CA2420065A1 - Methods for indentifying specifically cleavable peptides and use of such peptide sequences - Google Patents

Abstract

The invention relates to methods for detecting and identifying peptides that can be specifically cleaved, using a defined amino acid sequence based on a library of nucleic acid peptide fusion molecules. According to said methods, the variable part of the peptides is coded by the respective nucleic acid fused thereto, using proteolytically active solutions. The invention also relates to the use of said peptides for diagnosis.

Description

,f Methods for identifying specifically cleavable peptides and use of such peptide sequences Description The invention relates to methods for finding and identifying specifically cleavable peptides having a defined amino acid sequence by using proteolytically active solutions and to the use of said peptides for specific release of chemical active substances and for diagnostics.
US 5981200 describes a method for detecting specifically cleaved peptides by using a specific protein construct which fluoresces when cleaved proteolytically. The method is based on fluorescence resonance energy technique (FRET). This method has the advantage that it can also be used for in vivo detection of protein cleavage events. A disadvantage, on the other hand, is the fact that only specific peptides can be labeled or that, in the case of studying different labeled proteins, assigning the cleavage event precisely to the relevant sequence would be very complicated. Said method has another disadvantage in that relatively large amounts of labeled substrate have to be used in order to obtain a detectable signal.
Another method which may be used for selective identification of enryme substrates is phage display (documented, for example, in WO 97/47314).
This method has the disadvantage that the phage display peptide libraries, in contrast to other surface-represented peptide libraries, have a smaller number of possible different sequences and the subsequent work-up of the phage particles and determination of the identified substrates is relatively complicated.
It is the object of the present invention to develop a method for finding and identifying specifically proteolytically cleavable peptides and to provide substances which allow a targeted release of chemical active substances.
The object is achieved by a method for identifying specifically proteolytically cleavable peptides, which comprises the following method steps:

a) incubating a library of fusion molecules comprising a peptide and a nucleic acid encoding said peptide with a proteolytically active sample, b) isolating the proteolytically removed parts of said fusion molecules, c) determining the sequence of the nucleic acid part of the isolated fusion molecules.
The method of the invention is based on using fusion molecules which contain a phenotypic peptide part and a genotypic nucleic acid part which has a sequence encoding said peptide part. The peptide part is linked to the nucleic acid part via a suitable linker. The linker used is preferably a protein acceptor, for example puromycin, which is covalently bound to the nucleic acid part. The linker may comprise further components such as, for example, a noncoding nucleic acid sequence, preferably a polyA
sequence. In a preferred embodiment, the nucleic acid part contains further regions which are constant in its sequence and are located on one or both sides of the coding region. These constant nucleic acid regions may serve, for example, as primer binding sites for carrying out a PCR or as restriction enzyme cleavage sites. The coding region of the nucleic acid part of the fusion molecules contains a variable part which codes for at least two amino acids, preferably for at least four, particularly preferably for seven to twelve, amino acids. To this, further constant coding nucleotide sequences may be added. The constant nucleic acid sequences may encode peptide sections which allow the peptidic part of the fusion molecule to be readily attached to a solid surface or which generate interesting structural elements. Examples of such structural elements are epitopes for antibodies, tags for purifying or for detecting the constructs, or elements which determine particular tertiary structures.
Such fusion molecules may be prepared, for example, according to W098/31700 which describes a system in which a protein acceptor, for example a puromycin, is bound to the nucleic acid, preferably RNA, via a suitable linker. This makes it possible, shortly before in vitro translation of the RNA into the corresponding protein has finished, to bind the synthesized protein covalently to its encoding RNA and thus to characterize it in more detail. Examples of comparable systems which can be used for the present invention are described in DE 1964637201, W098/16636, US 5843701 and W094/13623.

