KR102285119B1 - Novel histone methyl transferase inhibitor and use thereof - Google Patents

Abstract

The present invention relates to novel histone methyl transferase inhibitors.
In addition, the present invention relates to a pharmaceutical composition or a food composition comprising the novel histone methyl transferase inhibitor.
The present invention also relates to a method of treating cancer using the novel histone methyl transferase inhibitor.

Description

NOVEL HISTONE METHYL TRANSFERASE INHIBITOR AND USE THEREOF

The present invention relates to novel histone methyl transferase inhibitors.

In addition, the present invention relates to a pharmaceutical composition or a food composition comprising the novel histone methyl transferase inhibitor.

The present invention also relates to a method of treating cancer using the novel histone methyl transferase inhibitor.

In the development of various cancers, epigenetic deregulation has been highlighted as a very important factor. NSD1, NSD2/MMSET/WHSC1, and NSD3/WHSC1L1 form a nuclear receptor binding SET domain (NSD) family and are classified into the histone methyltransferase (HMTase) KMT3 family. NSD is a histone modifier involved in the maintenance of chromatin. NSD mostly methylates histone H3 lysine 36 (H3K36) and histone H4 lysine 20 (H4K20) of chromatin, and in vivo NSD1, NSD2, and NSD3 act by different mechanisms. Specifically, it does not contribute substantially to the overgrowth phenotype in the case of NSD2 and NSD3, whereas it is observed in the NSD1 gene variant. However, it has not yet been revealed whether the NSD family members can replace each other. NSDs, particularly NSD2, play an important role in the development and progression of various cancers, and are considered to be very important drug targets. NSD1 amplification has been observed in multiple myeloma, lung cancer, neuroblastoma, and glioblastoma. Amplification of NSD1 and NSD2 triggers cell transformation into cancer cells. NSD3 plays a role in lung cancer and has been found to be amplified in breast cancer cell lines and primary breast carcinomas. NSD2/MMSET (multiple myeloma SET) is associated with tumor aggressiveness and prognosis in most cancer types, including prostate cancer and multiple myeloma. NSD2 is overexpressed in solid tumors, particularly breast cancer, myeloma and glioblastoma, resulting in an abnormal increase in the global level of H3K36me2. Overexpression of NSD2 in prostate cancer induces epigenetic aberration and induces a metastatic phenotype. NSD2/MMSET is an important focus in therapeutic intervention for multiple myeloma, especially t(4;14) myeloma, which has the poorest prognosis compared to other types. Multiple myeloma is the second most common blood cancer after non-Hodgkin's lymphoma. There is still no treatment or effective treatment method for multiple myeloma.

Chromatin remodeling inhibitors, targeting DNA methyltransferases (DNMTs), histone methyltransferases and histone deacetylases (HDACs), are being pursued in both chemotherapy and prophylactic chemotherapy. Several DNA-methylation and histone deacetylase inhibitors are currently in clinical trials. HMTase is a novel sample target, and HMTase inhibitors are still lacking and very few compounds have been reported. In particular, GlaxoSmithKline Inc. and Epizyme Inc. have made great strides in discovering potential inhibitors of HMTase DOT1L and EZH2. DOT1L does not contain a canonical SET domain, so it is distinguished from other HMTases. The HMTase inhibitors, BIX-01294 and BIX-01338, showed an effect on G9a with _{3 and 5 IC 50 , respectively.} In addition, Chaetocin inhibits Su(var)3-9 with _{an IC 50 of 0.8.} Importantly, Liu et al. completed the formation of several G9a inhibitors, such as UNC0224 with _{an IC 50 of 15 nM.} Epizyme Inc. and GlaxoSmithKline Inc's G9a and EZH2 researchers reported HMTase inhibitors with an IC50 of nanomolar levels. However, only one NSD2 inhibitor (MCTP39) was confirmed to have an IC50 of ~3 and was recently patented.

The catalytic mechanism of lysine-HMTase is carried out through a linear SN2 nucleophilic attack between the cofactor SAM (SAdenosylmethionine) and the Lysine-NH3 substrate. SAM binds into a small cavity adjacent to the histone-tailed large binding groove where the lysine substrate extends deep into a channel located at the interface between the two binding sites.

Previously, the present inventors confirmed that the SET domain of NSD1 accommodates a 7-amino acid peptide similar to that identified in SET8. We also demonstrated the mechanism of initiation of the SET domain of NSD1 through rotation of a small loop at the interface between the SET and postSET subdomains. This regulatory-loop acts like a seat belt for the lysine-air chamber and participates in both substrate recognition and catalytic mechanisms. This regulatory-loop is located on top of the lysine-substrate, which strongly anchors the histone-tail to the SET domain. Histone tail binding sites are associated with portions of both the SET and postSET subdomains. The sequence of the SET domain is highly conserved between NSDs, and therefore it is expected that the same mechanism described in NSD1 is performed in NSD2 and NSD3. In particular, NSDs are phylogenetically distant from other known HMTases (Fig. 2). Inhibition of HMTases is achieved by targeting the SAM binding pocket or histone-tail binding cleft. In EZH2, the inhibitor GSK343 has an inhibitory activity on EZH1 as a SAM competitor, and also has an inhibitory effect on SET7 and PRMT3. Interestingly, EZH1 and EZH2 have very close SET domain sequences for NSDs, which may be important for the selection of EZH2 inhibitors against NSDs, especially NSD2. MCTP39 was recently patented as a SAM competitor to NSD2. The SAM binding pocket is partially buried within the crystal structure of the SET domain within NSD1, and key residues stabilizing SAM are highly conserved among NSDs. In contrast, the larger and more accessible histone-tail binding site has greater sequence heterogeneity compared to the SAM binding pocket. Residues involved in stabilizing the binding of histone substrates are not conserved between NSDs, unlike the SAM binding site. This could be observed in the crystal structure of HMTase G9a-like protein and BIX-01294, a histone-tail mimicking inhibitor that targets the histone tail binding domain. In addition, the involvement of specific flexible regulatory loops in the binding of histone-tails to the NSDs SET domain suggests that histone-tail binding niches can be developed to achieve specific and selective inhibition in NSD2.

Therefore, in the present invention, the present inventors identified a novel NSD2 inhibitor against the histone-tail binding groove, which was performed using a hypothetical ligand screening for an open molecular model of the histone-tail binding domain of NSD2-SET. . In addition, the present inventors have developed a derivative of the novel NSD2 inhibitor.

One aspect of the present invention is to provide novel histone methyl transferase inhibitors.

In addition, one aspect of the present invention is to provide a pharmaceutical composition or a food composition comprising the novel histone methyl transferase inhibitor.

Another object of the present invention is to provide a method for treating cancer using the novel histone methyl transferase inhibitor.

One aspect of the present invention provides a compound consisting of the following formula: or a pharmaceutically acceptable salt thereof:

here,

The A group is selected from the group consisting of

The group B is selected from the group consisting of

The C group is selected from the group consisting of;

wherein R1 is selected from the group consisting of methyl, ethyl, di-methyl and i-Pr,

R2 and R3 are independently selected from the group consisting of hydrogen, a substituted or unsubstituted alkyl group, a substituted or unsubstituted cycloalkyl group, and a substituted or unsubstituted heterocycloalkyl group having O, N, S or P as a heteroelement; ,

R4 is selected from the group consisting of fluorine, chlorine, methyl, and OMe;

R5 is selected from the group consisting of methyl, ethyl, tert-butyl, substituted or unsubstituted phenyl, and a substituted or unsubstituted phenyl group having O, N, S or P as a heteroelement.

The compound may be an inhibitor of histone lysine-methyl transferase, and the histone methyl transferase may be preferably a nuclear receptor binding SET domain (NSD) family, more preferably NSD2/MMSET.

The inhibitor of the histone lysine-methyl transferase may be a compound, most preferably consisting of the following formula.

