Computer-Aided Discovery of Massonianoside B as a Novel Selective

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

Article

Computer-Aided Discovery of Massonianoside B as a Novel Selective DOT1L Inhibitor Jie Chen, and Hyun-Ju Park ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.8b00933 • Publication Date (Web): 05 Apr 2019 Downloaded from http://pubs.acs.org on April 8, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

1

Computer-Aided Discovery of Massonianoside B as a Novel

2

Selective DOT1L Inhibitor

3

Jie Chen and Hyun-Ju Park*

4

School of Pharmacy, Sungkyunkwan University, Suwon, Gyeonggi-do 16419, South Korea.

5 6

Corresponding author: Hyun-Ju Park

7

Email: [email protected]

8

T: +82-31-290-7759

9

1

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

GRAPHIC ABSTRACT

2

3

2


Page 2 of 29

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

ABSTRACT

2

Protein methyltransferases (PMTs) are involved in numerous biological processes and have been

3

studied as a promising target class in the field of oncology and other diseases. Disruptor of

4

telomeric silencing 1-like (DOT1L), a histone H3 lysine79 (H3K79) methyltransferase, plays an

5

important role in the progressions of mixed-lineage leukemia (MLL)-rearranged leukemias and

6

has been validated as a potential therapeutic target. Here, we report the discovery and

7

characterization of a novel DOT1L inhibitor, Massonianoside B (MA), by pharmacophore-based

8

in silico screening and biological studies. MA is a structurally unique natural product inhibitor of

9

DOT1L with an IC50 value of 399 nM. The compound displays high selectivity for DOT1L over

10

other S-adenosyl-methionine (SAM) dependent PMTs. Treatment of MLL-rearranged leukemia

11

cells with MA gives a dose-dependent reduction in cellular levels of histone lysine79 mono- and

12

di-methylation without affecting the methylation of other histone sites. Moreover, MA

13

selectively inhibits proliferation and causes apoptosis in MLL-rearranged leukemia cells, and

14

down-regulates the expression of MLL-fusion target genes, including HOXA9 and MEIS1.

15

Molecular docking analysis reveals that MA may bind to the SAM-binding site of DOT1L. We

16

identified MA as not only a novel DOT1L inhibitor with anti-leukemic activity, but also a

17

DOT1L-targeted molecular probe, which may serve as a useful chemical tool for investigating

18

the role of DOT1L in biological processes.

19

3



1

INTRODUCTION

2

Mixed lineage leukemia (MLL) is a genetically discrete form of acute leukemia that found in

3

over 70% of infant acute lymphoid and over 35% of infant acute myeloid leukemias.1 The

4

patients with this highly aggressive form of leukemia, have poor prognoses due to the lack of

5

currently accessible therapies. MLL-rearranged leukemia is associated with chromosomal

6

translocations that lead to produce an oncogenic fusion protein consisting of the amino terminal

7

region of MLL and the carboxy terminus of various fusion partners, including members of the

8

AF and ENL family of proteins.2,

9

different mechanisms: some are chromatin modifiers that introduce histone acetylation while the

10

others can interact with histone methyltransferase, disruptor of telomeric silencing 1-like

11

(DOT1L).4

12

DOT1L is the only protein lysine methyltransferase (PKMT) lacking a SET domain, and

13

catalyzes the mono-, di-, and trimethylation of the nucleosome histone H3 lysine 79 (H3K79)

14

which associated with active transcription of certain genes.5-8 Noticeably, DOT1L has been

15

known to be required in MLL fusion-mediated transformation.9-12 The MLL fusion partners in

16

leukemias, such as AF4, AF9, AF10 and ENL, mediate the recruitment DOT1L to specific

17

genomic loci leading to increased H3K79 methylation and transcriptional activation of

18

leukemogenic genes, including HOXA9 and MEIS1.13-20 Thus, DOT1L is necessary for the

19

occurrence and maintenance processes of MLL-rearranged leukemia. Recent studies have

20

demonstrated that inhibition of DOT1L by either genetic or pharmacological methods resulted in

21

selective cytotoxicity against MLL-AF4/9/10 translocation carrying cells.11, 21

3

The fusion partners activate the transcription through two

4


Page 4 of 29

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2

Figure 1. Chemical structures of representative DOT1L inhibitors.

3 4

A great deal of study has been conducted to develop selective DOT1L inhibitors (Figure 1). The

5

first selective inhibitor of DOT1L, EPZ004777, was discovered by scientists from Epizyme Inc.