In a preferred embodiment of the method of the invention, the fusion molecules are attached to a surface (support) via their peptidic part. The term "support" means in accordance with the present invention material which is present in solid or else gel-like form. Examples of suitable support materials are ceramic, metal, in particular semiconductors, noble metal, glasses, plastics, crystalline materials or thin layers of the support, in particular of said materials, and (bio)molecular filaments such as cellulose and structural proteins. However, suitable supports are not only flat materials but also particles such as, for example, materials for column chromatography, which can be loaded with proteins, or beads made of organic polymers. The support is generally loaded covalently, quasi-covalently, supramolecularly or physically.
The peptide part of the fusion molecules may be attached to the support, for example, via biotin-streptavidin binding but specific domains of the peptide part of the fusion molecules, such as, for example, metal-binding domains (e.g. His tag), terminal cysteine residues or domains containing epitopes which can be recognized by specific antibodies, can mediate such an attachment.
In the method of the invention, preference is given to libraries whose fusion molecules contain all possible permutations with respect to the variable part of the peptide sequence. Practically still manageable libraries of such fusion proteins are approx. 10'3 different sequences in size and thus are 103-104 times (e.g. phage display) or 106-10' times (classical, chemically synthesized libraries) larger than other known peptide libraries. In order to cover all permutations, preference is given to a variable peptide part of less than 11 amino acids, and particular preference is given to variable peptide parts composes of seven or eight amino acids. It is also possible, of course, to construct libraries whose fusion molecules contain longer variable peptide parts; in this case, however, the wide variety of sequences can no longer be covered completely. Since the proteases known to date, which cleave specific peptide bonds, require a recognition sequence of at least four amino acids, preference is given to libraries containing a variable peptide part of at least four amino acids.
The library of fusion molecules is exposed to a proteolytically active sample, either in solution or, preferably, attached to a support. A suitable proteolytically active sample, presample or comparison sample is especially tissue-specific extracts of animal, human or plant origin, extracts of particular cell compartments, such as, for example, cytosolic extracts, subcellular extracts, extracts of membrane components, extracellular extracts, or mixtures of such extracts. Of particular interest are also extracts of diseased tissue, for example of carcinomas. It is, however, also possible to use extracts which represent the entire proteolytic activity or parts thereof of an organism, in particular of viruses, microorganisms such as bacteria or protists. Proteolytically active samples, presamples or comparison samples which may be used are also solutions containing known proteases.
The library is incubated with the proteolytically active sample preferably under physiological conditions, preferably at temperatures between 0°C
and 45°C. For this purpose, the extracts to be tested for their proteolytic activity may be used directly or the extracts are taken up in a suitable solvent such as, for example, physiological saline and used for incubation.
During incubation the proteases contained in the sample cleave the variable peptide parts of the fusion molecules. This results in a specific proteolytic cleavage pattern which is characteristic for the sample extract used. The fusion molecule parts removed by cleavage are encoded by the nucleic acid part.
After incubating the library with the proteolytically active sample to be studied, the fusion molecule parts removed by proteolytic cleavage are isolated and the sequence of the coding nucleic acid part is determined.
This identifies the peptide sequences cleavable by said sample.
After incubation of library and proteolytically active extract in solution, the cut fusion proteins can be isolated as follows. The nucleic acid regions of the cut fusion proteins can be isolated from those of the uncut fusion proteins, for example, by making use of a constant structural feature, for example a known epitope for an antibody, in the particular peptide part which, due to proteolytic cleavage, is no longer linked to the rest of the fusion molecule including the nucleic acid part. For this purpose, the mixture is subjected to an affinity chromatography by utilizing said antibody, for example. The nucleic acid part of the cut fusion proteins cannot bind, is separated from the nucleic acids of the uncut fusion proteins and is thus present in the eluate. It is also possible to use other structural features such as, for example, a His tag or Strep tag for the removal. The fusion molecules retained on the affinity matrix may be eluted and used further in solution or, advantageously, directly in the attached state.
After incubation of the attached library with extract, different methods may be used to isolate the cut fusion proteins.
In a preferred embodiment of the method of the invention, after incubation of the attached library with the extract, the nucleic acid part of the cut fusion molecules are obtained directly in the eluate.
In the simplest case, the sequences of the cleaved peptides can be identified directly by PCR amplification of the nucleic acid parts of the cleaved fusion proteins, present in the eluate, followed by cloning and sequencing. It is, however, also possible first to purify the cleaved fusion proteins from other components present in the eluate by first another affinity chromatography and making use of specific structures in the peptide part or nucleic acid part of the cleaved fusion proteins. After eluting the cleaved fusion proteins from the chromatographic material, for example by incubation with a KOH solution, the cleaved peptides may be sequenced again by means of PCR, cloning anb sequencing.
In addition to this, a number of further chromatographic or electrophoretic methods for isolating the fusion molecule parts removed by proteolytic cleavage are possible. For this purpose, it is advantageous if the fusion molecules are labeled or modified for easier identification. Thus, for example, the nucleic acid part of the fusion molecules may be, for example, fluorescently labeled or radiolabeled. Magnetic labeling of the nucleic acid part of the fusion molecules is particularly preferred and may also be used in connection with fluoresence labeling or radiolabeling. The fusion molecule parts removed by cleavage may be magnetically isolated very easily and selectively. The proteolytic pattern is evaluated, for example, via hybridization of the isolated nucleic acid sequences on a biochip as is available, for example, from Affymetrix or Nanogen. Thus it is possible to analyze directly the isolated mixture of fusion molecule parts removed by cleavage and containing different nucleic acid sequences.

When preparing the extracts to be studied, attention must be paid to isolating the proteases in their active form. At the same time, however, DNA-degrading nucleases should be inactivated as far as possible.
For this purpose, the proteolytically active sample may be worked up mechanically as follows: the complete tissue is purified, where appropriate, by washing the outside in cold buffer (50 mM Tris-HCI, pH 7.5, 2 mM
EDTA, 150 mM NaCI, 0.5 mM DTT; Dignam, J.D., Preparation of Extracts from Higher Eukaryotes. From: Deutscher, M.P. (ed.) Guide to Protein Purification. Methods in Enzymology, Vol 182, Academic Press, (1990).
Page 194-202). The tissue is cut up in cold buffer and, where appropriate, is removed from components such as, for example, connective tissue, skin and blood vessels. The tissue pieces may be admixed with homogenization buffer according to Kobayashi et al. (Kobayashi, H., Fujishiro, S., Terao, T., Cancer Res. (1994) 54, 6539-6548): 0.01 M HEPES buffer, pH 7.2, 0.25 M
sucrose, 0.5% Triton X-100 in a volume ratio from 1:2 to 1:5 (w/w) and processed to a homogenate in an appropriate tissue homogenizes (see below) with ice cooling. The method must be adapted to the desired tissue type. Examples of suitable tissue homogenizers are:
- Mixer: liver, for example, can easily be mashed in a mixer and is then further strained through a close-meshed nylon net. A fine tissue homogenate is formed.
- "Potted' homogenizes: rotating glass rod which is inserted tightly in a fitting vessel so that the tissue between the vessel wall and the glass rod is crushed. Constant cooling must be ensured.
- Dispersers: homogenizers with rotating blades in a shaft.
The homogenized tissue is admixed with further homogenization buffer and incubated with shaking at 4 - 8°C for up to 45 min. The tissue homogenate is then centrifuged at 4°C and 16,000 g and the supernatant is used for analysis according to the method of the invention either directly or after further purification, for example via column chromatography or dialysis.
If frozen material is used instead of fresh tissue, the method should be modified. Frozen tissue should not just be left to thaw, since in this case proteases released from lysosomes inactivate the enzymes. Preferably, the tissue is pulverized with constant cooling in a mortar using liquid nitrogen and the powder is freeze-dried in a lyophilizes overnight. The dry powder can then be used further.