In addition, one aspect of the present invention provides a pharmaceutical composition for preventing or treating cancer, comprising the compound as an active ingredient. The cancer is lymphoma (lymphoid malignancies), acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), myeloma (myeloma), skin cancer, primary melanomas, yolk sac Yolk sac tumour, breast cancer, lung cancer, colorectal cancer, pancreatic cancer, ovarian cancer, bladder cancer, prostate cancer, kidney cancer, brain cancer, cervical cancer, leukemia, granulomatosis, and cellular transformation), and most preferably multiple myeloma.

In addition, one aspect of the present invention provides a food composition for preventing or improving cancer, comprising the compound as an active ingredient.

Another aspect of the present invention also provides a method of treating cancer, comprising administering to a subject a pharmaceutically effective amount of the compound.

Since the novel compound according to the present invention can effectively inhibit histone lysine-methyl transferase inhibitor, it can be used as a therapeutic agent in various tumors involving histone lysine-methyl transferase.

1 is a schematic diagram of the primary structures of NSD1, NSD2, and NSD3: all NSD members contain two PWWP domains, four PHD zinc finger domains and a catalytic domain. The catalytic domain has HMTase activity and includes pre-SET, SET, and post-SET domains. A regulatory loop covering the histone-binding portion is indicated in red.
2 is a diagram showing the HMTase activity of NSD1-CTD, NSD2-CTD, and NSD3-CTD on H3 and H4 substrates in vitro.
3 is a view confirming the open and closed structures and molecular modeling of the catalytic domain.
Left panel: The structure of the catalytic domain consists of one group of three groups of canonical -sheets and two closely adjacent -sheets with conserved -helices arranged in a triangular fashion, allowing for binding of lysine-histone ligands. form a gap Authorized AdoMet joins the CD into a suitable configuration through the channel. Fluid control loops are marked in red. Movement of the control loop is indicated by a curved inverted arrow.
Right panel: Model of NSD-CDs bound to H3-peptide (aa 32-38) after MD simulation. H3K36 was indicated. Electrostatic surfaces were colored as follows: blue: positive charge; red; negative charge. Units +5/-5 kb.T.ec -1.
4 is a diagram showing the molecular determinant of H3K4, H3K9 and H3K27 bonds.
Peptides H3K4 (aa1-7), H3K9 (aa 5-11), and H3K27 (aa 23-29) are aligned vertically and atomic details of residues are shown in purple. Interactions of amino acids from CDs of NSD are marked in brown and numbered. The green dotted line represents hydrogen bonding. Hydrophobic interactions were omitted. This drawing was constructed using the program Ligplot.
5 is a diagram showing the determinants of H3K36, H3K79 and H4K20 combinations. Peptides H3K36 (aa 32-38), H3K79 (aa 75-81), and H4K20 (aa16-22) are shown.
6 is a diagram illustrating a model of an open SET domain including the structure of NSD2/MMSET and LEM-14.
(A) Primary structure of NSD2: PWWP domain; PHD zinc finger domain; SET HMTase (histone methyl transferase), including preSET and postSET. The regulatory loop that closes the histone-binding portion is marked in red.
(B) The result of Coomassie staining of purified recombinant NSD2-SET on an SDS-PAGE gel.
(C) An open NSD2-SET model that is bound to LEM-14 in the most preferred structure through virtual ligand screening. Control loops are marked in red.
(D) A diagram showing the molecular surface of NSD2-SET bound to LEM-14. Electrostatic surfaces were colored as follows:
Blue: positive charge; Red: negative charge, unit +5/-5 kb.Te _c ^-1 .
7 is a diagram showing the sequence relationship between HMTases and SET domains of NSD.
(A) Unrooted phylogenic tree for protein sequences of the SET domains of human and yeast HMTases. The NSD family is far from other known HMTases (red). This phylogenetic tree was designed using the NJ (neighbor-joining) method after multiple sequence alignment using CLUSTAL 2.1. Hs: Homo sapiens ; Sc: Saccharomyces cerevisiae ; Sp: Saccharomyces pombe
(B) Sequence alignment results of preSET, SET, and postSET subdomains of NSD1, NSD2 and NSD3. The sequence in the blue box is involved in histone-tail binding. The sequence in the red box is reactive to S-Adenosylmethionine. Multiple sequence alignment was performed with CLUSTAL 2.1.
8 is a diagram showing a concentration-response curve of an inhibitor for NSD2-SET H3K36 mono-methylation activity. NSD2-SET H3K36 mono-methylation activity was displayed at various concentrations of candidate inhibitors. The concentration response curves and IC50 of LEM-06, LEM-14, and BIX-01294 were displayed using GraphPad Prism 6.0.
9 is the chemical structure of LEM-06, LEM-14 and BIX-01294.
10 is a diagram showing molecular details of inhibitor binding in the histone-tail binding groove of NSD2. (A) LEM-14, (B) LEM-06, (C) BIX-01294. The yellow dotted line represents hydrogen bonding.
11 is a diagram showing the binding of BIX-01294 and LEM-06 docked to the NSD2 histone-tail binding groove. Molecular surface model of the open structure of NSD2-SET coupled with BIX-01294 (A) and LEM-06 (B). Blue: positive charge; Red: negative charge, unit +5/-5 kb.Te _c ^-1 .

Best mode for carrying out the invention

One aspect of the present invention provides a compound consisting of the following formula: or a pharmaceutically acceptable salt thereof:

here,

The A group is selected from the group consisting of

The group B is selected from the group consisting of

The C group is selected from the group consisting of;

wherein R1 is selected from the group consisting of methyl, ethyl, di-methyl and i-Pr,

R2 and R3 are independently selected from the group consisting of hydrogen, a substituted or unsubstituted alkyl group, a substituted or unsubstituted cycloalkyl group, and a substituted or unsubstituted heterocycloalkyl group having O, N, S or P as a heteroelement; ,

R4 is selected from the group consisting of fluorine, chlorine, methyl, and OMe;

R5 is selected from the group consisting of methyl, ethyl, tert-butyl, substituted or unsubstituted phenyl, and a substituted or unsubstituted phenyl group having O, N, S or P as a heteroelement.

The compound may be an inhibitor of histone lysine-methyl transferase, and the histone lysine-methyl transferase may be preferably a nuclear receptor binding SET domain (NSD), more preferably NSD2.

The inhibitor of the histone lysine-methyl transferase may be a compound, most preferably consisting of the following formula.

The present invention includes optical isomers, enantiomers, diastereomers and geometric isomers of the compounds. The present invention also includes all tautomeric forms of the compounds.

Another aspect of the present invention provides N-oxides, hydrates, solvates, pharmaceutically acceptable salts, complexes and prodrugs of the compounds.

The compounds of the present invention also form salts that are within the scope of the present invention. It is understood that references to the above compounds of the present invention include references to salts thereof unless otherwise stated. As used herein, the term “salt(s)” refers to acidic and/or basic salts formed with inorganic and/or organic acids and bases.

Also, when a compound of the present invention contains both a basic moiety such as but not limited to pyridine or imidazole and an acidic moiety such as, but not limited to, a carboxylic acid, a zwitterion ("internal salt") ) may be formed and are encompassed within the term “salt(s)” as used herein. Other salts are useful, such as in separation or purification steps that may be applied during manufacture, but pharmaceutically acceptable (ie, non-toxic (with minimal or no undesired toxic effects)) salts are preferred. A salt of a compound of the present invention can be formed, for example, by reacting a compound of the present invention with a predetermined amount of an acid or base, such as an equivalent, in a medium, such as a medium or aqueous medium in which the salt is precipitated, and freeze-drying. In certain cases, greater than one equivalent of acid (base) may be used to provide, for example, di- or tri-salts.