6

in mid-2011.21 Shortly after, ligand/structure and mechanism-based approaches led to the

7

identification of dozens of DOT1L inhibitors.22-25 Based on their structural features, these

8

compounds can be divided into four classes: (i) S-adenosyl-homocysteine (SAH)-like inhibitors,

9

(ii) mechanism-based inhibitors, (iii) benzimidazole/urea-containing inhibitors, and (iv) non-

10

nucleoside DOT1L inhibitors. These compounds showed remarkable activities against the

11

enzyme, but they all have similar scaffolds, analogous to the cofactor SAM, that contain an

12

adenosine or a deazaadenosine. Among them, EPZ-5676 (pinometostat) has completed phase I

5



1

clinical trials in both adult and pediatric patients with MLL-rearranged leukemia (NCT01684150,

2

NCT02141828).26 In the adult trial, treatment of pinometostat led to emergent resistance due to

3

increased expression of the drug efflux transporter.27 A new group of DOT1L inhibitors which

4

are structurally distinct from SAM-based inhibitors, were identified by using the fragment

5

linking approach and virtual screening, but their poor pharmacokinetic properties precluded their

6

further development.28-31 Therefore, the discovery of DOT1L inhibitors with novel scaffolds is a

7

very attractive issue when investigating drug candidates which should have excellent drug-like

8

properties and lack the characteristics to be substrates for efflux pumps.

9

In the present study, we aimed to find novel non-nucleoside DOT1L inhibitors using

10

computational drug design approaches. Herein, we describe the identification of phenolic

11

glycoside Massonianoside B (MA), a potent, selective, and structurally novel inhibitor of

12

DOT1L, by conducting pharmacophore-based virtual screening of in-house chemical library.

13

DOT1L inhibitory activity of MA is validated by multiple biochemical assays. Moreover, MA is

14

highly selective for DOT1L over other SAM-dependent methyltransferases (SETD7, G9a, EZH2,

15

PRMT1, PRMT3, PRMT5, PRMT7 and CARM1). Treatment of MV4-11 leukemia cells with

16

MA results in decreased cellular levels of H3K79 methylation and downregulation of the

17

expression of MLL target genes, such as HOXA9 and MEIS1. We also showed that MA can

18

selectively repress the proliferation of MLL-rearranged MV4-11 cells by inducing apoptosis.

19

This compound may serve as a novel scaffold for the development of more potent and selective

20

DOT1L inhibitors with therapeutic potential. MA displays promising potency and molecular

21

properties suitable for further optimization to develop novel anticancer agents against MLL-

22

rearranged leukemia.

23

6


Page 6 of 29

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

RESULTS AND DISCUSSION

2

Discovery of MA by Ligand-Based Pharmacophore Screening and Similarity Search

3

Although all protein methyltransferases (PMTs) utilize the same cofactor SAM, the amino acid

4

residues that encompass the SAM-binding site are poorly conserved,32 resulting in a diversity of

5

SAM-binding modes and enzyme structure-based interactions among PMTs. Therefore, an

6

effective general strategy for discovering selective PMT inhibitors is targeting the SAM-binding

7

site. To identify selective non-SAH-like DOT1L inhibitors, we established a pharmacophore

8

model using non-nucleoside SAM-competitive inhibitors of DOT1L. Pharmacophore-based

9

virtual screening of in-house library containing natural product compounds (7,086 entities), was

10

conducted using the Phase program.33, 34 The 3D pharmacophore model of the lead compounds

11

was shown in Figure 2, containing two hydrogen bond acceptors and three aromatic rings. After

12

visual inspection, 35 out of 187 molecules that completely matched the pharmacophore features,

13

were subjected to biological evaluation. A bioluminescence-based S-adenosyl-homocysteine

14

hydrolase (SAHH)-coupled assay, which measures the conversion of the cofactor SAM to the

15

cofactor product SAH, was employed to evaluate the bioactivity of compounds. In the screening,

16

the compound S14, which contains a phenyl-dihydrobenzofuan moiety, showed a 46 %

17

inhibition against DOT1L at 100 μM (Table 1). To find more potential inhibitors, we next

18

performed a similarity-based analogue search using S14 as a template. Ten commercially

19

available analogues were purchased, and their DOT1L inhibition activities were assessed using a

20

bioluminescence-based SAHH-coupled assay. As shown in Table 1, several compounds

21

demonstrated modest activities against DOT1L. Among them, the compound MA exhibited

22

greater than 80% inhibition of DOT1L activity at 100 μM and showed the highest potency with

23

an IC50 value of 8.08 ± 2.63 μM.