An example of a suitable processing of brain tissue and separation thereof into nuclear fraction, cytosolic fraction, membrane fraction and cytoskeletal fraction is stated below. The extracts obtained may be used as proteolytically active samples, presamples or comparison samples either directly or with addition of further auxiliary substances.
Brains, for example from mice, of several preparations (approx. 50) are introduced into approx. 25 ml of solution buffer 1 (SB1 ) (20 mM Tris-HCI
pH 7.5; 150 mM NaCI, 1 mM EDTA, 1 mM EGTA) and homogenized for 1 min using an Ultraturrax at maximum speed. Alternatively, the Potter glass homogenizer is used in the case of less material.
A nuclear fraction may be isolated by centrifuging the homogenate at 1000 x g at 4°C for 10 min. The supernatant is collected for further preparations. The centrifugate (nuclear pellet) is resuspended in 12 ml of SB 1 and centrifuged as before. The nuclear pellet is resuspended in SB2 (20 mM Tris-HCI pH 7.5; 150 mM NaCI, 1 mM EDTA, 1 mM EGTA, 1 % NP-40).
The collected supernatants may be used for isolating and purifying a cytosolic fraction. For this purpose, they are centrifuged at 100,000 x g at 4°C for 1 h. The supernatant represents the extract containing cytosolic proteins.
The membrane fraction is isolated by washing the pellet from removing the cytosolic fraction three times in SB1 and centrifuging at 100,000 x g at 4°C
for 1 h. After the third washing step, the pellet is resuspended in 15 ml of SB2 and solubilized overnight, followed by centrifugation at 100,000 x g at 4°C for 1 h. The supernatant forms the membrane fraction.
A cytoskeletal fraction is obtained by washing the pellet from removing the membrane fraction three times in SB2 and centrifuging at 100,000 x g at 4°C for 1 h. After the third washing step, the pellet is resuspended in (12.5 mM Tris-HCI pH 7.5; 0.4% SDS) and solubilized at 95°C. This is followed by a final centrifugation at 20,800 x g at room temperature for 15 min. The supernatant contains the cytoskeletal fraction.

- $ -Proteolytically active extracts containing secreted proteases are obtained from cells and tissues by culturing, for example, human cells or cell lines in cell culture as monolayers. The cells are washed three times with cell culture medium. The medium is then replaced with fresh medium, with the secretory proteases being secreted into the medium at 37°C. A suitable secretion time is 1 to 3 h. Tissue pieces or tissue sections may be processed in the same way.
Tissue fluid can be obtained from tissue assemblages by carefully cutting up, for example, organs or tissue with scissors and scalpel. The tissue pieces are washed several times with tissue buffer. The tissue pieces or else small organs are incubated in physiological buffer or medium at 37°C
for 1-3 hours. After incubation, the tissue pieces are allowed to sediment in a 15 ml conical vessel, and the medium is removed and can be studied further.
Alternatively, the tissue may be transferred into centrifugation vessels and centrifuged gently. The supernatant contains the proteases of the interstitial fluid from the interstitial spaces.
Alternatively, it is also possible to obtain protein extracts according to the following method: cutting up the tissue in ice-cold RIPA buffer (50 mM Tris-HCI, pH 7.4; 1 % NP-40, 0.25% sodium deoxycholate, 150 mM NaCI, 1 mM
EGTA, 1 mM PMSF, 1 mM Na3V04, 1 mM NaF and in each case 1 ~glml aprotonin, leupeptin and pepstatin), freezing the cut-up tissue and subsequent crushing in a mortar. The tissue is homogenized by adding RIPA buffer in a volume ratio of 1 to 2 and the mixture is then thawed.
After 15 min, the lysate is centrifuged at 13,000 x g and 4°C for 10 min.
The lysate may be divided into aliquots and may be shock-frozen and stored in liquid nitrogen.
In a further modified method, the proteolytic activity of an unknown sample may be determined in comparison with a comparison sample. The method for differential determination of the proteolytic activity comprises the following method steps:
a) incubating a library of fusion molecules comprising a peptide and a nucleic acid encoding said peptide with a proteolytically active sample, _g_ b) incubating the same library of fusion molecules in at least one parallel mixture with at least one proteolytically active comparison sample, c) isolating the proteolyticaily removed parts of said fusion molecules, d) constructing a differential nucleic acid bank, e) determining the sequence of the nucleic acids determined.
Differential determination of the proteolytically removed nucleic acid parts of the fusion molecules may be carried out, for example, by applying an excess of single-stranded nucleic acid parts which have been isolated from the comparison sample to a suitable support, saturating the free binding sites of the support followed by hybridization with single-stranded nucleic acid parts which have been isolated from the sample. The removable not hybridized nucleic acid parts specifically represent only peptide sequences cut in the sample.
In another method of the invention, the proteolytic activity of a sample is determined in comparison with at least one presample. Successive incubations of the fusion molecule library with presample and sample make it possible to determine the differential proteolytic activity of said presample and sample. The method comprises the following method steps:
a) incubating a library of fusion molecules comprising a peptide and a nucleic acid encoding said peptide with at least one proteolytically active presample, b) removing the fusion molecule parts removed by cleavage, c) incubating the thus prepared library with a proteolytically active sample, d) isolating the fusion molecule parts removed by proteolytic cleavage, e) determining the sequence of the nucleic acid part of the removed fusion molecules.
In preferred embodiments of all three method variants of the invention, the nucleic acid parts of the proteolytically cleaved and isolated fusion molecule parts are multiplied by means of PCR. For this purpose, fusion molecules are used whose nucleic acid part have on both sides of the coding nucleic acid sequence constant nucleic acid sequences as primer binding sites.
In order to increase greatly both the amount of fusion molecule cut in the extract studied and the specificity of the selection step (cleavage in the proteolytically active extract), preference is given in the method of the invention to amplifying the isolated nucleic acid of the cleaved fusion proteins by means of PCR and transcribing it in vitro. Starting from the RNA prepared in this way, a new library of fusion proteins is prepared again by methods already described and is incubated with the same extract containing proteolytic activities. This cycle can be repeated several times.
In a further variant of the method of the invention, it is possible to carry out PCR, in vitro transcription and/or in vitro translation in one or more of the cycles described with a higher than normal error rate. This increases the available peptide sequences to be tested (in vitro protein evolution).
Likewise, the variant of directed in vitro protein evolution makes it possible to alter in a rapid and simple manner known cleavage sequences such that different kinetic properties with respect to the cutting proteases are produced.
Further additives such as, for example, specifically acting protease inhibitors or nuclease inhibitors or unspecific RNA and/or DNA for saturating possible nucleases may be added to the proteolytically active sample. The addition of an inhibitor specifically acting against said proteases, such as, for example, the addition of pepstatin A, is recommended in particular, if the sample extracts used contain unspecifically cutting lysosomal or endosomal proteases (acidic peptidases such as aspartate peptidases). Specific inhibition of exoproteases may also be advantageous in order to prevent degradation of the proteins contained in the sample, comparison sample or presample, in particular of the other proteases and peptidic cofactors thereof. The attached fusion molecules themselves need not be protected against exoproteases, since no free N-or C-terminal peptide ends are present. Degradation of the peptide moiety of the proteolytically cleaved fusion molecules is unimportant for the further course of the method.
The removal of nuclease activities from tissue extracts to protect the nucleic acid parts of the fusion molecules may be achieved by treating freshly prepared tissue homogenate with RNAse inhibitors or DNAse inhibitors already during work-up of said tissue and subsequent steps. In another embodiment, the nuclease inhibitors may be coupled covalently to particles or suitable column materials. Nucleases binding to surface-coupled nuclease inhibitors are thus removed from the remaining tissue homogenate. An alternative to nuclease inhibition is protection of the nucleic acid moiety of fusion molecules by auxiliary substances. The auxiliary substances enclose the DNA and RNA, form a complex and thus protect said DNA and RNA against nucleases (Katayose, S. (1998) J.
Pharm. Sci., 87: 160-163). Said auxiliary substances are in particular polyethyleneimine or PEG-PLL but may also be proteins or chaparones.
Protection against nucleases is likewise possible by using artificial nucleic acids such as, for example, PNA in the fusion molecule. In order to increase stability, the nucleic acids may also be chemically modified, for example by methylation.
The incubation times of library and extract depend on the proteolytic activity of the sample used. In an extended embodiment, the methods of the invention are carried out repeatedly with varying incubation times. The advantage of this is that it is possible to make kinetic statements regarding the sample and the peptidic substrates thereof which include the library of fusion molecules. Peptidic substrates which are proteolytically cleaved with short incubation times have a high rate constant with respect to proteolytically active sample components.
Another simple possibility to make kinetic statements about the proteolytic activity of the sample is either to vary the concentration of the library fusion molecules which constitute the peptidic substrates or, preferably, to vary the concentration of the sample intended for incubation.
A great advantage of the method of the invention is the fact that the proteolytic activity of the sample studied can be described phenomenotogically, i.e. without knowing all of the proteolyticaNy active components of the sample. When comparing various samples, it does not matter whether a different proteolytic activity is based on the presence of various proteases or on direct mutation of a protease, which is rather rare, or whether it is caused here by faulty regulation of a protease gene or a protease inhibitor or protease activator. The present method can also record alterations in proteolytic activity which are induced via extracellular signals. It is known that, for example, binding of ligands such as hormones to transmembrane receptors can alter cytosolic kinase and/or phosphatase activities in a cell. The degree of phosphorylation plays a decisive part in the regulation of enzyme activities in cells and thus has a decisive influence on protease activity. A deregulated intracellular phosphatase activity and, coupled thereto, protease activity play a decisive part, for example, in the formation of cancer.
Especially advantageous is the differential determination of specifically cleavable peptides with respect to extracts of healthy and diseased tissue and also of healthy tissue and tissue infected with pathogens. Finding specifically cleavable peptides when comparing individual organisms is likewise of particular interest.
Furthermore, the methods of the invention may be used to determine rapidly and easily peptide sequences which are cut, in particular specifically cut, by the sample studied. On the other hand, sequences which are not subgect to proteolytic cleavage by the sample studied can likewise be found.
Apart from identifying specifically proteolytically cleavable peptide sequences, it is also possible to use the above-described methods for finding specifically acting inhibitors or activators, without needing to know for the individual case the proteases to be inhibited. For this purpose, a potential inhibitor or activator is simply added to the sample solution and the differential proteolytic activity between sample and samples containing potential inhibitors or activators is determined. A reduction in the number of cleaved sequences indicates an inhibitor and an increase in the number of cleaved sequences indicates an activator.
In a preferred embodiment, inhibitors are identified by using a preselected library of fusion molecules which have been obtained after several selection cycles using the proteolytically active sample. Alternatively, it is also possible to use the specifically cleavable peptides identified by the methods of the invention directly for screening of inhibitors, for example in a FRET assay.
Thus a method for screening of protease inhibitors is available, which can be carried out rapidly and easily.
The rapid and simple screening of protease inhibitors makes it also possible to combine many effective inhibitors in a mixture of active substances, in order to prevent the appearance of resistances.

