WO2007138171A1 - Bioactive cyclic peptide - Google Patents

Abstract

The present invention relates to cyanobacterial cyclic peptides and particularly to a novel nostocyclopeptide. The invention further provides a compound and a pharmaceutical composition comprising said nostocyclopeptide. The invention also discloses said nostocyclopeptide for use as a medicament as well as the use of nostocyclopeptide for the manufacture of a medicine for the treatment against hepato toxins.

Description

Bioactive cyclic peptide

FIELD OF THE INVENTION

The present invention relates to cyanobacterial cyclic peptides and particularly to a novel nostocyclopeptide. The invention provides a sequence of nostocyclopeptide Ml. The invention further provides a compound and a pharmaceutical composition comprising nostocyclopeptide Ml. Nostocyclopeptide Ml for use as a medicament is provided as well as the use of nostocyclopeptide Ml for the manufacture of a medicine for the treatment against hepatotoxins.

BACKGROUND OF THE INVENTION

Cyanobacteria have an impressing capability to produce secondary metabolites. Equally fascinating is the broad range of biological activity of the peptides and peptide derivatives. A number of the cyanobacteria peptides or peptide derivates are toxic to higher organisms, including human. The most known, the microcystins and nodularin are potent hepatotoxins, and inhibit serine/threonine protein phosphatases at very low concentrations [MacKintosh et al. , 1990). During the last years, several novel peptide derivates with cytotoxic activity have been described (for thorough descriptions of cyanobacterial toxins, see (Sivonen and Jones, 1999; Burja et al., 2001; Rao et al, 2002)]. The mechanisms of action vary between the abovementioned protein phosphatase inhibitors to interference with tubulin assembly and disassembly [symplostatin 1 (Mooberry et al. , 2003)], inhibition of proteases [aerogunisin (Kodani et al., 1998), microviridin (Ishitsuka et al., 1990)] or other enzymes [anabaenopeptin (Itou et al., 1999)], protein synthesis inhibition [cylindrospermopsin (Runnegar et al., 2002)] and blocking of ion channels in nerve cells [anatoxin (Macallan et al. , 1988) and saxitoxin (Penzotti et al., 1998)]. In many cases, however, the compounds are only reported to have cytotoxic effects, without description of their specific mechanism(s) of action.

Peculiar for the prokaryotes and especially cyanobacteria is the large variety of peptides sometimes present in large amounts. From marine cyanobacteria it has been estimated that about 40% of the secondary metabolites are cyclic or linear lipopetides and 5% are pure amino acid compounds. (Burja et al, 2001). The peptides thus constitute almost 50% of the secondary metabolite pool isolated from cyanobacteria.

The cyclic peptides are of particular interest because of their generally high bioactivity and structural diversity, and the fact that they are phylogenetically very old (Rantala et al., 2004). The synthesis of cyclic peptides is nonribosomal and controlled by cassettes of enzymes encoded by gene clusters (Meissner et al., 1996). The gene clusters are subjected to natural recombination [imperfect repeats, gene loss and horizontal gene transfer (Mikalsen et al., 2003)], which can explain the large variation in cyclic peptide sequence. It has been postulated that the gene clusters coding the nonribosomal peptide synthesis pathways can be engineered to produce peptides with desired activity such as antibiotic, immunosuppressant or anti-cancer (Neilan et al., 1999).

WO 03/051910 discloses that a peptide may have one or more non-pep tide bonds such as imino, hydraxzide, semicarbazide or azo bonds. It also discloses viral protease inhibitors, particularly HCV protease inhibitors. WO 01/66565 discloses peptide libraries and cyclic peptide libraries. Nostocyclopeptides Al, A2 and A3 are known from the publication of Golakoti et al., 2001. They are cyclic heptapeptides possessing a unique imino linkage in the macrocyclic ring. Common features of the nostocyclopeptides are that they are cyclic, amino acids having tyrosine in position 1 and methylproline in position 6.

There is a growing need in the field of microbiology for peptides and compounds effective for protecting subjects against the toxic influence of cyanobacterial toxins.

SUMMARY OF THE INVENTION

The present invention is based on the surprising finding that a novel molecule nostocyclopeptide Ml is able to protect completely primary hepatocytes against the apoptogenic hepatotoxins microcystin and nodularin.

Contrary to previously described nostocyclopeptides no cytotoxicity was found. It was, however, found that nostocyclopeptide Ml inhibit uptake of microcystin and nodularin in primary rat hepatocytes, presumably by inhibiting organic anion transporter polypeptides (Oatp).

The present invention is related to a p eptide that is a nostocyclopeptide having the sequence Tyrl -Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7. The nostocyclopeptide comprises nostocyclopeptide Ml.

In an embodiment the invention is related to a peptide that has a formula (I):

In another embodiment of the invention the peptide is isolated from Nostoc sp. cyanobacteria.

The present invention is also related to a compound comprising a nostocyclopeptide having the sequence Tyrl-Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7.

In one embodiment the invention is related to a compound comprising a peptide that has a formula (I).

The present invention is also related to a pharmaceutical composition comprising a peptide or a compound of the invention. In a preferred embodiment the peptide comprises nostocyclopeptide Ml. In one embodiment the invention is related to a nostocyclopeptide for use as a medicament. The present invention is related to the use of the peptide of the invention for the manufacture of a medicine for the treatment against hepato toxins.

The present invention is further related to the use of the claimed peptide as a cell reagent. The present invention is also related to the use of the claimed peptide of in a screening method.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 depicts the chromatogram of semi-preparative HPLC (High performance liquid chromatography) -purification of nostocyclopeptide Ml. Nostocyclopeptide Ml was extracted from Nostoc sp. and subjected to anion exchange and Cl 8 reverse phase solid phase extraction as described in Example 1. The chromatogram shows semi- preparative Cl 8 reverse phase HPLC of the sample after solid phase extraction. M-I eluted at 18.9 minutes (peak 1), and a similar compound at 20.6 (peak 2). The grey dotted line designates the gradient as percent water in the pump. For details on extraction and HPLC instrumentation and protocol, see Example 1.

Figure 2 depicts the enantiomeric analysis of the amino acids in the cyclic peptide nostocyclopeptide Ml. The peptide was subjected to acid hydrolysis and derivatized as described in Example 1. The hydro lysate was then injected into a Chirex d-pencillamine column to be able to separate L and D forms of the amino acids.

Figure 3 depicts the structure of nostocyclopeptide Ml. The amino acid sequence is (Tyrl-Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7).

Figure 4 depicts the effect of nostocyclopeptide Ml on microcystin induced apoptosis in primary rat hepatocytes in suspension.

Figure 4 A depicts that hepatocytes were preincubated with 5μM nostocyclopeptide Ml for 15 minutes (open circles) or carrier (closed circles) before addition of microcystin (MC-LR), and incubated for another 2 hours. The cells were then fixed in buffered formaldehyde (2%) solution and the percent apoptotic cells determined by microscopic evaluation of surface membrane morphology.

Figure 4B depicts that nostocyclopeptide Ml inhibited ultra rapid apoptosis induced by nodularin. Hepatocytes were treated as described in the legend to Figure 4A, but were given 5μM of nodularin. Aliquots were fixed at the given time points and percent apoptotic cells scored as described in the legend to Figure 4A.

Figure 4C depicts that nostocyclopeptide Al and A2 and their demethyl-derivatives isolated from Nostoc sp. ATCC53789 protected against microcystin induced apoptosis.

The experiment was performed like described in the legend to Figure 4 A, but with 1.5 hours incubation with microcystin.

Data are means of 3-5 experiments and standard deviation except in Figure 4C, where only 2 experiments were run due to limited material.

Figure 4D - 41: Surface membrane (B, D and F) and nuclear (C, E and G) morphology of primary hepatocytes in suspension exposed to 10OnM nodularin for 2 h with or without pre- and co-incubation with 5μM of nostocyclopeptide Ml.

Figure 4D and 4E depict control cell.

Figure 4F and 4G depict ells treated with nodularin.

Figure 4H and 41 depict cells pretreated with nostocyclopeptide Ml for 15 minutes before addition of nodularin in the medium. Bar is 20 μm.

Figure 5 depict that nostocyclopeptide Ml did not protect against intracellularly delivered apoptosis inducers.

Figure 5 A depicts that normal rat kidney epithelial cells were micro injected with nodularin or a mixture of nodularin and nostocyclopeptide and the percentage of rounded cells were counted at the given time-points. The data show one experiment where 35-45 cells were injected.

Figure 5B depicts that freshly isolated primary rat hepatocytes in suspension were treated as described in the legend to Figure IA, but with okadaic acid instead of microcystin.

Figure 5C depicts that hepatocytes were pretreated with nostocyclopeptide Ml or the pan-caspase inhibitor zVAD-fmk for 15 minutes before the addition of TNF α (50ng/ml) and cycloheximide (2μg/ml), and incubated for 4 hours. The experiment was stopped and apoptotic cell death scored as described in the legend to Figure IA. The data in Figure 5 B and 5 C are average of 3 - 5 experiments and standard deviation.

Figure 6 depicts that nostocyclopeptide Ml inhibited nodularin-associated protein phosphorylation. Freshly isolated primary rat hepatocytes in suspension were incubated with [³²Pi] and either nostocyclopeptide Ml or carrier for 20 min. Nodularin (20OnM) was then added and the cells were further incubated for 20 min before stopping the experiment by adding TCA (10% final cone). The proteins were separated on a gradient gel (7-15% polyacrylamide) and phosphoproteins were visualised by autoradiography. Lane 1, 3, 5 and 7 are loaded with twice the protein content as compared to lane 2, 4, 6 and 8.