Compounds of the present invention containing a basic moiety such as, but not limited to, an amine or pyridine or imidazole ring can form salts with a variety of organic and inorganic acids. Exemplary acid addition salts include acetates (such as those formed with acetic acid or trihaloacetic acid, such as trifluoroacetic acid), adipate, alginate, ascorbate, aspartate, benzoate, benzenesulfonate. , bisulfate, borate, butyrate, citrate, camphorate, camphorsulfonate, cyclopentanepropionate, digluconate, dodecylsulfate, ethanesulfonate, fumarate, glucoheptanoate, glycerophosphate, hemi Sulfate, heptanoate, hexanoate, hydrochloride, hydrobromide, hydroiodide, hydroxyethanesulfonate (such as 2-hydroxyethanesulfonate), lactate, maleate, methanesulfonate, naphthalenesulfo nates (such as 2-naphthalenesulfonate), nicotinate, nitrate, oxalate, pectinate, persulfate, phenylpropionate (such as 3-phenylpropionate), phosphate, picrate, pivalate, propionates, salicylates, succinates, sulfates (such as those formed with sulfuric acid), sulfonates, tartrates, thiocyanates, toluenesulfonates such as tosylate, undecanoate and the like.

Compounds of the present invention containing acidic moieties such as, but not limited to, carboxylic acids can form salts with a variety of organic and inorganic bases. Exemplary basic salts include ammonium salts, alkali metal salts such as sodium, lithium and potassium salts, alkaline earth metal salts such as calcium and magnesium salts, salts with organic bases (eg organic amines) such as benzidine, dish chlorhexylamine, hydrabamine (formed from N,N-bis(dehydroabiethyl)ethylenediamine), N-methyl-D-glucamine, N-methyl-D-glycamide, t-butyl amine and arginine; and salts with amino acids such as lysine. Basic nitrogen-containing groups include lower alkyl halides (such as methyl, ethyl, propyl and butyl chloride, bromide and iodide), dialkyl sulfates (such as dimethyl, diethyl, dibutyl and diamyl sulfate), long chain halides (such as , decyl, lauryl, myristyl and stearyl chlorides, bromides and iodides), aralkyl halides (eg benzyl and phenethyl bromide), and the like.

As used herein, the term “pharmaceutically acceptable salts” refers to salts that retain the required biological activity of the aforementioned compounds and exhibit minimal or no toxic effects. Examples of such salts include, but are not limited to, acid addition salts formed with inorganic acids (eg, hydrochloric acid, hydrobromic acid, sulfuric acid, phosphoric acid and nitric acid) and organic acids such as acetic acid, oxalic acid, tartaric acid, succinic acid, malic acid, ascorbic acid, salts formed with benzoic acid, carbonic acid, pamoic acid, alginic acid, polyglutamic acid, naphthalenesulfonic acid, naphthalenedisulfonic acid, and polygalacturonic acid. The compounds may also be administered as pharmaceutically acceptable quaternary salts known to those skilled in the art, and they are particularly of the formula (-NR) ⁺ +Z ^- (wherein R is hydrogen, alkyl or benzyl, Z is chloride, bromide, Iodide, -O-alkyl, toluenesulfonate, methylsulfonate, sulfonate, phosphate or carboxylate (such as benzoate, succinate, acetate, glycolate, maleate, maleate, citrate, tartrate , ascorbate, benzoate, cinnamoate, mandeloate, benzyloate or diphenylacetate)).

The present invention also includes prodrugs of the compounds of the present invention. The term “prodrug” is intended to refer to a compound covalently bound to a carrier, and when the prodrug is administered to a mammalian subject, the prodrug is capable of releasing the active ingredient of the prodrug. Release of the active ingredient occurs in vivo. Prodrugs can be prepared by techniques known to those skilled in the art. These techniques generally modify the appropriate functional groups of a given compound. However, this modified functional group regenerates the original functional group either by routine manipulation or in vivo. Prodrugs of the compounds of the present invention include compounds in which a hydroxyl group, an amino group, a carboxyl group or a similar group is modified. Examples of prodrugs include, but are not limited to, esters (eg, acetate, formate and benzoate derivatives), carbamates of hydroxy or amino functional groups in the compounds of the present invention (eg, N,N-dimethylaminocarbonyl) , amides (eg, trifluoroacetylamino, acetylamino, etc.).

The histone methyl transferase and "HMTase" are enzymes that catalyze the transfer of one, two or three methyls to lysines of histones, and are histone-modifying enzymes. The term “histone” is meant to refer to any histone protein, including H1, H2A, H2B, H3, H4, and H5 from any species.

The term "histone methyl transferase inhibitor" means that it is capable of interacting with and inhibiting its enzymatic activity. The expression "inhibiting the enzymatic activity of histone methyl transferase" means reducing the ability to transfer methyl groups in histones. For example, inhibition of histone methyl transferase activity can be about 10% or greater. In some preferred embodiments of the invention, such reduction in histone methyl transferase activity is at least about 50%, more preferably at least about 75%, even more preferably at least about 90%. In another preferred embodiment, histone histone methyl transferase activity is reduced by at least 95%, more preferably by at least 99%. The IC ₅₀ value is the concentration of histone deacetylase inhibitor that reduces the activity of histone histone methyl transferase by 50%. The term “inhibitory effective amount” is intended to denote a dose sufficient to cause inhibition of histone methyl transferase activity.

The histone methyl transferase may be present in a cell, ie in a multicellular organism. A multicellular organism may be, for example, a plant, a fungus or an animal, preferably a mammal and more preferably a human. The fungus is capable of infecting a plant or mammal, preferably a human, so that it may be present in and/or on the plant or mammal. When the histone methyl transferase is in a multicellular organism, the method according to this aspect of the invention comprises administering to the organism a compound or composition according to the invention. Administration may be effected by any route including, without limitation, parenteral, oral, sublingual, transdermal, topical, intranasal, intratracheal or rectal. In certain particularly preferred embodiments, the compounds of the invention are administered intravenously in a hospital setting. In certain other preferred embodiments, it may be desirable for administration by the oral route.

In addition, one aspect of the present invention provides a pharmaceutical composition for preventing or treating cancer, comprising the compound as an active ingredient. The cancer is lymphoma (lymphoid malignancies), acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), myeloma (myeloma), skin cancer, primary melanomas, yolk sac Yolk sac tumour, breast cancer, lung cancer, colorectal cancer, pancreatic cancer, ovarian cancer, bladder cancer, prostate cancer, kidney cancer, brain cancer, cervical cancer, leukemia, granulomatosis, or cellular transformation), and most preferably multiple myeloma.

In addition, one aspect of the present invention provides a food composition for preventing or improving cancer, comprising the compound as an active ingredient.

Another aspect of the present invention also provides a method of treating cancer, comprising administering to a subject a pharmaceutically effective amount of the compound.

The pharmaceutical composition of the present invention may be formulated by any method well known in the art and by any route including, without limitation, parenteral, oral, sublingual, transdermal, topical, intranasal, intratracheal or rectal. may be prepared for administration. In certain preferred embodiments, the pharmaceutical compositions of the present invention are administered intravenously in a hospital setting. In certain other preferred embodiments, administration is preferably by the oral route. The pharmaceutical composition may be in any form including, but not limited to, a solution or a suspension. For oral administration, the formulation may be in the form of a tablet or capsule. For intranasal administration, the pharmaceutical composition may be in the form of a powder, nasal drop or aerosol. The pharmaceutical composition may be administered topically or systemically. The nature of the carrier will depend on the route of administration. As used herein, the term "pharmaceutically acceptable" refers to a non-toxic substance that is compatible with a biological system, such as a cell, cell culture, tissue, or organism, and does not interfere with the effect of the biological activity of the active ingredient(s). it means. Accordingly, the composition according to the invention may contain, in addition to the inhibitor, diluents, fillers, salts, buffers, stabilizers, solubilizers and other substances known in the art. Methods for preparing pharmaceutically acceptable formulations are described, for example, in Remington's Pharmaceutical Sciences, 18th Edition, ed. A. Gennaro, Mack Publishing Co., Easton, PA, 1990].