The present invention further relates to proteolytically cleavable substances which contain a peptide sequence obtainable using one of the methods described.
The specifically proteolytically cleavable peptide sequences may be used for synthesizing specifically acting inhibitors. For this purpose, for example, the peptide sequence is chemically modified or one or more a-amino acids of the peptide are replaced with ~i-amino acids or L-amino acids are replaced with D-amino acids in order to prevent proteolysis (see, for example, Werder M. & Hauser H. (1999) Helvetica Chimica Acta, 82, 1774-1783). The specifically proteolytically cleavable peptides which can be identified using the present method may be used for constructing drug delivery systems in order to achieve selective release of active substance at the target site. The target site may be, for example, a particular organ, tissue, a particular cell type, a subcellular compartment or may be diseased, compared with healthy, tissue or microbially infected, compared with uninfected, tissues or cells.
In the case of an active substance which can be activated by protease cleavage, the specifically proteolytically cleavable substances may be used as covalently bound inhibitors. The specifically proteolytically cleavable peptides may also be used as linkers in order to link an active substance to a targeting ligand such as, for example, a specific antibody or a specific receptor ligand. However, it is also possible to construct active substance loaded systems which are closed with specifically proteolytically cleavable substances (Minko T. et al., (2000) Int. J. Cancer, 86: 108-117).
The variant of directed in vitro protein evolution makes it possible to alter known peptidic cleavage sequences such that different cleavage kinetics are produced. This makes it possible to control the time profile of active substance release.
Consequently, it is possible, with the aid of the specifically proteolytically cleavable peptides and via a target-controlled release of active substance, to control more efficiently diseases in which, compared with the healthy state, altered proteolytic activities are present at the desired site of action or which are caused by a faulty regulation or mutation of specific proteases or activators or inhibitors thereof. Examples of diseases with disturbed protease activity are asthma, osteoporosis, cancers such as leukemia, breast cancer, bowel cancer, stroke, neuronal disorders such as Alzheimer's disease, arthritis, pancreatitis, hypertension, thromboses, colds and schistosomiasis.
The targeted release of active substances can minimize the side effects of said active substances and the amount of active substance to be used can be reduced, owing to the altered distribution profile in the organism.
Especially interesting are pharmaceutical active ingredients, for example against bacteria, fungi, viruses or against diseased tissue cells, but also herbicides, fungicides or insecticides may be used.
Another application of the method of the invention is the use as diagnostic assay. The assay may be based, for example, on a comparison of the profile of the proteolytic activity of extracts toward a fusion molecule library with defined composition between diseased tissue, for example from cancer cells, and a sample to be studied. However, preference is given to assays which, compared to a comparison sample or presample are based on peptides proteolytically cleavable exclusively by diseased tissue. In such a selective assay, the sequencing step which is intended to follow removal of the fusion molecule parts removed by cleavage can be dispensed with. In this case, the presence of a particular disease-specific protease activity may be identified using a hybridization probe. Such a method may be highly integrated by using in such an assay many cleavable peptides identified as disease-specific markers.
Alternatively, it is also possible to use the specifically cleavable peptides identified by the methods of the invention directly for the screening of inhibitors, for example in a FRET assay. The specifically cleavable sequences may also be used in vivo as substrate for FRET assays.
The peptides specifically cleavable by a proteolytically active sample may also be used for isolating the protease cleaving said peptide sequence (see, for example, Li Y.M. (2000) Nature, 405: 689-694).
US 5843701, for example, describes a method applicable for isolating serine proteases. Serine proteases are protein enzymes which catalyze hydrolysis of peptide bonds within proteins. Frequently, owing to the specific peptide linkage within the substrate, the target protein is selectively cut. The serine proteases include, inter alia, tissue plasminogen activator, trypsin, elastase, chymotrypsin, thrombin and plasmin. Many disease stages can be treated with serine protease inhibitors, such as, for example, blood clotting disorders. Elastase inhibitors can reduce the clinical progression of emphysemas.
Proteases may be isolated by coupling the specifically proteolytically cleavable peptides, determined according to the methods described above, to column material, for example, by using standard methods and carrying out an affinity chromatography of the proteolytically active samples on said column material. The buffer used during the binding cycles must not be denatured so as not to inactivate the protease. The proteases interacting with the peptides can additionally be bound to the attached peptides via common crosslinking methods, if required. For this purpose, it is also possible to use ~i-amino acid-containing cleavage sequences which can be specifically recognized but cannot be cleaved by proteases.
The invention is clarified by some exemplary embodiments stated below:
Example 1:
Preparation of fusion molecules comprising a peptide with a known protease cleavage site and a nucleic acid encoding said peptide.
General remarks:
2 types of fusion molecules were prepared, firstly fusion molecules containing a peptide sequence which is specifically cleaved by the protease thrombin ("thrombin fusion molecules"), and secondly fusion molecules containing a peptide sequence which is specifically cleaved by matrix metalloprotease-3 (MMP-3) ("MMP-3 fusion molecules"). The preparation of fusion molecules of this type is described, for example, in W 098/31700.
The fusion molecules prepared in this way have protease cleavage sites for MMP-3 or for thrombin.
The MMP-3 protease cuts between Glu and Leu, and thrombin protease cuts between Arg and Ser. The peptide sequences used and the corresponding nucleotide sequence are listed below.