Figure 7 depicts that nostocyclopeptide Ml did not inhibit CaM-KII-mediated phosphorylation in vitro. Rat liver cell lysates were incubated with [³²Py]-ATP and CaM-KII, nodularin, KN93 or Ml in various combinations. The reaction mixture were incubated at 25 °C and aliquots were taken at 15, 30 and 120 seconds, transferred to SDS-loading buffer, heated to 95°C for 5 minutes and subjected to SDS PAGE for separation of proteins. ³²P-labelled phosphoproteins were visualised by autoradiography. The figure shows phosphorylation for 30 seconds.

Figure 8 depicts accumulation of radiolabeled microcystin YR and glycodeoxycholate in primary rat hepatocytes treated with nostocyclopeptide Ml (Ml), sulfobromophthalein (SBP) or carrier (DMSO). Freshly isolated rat hepatocytes were preincubated with nostocyclopeptide Ml, SBP or carrier for 5 minutes and then added radiolabeled microcystin (A) or glycodeoxycholate (B). After 7 (microcystin) or 5 (glycodeoxycholate) minutes, the cells were washed and radioactivity measured as described in the materials and methods section. Figure 8A shows average of one representative experiment performed in duplicates and the sample deviation. Figure 8B shows the average of 3 - 5 experiments and standard deviation.

Figure 9 depicts that nostocyclopeptide Ml inhibits uptake of glycocholate in a competitive manner. Freshly isolated primary rat hepatocytes in suspension were incubated with Ml or vehicle before addition of radiolabeled glycocholate as described in the Example 1.

Figure 9A depicts Michalelis-Menten plot.

Figure 9B depicts Lineweaver-Burk plots.

Figure 9C depicts Lineweaver-Burk plots and is an enlarged portion of Figure 9B to show unaltered Vmax and decreasing Km.

In the following, the invention will be described in greater detail by means of embodiments with references to the attached figures.

DETAILED DESCRIPTION OF THE INVENTION

Abbreviations

CamKII Calcium/Calmodulin dependent protein kinase II

COSY Correlation Spectroscopy

HMBC Heteronuclear Multi-Bond Correlation

HSQC Heteronuclear Single Quantum Coherence

NOESY Nuclear Overhauser Effect Spectroscopy

Ntcp Na⁺-taurocholate cotransport proteins

Oatp Rat organic anion transporter polypeptide

PPl protein phosphatase 1

PP2A protein phosphatase 2A

SBP Sulfobromophthalein (also named bromosulfophthalein, BSP)

TOCSY TOtal Correlation Spectroscopy

Definitions

Unless otherwise specified, the terms used in the present invention, have the meaning commonly used in the art to which this invention belongs.

Some terms, however, may be used in a somewhat different manner and some terms benefit from additional explanation to be correctly interpreted for patent purposes. Therefore some of the terms are explained in more detail below. The term "peptide" refers to a molecule formed from the linking, in a defined order, of various amino acid residues, including derivatives. The peptide can be of any length, at least two, usually at least 3, 4, 5 or 6 to 7, often about 7, generally 7 to 8, normally about 9, 10, 11, 12, 13 or 14 or more amino acid residues. The peptide or polypeptide can be composed of amino acids joined to each other by peptide bonds or modified peptide bonds, i.e., peptide isosteres, and may contain amino acids other than the gene-encoded amino acids.

"Peptides" can include coded and non-coded amino acids, chemically or biochemically modified, for example post-translationally modified such as glycosylated or derivatized amino acids, polymeric polypeptides and polypeptides having modified peptide backbones. The "peptides" or "polypeptides" may be modified by either natural process, such as posttranslational processing, or by chemical modification techniques which are well known in the art. Such modifications are well described in basic texts and in more detailed monographs, as well as in a voluminous research literature. Modifications can occur anywhere in a peptide or polypeptide, including the peptide backbone, the amino acid side-chains and the amino or carboxyl termini. It will be appreciated that the same type of modification may be present in the same or varying degrees at several sites in a given peptide or polypeptide. Also, a given peptide or polypeptide may contain many types of modifications. Peptides or polypeptides may be branched, for example, as a result of ubiquitination, and they may be cyclic, with or without branching. Cyclic, branched, and branched cyclic polypeptides may result from posttranslation natural process or may be made by synthetic methods. Modifications include acetylation, acylation, ADP-ribosylation, amidation, covalent attachment of flavin, covalent attachment of a heme moiety, covalent attachment of a nucleotide or nucleotide derivative, covalent attachment of a lipid or lipid derivative, covalent attachment of phosphotidylinositol, cross-linking, cyclization, disulfide bond formation, demethylation, formation of covalent cross-links, formation of cysteine, formation of pyroglutamate, formylation, gamma-carboxylation, glycosylation, GPI anchor formation, hydroxylation, iodination, methylation, myristoylation, oxidation, pegylation, proteolytic processing, phosphorylation, prenylation, racemization, selenoylation, sulfation, transfer-RNA mediated addition of amino acids to proteins such as arginylation, and ubiquitination. (See, for instance, Proteins — structure and molecular properties, 2nd Ed., T. E. Creighton, W. H. Freeman and Company, New York (1993); Postranslational covalent modification of proteins, B. C. Johnson, Ed., Academic Press, New York, pgs. 1-12 (1983); Seifter et al, Meth Enzymol 182:626- 646 (1990); Rattan et al., Ann NY Acad Sci 663:48-62 (1992).)

The term "cyclic peptide" or cyclic protein refers to a peptide or polypeptide whose amino and carboxy termini are themselves linked together with a peptide bond, forming a circular chain. A number of cyclic peptides have been discovered in nature and they can range from a few amino acids in length to hundreds. Cyclic peptides tend to be extremely resistant to digestion, allowing them to survive intact in the human digestive tract. This trait makes cyclic peptides attractive to protein based drug designers for use as scaffolds which, in theory, could be engineered to incorporate any arbitrary protein domain of medicinal value, in order to allow those components to be delivered orally.

Proteins are polypeptide molecules (or consist of multiple polypeptide subunits).

General Description of the invention

The present invention is based on the surprising finding that a novel molecule nostocyclopeptide Ml is able to protect completely primary hepatocytes against the apoptogenic hepatotoxins microcystin and nodularin. It was found that nostocyclopeptide Ml inhibit uptake of microcystin and nodularin in primary rat hepatocytes, presumably by inhibiting organic anion transporter polypeptides (Oatp).

The present invention comprises nostocyclopeptide Ml having the sequence Tyrl- Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7.

Nostocyclopeptide Ml comprises a peptide that has a formula (I):

The peptide of the present invention is isolated from Nostoc sp. cyanobacteria.

The present invention provides a significant improvement for protecting subjects, especially humans and animals against the lethal influence the cyanobacterial toxins, which exists in increasingly amounts in the algal blooms around the world.

The present invention discloses for the first time a novel cyclic peptide, nostocyclopeptide Ml, with anti-apoptotic activity. Contrary to previously described nostocyclopeptides (Golakoti et al , 2001) no cytotoxicity was found. It was, however, found that nostocyclopeptide Ml inhibit uptake of microcystin and nodularin in primary rat hepatocytes, presumably by inhibiting organic anion transporter polypeptides (Oatp). The present invention shows that other activities than cytotoxicity must be taken into account when looking for bioactive compounds, and might also shed new light on the role of cyclic peptides in the cyanobacteria.

This present invention relates to biological activities of cyanobacterial cyclic peptides against eucaryotic cells. A family of cyclic peptides found from Nostoc sp cyanobacteria protects completely hepatocytes against the lethal influence the cyanobacterial toxins microcystin and nodularin which exists in increasingly amounts in the algal blooms The present invention is based on the surprising finding that a novel molecule nostocyclopeptide Ml is able to protect completely primary hepatocytes against the apoptogenic hepatotoxins microcystin and nodularin around the world.

The invention concerns a bioactivity of a family of molecules which are isolated from Nostoc sp. cyanobacteria. Common features of these molecules are that they are cyclic, amino acids tyrosine is in position 1 and methylproline in position 6 and above all instead of an ordinary peptide lingake an imino lingake novel for cyclic cyanobacterial peptides connects two amino acids together.

A new member of this peptide family called nostocyclopeptides was isolated from a benthic Baltic Sea Nostoc sp. cyanobacterium. The compound was named as nostocyclopeptide Ml (amino acid sequence: Tyrl-Tyr2-(D-HSer)3-L-Pro4-L-Val5- ((2S,4S)-4-MePro)6-Tyr7). This novel molecule nostocyclopeptide Ml was able to protect completely primary hepatocytes against the apoptogenic hepatotoxins microcystin and nodularin. Mechanism of this protective effect was shown to be the selective inhibition of the intake of the toxins microcystins and nodularin. Presumably this inhibition was mediated through a small subset of the organic anion transporters. Nostocyclopeptide Ml did not show any kind of toxic character against several mammalian cell lines tested and appears therefore to be a non-toxic inhibitor of microcystin-induced hepatocyte death.

General non- toxicity of nostocyclopeptide Ml makes it a molecule which could be applied for preventing/minimizing the necrosis of liver tissue in cases of microcystin and nodularin exposures. More generally nostocyclopeptide Ml can be used to block the intake of all the molecules into the hepatocytes which are using the same intake system. Nostocyclopeptide Ml can be used also as a cell reagent in research purposes to selectively inhibit the Microcystin/nodularin intake machinery.

A novel bioactive nostocyclopeptide was isolated from a benthic Baltic Sea Nostoc sp. cyanobacterium. The compound nostocyclopeptide Ml (Ml: Tyrl-Tyr2-(D-HSer)3-L- Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7) had an imino bond between Tyrl and Tyr7. It was able to protect completely primary hepatocytes against the apoptogenic hepatotoxins microcystin and nodularin, by inhibiting their cellular entry. Nostocyclopeptide Ml did not protect against intrecellularly delivered phosphatase inhibitors or against apoptogens acting via membrane receptors, like TNFα and TGFβ. Moreover, Ml did not interfere with intracellular microcystin- induced death-signalling events like CaMKII-dependent protein hyperphosphorylation. Ml appeared to block only partially bile acid uptake into hepatocytes. Compared to sulfobromophthalein, Ml showed much higher inhibition of microcystin uptake than glycodeoxycholic acid uptake. Nostocylopeptide Ml is a selective inhibitor of the uptake of microcystin into hepatocytes, presumably via inhibition of a small subset of the organic anion transporters. Ml was tested against several mammalian cell lines at concentration up to 0.1 mM without showing any sign of toxicity and appears therefore to be an apparently non-toxic inhibitor of microcystin- induced hepatocyte death.