The compound, N-oxide, hydrate, solvate, pharmaceutically acceptable salt, complex, prodrug or mixture, or racemic mixture, diastereomer, enantiomer or tautomer thereof does not cause serious toxic effects in the patient being treated. and included in a pharmaceutically acceptable carrier or diluent in an amount sufficient to deliver a therapeutically effective amount to a patient. A typical topical dosage will be in the range of 0.01-3% wt/wt in a suitable carrier. An effective dosage range of a pharmaceutically acceptable derivative can be calculated based on the weight of the parent compound to be delivered. Where the derivative exhibits intrinsic activity, the effective dosage can be calculated as above using the weight of the derivative, or by any means known to those skilled in the art.

Depending on the particular disease or condition being treated, additional therapeutic agents that would normally be administered to treat the disease or condition may also be present in the compositions of the present invention. Alternatively, the administration of such agents may be performed sequentially or concurrently with the administration of the composition according to the invention. In other words, the compounds of the present invention may be administered as a single pharmaceutical agent or in combination with one or more other additional therapeutic (pharmaceutical) agents in which the combination does not cause any unacceptable side effects. It may be particularly suitable for the treatment of hyperproliferative diseases such as cancer. In such instances, the compounds of the present invention may be combined with known anticancer agent(s) as well as mixtures and combinations thereof. As used herein, an additional therapeutic agent generally administered to treat a particular disease or condition is known as "suitable for the disease or condition being treated." As used herein, “additional therapeutic agents” include, for example, chemotherapeutic agents and other anti-proliferative agents. In certain preferred embodiments of the invention, the composition comprises one or more compound(s) according to the invention. In certain preferred embodiments of the invention, the composition comprises one or more compound(s) according to the invention and/or another HMTase inhibitor known or to be discovered by the person skilled in the art. The active ingredients of such compositions preferably act synergistically to produce a therapeutic effect.

Modes for carrying out the invention

Example 1: Histone Lysine Methylation by NSD1, NSD2/MMSET/WHSC1, and NSD3/WHSC1L

The present inventors investigated the recognition and methylation of histone H3 by NSDs. The present inventors focused on examining the carboxyl terminal domain (CTD) of three NSDs through a consistent experimental approach. NSDs-CTDs contain pre-SET, SET, post-SET, and PHD4 domains within a catalytic core and histone-recognition module. The present inventors investigated the methylation properties of CTDs of NSD1, NSD2, and NSD3 in H3K4, H3K9, H3K27, H3K36, H3K79, and H4K20 in vitro. Next, the present inventors demonstrated the binding of histone tails to the NSD catalytic SET domain using molecular dynamic simulations.

1.1 Cloning

The primary structures of NSD1, NSD2, and NSD3 belonging to the NSDs family, which are HMTases, are shown in FIG. 1A.

1 is a schematic diagram of the primary structures of NSD1, NSD2, and NSD3: all NSD members contain two PWWP domains, four PHD zinc finger domains and a catalytic domain. The catalytic domain has HMTase activity and includes pre-SET, SET, and post-SET domains. A regulatory loop covering the histone-binding portion is indicated in red.

The carboxyl terminal domains (CTDs) of NSD1, NSD2 , and NSD3 were cloned as follows;

NSD1-CTD (1338 bp, 5325-6663 nt; encoding 446 aa, 1775-2221 aa), NSD2-CTD (1287 bp, 2808-4095 nt; encoding 429 aa, 936-1365 aa), and NSD3-CTD (1344) bp, 2967-4311 nt; encoding 484 aa, 989-1437 aa) was amplified by PCR using a human liver cDNA library (TAKARA, Japan) as a template.

The forward and reverse primers used are as follows.

For NSD1-CTD ,

5- GCTGAGGTCGAC( Sal I)CATCCTCGAGCTGTTCCTTCC-3 and

5- CGGTACGCGGCCGC( Not I)GCACATACTCACGGATCTCCCC-3;

For NSD2-CTD ,

5-CAGGCGCATATG ( Nde I)TTCCCGTACATGGAGGGGGAC-3 and

5-CGCCTGCTCGAG( Xho I)TTTGCCCTTGTGACTCTCCG-3;

For NSD3-CTD

5-CTGAACGTCGAC( Sal I)GGCCTTAAACATGACTTGGGG-3 and

5-GACACCGCGGCCGC( Not I)ATTCTTTTACTTC TTCTCCATG-3.

The amplified NSD1-CTD and NSD3-CTD strands ( containing Sal I and Not I restriction sites) and NSD2-CTD strands ( containing Nde I and Xho I restriction sites) were subjected to protein expression intein-tagged The vector, pTYB2 or pTYB12 (New England Biolabs, USA) was inserted into the restriction site of the polyclonal site.

1.2 Protein Expression and Purification

Escherichia coli expression lines (ER2566 or BL21) were transformed with pTYB2 or pTYB12 plasmids containing NSD1-CTD , NSD2-CTD , and NSD3-CTD ( NSD-CTDs). The transformed cell line was cultured in LB medium containing 100 g/mL ampicillin, and the expression of recombinant NSD-CTDs was induced with 250 M IPTG (isopropyl 1-thio-D-galactopyranoside) at 15 °C for 4 hours.

E. coli cells were recovered and lysed by a freeze-and-thaw method, and buffer A containing 0.1% Triton X-100 and 1 mM PMSF (phenylmethanesulfonylfluoride) [20 mM Tris (pH 8.0), 500 mM NaCl, and 0.1 mM EDTA] followed by sonication on ice 20 times.

The cell extract containing the NSD-CTD-Intein-CBD (chitin-binding domain) fusion protein was loaded on an affinity column of chitin beads, and washed with 100-column volume of buffer A containing 0.1% Triton X-100. Then, it was washed with a 20-column volume of buffer A without Triton X-100. To remove the bacterial chaperone bound to the recombinant protein, the recombinant NSDCTD-bound chitin beads were washed with a 10-bead volume of buffer A containing _{10 mM ATP (adenosine triphosphate) and 2.5 mM MgCl 2 .} The affinity column was then washed with 20 bead volumes of Buffer A to remove bacterial chaperones. NSD-CTD protein was excised from chitin beads by incubation for 48 hours at 4 in buffer A containing 50 mM 2-mercaptoethanol. After eluting with buffer A, the mixture was concentrated using an Amicon Ultra centrifugal filter, and then used for a methyltransferase assay. Some of the purified NSD-CTDs were analyzed by SDSPAGE. Soluble, pure NSD-CTDs of expected molecular weights of 50.8 kDa (NSD1-CTD), 48.9 KDa (NSD2-CTD) and 51.1 KDa (NSD3-CTD) were identified in Coomassie stained gels ( FIG. 1B ). Recombinant NSD2-SET was stable thereafter and maintained catalytic activity when stored at 80 °C.

1.3 Histone Methyltransferase Assay

The histone methyltransferase activity of NSD-CTDs in H3K4, K9, K27, K36, K79, and H4K20 was measured using a colorimetric quantification kit (Epigentek, USA) according to the manufacturer's protocol. For H3K36, H3K79, and H4K20, purified recombinant NSD-CTDs (0.2 g and 2.0 g) were incubated with recombinant histones H3 or H4 (Epigentek, USA) and methyl group donors in strip wells coated with specific antibodies. (Adomet) and incubated at room temperature for 60-90 min. Specific antibodies bound to the bottom of the strip wells captured the methylated substrate.

