterium exchange kinetic profile of the target protein in an unliganded state and in a ligand bound state in the presence of a natural ligand for the binding site of the target protein;

(b) determining an optimal labeling time (TY) of deuterium exchange by measuring the relative deuteration exchange values for the unliganded state and the ligand bound state determined in step (a) for peptide regions within the protein which show a maximum difference in relative deuteration exchange values between the unliganded state and the ligand bound state, wherein the peptide regions within the protein which show a maximum difference are identified as test peptides;

c) incubating small molecule fragments having a low binding affinity to the target protein, with the target protein for the period of time TL, determined in the step (b);

d) assigning a Ligand-Peptide Binding Index (LPBI) value for each of the test peptides as a ratio of change in number of deuterons (D) exchanged upon fragment binding to the number of heavy atoms (HA) of the small molecule fragments (AD/HA); e) forming a composite LPBI value for each of the fragments by summing the individual LBPI values of step (d) for all of the test peptides; and

f) identifying at least one small molecule fragment with a composite LPBI value of interest, thereby identifying a small molecule fragment that binds to the target protein at a site, and thereby identifying a binding site of interest within the target protein.

11. The method of Claim 10, wherein the small molecule fragments of step (c) are obtained by screening a library of fragments for binding characteristics to a binding site of the target protein.

12. The method of Claim 10, wherein the binding site of interest is a catalytic active site.

13. The method of Claim 10, wherein the binding site of interest is an allosteric site.

14. The method of Claim 10, wherein a ratio of step d) is 2000:1.

15. The method of Claim 10, wherein the concentration of the target protein is 10- 50 μΜ.

16. A method of identifying a binding site of interest within a target protein, the method comprising:

a) incubating small molecule fragments having a low binding affinity to the target protein, with the target protein for the period of time TL, wherein TL is obtained from a standardized chart of TL values for the target protein; b) assigning a Ligand-Peptide Binding Index (LPBI) value for peptide regions within the protein which show a maximum difference in the relative deuteration values between the unliganded state and the ligand bound state, wherein the peptide regions within the protein which show a maximum difference are identified as test peptides, as a ratio of change in number of deuterons (D) exchanged upon fragment binding to the number of heavy atoms (HA) of the small molecule fragments (AD/HA);

c) forming a composite LPBI value for each of the small molecule fragments by summing the individual LBPI values of step b) for all of the test peptides; and

d) identifying at least one small molecule fragment with a composite LPBI value of interest, thereby identifying a small molecule fragment that that binds to the target protein at a site, and thereby identifying a binding site of interest within the target protein.

17. A computer-based apparatus screening small molecule fragments having a low binding affinity to a target protein, to identify small molecule fragments of interest, the apparatus comprising:

a digital processor; and

a computer memory area, which when executed by the processor, is configured to:

a) assign a Ligand-Peptide Binding Index (LPBI) value for each test peptide in a set of test peptides as a ratio of change in number of deuterons (D) exchanged upon fragment binding to a number of heavy atoms (HA) of the small molecule fragments (AD HA), wherein a deuterium exchange kinetic profile of the target protein is generated in an unliganded state and in a ligand bound state in the presence of a natural ligand/substrate or substrate analog, and wherein an optimal labeling time (TL) of deuterium exchange is determined by measuring relative said generated deuteration exchange values for the unliganded state and for the ligand bound state for peptide regions within the protein which show a maximum difference in relative deuteration exchange values between the unliganded state and the ligand bound state, the peptide regions within the protein which show a maximum difference being identified as the set of test peptides, and each of the small molecule fragments having a low binding affinity to the target protein being incubated with the target protein for the determined period of time TL ; b) form a composite LPBI value for each of the small molecule fragments by summing the assigned individual LBPI values for the test peptides; and

c) identify at least one small molecule fragment with a composite LPBI value of interest, thereby identifying a small molecule fragment of interest, wherein the processor on output displays to a user an indication of the identified small molecule fragments of interest.

18. Computer apparatus as claimed in Claim 17 wherein the computer memory area is further configured to order the identified small molecule fragments of interest.