MSl data of nostocyclopeptide Ml was interpreted so that m/z 882 is [M+H]⁺ and m/z 880 is [M-H]^" indicating an exact molecular mass of 881. Higher m/z ions were water, sodium, potassium or/and methanol adducts of the molecular ion. MS/MS analysis supported this interpretation (Table 1). Incorporation of water to the molecule especially in acidic conditions indicated that the molecule contains an easily hydrolysable structure. In acidic conditions all positive ions detected except the low intensity [M+H]⁺ ion were M+18 forms of the molecule. In neutral conditions ions from the native molecule dominated. ¹H-, ¹³C- and ¹⁵N-NMR imino linkage signals from nostocyclopeptide Ml were almost identical with the imino linkage signals of nostocyclopeptides described by Golakoti and co-workers. (Golakoti et al, 2001). This together with the Q-tof fragmentation data (Table 3) and 2,4-dinitrophenylhydrazine and acidic acetylation derivatizations showed that nostocyclopeptide Ml was a new member in the nostocyclopeptide family. The amino acid sequence of nostocyclopeptide Ml (Tyrl-Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7) was quite different from the earlier nostocyclopeptides Al (Tyrl-Gly2-Gln3-Ile4-Ser5-(4-MePro)6-Leu7) and A2 (Tyrl-Gly2-Gln3-Ile4-Ser5-(4-MePro)6-Phe7) The common amino acids in Ml were Tyrl and MeProό. Nostocyclopeptides also have the typical cyanobacterial -CH₂- unit variation in the secondary metabolite peptides. Nostocyclopeptides Al and A2 contained serine in position 5 whereas nostocyclopeptide Ml had homoserine in position 3. Nostocyclopeptide Ml inhibited all the hallmark of microcystin or nodularin induced apoptosis, including polarised membrane budding and chromatin distortion (Fig. 3), actin aggregation (not shown) and excessive protein phosphorylation (Fig. 5 and 6). Moreover, cells that were treated with Ml and nodularin attached to the substratum, proving that the cells were viable (not shown).

However, no evidence for an intracellular target of Ml -mediated protection of cell death was found (Fig. 5). Intracellular apoptosis antagonists such as CaMKII inhibitors and caspase inhibitors have previously been shown to inhibit or at least delay cell death caused by microinjected nodularin (Fladmark et ah, 1999; Fladmark et a , 2002). Ml gave no protection when co-injected with nodularin into NRK epithelial cells (Fig. 5A). This could not be explained by lack of pretreatment of the cells with Ml, since hepatocyte cell death could not be inhibited completely when Ml and nodularin were given simultaneously (not shown). Still, the anti-apoptotic effect could be through intracellular mechanisms specific for hepatocytes. However, Ml had no inhibitory effect against the PPl and 2A inhibitor okadaic acid (OA) induced apoptosis in HL-60 cells (not shown), and only modestly inhibited apoptosis in primary hepatocytes (Fig. 5B). The modest protection observed could be caused by inhibition of OA transport through organic anion channels. This would reduce the rate of OA uptake and moderately inhibit apoptosis. Our hypothesis was further supported by the findings that M-I abolished all protein phosphorylation associated with microcystin or nodularin- induced cell death (Fig. 6). Much of this phosphorylation can be blocked by inhibition of CaMKII (Fladmark et ah, 2002). However, it was found that Ml did not inhibit CamKII-mediated phosphorylation of cell lysate proteins (Fig. 7) or a synthetic peptide substrate (not shown).

The anti-apoptotic activity of nostocyclopeptide Ml was solely due to inhibition of microcystin and nodularin transport trough the hepatocyte membrane (Fig. 8). This transport is mediated via the organic anion transporter polypeptides (Oatps (Hagenbuch and Meier, 2003; Fischer et al, 2005)). The oatps are responsible for the transport of various organic anions such as bile acids, hormones and their conjugates, peptides and toxins into cells. Liver cells express more Oatps than other tissues (Hagenbuch and Meier, 2003), as part of their detoxification capabilities. The Oatps are particularly interesting from a pharmaceutical point of view since liver cells are responsible for detoxification of various harmful compounds (Sekine et al , 2000). It is shown that of the eleven Oatps, only Oatplb2 is able to transport microcystin into cells (Hagenbuch and Meier, 2003; Fischer et al, 2005).

SBP is widely used to study Oatps, since it is an efficient blocker of all the uptake mediated by Oatps (Hagenbuch and Meier, 2003). In line with previous findings (Eriksson et al, 1990; Runnegar et al, 1991), it was found that SBP could inhibit the uptake of microcystin (Fig. 8A). Moreover, SBP was a more efficient inhibitor of bile acid transport than Ml (Fig. 8B). In line with this, it was found that SBP inhibited apoptosis induced by both nodularin and the bile acids cholic acid and glycochenodeoxycholic acid. Ml, however, only protected against nodularin-induced cell death (Hagland, Lied Larsen and Herfindal, 2006, unpublished observations). This points to a conclusion that Ml only inhibits one or a few of the Oatps. It should be noted however, that [¹⁴C]-glycodeoxycholate was used as a measure of bile acid uptake. In their review, Hagenbuch and Meier do not mention this as a substrate for Oatplb2 (Hagenbuch and Meier, 2003). However, in an earlier study, they show that Oatplb2 has similar to affinity to various substrates as other liver Oatps (Cattori et al., 2001).

All Oatps expressed in liver are described as multispecific transporters with high affinities for an array of molecules, including anionic peptides (Trauner and Boyer, 2003). The fact that peptides are substrates to Oatps, but not the Na⁺-taurocholate cotransport proteins (Ntcps), makes us suspect that small (<10 amino acids) cyanobacterial cyclic peptides such as microcystin, nodularin and the nostocyclopeptides can be transported through, or inhibit the former. An alternative explanation is therefore that all the Oatp's mediates microcystin uptake, whereas both the Oatp's and the Na⁺-taurocholate cotransport proteins (Ntcp) transport bile acids. If the latter is true, Ml could inhibit all Oatp's, but none of the Ntcp's. Further experiments are needed to test if nostocyclopeptide Ml inhibits a broad range of Oatps, and if it inhibits Oatp-mediated microcystin-uptake by binding and permanently blocking the Oatp's, or by competition of the transporter.

It is concluded that nostocyclopeptide Ml, together with the previously published analogs Al and A2 (Golakoti et al., 2001) are inhibitors of channels responsible for the transport of certain organic anions through the membrane. Of interest is that the Ml was non-toxic, and was a potent inhibitor of apoptosis induced by cyclic peptide protein phosphatase inhibitors (microcystin and nodularin). Such activity has previously not been reported in cyanobacteria. The purpose of such inhibitors, and of cyclic peptide generally is unclear, and cyanobacteria biologists still argue for different views. The nostocyclopeptides might play a role in controlling the transport of organic molecules across the cyanobacterial membrane. Also, they can be directed to targets in grazers or competing organisms. The abundance of the nostocyclopeptides (approx. 1000 ppm of total DW) suggests that it is accumulated or stored by cyanobacteria, perhaps for rough situations with heavy grazing pressure or competition. Further investigation will enlighten the purpose these molecules have in cyanobacterial ecology and physiology, and can also shed light on the origin of cyclic peptides.

The present invention provides a compound comprising a nostocyclopeptide having the sequence Tyrl-Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7. The present invention also provides a compound comprising a peptide that has a formula

(I)-

The present invention is also related to a pharmaceutical composition comprising a peptide or a compound of the invention. In a preferred embodiment the peptide comprises nostocyclopeptide Ml. In one embodiment the invention is related to a nostocyclopeptide for use as a medicament. The present invention is related to the use of the peptide of the invention for the manufacture of a medicine for the treatment against hepato toxins.

The present invention teaches the use of the peptide of claim 1 as a cell reagent. The present invention is also related to the use of the peptide of claim 1 in a screening method.

There is a great need for novel peptides and compounds effective for protecting subjects against the toxic influence of cyanobacterial toxins, especially those which exist in increasingly amounts in the algal blooms around the world. T he present invention accordingly relates to a novel peptide useful in protecting hepatocytes against lethal influence of the cyanobacterial toxins microcystin and nodularin.

Having now generally described the invention, the same will be more readily described through reference to the following examples, which are provided by way of illustration and are not intended to be limiting of the present invention.

EXAMPLES

EXAMPLE 1 Materials and methods

Strains

Cyanobacterium (strain XSPORK 13A) classified as a Nostoc sp. according to the morphological characteristics and 16S sequencing was isolated from a gastropod collected 22 June.1999 from shallow seawater at the cape of Porkkala, Finland. Isolation of the strain from the extraction solution, strain purification, mass cultivation and freeze-drying was as previously described (Herfindal et al. , 2005). The reference strain Nostoc sp. ATCC53789 producing nostocyclopeptides Al and A2 (Golakoti et al., 2001 ) was also cultivated and freeze-dryed according Herfindal et al. , 2005. This strain produced mainly demethyl variants from the nostocyclopeptides Al and A2 under the cultivation conditions used.