For H3K4, H3K9, and H3K27, a biotinylated histone substrate was used instead of the antibody-coated strip wells and stably captured on the strip wells at 37°C for 60 minutes. The captured photinylated histone substrate was then incubated with the high-affinity antibody for 60 min at room temperature with agitation of 100 rpm. Excess purified NSD-CTDs, histones, Adomet, and antibody were washed away and the antibody-bound or biotinylated H3/H4me-bound strip wells were treated with labeled detection antibody for 60 minutes at 100°C in the dark. Incubated in vortex conditions at rpm. After washing the wells, the degree of methylation was quantified at 450 nm with an ELISA plate reader using an HRP-conjugated secondary antibody-chromogenic system. It was confirmed that cross-reaction between substrates did not occur using the antibody provided by Epigentek. The assay was performed in duplicate. Results were normalized based on the control group without enzyme (NC in FIG. 2 ).

1.4 Molecular Modeling, Docking, and Energy Minimization

Crystal structure of the catalytic domain (CD) of NSD1 to construct a homology model of NSD2-CD (aa 971-1202) and NSD3-CD (aa 1044-1283) (PDB: 3OOI - aa 1850-2080) was used as a template. NSD2-CD (aa 971-1202) shares 75.9% identity and 90.1% similarity with NSD1-CD (PDB: 3001), and NSD3-CD (aa 1044-1283) with NSD1-CD 72.9% identity and 85% similarity were shown.

In order to accurately construct the molecular model, the guidelines and methods previously described by the present inventors were followed. After multiple-sequence alignment using ClustalW V2.0.9, 100 models were generated using Modeler V9.5. Side-chains positioning was performed using the SCWRL4 program for the optimal model according to the unique Modeler function. Stereochemistry was confirmed using PROCHECK. The model was manually checked using COOT. Histone tail peptides were identified as H3K4 (aa 1-7), H3K9 (aa 5-11), H3K27 (aa 23-29), H3K36 (aa 32-38), H3K79 (aa 75-81), and H4K20 (aa 16- 22), manually constructed in COOT. Using the SSM algorithm in COOT, the NSD1-CD (PDB: 3OOI) and histone H4 peptide (NSD1-CD (PDB: 3OOI)) and histone H4 peptide ( Structural overlay with the crystal structure of SET8 combined with PDB: 3F9Y) was used. Electrostatic calculations were performed using APBS V1.2.1, and the electrostatic properties of the molecular surface were expressed as VMD V1.8.7 and PyMOL V1.2.

1.5 Energy Minimization and Molecular Dynamics Simulation

The peptide-NSD-CD complex was fully solvated in a water bath (140 x 100 x 110) using VMD 1.9. Molecular systems used for energy minimizations (EM) and molecular dynamics (MD) contain the following number of atoms: NSD1-CD-H3K4(1-7), 42,121 atoms; NSD1-CD-H3K9(5-11), 42,104 atoms; NSD1-CD-H3K27(23-27), 42,095 atoms; NSD1-CD-H3K36(32-38), 42,106 atoms; NSD1-CD-H3K79 (75-81), 42,120 atoms; NSD1-CDH4K20 (32-38), 42,130 atoms; NSD2-CDH3K4(1-7), 44,848 atoms; NSD2-CD-H3K9(5-11), 44,840 atoms; NSD2-CD-H3K27 (23-27), 44,828 atoms; NSD2-CD-H3K36(32-38), 44,830 atoms; NSD2-CD-H3K79 (75-81), 44,844 atoms; NSD2-CD-H4K20 (32-38), 44,860 atoms; NSD3-CD-H3K4(1-7), 44,367 atoms; NSD3-CDH3K9 (5-11), 44,353 atoms; NSD3-CDH3K27 (23-27), 44,341 atoms; NSD3-CDH3K36 (32-38), 44,349 atoms; NSD3-CDH3K79 (75-81), 44,366 atoms; NSD3-CDH4K20 (32-38), 44,379 atoms. About 90% of the atoms are derived from water molecules. EM and MD were performed with NAMD v2.8 using a CHARMM force field. The constant temperature was maintained using Langevin damping at 300 K, and the pressure was kept constant at 1 pressure using a Langevin piston. Electrostatic force was considered as a 10 cut-off distance using the particle mesh Ewald (PME) method. The solvated complex, histonepeptide-NSD-CD, was first treated with 5 repetitions of 5,000 steps (5 ns) conjugate gradient EM, followed by 20,000 steps (20 ns) molecular kinetics and 5,000 steps (5 ns) EM. equilibrated. The last step was 5ns of EM. A total of 100,000 steps (100 ns) of MD and a total of 30,000 steps (30 ns) of EM were performed on each of the solvated histone-peptide-NSD-CD complexes. After MD simulations, the solvated complexes were manually identified by COOT for ideal stereochemistry. The MD trajectory was analyzed with VMD 1.9.

1.6 Determination of Binding Energy

The binding energy was calculated as follows using the total energy calculation of NAMD 54.

binding = complex (peptide + NSD).

The model of the final MD phase was separated from the solvent molecules and used to calculate the binding energy of the complex. All models of binding complexes, peptides, and NSD-CD apo were subjected to the same EM procedure, and then binding energy calculations were performed. The model was energy-minimized with 4,000 steps (4 ns) of conjugate gradient EM, followed by 2,000 steps (2 ns) of MD, and 6,000 steps (6 ns) of EM (NAMD v2.8). The EM phase was long enough to reach a local minimum of <1 kJ/mol. The constant temperature was maintained using Langevin damping at 300 K, and the pressure was kept constant at 1 pressure using a Langevin piston.

1.7 Model Analysis and Verification

The details of the interaction between NSD-CD and H3- or H4-peptides were analyzed by Ligplot. The interactive structure was manually mapped through a visual review of the model within the COOT.

1.8 Confirmation that CTD of NSD member can methylate multiple histone lysines in vitro

To determine whether the CTDs of NSD1, NSD2, and NSD3 can specifically recognize and methylate multiple different lysines in H3 and H4 in vitro , we present H3K4, H3K9, H3K27, H3K36, H3K79, and The pan-methylation activity of NSD1-CTD, NSD2-CTD, and NSD3-CTD on H4K20 was quantified.

The results are shown in FIG. 2 . 2 is a diagram showing the HMTase activity of NSD1-CTD, NSD2-CTD, and NSD3-CTD on H3 and H4 substrates in vitro.

As shown in FIG. 2 , NSD1-CTD, NSD2-CTD, and NSD3-CTD showed similar substrate recognition and methylation properties, but decreased methylation activity was confirmed in NSD3-CTD ( FIG. 2 ). In NSD1-CTD and NSD2-CTD, more H3K4me3 and H3K27me3 species were detected than me1 and me2 species. However, H3K9-me1 & -me2 and H3K79-me1 & -me2 species were more preferred than me3 species. Compared with NSD1-CTD and NSD2-CTD, NSD3-CTD showed significant methylation activity (~1.5 fold) except for insignificant activity on H3K4 and H3K9. For specific substrates, the methylation activity of NSD3-CTD to H3K9, H3K27, and H3K79 was nearly identical to that of NSD1-CTD and NSD2-CTD.

However, H4K20 methylation varied among NSD-CTDs. Nevertheless, significant mono-/tri-methylation activity of NSD1-CTD, di-/tri-methylation activity of NSD2-CTD, and di-/tri- methylation activity of NSD3-CTD were confirmed in H4K20, indicating that H4K20 is in It is the target of all NSD-CTDs in vitro. H3K36 was methylated by both NSD-CTDs and was the preferred substrate for NSD1-CTDs.

Taken together, the above results show that the CTD of NSD1, NSD2, and NSD3 can recognize and methylate H3K4, H3K9, H3K27, H3K36, H3K79, and H4K20 in vitro with very different catalytic activities depending on the substrate. indicates Although some HMTases (eg SET9) do not possess significant methyltransferase activity in vitro , they are not members of the NSD for which significant catalytic activity has been identified.