7



1 2

Figure 2. (Left) Scheme of ligand-based virtual screening, consisting of pharmacophore-based

3

screening and similarity-based searching. (Top right) The pharmacophore query derived from

4

selected non-nucleoside SAM-competitive inhibitors of DOT1L. The model contains 2 hydrogen

5

bond acceptors (pink) and 3 aromatic centers (orange). (Bottom right) Overlay of the

6

pharmacophore query AARRR with active compound S14.

7 8 9 10 11 12 13

8


Page 8 of 29

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2


Table 1. Structures of the S14 analogues and their inhibitory activity against DOT1L.

3 DOT1L Inhibitory Activity a Compound

R1

R2

R3

R4

Inhibition Ratio% (100 μM)

1

S14 (2S,3R)

H

A

methoxy

hydroxypropyl

46.25 ± 3.60

2

S15 (2R,3S)

H

B

methoxy

hydroxypropyl

10.41 ± 2.46

3

S16 (2S,3R)

H

H

methoxy

hydroxypropyl

NA b

4

S17 (MA) (2S,3R)

A

H

hydroxy

hydroxypropyl

81.16 ± 0.69

5

S18 (2R,3S)

A

H

methoxy

hydroxypropyl

NA

6

S19 (2R,3S)

A

H

-

hydroxypropyl

16.39 ± 0.43

7

S20 (2R,3S)

B

H

hydroxy

hydroxypropyl

78.84 ± 6.98

8

S21 (2R,3S)

B

H

methoxy

hydroxypropyl

15.17 ± 1.20

9

S22 (2S,3R)

H

H

methoxy

C

NA

10

S23 (2R,3S)

Me

H

methoxy

C

NA

EPZ5676

-

-

-

-

-

　

IC50 (μM)

8.08 ± 2.63

82.2 ± 1.09

0.0266 ± 0.0031

4 5 6 7

a

8

MA is a Selective and SAM-competitive Inhibitor of DOT1L

9

The inhibitory activity of MA was further evaluated using a radioisotope-based filter-binding

10

assay.35 SAH, which has broad activity against all PMTs, was used as a positive control. MA

11

exhibits ~four-fold higher potency at inhibiting DOT1L (IC50 = 0.40 ± 0.05 μM) than SAH (IC50

12

= 1.54 ± 0.21 μM) (Figure 3B). We also employed multiple approaches to test whether MA is an

13

authentic inhibitor of DOT1L. To rule out the possibility that MA is a promiscuous inhibitor

14

which forms an aggregate, we conducted a detergent-based assay.35 Addition of nonionic

15

detergent (0.01% (v/v) Triton-X 100) had no effect on the inhibitory activity of MA

b

DOT1L inhibitory activity is determined by bioluminescence-based SAHH-coupled assay. No activity.

9



1

(Supplementary Figure S1), demonstrating that MA is not an aggregator. We next investigated

2

whether MA acts as a covalent inactivator. The activity of DOT1L was recovered after buffer

3

exchange (Supplementary Figure S2), indicating that MA does not covalently inactivate the

4

enzyme, and that it is a reversible inhibitor of DOT1L.

5

To further examine the biochemical mechanism of the DOT1L inhibition by MA, we measured

6

the apparent IC50 values of MA as a function of the cofactor SAM concentrations using

7

bioluminescence-based DOT1L inhibition assays. As illustrated in Figure 3C, the IC50 values of

8

MA increases in proportion to SAM concentrations, indicating that the drug binds to the DOT1L

9

enzyme competitively with cofactor SAM. We also examined the inhibition profile of MA

10

against several representative human PMTs, including three SET-domain PKMTs as well as five

11

protein arginine methyltransferases (Table 2). Notably, MA is not active against protein lysine

12

methyltransferase G9a and protein arginine methyltransferases (PRMTs) PRMT1, PRMT3,

13

CARM1, PRMT5 and PRMT7. MA has weak but measurable activity against SETD7 (H3K4;

14

IC50 = 40.1 μM) and EZH2 (H3K27; IC50 = 50.4 μM). Nevertheless, MA exhibits outstanding

15

selectivity for DOT1L over the other PMTs that were tested (>2000-fold for PKMTs; >5000-fold

16

for PRMTs). The metabolic stability of MA was also evaluated in human and mouse liver

17

microsomes. MA shows a very high metabolic stability in both human and mouse liver

18

microsomes, with CLint values of < 4 μL/min/mg protein (Supplementary Table S2).

19

10


Page 10 of 29

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2

Figure 3. MA is a potent SAM-competitive inhibitor of DOT1L. (A) Chemical structure of MA.