Methods

Purification and chemical analyses

Freeze-dried biomass (16.4 g) was extracted 3 times for 45 min with 400 ml of methanol. The extract was evaporated to dryness, and the residue dissolved into 100 ml of methanol and filtered before passing through two solid-phase extraction cartridges (StrataX Polymeric Sorbent, 30 mg, Phenomenex, Torrance, CA, USA). The nostocyclopeptide eluted with methanol, which was reduced to 15 ml. The methanol extract was then extracted 3 times with 15 ml of heptane, then evaporated to dryness, and dissolved in 3.5 ml of 10% aqueous methanol. Purification was done with HP 1100 Series modular HPLC (Agilent Technologies, Palo Alto, USA) with diode array detector with a monitoring wavelength of 222 nm. The sample was injected into the Luna C18 column (150 x 4.6 mm, 5 μm, Phenomenex^®) in 0.3 - 0.5 ml batches. Nostocyclopeptide was eluted isocratically with a mixture of 75% of water (HPLC-grade, Rathburn) and 25 % of acetonitrile (LiChrosolv^®, Merck) at 40°C, and washed with 95% aqueous acetonitrile between the injections. The fractions containing cyclic peptides (identified by UV-spectra) were pooled and evaporated to dryness under a jet OfN₂.

Liquid chromatography - mass spectrometry (LC-MS)

HP 1100 Series modular HPLC (Agilent Technologies, Palo Alto, USA) interfaced with an esquire3000 plus ESI-ion trap mass spectrometer (Bruker Daltonics, Bremen, Germany) containing diode array detector was used for the mass spectrometric analysis of the nostocyclopeptides and amino acid derivatives. Luna Cl 8 column (150 x 4.6 mm, 5 μm, Phenomenex^®) was used when sample components were separated prior the mass spectrometric analyses. Column was eluted with different water or 0.1 % HCOOH and acetonitrile or methanol gradients at 40 °C. Flow injection analysis was used for the purified nostocyclopeptide Ml to get the MSⁿ spectra.

Accurate mass measurements of nostocyclopeptide Ml dissolved in 1% formic acid: acetonitrile (1:1) were made with ESI-MS Micromass Q-tof Ultima tandem mass spectrometer (Micromass, Manchester, UK). Nostocyclopeptide solution was infused directly into the mass spectrometer at a flow rate of 0.5 μL/min. The mass spectrometer was operated in positive ion electrospray mode with a source temperature of 80 °C and a potential of 3.5 kV applied to the nano-LC probe. In MSMS mode collision energy of 30 eV was used.

NMR spectra were recorded in DMSO at RT. HSQC experiments were optimized for ¹JcH = 150 Hz and HMBC experiments for ⁿJ_CH = 7 Hz. IR

Spectrum from 500 to 4000 cm^"1 was recorded on a Bruker Vertex 70 FTIR spectrometer (Bruker Optics, Karlsruhe, Germany) equipped with a microplate HTS-XT accessory unit. Dissolved peptide was placed on a silicon plate, evaporated and transmission spectrum measurement was read in HST-XT unit.

Partial acid hydrolysis

100 μg of Ml in an 300 μl ampoule was filled with 1.2 M HCl and hydrolyzed 70 min at 105 C. Acid was vacuum evaporated and residue dissolved in water. Peptide fragments were analyzed from the preparate with the LC-MS.

Derivatization of Nostocyclopeptide Ml

Acetylation: 30 μg of purified nostocyclopeptide Ml was dissolved to 400 μl of 50 mM ammoniumbicarbonate solution. 1 ml of acetylation reagent (1.2 ml of methanol and 400 μl acetic anhydride) was added, solution was incubated 1.5 h at room temperature, then vacuum evaporated to dryness and then dissolved to 0.1 % HCOOH/acetonitrile (1:1) prior to mass spectrometric flow injection analysis.

DNPH derivative: 20 μg of purified nostocyclopeptide Ml was dissolved to 50 μl of 0.5 % trifluoro acetic acid solution. 50 μl of 2,4-dinitrophenyl hydrazine saturated (solution was sonicated and shaked 10 mins) acetonitrile/water (1:1) solution containing 0.5 % trifluoroacetic acid was added, reaction mixture was incubated 30 min at 37 ⁰C and after adding of 25 μl of acetonitrile the sample was analyzed with LC-MS. FDAA derivative: FDAA derivative of nostocyclopeptide Ml (50 μg) was prepared at the same way as the FDAA (Marfey) derivatives of the nostocyclopeptide Ml amino acid, nostocyclopeptide Ml FDAA derivative was purified by adding 800 μl of water to 200 μl of reaction solution (50 % acetonitrile in water) and then passing the solution through a solid-phase extraction cartridge (StrataX Polymeric Sorbent, 30 mg; particle size, 33 _m; pore size, 85 A; Phenomenex, Torrance, CA, USA) which was equilibrated with 1 ml of 85% ACN. Cartridge was washed with 2 x 200 μl of 30 % acetonitrile and 200 μl of 60 % acetonitrile. FDAA derivative was eluted from the cartridge with 2 x 200 μl of 100 % methanol. After evaporation of methanol, residue was acetylated as explained earlier but using 10 times smaller acetylation reagent volumes. Enantiomeric analysis of amino acid residues

Nostocyclopeptide Ml (0.1 mg) was added 0.3 ml of 6 N HCl and left for acid hydrolysis at 110 ⁰C for 22 h. After hydrolysis the acid solution was evaporated in a vacuumed centrifuge (Heto Maxi Dry Plus, Denmark). The residue was dissolved into 50 μl of water (sample for Chirex d-pencillamine column), which was added 20 μl of 1 M NaHCO₃ and 100 μl of 1% FDAA (l-Fluoro-2,4-dinitrophenyl-5)-L-alaninamide, Marfey reagent, Pierce) in acetone. After incubation of 1 h at 37 ⁰C, the reaction was terminated with 20 μl of 1 N HCl. Sample containing Marfey derivatives of the Nostocyclopeptide Ml amino acids was diluted with 200 μl of acetonitrile and analyzed with LC-MS as described earlier. Reference L- and D-amino acid Marfey derivatives were prepared accordingly with 50 μl of 50 mM amino acid solutions. Reference compound (2S,4S)-4-MePro not available commercially was obtained by hydrolysing spumigins A, Bl, B2 and C purified from Nodularia spumigena AVl (Fujii et ah, 1997) at 112°C for 24 h. (2S,4S)-4-MePro was located from the acid hydrolysate chromatograms of spumigins A, Bl and B2 by mass spectrometry and chromatographic comparison to spumigin C hydrolysate not containing methylated proline.

Extraction of Ml for biological studies

A large-scale extraction protocol was developed to obtain sufficient amount of nostocyclopeptide Ml for biological studies. Between 1.5 and 2 grams of freeze-dried cyanobacteria were added to 200ml of 1/10 me thano I/water (v/v), sonicated for 10 minutes and left on ice in darkness for 1 hour. After centrifugation (12 000 x g) and two washes, the supernatants were combined and subjected to two solid phase extraction steps. First, proteins and other anionic molecules were removed by a QMA anion- exchange SPE cartridge (35cc, Waters). After wetting and equilibration of the cartridge (100 ml MeOH followed by 100 ml of water), extract from approximately lgram of freeze dried biomass was loaded onto the cartridge, followed by a wash with 1.5 column volumes of low ionic strength buffer (1OmM Hepes) or MQ-water. Whereas most proteins, DNA and pigments were retained on the column, nostocyclopeptide Ml was in the flow-through and wash. Further elution of higher strength buffers yielded no nostocyclopeptide Ml (measured by HPLC, not shown). Secondly, the QMA-extract was concentrated and desalted by loading it onto a 35cc C18 SPE cartridge (Waters). The cartridge was washed with three column volumes of 10% aqueous methanol and nostocyclopeptide Ml was eluted with 100% methanol.

The methanol elute was evaporated to dryness and the bioactive compound was isolated by semi-preparative reversed phase HPLC (column, Kromasil 100-10, C- 18, 250 x 10 mm ID) using deionised water (Milli-Q quality, Millipore systems) and acetonitrile (HPLC-grade, Rathburn, UK) as mobile phases. The gradient was from 5 to 100% acetonitrile during 25 minutes and the flow was 2.7ml/min. The peaks eluting after 18.9 and 20.6 minutes were Ml and an uncharacterised cyclic peptide respectively (Fig. 1). With this method, we obtained 2.6 mg Ml from 2.2 g of freeze-dried cyanobacteria. Nostocyclopeptide Ml was lyophilised and kept at -80°C, and dissolved in DMSO for use in biological studies.

Cell handling and experimental conditions

Isolation of rat hepatocytes by in vitro collagenase perfusion was done as previously described (Mellgren et al, 1995). Primary rat hepatocytes were resuspended in pre- gassed (5% CO₂/95% O₂) 10 mM Hepes (pH 7.4) with 5 mM lactate, 5 mM pyruvate, 8 mM glucose, 120 mM NaCl, 5.3 mM KCl, 0.01 mM KH₂PO₄, 1.2 mM MgSO₄, and 1.0 mM CaCl₂. For short-term suspension cultures the cells were incubated in 48-well tissue culture dishes at 37°C in 5% atmosphere with gyratory shaking (Herfindal et al., 2005 #915). For viability assay, primary hepatocytes treated with nodularin with or without nostocyclopeptide Ml were cultured in collagen-treated 24-well tissue culture dishes in a hepatocyte-specific modified Dulbecco's medium (Mellgren et al, 1995). Insulin and dexamethasone (0.2 nM and 5nM respectively) were added 2 hours after culturing. 18 hour after seeding, viable cells had attached to the substratum, whereas apoptotic or necrotic cells were in suspension.

HL-60 human AML leukaemia cells (ATCC: CCL-240) and NRK rat kidney epithelial cells (ATCC: CRL-6509) were cultured in RPMI medium enriched with 10% foetal calf serum in a humified atmosphere (37°C, 5% CO2). Toxicity assays were performed as described previously (Herfindal et al, 2005; Selheim et al, 2005). Microinjection of NRK-cells was performed with an Eppendorf 5171 micromanipulator and a 5246 microinjector mounted on a Nikon Diaphot 300 inverted microscope. Microcapillaries and puller were from Sutter intreument Co (Novato, CA). The methodology is described in detail in (Fladmark et ah, 2002). Determination and quantisation of cell death was done by microscopic evaluation of surface membrane and chromatin morphology.