Protease cleavage sites:
MMP-3 cleavage site (Nagase, H., 1995, Methods Enzymol. 248,449-470) Protein (AS): Arg-Pro-Lys-Pro-Val-Glu- ; Leu-Trp-Arg-Lys (Seq.
IDNo.1) DNA (bp): AGA CCA AAA CCC GTT GAG CTC 'IGG AGA AAG (Seq.
ID No. 2) Thrombin cleavage site (Le Bonniec, B.F., 1996, Biochemistry 35, 7114-7122) Protein (AS): Val-Pro-Arg- ; Ser-Phe-Arg (Seq. ID No. 3) DNA (bp): GTT CCA AGA AGC TTC AGG (Seq. ID NO. 4) 1.1 Preparation of MMP-3 template DNA and thrombin template DNA
via PCR
250 ng of DNA template, 25 pmol of 5' and 3' primers, 10 mM dNTPs, 5 p.1 of 10 x PCR buffer and 5 U/wl Taq DNA polymerase were combined in a 50 lul reaction mixture (buffer and enzyme from Promega, Madison, USA). The PCR was carried out in a Biometra T gradient cycler: 1 min 94°C, 21 x (0.5 min 94°C, 0.5 min 55°C, 0.5 min 72°C).
Primer and template:
5'Strep Tag primer (90 bp):
5'-taa tac gac tca cta tag gga caa tta cta ttt aca att aca atg tgg tcc cac ccc cag ttc gag aag agt ggc tca agc tca gga tca-3' (Seq. ID No. 5) 3'MMP-3 primer (44 bp):
5'-ttt taa ata gcg gat get act agg cta gac cca gag cta ccc ga-3' (Seq. ID No.
6) 3'-thrombin primer (44 bp):
5'-ttt taa ata gcg gat get act agg cta gac cca gag gat cct ga-3' (Seq. ID No.
7) MMP-3 template:
5'-agt ggc tca agc tca gga tca gga tct ggt aga cca aaa ccc gtt gag ctc tgg aga aag cac cat cac cat cac cat gga agt ggc tcg ggt agc tct ggg tct agc-3' (Seq. ID No. 8) Thrombin template:
5'-agt ggc tca agc tca gga tca gga tct ggt gtt cca aga agc ttc agg cac cat cac cat cac cat gga agt ggc tca gga tcc tct ggg tct agc-3' (Seq. ID No. 9) . 17 .
Fig. 1 shows the PCR products on a 2% agarose ethidium bromide gel.
Lane 1: 50 by DNA marker, Lane 2: MMP-3 DNA, Lane 3: thrombin DNA, Lane 4: 100 by DNA marker.
As Fig. 1 shows, MMP-3 DNA and thrombin DNA were successfully amplified. The PCR products have the expected molecular weight of 199 by for MMP-3 DNA and 189 by for thrombin DNA.
1.2 In vitro transcription of MMP-3 and thrombin DNA
100 pmol of DNA were used in the transcription reaction (Ribomax T7/Promega, Madison, USA). The DNA was incubated with 5 x T7 buffer, rNTPs and T7 RNA pofymerase in a 500 w1 mixture at 37°C for 4 hours.
RNA was purified by means of phenol/chloroform extraction and analyzed on a 6% urea gel (Novex GeU Invetrogen, Groningen, Netherlands).
Fig. 2 shows the RNA obtained after transcription on a 6% urea gel.
Lane 1: MMP-3 RNA, Lane 2: thrombin RNA
As Fig. 2 shows, RNA can be detected, as expected, after the transcription reaction. The RNA is not degraded.
1.3 Ligation of RNA with puromycin linker via UV crossiinking 3 nmol of RNA, 4.5 nmol of puromycin (Pu) linker (PEG = polyethylene glycol, Pso = psoralen) having the sequence 5'-(Pso)2'OMe-U AGC GGA UGC AAA AAA AAA AAA AAA AAA PEG
PEG CC-(Pu)-3' (Seq. ID No. 10 represents nucleotides 1 to 28) (lntreractiva/Ulm), 12 w( of 10x ligase buffer (1 M NaCI, 250 mM Tris pH
7.5) were combined in a 120 ~I mixture and incubated at 85°C for 5 minutes and then at room temperature for 10 minutes. UV crosslinking was carried out at 366 nm for 15 minutes (handheld UV tamp by LTF-Labortechnik, 12 W).
Ligation efficiency was checked on a 5% urea-agarose gel.