1.9 Identification of open and closed structural and molecular modeling of catalytic domains

The above results indicate that the CTD of NSD1, NSD2, and NSD3 can recognize and methylate multiple histone lysines (Fig. 2). To examine the methylation of histone lysine by NSD members in more detail, the present inventors attempted to access the lysine-binding pocket by structurally modifying the catalytic domain (CD).

The results are shown in FIG. 3 .

Left panel: The structure of the catalytic domain consists of one group of three groups of canonical -sheets and two closely adjacent -sheets with conserved -helices arranged in a triangular fashion, allowing for binding of lysine-histone ligands. form a gap Authorized AdoMet joins the CD into a suitable configuration through the channel. Fluid control loops are marked in red. Movement of the control loop is indicated by a curved inverted arrow.

Right panel: Model of NSD-CDs bound to H3-peptide (a.a. 32-38) after MD simulation. H3K36 was indicated. Electrostatic surfaces were colored as follows: blue: positive charge; red; negative charge. Units +5/-5 kb.T.ec -1.

As shown in FIG. 3 , the crystal structure of NSD1-CD (PDB: 3OOI) has no peptide substrate, and the histone-lysine binding moiety is occluded with a regulatory loop at the interface of the SET and post-SET moieties. The molecular determinant of this regulatory loop extends to the histone binding region and sterically interferes with the binding of other substrates. The H3K4 HMTase MLL1 shares similar properties with the flexible post-SET responsible for the open- and closed structures of the lysine-airbox binding moiety. To construct CDs of NSD2 and NSD3, the crystal structure of NSD1-CD (PDB: 3OOI - aa 1850-2080) was used, which shared 40% or more sequence identity and was performed through homology modeling as it had a similar structure. did.

The pair-wise sequence alignment between the template and the target showed that NSD2-CD (aa 971-1202) shares 75.9% identity and 90.1% similarity with NSD1-CD (PDB: 3001), and , showed that NSD3-CD (aa 1044-1283) shares 72.9% identity and 85% similarity with NSD1-CD. NSD1-CD, NSD2-CD, and NSD3-CD share very high base homology, indicating that they share a very similar structure ( FIG. 1 ).

Then, the present inventors investigated the docking of peptides including histone-lysine (H3K4, H3K9, H3K27, H3K36, H3K79, H4K20) and NSD1-CD, NSD2-CD, NSD3-CD using molecular dynamics (MD). did. The original model was based on the crystal structure of the closed, substrate-free NSD1-CD that had been revealed. Thus, the model of apo CD of NSD1, NSD2, and NSD3 docked to peptides intentionally results in a sterically unstable structure. By performing energy minimization and long-term MD simulations in a water box, we focused on ensuring that there are no steric collisions and that the CDs become open structures that accommodate substrates.

4 is a diagram showing the molecular determinant of H3K4, H3K9 and H3K27 bonds.

Peptides H3K4 (a.a.1-7), H3K9 (a.a. 5-11), and H3K27 (a.a. 23-29) are aligned vertically and atomic details of residues are shown in purple. Interactions of amino acids from CDs of NSD are marked in brown and numbered. The green dotted line represents hydrogen bonding. Hydrophobic interactions were omitted. This drawing was constructed using the program Ligplot.

5 is a diagram showing the determinants of H3K36, H3K79 and H4K20 combinations. Peptides H3K36 (a.a. 32-38), H3K79 (a.a. 75-81), and H4K20 (a.a.16-22) are shown.

Long-term MD simulations were applied to equilibrate the complex to a stable structure with small RMSD oscillations observed in the post-SET portion (overall c-RMSD < 0.6 A). The resulting model may approximate the structure of an open CD resulting from the binding of nucleosome DNA as an allosteric effector. Following the MD simulations, we observe structural deformations on the CD of NSD2 and NSD3, especially on the loop of the interface between the SET and post-SET portions (aa 1178 to 1191 for NSD2-CD, and aa 1260-1273 for NSD3-CD). did. This loop was rotated ~40, and the tip was translated ~6, which is a very significant displacement (Fig. 3). The rest of the backbone did not undergo significant structural transformation with a total c-RMSD < 0.6.

1.10 Molecular details of binding of H3K4, H3K9, H3K27, H3K36, H3K79, and H4K20 peptides.

CDs of NSD members share similar functional domains and also share general structural properties ( FIG. 1A ). The histone tail-defective portion is removed when the regulatory loop of the interface between the SET and post-SET portions is rotated and displaced so that the negatively charged surface regions are exposed so that the positively charged H3 and H4 peptide tails are suitable for docking. Only access is possible (FIG. 3).

To understand the molecular details of substrate recognition by NSD-CDs, we performed long-term MD simulations in a water bath to test the binding properties of H3K4, H3K9, H3K27, H3K36, H3K79, and H4K20 peptides.

Finally, we found that all histone H3 and H4 peptides tested (H3K4 aa1-7; H3K9 aa5-11; H3K27 aa23-29; H3K36 aa32-38; H3K79 aa75-81; H4K20 aa16-22) ) share a similar steric hindrance and have a similar positive charge on the electric field and molecular surface. These properties suggest the preferred binding energies of NSD-CDs (Table 1).

Table 1 is a table showing the binding energies of peptides when they contain lysine-histone substrates on the catalytic domains of NSD1, NSD2, and NSD3. The peptides are: H3K4 (a.a. 1-7); H3K9 (a.a. 5-11); H3K27 (a.a. 23-29); H3K36 (a.a. 32-38); H3K79 (a.a. 75-81); H4K20 (a.a. 16-22). In the docking test, polyE was used as a control test. polyE is a 6-mer poly-glutamic acid peptide.

The total energies of the complexes after MD simulations were very similar between the tested substrates (0.12% standard deviation for NSD1 and NSD2, 0.11% standard deviation for NSD3), indicating that all peptides tested similarly bind to NSD members. (Table 1). In addition, the regulatory loop acts like a safety belt, anchoring a single histone tail to the binding domain, allowing the lysine substrate to extend to the cofactor-binding moiety and catalyze it ( FIG. 3 ). The peptide-binding portion is large enough to accommodate the 7-amino acid peptide. Interestingly, 7-amino acid peptides (including H3K4, H3K9, H3K27, H3K36, H3K79, and H4K20) share very similar steric and electrostatic properties (Figure 1). All H3 and H4 histone peptides tested were very tightly stabilized in the histone-tail binding cleft of NSD1, NSD2, and NSD3 via an extensive network of hydrogen bonds (Figs. 4 & 5).

These observations are consistent with the calculated binding energies, indicating favorable steric and electrostatic compatibility between histone peptides and CDs (Table 1).

Docking of all histone peptides in CD correlates with significantly favorable docking energies, highlighting good substrate compatibility. As a control experiment, we modeled the docking of a highly negatively charged 7-mer polyE (poly-glutamic acid) peptide that induces significant steric hindrance in the protein binding-bag. Docking of polyE peptides is associated with highly dissociated binding energies, highlighting that this indicates poor coordination with any NSD-CDs (Table 1).

Taken together, these results are equally compatible with histone peptides in which histone-binding pockets of NSD members are tested, providing a steric and electrostatic match for the stability of H3 and/or H4 histone tails.

Example 2: Small molecule inhibitor of HMTase, NSD2

We hypothesized that selective inhibition of NSDs could be achieved by small molecules targeting histone-tail sites on conserved SAM binding sites. In order to secure the inhibition method targeting the existing SAM binding pocket, the present inventors screened a molecule capable of binding to the histone-tail binding portion of NSD2-SET.