Study of protein phosphorylation

³²P-labelled hepatocytes exposed to nodularin with or without nostocyclopeptide Ml were subjected to 7% trichloroacetic acid (TCA, final concentration) to precipitate proteins. The protein pellet was washed twice with 5% TCA followed by 3 washes with ether saturated with water. After evaporation of the ether, the protein pellet was resuspended in SDS loading buffer and separated on a gradient polyacrylamide gel (7- 15% polyacrylamide). Phosphoproteins were visualised by gel exposure to Kodak bio max films.

Rat liver cytosol fraction was prepared by ultracentrifugation (100 000 g, 1 h) of frozen rat liver powder (liquid nitrogen) dissolved in homogenization buffer (50 mM Hepes pH 7.20, 120 mM KCl, 5 mM EDTA, 3 mM EGTA, 2 mM dithioerythritol, 10 mM benzamidine, 50 μg ml^"1 soya bean trypsin inhibitor, 50 μg ml^"1 aprotinin, 7 μg ml^"1 chymostatin, 6 μg ml^"1 antipain, 48 μg ml^"1 leupeptin, 14 μg ml^"1 pepstatin). The cytosol fractions (25 mg ml^"1 protein) were then kept at -80 °C until used.

Rat liver cytosol was incubated in the presence or absence of the following; 1 μg ml -1

CaM-KII (Upstate, USA), 1 μM Nodularin, 0.5 Abs m τl-^"1¹ matti-1, 10 μM KN-93. AU samples were incubated with 20 nM protein kinase A inhibitor, 5 mM MnCl₂, 50 μM ATP, 50 mM Hepes pH 7.20, 50 μg ml^"1 SBTI, 1 mM Ca²⁺, 10 μg ml^"1 calmodulin and 5 mg ml^"1 cytosol protein, at 25 °C for 15 s to 6 min.

Phosphorylation of syntide-2 (PLARTLSVAGLPGKK) by CaM-KII was monitored in the presence or absence of nostocyclopeptide Ml or KN93 at the following conditions; 10 mM MgCl₂, 0.5 mM ATP, 15 mM Hepes pH 7.40, 1 mM Ca²⁺, 10 μg ml^"1 calmodulin, 10 μM syntide 2, 1.5 μg ml^"1 CaM-KII, 30 °C.

Accumulation of radiolabeled microcystin and glycodeoxycholate in primary hepatocytes

Microcystin YR (Calbiochem (San Diego, CA, USA)) was iodinated with [¹²⁵I] (Amersham Little Chalfont, UK) as previously described (Serres et al. , 2000; Herfϊndal and Selheim, 2006). Freshly prepared primary rat hepatocytes in suspension (800 000 cells/ml) were incubated with inhibitors of bile acid uptake or vehicle for 5 minutes (37°C in a humified atmosphere containing 6% CO₂). A mixture of radiolabeled microcystin YR (8nM) and microcystin LR (10OnM) was then added and the cells were incubated for another 7 minutes before rapid separation from the medium as described by Ueland and coworkers (Ueland et al, 1982). Briefly, 0.3 ml of cell suspension was carefully layered onto a mixture (0.4 ml) of dinonyl phthalalte and dibutyl phthalate (1:3, v/v) in 2-ml eppendorf tubes. After rapid centrifugation (15 s, 7000 rpm), the tubes were quick-frozen on dry ice, and the bottom, containing the cells were cut off and placed in scintillation tubes (Packard Biosciences, Groeningen, the Netherlands) containing ImI of 2% SDS. After 1 hour, 8ml of scintillation fluid (Emulsifier Safe, Packard Biosciences) was added, and levels of radioactivity measured by a TriCarb 3100TR liquid scintillation analyzer (Packard Bioscienses).

Accumulation of [¹⁴C]glycodeoxycholate (Amersham Little Chalfont, UK) was measured similarly, except that the cells were separated from the medium by gentle filtration (0.45 μm HA filters, Millipore Billerica, MA, USA). The cells were washed twice with 2 ml of medium (37°C) and the filters were placed in scintillation vials containing 2 ml 2%SDS. This method was not applicable with [¹²⁵I]microcystin, since the microcystin in the medium binds to the filters, causing a background of more than 60%. With [¹⁴C]glycodeoxycholate, the background was less than 5%. The addition of 1% DMSO in the medium did not affect the uptake of [¹⁴C]glycodeoxycholate.

EXAMPLE 2

Chemical characterisation of nostocyclopeptide Ml

Nostocyclopeptide Ml purified from a cyanobacterial strain XSPORK 13A was dissolved to neutral and acidic aqueous solvents, then directly infused to the electrospray ion trap mass spectrometer (Table 1). Table 1. The effect of different solvent systems (in 1:1 mixtures) on the formation of positive and negative ions from Nostocyclopeptide Ml directly infused into the electrospray-ion trap mass spectrometer.

H_: O/ACN H₂O/MeOH 0,1 % HCOOH/ACN³ 0,1 % HCOOH/MeOH Interpretation'⁵ m/z Rl (%) m/z Rl (%) m/z Rl (%) m/z Rl (%) Positive ions

[M₊H]⁺ < 1

882 8 882,6 20 882,6 16 882,6 9

900,7 7 900,6 50 900,7 41 [M+H+H₂O]⁺

[M₊Na]⁺ _<

904 100 904,6 100

920 41 920,4 8 [M₊K]⁺

B22 22 B22,O 15 922,5 100 B22, 5 2i

936,4 5 [M₊MeOH₊Na]⁺

938 8 [M₊H₂O₊K]⁺

940 5 940,5 4 940,4 38 940,5 6 [M+Na+2H₂O]

954,5 24 954,4 8 954,6 100 [M₊Na₊H₂O₊MeOH]⁺

Negative ions

880 100 880 100 880 12 [M-H]^"

898 6 898 16 898 100 898 100 [M-H₊H₂Of

901 54

914 12

920 25 920 24 [M-2H₊H₂O+Na]^~

952 6 [M-2H₊H₂O₊Na₊MeOH]^{~ a} = contained traces of MeOH originating from the sample preparation ^b = Arrows show parent — > product ion relationships from MS/MS experiments

Nostocyclopeptide Ml dissolved to water/acetonitrile or methanol solvent mixtures (50/50) gave positive base peak at m/z 904 and negative base peak at m/z 880. When water was replaced with 0.1 % formic acid, m/z 904 peak was absent and base peaks appeared ≥18 mass units higher forming peaks which were already present in neutral solvent with lower intensities.

In neutral solvent environment, soft ionization mass spectra of nostocyclopeptide Ml preparate contained four major smaller ions of m/z 259, 624, 774 and 8Ox. Protonated ions 259 and 624 are suitable to be two more stable units of the molecule connected together with less stable bond(s). In acidic solvent environment, mass spectrum contained same ions but m/z 259 was replaced with the m/z 277 ion, 18 units larger than the ion 259 in neutral environment indicating that the smaller structural unit of the molecule contains the additional water already seen in the analysis of the molecular peak. ID ¹H-NMR spectrum, strong amide signal (1640 cm^"1) in IR spectrum and MS/MS fragmentation patterns acquired from the protonated molecular ions [M+H]⁺ and [M+ 18+H]⁺ indicated a peptidic structure of the compound nostocyclopeptide Ml. Due to the low amount of purified analyte, ID ¹³C-NMR spectra were not acquired. In ¹H-

NMR spectrum, four 2°-amide type doublet proton signals at 8.54, 7.81, 7.76 and 7.46 were found. Correlations of these ^-signals were not seen in ¹³C HSQC but were present in ¹⁵N HSQC spectrum within a range typical for amino acids in peptidic structures. Analysis of ¹R-¹R COSY, ¹R-¹R TOCSY, ¹³C HSQC, ¹⁵N HSQC, ¹H-¹³C HMBC and ¹H-¹⁵N HMBC spectra led to the construction of seven partial structures, namely Tyr I, Tyr II, HSe, Pro, VaI, MePro and modified Tyr III (Table 2).

Table 2. l ¹τHτ, „

a „nd j 15 N NMR Spectral Data for Nostocyclopeptide Ml in DMSO-d6.

For additional information, see Example 3.

Six typical amino acid -methine correlations and 3 methyl correlations were assigned from the HSQC spectrum. Three p-hydroxyphenyl ring structures with typical signal patterns in ¹H-NMR spectrum were assigned from the HMBC spectrum containing characteristic aromatic proton correlations to the oxygen substituted aromatic quaternary carbons. For a detailed description of the NMR-analyses, see Table 2 and Example 3.

Amino acid sequence of Ml peptide was further confirmed with mass spectrometry. As typical for cyclic peptides producing complex mass spectra also protonated native Ml peptide (m/z 882) produced ions originating from multiple ring opening reactions. Ions from the peptide bond cleavages from proline nitrogen ([MPr-Tyral-Tyr-Tyr-Hse]⁺) and also from the imino bond cleavage ([Tyr-Tyr-HSe-Pro-Val]⁺, [Pro-Val-MPr-Tyral]⁺, [Tyr-Tyr-HSe]⁺, [MPr- Tyral]⁺) dominated the MS² spectrum and supported the sequence introduced for the Ml peptide. However the overall interpretation of the MSⁿ spectra from ion m/z 882 turned out to be complicated. Therefore, the Ml peptide was linearized by acetylating the amino group of Tyr I residue. In acidic conditions the backbone imino nitrogen of Tyr I becomes accessible for an acetyl group keeping Ml peptide linear, which leads less complicated fragmentation. Acetylated Ml peptide formed intense sodium ion complex (m/z 964) which MS² produced continuous series of a, b and y-type fragment ions down to dipeptide level (Table 3).