Fig. 3 shows analysis of the ligation reaction on a 5% urea-TBE gel.
Lane 1: MMP-3 RNA, Lane 2: MMP-3 RNA with puromycin linker, Lane 3:
thrombin RNA, Lane 4: thrombin RNA with puromycin linker. (The signal L
is caused by the color marker).
After ligation of the puromycin linker to the RNA, a distinct increase in the molecular weight is observed, indicating that thrombin RNA and MMP-3 RNA has been linked successfully to puromycin.
1.4 In vitro translation 8 ~I of linker-RNA were mixed with 200 ~I of reticulocyte lysate (Promega, Madison, USA L416X), 5 ~I of ~S-methionin (Hartmann, 10.8 p.M, specific activity: the specific activity is 72181 dpm/pmol)), 6 p1 of amino acid mix (without methionin, 1 mM, Promega) and 80 ~I of H20 were mixed and then incubated at 30°C for 30 minutes. Addition of 130 ~I of 2 M KCI and 75 ~,I of MgCl2 was followed by another incubation at 30°C for 30 minutes.
The fusion molecules obtained were purified using oligo dT cellulose (Amersham Pharmacia, Freiburg, Germany) (removal of all those components of the in vitro translation mixture which contain no polyA RNA
sequences).
1.5 Reverse transcriptase reaction The purified 200 ~.I fusion molecules were mixed with 2.5 p1 of 100 ~,M
reverse primer and incubated at 80°C for 5 minutes. After cooling the reaction mixture on ice, 50 p,1 of 5x strand buffer, 20 p1 of dNTPs (in each case 10 mM), 2.5 p.1 of RT-Superscript II (all from Promega, Madison, USA) were added and the mixture was incubated at 42°C for 40 minutes.
Example 2:
Proteolytic cleavage of the fusion molecules by thrombin and MMP-3 in solution The fusion molecules prepared were incubated with the corresponding proteases and the reaction products were analyzed (Fig. 4). 10 nM MMP-3 (Sigma, Deisenhofen, Germany) were added to 5 pmol of fusion molecule containing the MMP-3 cleavage site and the mixture was incubated in MMP-3 buffer (50 mM Tris pH 7, 150 mM NaCI, 10 mM CaCl2) in a total volume of 15 ~I at 37°C for 45 minutes. 1 U of thrombin (1 NIH unit =
0.324 pg, Sigma, Deisenhofen, Germany) was added to 5 pmol of fusion molecule containing the thrombin cleavage site and incubated in thrombin buffer (50 mM Tris pH 8, 150 mM NaCI) in a total volume of 15 p1 at 37°C
for 45 minutes. The starting fusion molecules and the fusion molecules after proteolytic cleavage were analy2ed using a phosphoimager, after SDS PAGE (Fig. 4) Fig. 4 shows the fusion molecule digest with thrombin and MMP-3 in solution:
Phosphoimager image after 4-20°l° Tris-glycine SDS PAGE.
Lane 1: "MMP-3 fusion molecules" prior to digest;
Lane 2: "thrombin fusion molecules" prior to digest;
Lane 3: "thrombin fusion molecules" + thrombin Lane 4: "thrombin fusion molecules" prior to digest Lane 5: "MMP-3 fusion molecules" + MMP-3 Lane 6: "MMP-3 fusion molecules" prior to digest.
(The bands i show fusion molecules, the bands ii show peptidic by products and the bands iii are caused by N-terminal fusion molecule fragments.) The "MMP-3 and thrombin fusion moiecutes» were proteolytically cleaved by MMP-3 protease and thrombin protease, respectively.
Example 3 Isolation and detection of the proteolytically cleaved "thrombin fusion molecules" and amplification of the nucleic acid moiety Specifically proteolytically cleaved fusion molecules were isolated and identified by carrying out three successive method steps. Fig. 5 shows diagrammatically the method steps according to this example for identifying specifically cleavable peptides:
Step 1. Coupling of the fusion molecules to Streptactin-Sepharose (support material) via an N-terminal Strep tag Step 2. Proteolytic cleavage of the "thrombin fusion molecules" on the support material by thrombin.