2.1 Cloning

PCR was performed using a human liver cDNA library (TAKARA, Japan) as a template to amplify the human NSD2 gene (NSD2-SET , 873 bp, 2844-3717 nt; 291 aa, 948-1239 aa). The forward and reverse primers were PK162, 5- GGCAGCCATATG( Nde I)CAGGGGGTCAGAGGGATCGGAAGAG-3 and PK163, 5-GAAGCACTCGAG( Xho I)CTTCGACTGCCTCTTCCCTTCCCC-3, respectively. The PCR amplified NSD2-SET DNA manipulation was digested with Nde I and Xho I and inserted into the multiple cloning site of a protein expression Intein-tagged vector, pTYB2 (New England Biolabs, USA). The sequence was confirmed through sequencing.

2.2 Protein Expression and Purification

with NSD2-SET The Escherichia coli expression strain, BL21, transformed with the pTYB2 plasmid was cultured in LB medium containing 100 g/mL ampicillin. Expression of recombinant NSD2-SET was induced with 250 M IPTG (isopropyl 1-thio-D-galactopyranoside) for 15 4 hours. E. coli cells were recovered and lysed by a freeze-and-thaw method, and buffer A containing 0.1% Triton X-100 and 10 mM PMSF (phenylmethanesulfonylfluoride) [20 mM Tris (pH 8.0), 500 mM NaCl, and 0.1 mM EDTA] followed by sonication on ice 20 times. The cell extract containing the NSD2-SET-intein-chitin-binding domain fusion protein was loaded on an affinity column of chitin beads, washed with 100-column volume of buffer A containing 0.1% Triton X-100, Wash with 20-column volume of buffer A without Triton X-100. To remove the bacterial chaperone bound to the recombinant protein, the recombinant NSD2-SET-bound chitin beads were washed with a 10-bead volume of buffer A containing _{10 mM ATP (adenosine triphosphate) and 2.5 mM MgCl 2 .} NSD2-SET protein was excised from chitin beads by incubation for 4 to 48 hours in buffer A containing 50 mM 2-mercaptoethanol, eluted with buffer A, and concentrated using an Amicon Ultra centrifugal filter. After that, it was used in a methyltransferase assay. A portion of the purified NSD2-SET was analyzed by SDSPAGE. A soluble, pure NSD2-SET with an expected molecular weight of 33.2 KDa was confirmed on a Coomassie-stained gel ( FIG. 6B ). Recombinant NSD2-SET was stable thereafter and maintained catalytic activity when stored at 80 °C.

2.3 Compounds

The compound was purchased from MolPort (http://www.molport.com/) (Table 2), and was prepared with >99.9% purity in DMSO (dimethyl sulfoxide) (D8418 Sigma).

2.4 Histone Methyltransferase Inhibitor Assay

The histone methyltransferase activity of NSD2-SET on H3K36 was measured using a colorimetric quantification kit (Epigentek, USA) according to the manufacturer's protocol. Purified recombinant NSD2-SET (4.0 g, final 2.42 M) was preincubated with recombinant human histone H3.1 (5 M) (New-England Biolabs NEB, USA), in C2 buffer of the assay kit. The concentration of inhibitor/DMSO or DMSO alone (0 mM inhibitor) was measured. The recombinant histones used in this assay did not undergo any potential modifications in production in E. coli and were confirmed using ESI-TOF mass spectrometry according to the manufacturer's (NEB) instructions. Methyl group donor (50 M) was added to the preincubated mixture, and immediately transferred to strip wells and incubated in the dark at room temperature for 120 minutes. Mono-methylated H3K36 (H3K36me1) was captured using anti-mono-H3K36 antibody attached to the bottom of the strip wells. Excess purified NSD2-SET, histone H3, inhibitor, and Adomet were washed away, followed by a labeled detection antibody, followed by a color development reagent to detect the captured antibody-H3K36me1. Strip wells were analyzed at 450 nm with an ELISA plate reader. The level of H3K36me1 is proportional to the intensity of absorbance. Assays were performed independently in triplicate. The results were normalized to the control containing no enzyme.

2.5 Molecular Modeling of the Closed Structure of NSD2-SET

Using homology modeling, the crystal structure of the SET domain of NSD1 (PDB: 3OOI - a.a. 1850-2080) was used as a template for constructing the SET domain of NSD2 (a.a. 971-1202). The sequence of NSD2-SET (a.a. 971-1202) shares 75.9% identity and 90.1% similarity with template NSD1-SET (PDB: 3001). After careful multiple-sequence alignment with ClustalW V2.0.9, 100 models were made using Modeler V9.5. According to the intrinsic modeler function, using the SCWRL4 program, side-chains positioning was performed for the optimal model. Stereochemistry was confirmed using PROCHECK. The model was manually checked using COOT. The model of NSD2-SET had an H-factor of 18.1%. The quality and accuracy of the model set were evaluated as H-factors. This is a new quality metric in homology modeling. Due to the high sequence conservation of the SET domain between NSD1 and NSD2, modeling of NSD2-SET represented an ideal case.

2.6 Molecular Modeling of the Open Structure of NSD2-SET

Histone tail H3K36 (a.a. 32-38) peptides were manually constructed at COOT. Structural overlay between the crystal structures of NSD1-SET (PDB: 3OOI) and the crystal structure of SET8 bound to histone H4 peptide (PDB: 3F9Y) was obtained by placing the H3K36 (aa 32-38) peptide within the histone binding site of the closed structural model of NSD2. It was used for manual positioning and fitting. Structural overlay was performed using the SSM algorithm in COOT. The peptide-NSD2-SET complex was fully solvated in a water bath (140 x 100 x 110) using VMD 1.9. Energy minimization and molecular dynamics (MD) were performed with NAMD v2.8 using a CHARMM force field. The constant temperature was maintained using Langevin damping at 300 K, and the pressure was kept constant at 1 pressure using a Langevin piston. Electrostatic force was considered as 10 cut-off distances using the particle mesh Ewald (PME) method. A total of 100,000 steps of (100 ns) MD and a total of 30,000 steps of (30 ns) energy minimization were performed for the solvated peptide-NSD2-SET complex. After MD simulations, the solvated complexes were manually identified by COOT for ideal stereochemistry. The MD trajectory was analyzed with VMD 1.9. Peptide H3K36 (a.a. 32-38) and water molecules were removed from the peptide-NSD2-SET complex, leaving the open structure of NSD2-SET suitable for hypothetical ligand screening.

2.7 Hypothetical Ligand Screening and Selection of Compounds

AutoDock Vina was used to perform docking with Clean Drug-Like compounds downloaded from ZINC docking database version 11 42. The missing hydrogen atoms on the open NSD2-SET were added using the PDB2PQR program and the CHARMM force field together. The ligand was prepared with AutoDock tools (ADT) v1.5.4 r29. The docking grid size and position were set manually using ADT. The size of the grid (22x20x20) was large enough to cover the open NSD2-SET and was the center of the enzymatic channel identified in NSD1-SET. A fraction of 40,325 molecules (190 Da < MW < 500 Da) was docked, and the top 3,000 results were analyzed and filtered with Cresset's FieldView (Herts, AL7 3AX, United Kingdom). In each molecule, the top 9 binding structures were preserved. The upper docking solution was checked manually using COOT and PyMOL.

2.8 Results

NSD2/MMSET is a major therapeutic target for multiple myeloma. Multiple myeloma is a malignancy with a very low survival rate and no treatment. HMTase inhibitors are very rare and only one NSD2 inhibitor, MCTP39, a SAM competitor has been identified so far. In the present invention, using a structure-based discovery approach, we identified a small molecule that inhibits the H3K36 HMTase activity of NSD2 in vitro.

The structure of NSD2 has not yet been elucidated, and the closest confirmed crystal structure is the apoSET domain of NSD1. The present inventors constructed an NSD2-SET model utilizing the NSD1-SET crystal structure using homology modeling, and confirmed that they share very high sequence identity (75.9%) and similarity (90.1%). Previously, the present inventors studied the motion and role of a regulatory loop located at the interface between the SET and postSET domains in NSD1-SET.