Table 3. Interpretation of the MSMS fragment ions of the acetylated Ml peptide m/z 964 [Ac-Tyr-Tyr-HSe-Pro-Val-MPr-Tyr-CHO + Na]⁺. Arrows show the fragmentation of MSMS fragments.

Fragment ion interpretation confirmed with MSⁿ fully supported Ml peptide sequence obtained by NMR. For a detailed description of NMR-data on the various amino acid residues, see Table 2 and Example 3.

Nostocyclopeptide Ml was also partially hydro lyzed in mild acidic and temperature conditions in order to further check the sequence interpretation obtained from the MSMS fragmentations and NMR shifts of the Ml. Peptides m/z 215 [Pro-Val +H]⁺, m/z 229 [Val-MePro +H]⁺, m/z 326 [Pro-Val-MePro +H]⁺, m/z 345 [Tyr-Tyr +H]⁺, m/z 428 [Tyr-Tyr-HSe-H₂O +H]⁺, m/z 446 [Tyr-Tyr-HSe +H]⁺ and m/z 473 [Pro- Val-MePro-Tyr aldehyde +H]⁺ identified from hydrolyzate by the LC-MS analysis also fully supported the Ml sequence. Imino bond containing peptide -Tyr aldehyde-Tyr- was missing due to the acid labile character of the imino bond. Interestingly, a peptide containing HSe -Pro sequence was not found from the acid hydrolyzate and also HMBC correlations could not be found between these amino acids. The existence of the special amino acid tyrosine aldehyde linked via imino bond to another amino acid in the Ml molecule was showed also by derivatization of Ml with aldehyde/ketone specific reactant 2,4-dinitrophenylhydrazine. Protonated reaction product had m/z of 1080 which is in accordance with the 2,4-dinitrophenylhydrazone structure of Ml formed after hydrolysis of the nostocyclopeptide Ml imino bond in acidic derivatization conditions. MSMS analysis of the derivative showed that 2,4- dinitrophenylhydrazine has been reacted with Tyr III demonstrating it as an aldehyde.

Together with the Marfey derivatization of the amino acids of the Ml hydro lyzate, also the native Ml peptide was derivatized using the standard conditions for amino acid derivatization. All free amino and hydroxyl functionalities react with Marfey's reagent (l-Fluoro-2,4-dinitrophenyl-5-L-alanine amide, FDAA, MW 272) (Bhushan and Bruckner, 2004) increasing the molar mass of the compound with 252 units. Derivatization increased the m/z of the protonated Ml from 882 to 1638 meaning that native Ml peptide contained three reactive functionalities. MSMS analysis showed that FDAA has been reacted with the amino acids Tyr I, II and III, probably with the hydroxyl groups since reaction of α-amino group in peptides with FDAA requires long (12-24 hr) reaction time (Szabό et ah, 2001). Acetylation of the FDAA derivative of the Ml peptide further increased m/z from 1638 to 1698 indicating addition of water and acetyl group to the Ml peptide. MSMS analysis showed that the acetyl group situated in Tyr I. These results further confirmed the TyrIII=TyrI-TyrII structure in Ml peptide.

After total hydrolysis of nostocyclopeptide Ml with 6 M HCl, stereochemistry of the hydrolysate amino acids were analyzed by chiral HPLC using a Chirex d-pencillamine column. Chromatogram consisted six peaks from which four peaks were identified as amino acids L-Pro, L-VaI, L-Tyr and D-Tyr by comparing retention times to commercial amino acid standards (Figure 1). In order to identify the stereochemical structure of the unidentified amino acids, LC-MS was used together with the Marfey derivatization of the amino acids in Ml hydrolyzate. Molecular weights of the amino acids were solved from the protonated molecular ions of the derivatives by substracting the weight of the FDAA residue from of the amino acid FDAA-derivatives. Identification of the amino acids L-Pro, L-VaI, L-Tyr and D-Tyr was confirmed. Molecular weight of one of the unknown amino acids was 119, same as the molecular weight of threonine, but there was no congruence between retention times of the FDAA- derivatives. Screening of commercial threonine isomers showed that the D-homoserine- FDAA had an identical chromatographic behaviour and mass spectrometric characteristics with the unknown amino acid-FDAA . The FDAA-derivative of the second unknown amino acid had nearly identical mass spectrum with proline-FDAA except that all peaks were 14 mass units larger suggesting that the amino acid is methylproline. Identical retention times in the two different chiral chromatographic systems proved that the unknown amino acid was (2S,4S)-4-MePro (Table 2). In conclusion, the compound Ml is a cyclic peptide with the sequence (Tyrl-Tyr2-(DH- Ser)3-Pro4-Val5-((2S,4S)-4-MePro)6-Tyr7). Similar to previously described nostocyclopeptides (Golakoti et al, 2001), there is an imino linkage, in this case between Tyrl and TyrIII. The novel compound is further referred to as nostocyclopeptide Ml (abbreviated Ml).

EXAMPLE 3

NMR-analyses of nostocyclopeptide Ml

In Tyr I p-hydroxyphenyl group bonding to C-3 was assigned from the H-5/9 (δ 6.82) to C-3 (δ 41.8) and H-3 (δ 2.38; δ 2.91) to C-4 (δ 130.0) and C-5/9 (δ 132.8) correlations in HMBC. TOCSY correlations showed that without 2°-amide proton signal correlations, H-3 signals formed a spin system with the signal δu 3.43 which correlated with carbon signal δ 76.4 in HSQC. In HMBC signal δ_H 3.43 correlated with C-4 of the p-hydroxyphenyl group and with C-I carbonyl signal δ 173.4 indicating that atypical signals δu 3.43 and δc 76.4 indeed were from α-methine of Tyr I.

In Tyr II 2°-amide proton signal (δ 7.81) correlated only with H-2 (δ 4.53) and H-3 (δ 2.82; δ 3.06) TOCSY signals. Bonding of p-hydroxyphenyl group to C-3 was seen from the H-2 and H-3 to C-4 (δ 131.1), H-3 to C-5/9 (δ 133.1) and H-5/9 (δ 6.93) to C-3 (δ 39.4) correlations in HMBC spectrum.

Homoserine (HSe) partial structure was solved from the TOCSY 2°-amide type proton (δ 8.54) correlations to a H-2 (δ 4.47) type proton and two other proton signals (δ 1.78, δ 3.52) which correlated to three different carbons in HSQC spectrum. From HMBC only one additional carbon signal (carbonyl; δ 174.4) was found which correlated with H-2 (δ 4.47) proton and with proton signal δ 1.78. In HSQC δu 3.52 signal correlated with the other signals (δu 3.46; δc 59.8) typical for a methylene group bonded to an oxygen like electronegative atom. COSY correlations linked this -CH₂X group to the δ 1.78 proton signal. As an outcome from these signals is a δ-amino acid with a linear two carbon side chain preferentially having a hydroxy end group which structure was definitively shown to be homoserine using LCMS.

Proline spin system in TOCSY spectrum characteristically did not have H-2 (δ 4.41) type proton correlation with 2°-amide type proton but had correlations with protons δ 2.01, δ 1.79, δ 1.96, δ 3.66 and δ 3.90. In HSQC aforementioned protons correlated with four carbons δ 62.8 δ 32.2, δ 26.0 and δ 48.7, respectively, showing the presence of a methylene group bonded to a nitrogen like electronegative atom. As COSY correlations positioned this methylene group to C -5 and HMBC spectrum showed no additional correlations than α-carbonyl carbon (δ 173.5), signals were assigned as proline structure.

Valine was identified from a typical spin system in TOCSY spectrum containing strong 2°-amide proton (δ 7.79) couplings to protons of two methyl groups. Protons of both methyl groups were coupled to only two other kinds of protons, H-I (δ 4.55) and H-2 (δ 2.47) type protons. Together with the corresponding carbon signals from HSQC spectrum, this partial structure was assigned as valine.

In 4-MePro methyl group protons correlated to a H-2 (δ 4.43) type proton and to protons δ 4.06, δ 3.12, δ 2.21 and δ 1.68 of two methylene groups (from HSQC) without 2°-amide type proton correlation in TOCSY spectrum. HMBC signals showed that this partial structure contained one more methine group which altogether build up a methylproline structure having a methyl group in position C-4 based to the strong correlation of H-2 (δ 4.43) to the methylene proton δ 1.68 and to the missing correlations between methylene protons δ 3.12, δ 4.06 and δ 1.68 in COSY spectrum. Modified Tyr III TOCSY spin system contained typical 2°-amide proton (δ 7.46), H-2 (δ 4.40) and H-3 (δ 2.22; δ 2.52) correlations but also an additional 5_H 6.94 signal. In ¹³C HSQC signal δ_H 6.94 correlated to δ_c 165.9 and in ¹⁵N HSQC correlations were not detected. Instead of typical carbonyl carbon correlations in HMBC spectrum, 2°-amide proton, H-2 and H-3 correlated with δc 165.9 and δu 6.94 signal correlated with Tyr I C-2 (δ 76.4). In ¹⁵N HMBC a correlation of signal δ_H 6.94 with S_N 315.0 signal was present and also Tyr I H-3 signal δ 2.91 correlated with the signal δ^ 315.0. These chemical shifts showed that instead of normal peptide bond between amino acids Tyr III and I, an imino structure -CH=N- connected Tyr III and I to each other. Bonding of p- hydroxyphenyl group to C-3 was seen from the H-2 and H-3 to C-4 (δ 129.2), H-3 to C- 5/9 (δ 132.8) and H-5/9 (δ 6.49) to C-3 (δ 40.2.4) correlations in HMBC spectrum.

¹³C HMBC, ¹⁵N HMBC, ¹R NOESY and ¹³C NOESY techniques were used in sequencing of the identified amino acids.