Step 3. Isolation of the cleaved C-terminal fusion molecule fragments and detection of said fragments after PCR amplification of the nucleic acid moiety.
Optionally, the isolated fragments may be purified via their His tag by means of affinity chromatography to nickel particles prior to analysis.
3.1 Coupling of "thrombin fusion molecules" to Streptactin Sepharose 50 ~I of "thrombin fusion molecules" (corresponding to approx. 5 pmol) were [lacuna] with 50 ~.I of Streptactin Sepharose, washed 5x in Streptactin buffer (150 mM NaCI, 100 mM Tris pH 8) and then incubated at 4°C for 2 hours. Unbound fusion molecules were removed by washing five times with Streptactin buffer.
3.2 Proteolytic digest of the "thrombin fusion molecules" by thrombin The fusion molecules bound to Streptactin Sepharose were resuspended in 80 ~I of water and incubated with 10 p1 of 10x thrombin buffer (1.5 M NaCI, 500 mM Tris pH 8) and 10 ~.I of 100 U/~,I thrombin at 37°C for 45 minutes.
As a control, the same reaction was carried out without thrombin.
3.2.1 Isolation and detection of the bound "thrombin fusion moleculesn After the incubation period, the reaction mixture was centrifuged (1 minute at 3000 rpm). The pelleted Streptactin Sepharose containing the N-terminal fragment of the cleaved fusion molecules was washed three times with Streptactin buffer and then resuspended in 80 ~I of water.
The "thrombin fusion molecules" bound to Streptactin Sepharose (control mixture) or the N-terminal fragments bound after incubation with thrombin were fractionated by SDS PAGE and visualized by means of phosphoimaging (Fig. 6).
Fig. 6 shows "thrombin fusion molecules" and fusion molecule fragments bound to Streptactin Sepharose. 15 ~I of resuspended Streptactin Sepharose were mixed in each case with 5 ~.I of Lammli loading buffer, heated at 82°C for 5 min and fractionated by means of 4-20% Tris glycine SDS PAGE. The gel was then dried at 80°C for 2 hours and visualized by means of phosphoimaging.

Lane 1: "thrombin fusion molecules" after incubation with thrombin. Lane 2:
"thrombin fusion molecules" without incubation with thrombin (control).
(Band i shows the infact fusion molecule, bands ii are from peptidic by-products and band iii is caused by the N-terminal fusion molecule fragment.).
After incubating the "thrombin fusion molecules" with thrombin, only the N-terminal fragments (labeled with 35S-methionin) remain bound to Streptactin Sepharose.
3.2.2 Detection of the nucleic acid moieties of the "thrombin fusion moleculesn by means of PCR analysis After removing Streptactin Sepharose, 15 ~I of the supernatant or 15 ~,I of resuspended Streptactin Sepharose were used for the PCR. Apart from the 15 ~I samples, the 50 ~,l PCR mixtures also contained 5 ~I of 5' and 3' thrombin primers (5 pmoU~.l), 10 mM dNTPs, 5 ~I of 10 x PCR buffer and 5 U/~I Taq DNA polymerase (buffer and enzyme from Promega, Madison, USA). The amplification was carried out under the following conditions:
1 min 94°C, 21 x (0.5 min 94°C, 0.5 min 55°C, 0.5 min 72°C). The PCR
products obtained are depicted in Fig. 7.
Fig. 7 shows the PCR products of the "thrombin fusion molecules" after 2%
agarose gel electrophoresis. Lanes 1 and 2 show the products as black signals on a light background and lanes 3 and 4, owing to a different imaging technique, show the products as white signals on a dark background.
Lane 1: °thrombin fusion molecules" + thrombin; supernatant.
Lane 2: "thrombin fusion molecules" without thrombin (control);
supernatant.
Lane 3: "thrombin fusion molecules" + thrombin; bound to Streptactin Sepharose.
Lane 4: "thrombin fusion molecules" without thrombin; bound to Streptactin Sepharose.
Consequently, it is shown that "thrombin fusion molecules" are cut by incubation with thrombin. After removing the N-terminal fragment which contains no nucleic acid moiety, the C-terminal moiety remains in the supernatant. Consequently, a strong PCR amplicon is obtained in the supernatant (see, for example, lane 1, Fig. 7) but in the pellet only a very weak PCR amplicon (e.g. lane 3, Fig. 7) is obtained. In the control reaction, the "thrombin fusion molecules" were not incubated with thrombin.
Consequently, the fusion molecules remain bound to Streptactin Sepharose and a strong PCR amplicon is obtained in the pellet (e.g. lane 4, Fig. 7), while the PCR amplicon of the supernatant (e.g. lane 2, Fig. 7) is only very weak.

r CA 02420065 2003-02-19 SEQUENCE LISTING
<110> Xzillion GmbH & Co KG
<120> Methods for identifying specifically cleavable peptides and use of such peptide sequences <130> 200at19.wo <140>
<141>
<150> 10041238.6 <151> 2000-08-22 <160> 10 <170> Patentln Ver. 2.1 <210> 1 <211> 10 <212> PRT
<213> Artificial sequence <220>
<223> Description of artificial sequence: MMP-3 cleavage site <400> 1 Arg Pro Lys Pro Val Glu Leu Trp Arg Lys <210> 2 <211> 30 <212> DNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: MMP-3 cleavage site-encoding DNA
<400> 2 agaccaaaac ccgttgaqct ctggagaaag 30 <210> 3 <211> 6 <212> PRT
<213> Artificial sequence <220>
<223> Description of artificial sequence: thrombin cleavage site <400> 3 Val Pro Arg Ser Phe Arg <210> 4 <211> 18 <212> DNA
<213> Artificial sequence <220> ' <223> Description of artificial sequence: thrombin cleavage site-encoding DNA
<400> 4 gttccaagaa gcttcagg <210> 5 <211> 90 <212> DNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: 5' Strep tag primer <400> 5 taatacgact cactataggg acaattacta tttacaatta caatgtggtc ccacccccag 60 ttcgagaaga gtggctcaag ctcaggatca g0 <210> 6 <211> 44 <212> DNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: 3' 1~IP-3 primer <400> 6 ttttaaatag cggatgctac taggctagac ccagagctac ccga qq <210> 7 <211> 44 <212> DNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: 3' thrombin primer <400> 7 ttttaaatag cggatgctac taggctagac ccagaggatc ctga 44 <210> 8 <211> 108 <212> DNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: 1~IP-3 template <400> 8 agtggctcaa gctcaggatc aggatctggt agaccaaaac ccgttgagct ctggagaaag 60 caccatcacc atcaccatgg aagtggctcq qgtagctctg ggtctagc 108 <210> 9 <211> 96 <212> DNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: thrombin template <400> 9 agtggctcaa gctcaggatc aggatctggt gttccaagaa gcttcaggca ccatcaccat 60 caccatggaa gtggctcagg atcctctggg tctagc 96 <210> 10 <211> 28 <212> RNA
<213> Artificial sequence <220>
<223> Description of artificial sequence: puromycin-linker-RNA
part <400> 10 uagcggaugc aaaaaaaaaa aaaaaaaa 2g

Claims (34)

claims:

1. A method for identifying specifically proteolytically cleavable peptides, comprising the method steps:
a) incubating a library of fusion molecules comprising a peptide and a nucleic acid encoding said peptide with a proteolytically active sample, b) isolating the proteolytically removed parts of said fusion molecules, c) determining the sequence of the nucleic acid part of the removed fusion molecules.