The results are shown in FIG. 6 .

6 is a diagram illustrating a model of an open SET domain including the structure of NSD2/MMSET and LEM-14.

(A) Primary structure of NSD2: PWWP domain; PHD zinc finger domain; SET HMTase (histone methyl transferase), including preSET and postSET. The regulatory loop that closes the histone-binding portion is marked in red.

(B) The result of Coomassie staining of purified recombinant NSD2-SET on an SDS-PAGE gel.

(C) An open NSD2-SET model that is bound to LEM-14 in the most preferred structure through virtual ligand screening. Control loops are marked in red.

(D) A diagram showing the molecular surface of NSD2-SET bound to LEM-14. Electrostatic surfaces were colored as follows:

Blue: positive charge; Red: negative charge, unit +5/-5 kb.T.ec ^-1 .

6 , in the closed structure, binding of the H3 or H4 tails was sterically prevented. Accordingly, NSD2-SET of the open structure was modeled using long-term MD simulation. We inserted a 7-mer H3 peptide (a.a. 32-38) below the regulatory loop at the interface of the SET and postSET sub-domains of the closed structure of NSD2-SET. The NSD2-SET-H3K36 (a.a. 32-38) complex was intentionally rendered sterically unstable. Energy minimization and long-term MD simulations relieved steric constraints and allowed the SET domain to accommodate substrates (Figures 6C and D). MD simulations allowed the composite to stabilize into a stable structure. The regulatory loop of NSD2-SET underwent significant displacement with ~45 rotations and ~6 translations at the ends, which are highly negatively negatived binding grooves suitable for docking of H3 or H4 tails. to open (Fig. 6D). Both NSD1-SET and NSD2-SET have recognition sequences comprising at least 7 amino acids in a portion of the H4K20 HMTase SET8.

BIX-01294 is a histone-tail mimic targeting the SET domain of the G9a-like protein HMTase. By confirming the crystal structure of BIX-01294 bound to the histone tail cleft of GLP, the present inventors designed a virtual ligand screening method focusing on the open structure histone-tail binding pocket of NSD2-SET. The HMTase inhibitors BIX-01294, BIX-01338, and ketocin (Chaetocin) have molecular weights of 492 Da, 535 Da, and 696 Da, respectively ( FIG. 9 ). BIX-01294, BIX-01338 and ketosine have positive electrostatic fields with steric hindrance matching the binding grooves of histone tails. However, molecules above 500 Da have lowered pharmacokinetic parameters. Thus, we screened compounds with molecular weights less than 500 Da, forming a subset of 40,325 molecules for use in docking. 40,325 docking results were sorted according to the optimal docking energy of the optimally docked conformer. The present inventors investigated whether the selected compound has a positive electromagnetic field matching the histone-tail binding groove of NSD2-SET (priority #1), and whether it has a suitable conformational coordination and binding interaction in the SET domain according to a passive check using COOT. Whether or not (priority #2) follows Lipinski's law of 5 (priority #3), total polar surface area < 100 ² (priority #4), molecular refractivity (priority #5) ), a screening filter was applied to the top 3,000 compounds. Molecules with heavy atoms are not considered. Of the 20 selected molecules, 9 molecules were subjected to in vitro assays for H3K36 monomethylation (H3K36me1). The results are shown in Table 2.

The nine selected molecules and BIX-01294 were tested for inhibitory activity of H3K36me1 using recombinant NSD2-SET and recombinant unmodified human histone H3.1. Compounds LEM-06, LEM-14, and BIX-01294 inhibited NSD2-SET with _{IC 50 of 890 M, 11.2 M, and 9.4 M, respectively.} LEM-14 in vitro IC ₅₀ corresponded to _{the IC 50} range of other HMTase inhibitors such as MCTP39, TMDC 1c, AMI-1, AMI-5, AMI-6, AMI-9 and GSK343.

8 is a diagram showing a concentration-response curve of an inhibitor for NSD2-SET H3K36 mono-methylation activity. NSD2-SET H3K36 mono-methylation activity was displayed at various concentrations of candidate inhibitors. The concentration response curves and IC50 of LEM-06, LEM-14, and BIX-01294 were displayed using GraphPad Prism 6.0.

9 is the chemical structure of LEM-06, LEM-07 and BIX-01294.

Weaker stabilization of LEM-06 than LEM-14 and BIX-01294 resulted from LEM-15 and BIX-01294 having 5 hydrogen bonds, whereas LEM-06 had 3 hydrogen bonds in the SET domain. is described as (FIG. 9). It is not clear why other LEMs do not show significant H3K36me1 HMTase inhibitory effects in vitro. However, a possible explanation lies in the molecular refraction of the tested molecules, which effectively competes for binding to histone H3.1. Both LEM-14 and BIX-01294 had the strongest positive electrostatic field compared to other LEMs, which could be related to the strength of the interaction with the negatively charged surface of the histone-tail binding domain (Fig. 6D and Fig. 11). ). BIX-01294 binds to the histone-binding groove of HMTase G9a and inhibits its activity with _{IC 50 of 3M.} In addition, the derivative of BIX-01294 can inhibit the H3K9 Jumonji demethylase family with an IC ₅₀ of 0.05-75M.

Binding of LEM-07 and BIX-01294 instead of histone-tails into the SAM binding pocket is not excluded. Therefore, to evaluate the possibility of binding, we docked LEM-14 and BIX-01294 into the SAM-binding pocket of NSD2-SET. LEM-14 and BIX-01294 bind with docking energies of -9.2 Kcal/mol and -7.6 Kcal/mol, respectively. This is in stark contrast to the more favorable binding of LEM-14 and BIX-01294 to the histone-tail pocket with docking energies of -10.7 Kcal/mol and 9.7 Kcal/mol, respectively (Table 2). In addition, both LEM-14 and BIX-01294 are bulkier than MCTP39 or AdoMet, and sterically constrained binding of electrons into the SAM binding bag of NSDs consumes greater energy and reduces the likelihood of occurrence.

In the present invention, the present inventors developed LEM-07, a novel NSD2 inhibitor for H3K36 methylation, which exhibited an in vitro IC ₅₀ =11.2M. In addition, the present inventors confirmed that BIX-01294 inhibited NSD2 (IC ₅₀ =9.4M) with binding properties similar to those observed in G9a.

Collectively, we found that derivatives of LEM-14 can be used as specific NSD2/MMSET inhibitors suitable for various tumors including multiple myeloma.

Claims (10)

As an inhibitor of histone lysine-methyl transferase, a pharmaceutical composition for the prevention or treatment of cancer comprising the following compound or a pharmaceutically acceptable salt thereof as an active ingredient:

delete The method according to claim 1, wherein the histone lysine-methyl transferase is a human NSD (nuclear receptor binding SET domain) family, the pharmaceutical composition for the prevention or treatment of cancer. The pharmaceutical composition for preventing or treating cancer according to claim 3, wherein the NSD is NSD2. delete delete The method according to claim 1, wherein the cancer is lymphoma (lymphoid malignancies), acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), myeloma (myeloma), skin cancer, primary melanoma (primary) melanomas), Yolk sac tumour, breast cancer, lung cancer, colon cancer, pancreatic cancer, ovarian cancer, bladder cancer, prostate cancer, kidney cancer, brain cancer, cervical cancer, leukemia, granulomatosis, and A pharmaceutical composition for the prevention or treatment of cancer, which is selected from the group consisting of cellular transformation. The pharmaceutical composition for preventing or treating cancer according to claim 1, wherein the cancer is multiple myeloma. As an inhibitor of histone lysine-methyl transferase, a food composition for the prevention or improvement of cancer comprising a compound of the following formula or a pharmaceutically acceptable salt thereof as an active ingredient:

delete

Images