Tyr I was linked to Tyr III through several HMBC correlations which were from Tyr III H-I to Tyr I C-2, N-2 and carbonyl-C, from Tyr I H-2 to Tyr III C-I and through NOESY correlations from Tyr III H-I to Tyr I C-3, H-2 and H-3 due to the extra proton H-I belonging to the structure of modified Tyr III. Correlation between Tyr III H-I and Tyr I H-2 also indicated that imino double bond geometry was E (Golakoti et ah, 2001).

Between Tyr I - Tyr II typical HMBC correlations from Tyr II H-2 and 2°-amide proton to Tyr I carbonyl-C were found. In NOESY Tyr II 2°-amide proton signal δ 7.81 was fused with the Pro 2°-amide proton signal δ 7.79 making the detected correlations too uncertain to be used in sequence determination.

One strong HMBC correlation, from HSe 2°-amide proton to TyrII carbonyl-C, linked the amino acids Tyr II - HSe to each other. In ¹H NOESY strong HSe 2°-amide proton correlations to Tyr II H-2, H-3/3', weak correlations to Tyr II 2°-amide proton, H-5/9 and in ¹³C NOESY HSe 2°-amide proton correlations to Tyr II C-2, C-3 further evidenced the Tyr II - HSe peptide linkage. Despite long data accumulation times and variable accumulation parameters, HMBC correlations between amino acids HSe and Pro could not be obtained. However, in ¹H NOESY strong Pro H-5/5' correlations to HSe H-2 and Pro H-5 correlation to HSe H-3 were seen as in ¹³C NOESY medium intensity Pro H-5/5' correlations to HSe C-2 and HSe H-2 correlation to Pro C-5. These NOESY signals were also much more intense than the few other signals assigned between the aforementioned H's and Cs of Pro and HSe and the other amino acids.

Two HMBC correlations from VaI 2°-amide proton and H-2 to Pro carbonyl-C were identified. However, these signals were almost identical with the correlations of Tyr II 2°-amide proton and H-2 to Tyr I carbonyl-C making the interpretation somewhat uncertain.

Two HMBC correlations from 4-MePro H-2 and H-5 to VaI carbonyl-C linked 4-MePro and VaI together. ¹H NOESY 4-MePro H-5/5' correlations to VaI H-2 (strong) and H-3 (medium and weak) and ¹³C NOESY correlations from 4-MePro H-5/5' to VaI C-2, 4- MePro H-5 to VaI carbonyl-C, 4-MePro H-5' to VaI C-4 and VaI H-2 to 4-MePro C-5 confirmed 4-MePro - VaI linkage.

HMBC correlations from modified Tyr III 2°-amide proton (strong) and H-2 to 4-MePro carbonyl-C showed the Tyr III - 4-MePro bonding. ¹H NOESY correlations from Tyr III 2°-amide proton to 4-MePro H-3' and H-5 and ¹³C NOESY correlations from Tyr III 2°-amide proton to 4-MePro carbonyl-C and C-2 (weak) were also present.

EXAMPLE 4

Nostocyclopeptide Ml was not cytotoxic, but inhibited apoptosis induced by microcystin or nodularin.

Contrary to a previous report on nostocyclopeptide Al and A2 (Golakoti et a , 2001), we did not find any cytotoxicity of Ml in any of the cell lines tested at doses up to 0.1 mM (IPC-81 rat leukaemia cells, mouse fibroblasts and primary rat hepatocytes in monoculture, 24 hour incubation, data not shown). However, Ml protected primary rat hepatocytes against microcystin-induced cell death (Fig. 4A). Pre-incubation of hepatocytes Ml 15 minutes prior to addition of up to 20OnM of microcystin, gave normal morphology more than 2 hours later (Fig. 4A and 4D-I). Whereas microcystin- treated cells had polarised budding and a distorted chromatin (Fig. 4F and 4G), cells incubated with Ml prior to microcystin treatment had a smooth membrane and the chromatin appeared normal (Fig. 4H and I). When washed and plated, the cells that had been co -treated with Ml and microcystin attached to the substratum similar to untreated cells whereas cells that were treated with microcystin failed to attach (not shown). The apoptosis-inhibition was very efficient as lOμM of Ml could inhibit completely the rapid hepatocyte apoptosis induced by 5μM of nodularin (Fig. 4B).

It was tested if the other nostocyclopeptides could inhibit microcystin-induced apoptosis. Both nostocyclopeptides Al and A2 and the deMe- variants (MePro was replaced with Pro) could protect the hepatocytes against cell death, with nostocyclopeptid Al being superior to A2 (Fig. 4C). Unfortunately, we were unable to obtain enough of these nostocyclopeptides to determine their concentration accurately by HPLC, or to do further cell experiments. We therefore focused on the activity of nostocyclopeptide Ml.

Microcystin and nodularin are potent inhibitors of serine/threonine protein phosphatases (PP), and we therefore wanted to investigate if Ml also acted through intracellular targets.

EXAMPLE 5

Nostocyclopeptide Ml did not protect against cytosolic phosphatase inhibitors or receptor-mediated apoptosis

Previous reports have shown that uptake of microcystin in primary hepatocytes depend on organic anion transporter polypeptides (Oatp) (Eriksson et al., 1990; Runnegar et al., 1991; Fischer et al., 2005). Hence, we wanted to know if Ml could protect cells against phophatase inhibition in the cytosol. When a solution containing nodularin (lOOμM) and nostocyclopeptide Ml (200μM) was microinjected into NRK-cells, we did not observe any inhibition of apoptosis compared to cell microinjected with nodularin alone (Fig. 5A). In line with this, we found that Ml was a poor antagonist of okadaic acid induced hepatocyte apoptosis (Fig. 5B). In contrast to microcystin and nodularin, okadaic acid is a membrane permeable phosphatase inhibitor, which induces apoptosis in other cells than primary hepatocytes. We noted that Ml protected the hepatocytes slightly (between 10 and 15% reduction) at low concentrations of okadaic acid (Fig. 5B). The findings that Ml did not act via intracellular inhibition of apoptosis were supported when we exposed primary hepatocytes to apoptosis -inducers that act via membrane receptors. Tumour necrosis factor α (TNF α) and transforming growth factor β (TGF β) are ligands to membrane receptors that activate the apoptosis machinery within the cells. However, neither TNF α- nor TGFβ-induced cell death was inhibited by the presence of Ml (Fig. 5 C and data not shown).

To support the findings that Ml did not have its anti-apoptotic effect inside the cells, the protein phosphorylation status of cells treated with 15 μM Ml, 200 nM of nodularin, or pretreated with Ml for 15 min before addition of nodularin was examined (Fig. 6). Nodularin and microcystin induces massive protein phosphorylation (Fig. 6, lane 5 and 6), mediated amongst others through Ca²⁺/Calmodulin dependent protein kinase II (CaMKII). Inhibition of CaMKII antagonises nodularin and microcystin-induced apoptosis and protein phosphorylation (Fladmark et al, 2002). Nostocyclopeptide Ml did not induce any visible changes in protein phosphorylation (Fig. 6, lane 3 and 4), but abolished the excessive phosphorylation induced by nodularin Fig. 6, lane 7 and 8). However, we could exclude nostocyclopeptide Ml as being a CaMKII inhibitor since it did not protect against microinjected nodularin (Fig. 5A) and did not inhibit CaMKII- mediated in vitro phosphorylation of either cell lysate (Fig. 7) or a synthetic substrate (syntide-2, data not shown). We obtained a syntide-2 phosphorylation rate of 92 and 82 nmol min^"1

in the presence and absence of nostocyclopeptide Ml, respectively. The CaM-KII activity was readily inhibited by KN-93.

EXAMPLE 6

Nostocyclopeptide Ml inhibited hepatocellular uptake of microcystin and bile acids

Previous reports have shown that uptake of microcystin in primary hepatocytes depend on organic anion transporter polypeptide (Oatp). In light of the results presented above, the mechanism of Ml -inhibition is consistent with inhibition of uptake. To confirm this hypothesis, we measured the accumulation of radiolabeled microcystin in hepatocytes incubated with vehicle, 30μM Ml or 50μM of the known Oatp-inhibitor sulfobromophthalein (SBP, see (Eriksson et al, 1990) (Runnegar et al, 1991)). We found that Ml inhibited accumulation of [¹²⁵I] -labelled microcystin- YR equally to SBP (Fig. 8A). It has been published that of the 11 Oatp's found in rat, only one is responsible for the uptake of microcystin (Hagenbuch and Meier, 2003; Fischer et ah, 2005). To test if nostocyclopeptide Ml was a general inhibitor of Oatp's or only inhibited a subset, we tested its ability to inhibit accumulation of radiolabeled bile acids. Even though nostocyclopeptide Ml significantly reduced the accumulation of [14C] glucodeoxycholic acid (by 42%, p<0.005 ANOVA), SBP was a more potent inhibitor. At a concentration of 50μM, SBP blocked 87% of the accumulation (p>0.001 ANOVA). Whereas SBP seemed to inhibit all uptake of glycodeoxycholic acid, nostocyclopeptide Ml appeared to be a potent inhibitor of the subpopulation of transporters responsible for uptake of microcystin and nodularin, but a less potent inhibitor of other liver Oatp's.

EXAMPLE 7

Nostocyclopeptide Ml inhitits uptake of glycocholic acid in a competitive manner

To learn more about the nature of nostocyclopeptide Mi 's inhibition of bile acid uptake, we incubated primary hepatocytes in suspension with several dilution series of both the inhibitor (Ml) and the substrate (glycocholic acid) and washed them as described previously (Fig. 9A). When plotted in a Lineweaver-Burk diagram (Fig. 9B and 9C), we found that Ml inhibited the bile acid transporters in a competitive manner. Competitive inhibition is characterised by increased Km (where the extrapolated trendlines crosses the X-axis), but no change in the Vmax (where the trendlines crosses the Y-axis Fig. 9C).