2. A method for identifying specifically. proteolytically cleavable peptides, comprising the method steps:
a) incubating a library of fusion molecules comprising a peptide and a nucleic acid encoding said peptide with a proteolytically active sample, b) incubating the same library of fusion molecules in a parallel mixture with at least one proteolytically active comparison sample, c) isolating the proteolytically removed parts of said fusion molecules, d) constructing a differential nucleic acid bank, e) determining the sequence of the nucleic acids determined.

3. A method for identifying specifically proteolytically cleavable peptides, comprising the method steps:
a) incubating a library of fusion molecules comprising a peptide and a nucleic acid encoding said peptide with at least one proteolytically active presample, b) removing the fusion molecule parts removed by cleavage, c) incubating the thus prepared library with a proteolytically active sample, d) isolating the fusion molecule parts removed by proteolytic cleavage, e) determining the sequence of the nucleic acid part of the removed fusion molecules.

4. The method as claimed in any of the preceding claims, wherein a library of fusion molecules which comprise a peptide and a nucleic acid encoding said peptide and flanked on one or both sides by known nucleotide sequences is used.

5. The method as claimed in any of the preceding claims, wherein a library of fusion molecules which comprise a peptide linked via a puromycin molecule to a polyA nucleic acid sequence which is followed by the peptide-encoding region and one or more known nucleic acid sequences is used.

6. The method as claimed in any of the preceding claims, wherein fusion molecules comprising peptides composed of at least 4 variable amino acids are used.

7. The method as claimed in any of the preceding claims, wherein fusion molecules comprising peptides composed of 7 to 11 variable amino acids are used.

8. The method as claimed in any of the preceding claims, wherein fusion molecules are used, which contain peptides having at least one constant sequence region in addition to the variable sequence region.

9. The method as claimed in any of the preceding claims, wherein fusion molecules having a labeled nucleic acid part and/or peptide part are used.

10. The method as claimed in claim 9, wherein fusion molecules having a magnetic label, a radiolabel, a fluorescent label of the nucleic acid part and/or peptide part or having a label containing a specific known nucleic acid sequence and/or a known specific amino acid sequence are used.

11. The method as claimed in any of the preceding claims, wherein the fusion molecules of the library are attached to a solid surface.

12. The method as claimed in any of the preceding claims, wherein the fusion molecules of the library are attached to solid particles.

13. The method as claimed in any of the preceding claims, wherein the known nucleic acid sequences are cut by restriction enzymes and only the nucleic acid part which has been cut off or cut out is used further.

14. The method as claimed in any of claims 4 to 13, wherein, after isolating the fusion molecule parts cut by the proteolytically active sample, a PCR is carried out in order to multiply the nucleic acid parts.

15. The method as claimed in claim 14, wherein the amplified nucleic acid is used for constructing a new library of fusion molecules and a method cycle as claimed in claims 1 to 3 is carried out again using said library.

16. The method as claimed in claim 14, for in vitro protein evolution, wherein the PCR, the in vitro transcription and/or the in vitro translation proceed at an increased error rate.

17. The method as claimed in any of the preceding claims, wherein the proteolytically active sample, presample and comparison sample used is a total cell extract, a cytosolic extract, a membrane extract, an extracellular extract, a subcellular extract, a combination of said extracts or a solution of known proteases.

18. The method as claimed in any of the preceding claims, wherein the sample, presample or comparison sample used is disease-specific and/or tissue-specific and/or organ-specific and/or organism-specific extracts.

19. The method as claimed in any of the preceding claims, wherein the proteolytically active sample and/or presample are used in the method under physiological conditions.

20. The method as claimed in any of the preceding claims, wherein further auxiliary substances are added to the sample and/or presample.

21. The method as claimed in claim 20, wherein the auxiliary substances added are specifically acting protease inhibitors and/or nuclease inhibitors.

22. The method as claimed in any of the preceding claims, wherein potential protease inhibitors or protease activators are added to the sample.

23. The method as claimed in any of the preceding claims, wherein the method steps are repeated or are repeated with different incubation periods and/or different sample concentration.

24. The use of the nucleic acid sequences determined according to the method as claimed in claims 1 to 23 for preparing specifically proteolytically cleavable substances.

25. A fusion molecule comprising a proteolytically cleavable peptide and a nucleic acid encoding said peptide.

26. A specifically proteolytically cleavable peptide obtainable by a method as claimed in any of claims 1 to 23, comprising a chemical active substance.

27. The specifically proteolytically cleavable peptide as claimed in claim 26, wherein the chemical active substance can be activated by proteolytic release.

28. The specifically proteolytically cleavable peptide as claimed in either of claims 26 and 27, wherein further auxiliary substances and additives are added to the substance.

29. The use of the fusion molecules comprising a proteolytically cleavable peptide and a nucleic acid encoding said peptide for -26a-identifying specifically proteolytically cleavable peptides for preparing medicaments for asthma, osteoporosis, cancers, stroke, neuronal disorders, arthritis, pancreatitis, hypertension, thromboses, colds or schistosomiasis.

30. The use of fusion molecules comprising a proteolytically cleavable peptide and a nucleic acid encoding said peptide for identifying specifically proteolytically cleavable peptides for preparing diagnostic markers.

31. A method for detecting specifically acting proteases as claimed in claims 1 to 23, comprising a diagnostic marker as claimed in claim 30.

32. The use of fusion molecules comprising a proteolytically cleavable peptide and a nucleic acid encoding said peptide for identifying specifically proteolytically cleavable peptides according to a method as claimed in claims 1 to 23 for identifying the proteases cleaving said peptides.

33. The use of fusion molecules comprising a proteolytically cleavable peptide and a nucleic acid encoding said peptide for identifying specifically proteolytically cleavable peptides according to a method as claimed in claims 1 to 23 for identifying protease inhibitors.

34. The use of fusion molecules comprising a proteolytically cleavable peptide and a nucleic acid encoding said peptide for identifying specifically proteolytically cleavable peptide sequences according to a method of claims 1 to 23 for preparing inhibitors.