3

(B) DOT1L Inhibition activity of MA. The concentration-dependent DOT1L inhibition by SAH

4

(●) and MA (■) was investigated using radioisotope-based filter-binding assays containing 1μM

5

SAM. Fitting the data gives an IC50 of 1.58 μM for SAH and 0.40 μM for MA. (C) MA is

6

competitive with the cofactor SAM, since the IC50 values increased in proportional to the SAM

7

concentrations. Assays contained 60 nM human recombinant DOT1L, 0.08 mg/ml HeLa

8

oligonucleosomes, and SAM was titrated in a concentration range between 0.125 - 10 μM.

9 10 11

12 13 14 15

Table 2. Selectivity of MA for DOT1L over other human PMTs. Enzyme (Methylation Site)

IC50(M)a

Fold Selectivity b

DOT1L (H3K79)

2.31E-08

1

SETD7 (H3K4)

4.01E-05

1735

G9a (H3K9)

>2.00E-04

>8000

EZH2 (H3K27) c

5.04E-05

2181

PRMT1

>1.00E-04

>5000

PRMT3

>2.00E-04

>8000

CARM1

>1.00E-04

>5000

PRMT5d

>2.00E-04

>8000

PRMT7

>1.00E-04

>5000

a

IC50 is determined by Reaction Biology Corporation using HotSpot PMT activity assay. Fold selectivity is calculated as the ratio of the IC50 for the enzyme under study over the IC50 for DOT1L. c Measured using PRC2 multiprotein complex. d Measured using PRMT5/MEP50 complex. b

11



1

Binding Mode Analysis

2

To explore the molecular basis of interaction between MA and DOT1L, we examined a possible

3

binding mode of MA by docking modeling using Surflex-Dock.36 It has been previously reported

4

that docking modeling using the X-ray structure of DOT1L:SAM complex failed in predicting

5

the binding mode of DOT1L inhibitors, because the flexible activation loop of DOT1L

6

undergoes large conformational changes. Considering the integral nature of the activation loop,

7

the crystal structure of the urea-containing inhibitor FED2:DOT1L complex (PDB ID:4EQZ)25

8

was selected as the receptor for docking in our study. As illustrated in Figure 4, MA extends

9

deeply into a channel-like cofactor binding pocket, but leaves the adenine binding site

10

unoccupied. The dihydrobenzofuran core of MA occupies the binding site where the isopropyl

11

ammonium substituted urea linker of co-crystalized ligand FED2 binds. Previous work indicated

12

that the urea linker is critical for the DOT1L inhibition potency, and its functionality is well

13

recognized by key residues, such as Asp161 and Asn241, forming hydrogen bond interactions in

14

the X-ray structure.25 Similar to the intermolecular interactions revealed in the FED2:DOT1L

15

complex, the oxygen atom in the dihydrobenzofuran ring of MA acts as a hydrogen bond

16

acceptor interacting with the backbone of Asn241. In addition, the 3,7-substituted

17

hydroxymethyl group and hydroxyl group serve as hydrogen donors for the backbone carbonyl

18

oxygen of Phe131 and Lue162, respectively. Another hydrogen bond is formed between the

19

hydroxypropyl OH and the backbone oxygen of Tyr128. Additionally, the oxygen atom of

20

methoxyphenyl moiety involves in a hydrogen bond with sidechain of Ser140. The α-L-

21

rhamnose (Rha) end of MA is ideally positioned into a narrow internal cavity, contributing

22

significantly to binding by forming multiple hydrogens bonds with the Glu134, Val240, Val267,

23

Ser268, and Tyr312 residues. The interactions between MA and DOT1L are similar to those of

12


Page 12 of 29

Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

the FED2:DOT1L complex in the inhibitor-bound DOT1L structure, suggesting that MA can fit

2

into the cofactor SAM binding site of DOT1L.

3

Subsequently, we investigated the preliminary structure-activity relationships (SAR) of MA and

4

its analogues (Table 1). All the compounds share the same scaffold of 2-phenyl-2,3-

5

dihydrobenzofuran substituted with different groups. Replacing the hydroxyl group with a

6

methoxy group in R3 (Table 1, S18) or removal of hydroxyl group (Table 1, S19) reduced

7

DOT1L inhibition potency, demonstrating that the hydrogen donor in R3 is crucial for inhibition

8

activity. This observation is highly consistent with our docking simulation result where the 7-

9

substituted hydroxyl group in MA serves as a hydrogen donor for the backbone carbonyl oxygen

10

of the key residue Lue162. Another important finding is that the sugar moiety in R1 appears to