This finding suggests that nostocyclopeptide Ml binds to the same site as glycocholic acid. Whether nostocyclopeptide Ml is transported into the hepatocytes or stays bound to the transporters is not known. The nature of the broad-substrate transporters suggests that nostocyclopeptide Ml is transported into the hepatocytes, but further studies with radiolabeled nostocyclopeptide Ml is needed to find if this is the case.

REFERENCES

Bhushan, R., Bruckner, H., (2004). Marfey's reagent for chiral amino acid analysis: a review. Amino Acids 27, 231-247. Burja, A.M., Banaigs, B., Abou-Mansour, E., Burgess, J.G., wright, P. C, (2001). Marine cyanobacteria - a prolific source of natural compounds. Tetrahedron 57, 9347- 9377.

Cattori, V., van Montfoort, J.E., Stieger, B., Landmann, L., Meijer, D.K.F., Winterhalter, K.H., Meier, PJ. , Hagenbuch, B., (2001). Localization of organic anion transporting polypeptide 4 (Oatp4) in rat liver and comparison of its substrate specificity with Oatpl, Oatp2 and Oatp3. Pflugers Arch 443, 188-195.

Eriksson, J.E., Gonberg, L., Nygard, S., Slotte, J.P., Meriluoto, J.A.O., (1990). Hepatocellular uptake of 3H-dihydromicrocystin-LR, a cyclic peptide toxin. Biochim Biophys Acta 1025, 60-66.

Fischer, W.J., Altheimer, S., Cattori, V., Meier, P.J., Dietrich, D.R., Hagenbuch, B., (2005). Organic anion transporting polypeptides expressed in liver and brain mediate uptake of microcystin. Toxicol Appl Pharmacol 203, 257-263.

Fladmark, K.E., Brustugun, O. T., Hovland, R., Boe, R., Gjertsen, B. T., Zhivotovsky, B., Døskeland, S.O., (1999). Ultrarapid caspase-3 dependent apoptosis induction by serine/threonine phosphatase inhibitors. Cell Death Differ 6, 1099-1108.

Fladmark, K.E., Brustugun, O. T., Mellgren, G., Krakstad, C, Boe, R., Vintermyr, O.K., Schulman, H., Døskeland, S.O., (2002). Ca2+/calmodulin-dependent protein kinase II is required for microcystin- induced apoptosis. J Biol Chem 277, 2804-2811.

Fujii, K., Sivonen, K., Adachi, K., Noguchi, K., Sano, H., Hirayama, K., Suzuki, M., Harada, K.-L, (1997). Comparative study of toxic and non-toxic cyanobacterial products: Novel peptides from toxic nodularia spumigena AVl. Tetrahedron Letters

38, 5525-5528.

Golakoti, T., Yoshida, W.Y., Chaganty, S., Moore, R.E., (2001). Isolation and structure determination of nostocyclopep tides Al and A2 from the terrestrial cyanobacterium Nostoc sp. ATCC53789. J Nat Prod 64, 54-59. Hagenbuch, B., Meier, PJ. , (2003). The superfamily of organic anion transporting polypeptides. Biochim Biophys Acta 1609, 1-18.

Herfmdal, L., Oftedal, L., Selheim, F., Wahlsten, M., Sivonen, K., Døskeland, S.O., (2005). A high proportion of Baltic Sea benthic cyanobacterial isolates contain apoptogens able to induce rapid death of isolated rat hepatocytes. Toxicon 46, 252- 260.

Herfindal, L., Selheim, F., (2006). Microcystin produces disparate effects on liver cells in a dose dependent manner. Mini Reviews in Medicinal Chemistry 6, 279-285.

Ishitsuka, M.O., Kusumi, T., Kakisawa, H., (1990). Microviridin: A novel tricyclic depsipeptide from the toxic cyanobacterium Microcystis viridis. J Am Chem Soc 112, 8180-8182.

Itou, Y., Suzuki, S., Ishida, K., Murakami, M., (1999). Anabaenopeptins G and H, potent carboxypeptidase A inhibitors from the cyanobacterium Oscillatoria agardhii (NIES-595). Bioorganic & Medicinal Chemistry Letters 9, 1243-1246.

Kodani, S., Ishida, K., Murakami, M., (1998). Aeruginosin 103-A, a thrombin inhibitor from the cyanobacterium Microcystis viridis. J Nat Prod 61, 1046-1048.

Macallan, D.R.E., Lunt, G. G., Wonnacott, S., Swanson, K.L., Rapoport, H., Albuquerque, E.X., (1988). Methyllycaconitine and (+)-anatoxin-a differentiate between nicotinic receptors in vertebrate and invertebrate nervous systems. FEBS Lett 226, 357-363.

MacKintosh, C, Beattie, K.A., Klumpp, S., Cohen, P., Codd, G.A., (1990). Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants. FEBS Lett 264, 187- 192. Meissner, K., Dittmann, E., Borner, T., (1996). Toxic and non-toxic strains of the cyanobacterium Microcystis aeruginosa contain sequences homologous to peptide synthetase genes. FEMS Microbiol Lett 135, 295-303.

Mellgren, G., Vintermyr, O.K., Døskeland, S.O., (1995). Okadaic acid, cAMP, and selected nutrients inhibit hepatocyte proliferation at different stages in Gl: modulation of the cAMP effect by phosphatase inhibitors and nutrients. J Cell Physiol 163, 232- 240.

Mikalsen, B., Boison, G., Skulberg, O.M., Fastner, J., Davies, W., Gabrielsen, T. M., Rudi, K., Jakobsen, K.S., (2003). Natural variation in the microcystin synthetase operon mcyABC and impact on microcystin production in Microcystis strains. Journal of Bacteriology 185, 2774-2785.

Mooberry, S.L., Leal, R.M., Tinley, T.L., Luesch, H., Moore, R.E., Corbett, T.H., (2003). The molecular pharmacology of symplostatin 1 : a new antimitotic dolastatin 10 analog. Int J Cancer 104, 512-521.

Neilan, B.A., Dittmann, E., Rouhiainen, L., Bass, R.A., Schaub, V., Sivonen, K., Borner, T., (1999). Nonribosomal peptide synthesis and toxigenicity of cyanobacteria. Journal of Bacteriology 181, 4089-4097.

Penzotti, J.L., Fozzard, H.A., Lipkind, G.M., Dudley, S.C., Jr., (1998). Differences in saxitoxin and tetrodotoxin binding revealed by mutagenesis of the Na+ channel outer vestibule. Biophys J 75, 2647-2657.

Rantala, A., Fewer, D.P., Hisbergues, M., Rouhiainen, L., Vaitomaa, J., Borner, T., Sivonen, K., (2004). Phylogenetic evidence for the early evolution of microcystin synthesis. Proc Natl Acad Sci U S A 101, 568-573.

Rao, P.V.L., Gupta, N., Bhaskar, A.S., Jayaraj, R., (2002). Toxins and bioactive compounds from cyanobacteria and their implications on human health. J Environ Biol 23, 215-224. Runnegar, M. T., Gerdes, R.G., Falconer, I.R., (1991). The uptake of the cyanobacterial hepatotoxin microcystin by isolated rat hepatocytes. Toxicon 29, 43- 51.

Runnegar, M.T., Xie, C, Snider, B.B., Wallace, G.A., Weinreb, S.M., Kuhlenkamp, J., (2002). In vitro hepato toxicity of the cyanobacterial alkaloid cylindrospermopsin and related synthetic analogues. Toxicol Sci 67, 81-87.

Sekine, T., Cha, S. H., Endou, H., (2000). The multispecific organic anion transporter (OAT) family. Pfαigers Arch 440, 337-350.

Selheim, F., Herfmdal, L., Martins, R., Vasconcelos, V., Døskeland, S.O., (2005). Neuro-apoptogenic and blood platelet targeting toxins in benthic marine cyanobacteria from the Portuguese coast. Aquatic Toxicology 74, 294-306.

Serres, M. H., Fladmark, K.E., Døskeland, S.O., (2000). An ultrasensitive competitive binding assay for the detection of toxins affecting protein phosphatases. Toxicon 38, 347-360.

Sivonen, K., Jones, G., (1999). Cyanobacterial toxins. In Chorus L, Bartram J. (Ed.), Cyanobacterial toxins. E & FN Spon, London, pp. 41-111.

Szabό, S., Szόkan, G., Khlafulla, A.M., Almas, M., Kiss, C, Rill, A., Schon, L, (2001). Configuration and racemization determination of cysteine residues in peptides by chiral derivatization and HPLC: application to oxytocin peptides. J Pept Sci 7, 316-

322.

Trauner, M., Boyer, J.L., (2003). Bile salt transporters: molecular characterization, function, and regulation. Physiol Rev 83, 633-671.

Ueland, P.M., Aarbakke, J., Bessesen, A., (1982). Characterization of S- adenosylhomocysteine binding to isolated rat hepatocytes and purified rat liver plasma membranes. Effect of analogues of S-adenosylhomocysteine. Molecular Pharmacology 21, 108-113.

Claims

Claims

1. A peptide, characterized in that it is a nostocyclopeptide having the sequence Tyrl-Tyr2-(D-HSer)3-L-Pro4-L-Val5-((2S,4S)-4-MePro)6-Tyr7.

2. The peptide of claim 1, characterized in that it has a formula (I)

3. The peptide of claim 1, characterized in that it is obtained from cyanobacterium.

4. The peptide of claim 3, characterized in that the cyanobacterium is Nostoc sp.

5. A compound comprising a peptide of any of claims 1 to 4.

6. A pharmaceutical composition comprising a peptide of any of claims 1 to 4 or a compound of claim 5.

7. The peptide of any of claims 1 to 4 for use as a medicament.

8. Use of the peptide of any of claims 1 to 4 for the manufacture of a medicine for the treatment against hepatotoxins.

9. Use of the peptide of any of claims 1 to 4 as a cell reagent.

10. Use of the peptide of any of claims 1 to 4 in a screening method.