11

be indispensable for DOT1L inhibition. The analogues that lack sugar groups lost their activity

12

against DOT1L (S16, S22 and S23). Notably, compound S20 with β-D-glucopyranose (Glc)

13

moiety in R1 exhibited much lower potency (a 10-fold decrease), suggesting that the bulkier

14

substitution in R1 is not favorable for binding in the active site. The predicted binding mode of

15

S20 supports our experimental findings: the decrease in activity may be caused by steric effects

16

interfering with fitting of the molecule into the narrow internal cavity of DOT1L (Figure 4C). In

17

summary, a 7-substituted hydroxyl group at dihydrobezofuran ring and a less bulky sugar group

18

at the benzene ring are essential for inhibition of DOT1L.

19 20 21 22 23

13



1 2 3

Figure 4. Putative binding mode of MA in the cofactor SAM binding site of DOT1L (PDB

4

ID:4EQZ). (A) A close-up view of binding mode of MA (cyan) in the SAM binding pocket (grey

5

ribbon). Key residues are shown as grey sticks and labeled. Hydrogen bonds are depicted as

6

yellow dash lines. (B) Binding mode of MA (cyan) aligned with the co-crystalized ligand FED2

7

(violet). (C) MA (cyan) or S20 (orange) is aligned with FED2 (violet) in the cofactor binding site

8

of DOT1L.

9

14


Page 14 of 29

Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

MA Selectively Inhibits Cellular H3K79 Methylation and Suppresses MLL Fusion Target

2

Gene Expression

3

We next examined the efficiency of MA on DOT1L cellular function in MLL cell lines. After

4

treating human biphenotypic leukemia cells (MV4-11, MLL-AF4) with inhibitors for 4 days,

5

immunoblot analyses of extracted histones were conducted. For comparison, the clinical

6

candidate DOT1L inhibitor EPZ5676 was also evaluated. Treatment of MV4-11 cells with 1 μM

7

EPZ5676 or 2 μM MA significantly reduced H3K79me2 mark (Figure 5C). Moreover, MA

8

showed dose-dependent decreases in global H3K79me2 levels in MV4-11 MLL leukemia cells.

9

The result was further quantified by measuring the relative ratios of H3K79me2 levels versus

10

internal control histone H3. The cellular level of H3K79Me2 was reduced by about 94 % after

11

being treated with 10 μM of MA (Figure 5C). Histone H3K79me1 states were also examined and

12

found to be reduced by MA treatment (Figure 5B). Considering that there is no known histone

13

demethylase enzyme specific for H3K79 available and DOT1L is the only HMT capable of

14

methylating H3K79, the decrease in methylation at H3K79 is most likely related to DOT1L

15

inhibition.

16

To test the specificity of MA inhibitory activity in cells, immunoblot analysis were also

17

conducted using a panel of methyl-lysine residue specific antibodies. With the exception of

18

H3K79me1 and H3K79me2, the methylation state of other lysine residues (H3K9 and H3K27)

19

was not affected by MA treatment, indicating that MA is a highly specific DOT1L inhibitor at

20

the cellular level.

21

DOT1L can be recruited to a multi-protein complex by aberrant MLL fusion proteins, resulting

22

in the hypermethylation of histone H3K79 at MLL target genes, and ultimately, enhanced the

23

expression of critical genes involved in leukemogenesis, such as HOXA9 and MEIS1.

15



1

Furthermore, both genes are downregulated by DOT1L depletion in MLL-r leukemia cells. 10, 12,

2

16, 18, 21

3

The qRT-PCR analysis was performed to measure mRNA transcript abundance for the MLL

4

fusion target genes HOXA9 and MEIS1. As illustrated in Figure 5D, after a seven-day treatment

5

of either EPZ5676 or MA, the expression of both HOXA9 and MEIS1 genes was decreased.

6

Thus, the altering the expression of MLL target genes in MLL-rearranged cells by MA treatment

7

further ensured the on-target DOT1L inhibition effect of MA.

We then examined the gene expression consequences of MA treatment in MV4-11 cells.

8

9 10

Figure 5. MA selectively inhibits cellular H3K79 methylation and MLL fusion target gene

11

expression in MLL-rearranged leukemia cells. (A) Selectivity of MA over a panel of human PMT

12

enzymes. (B) Immunoblot analysis of MA (10 μM) treated MV4-11 cells with a panel of methyl-

16


Page 16 of 29

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

lysine specific antibodies. (C)Immunoblot analysis of H3K79me2 levels after treatment of MV4-

2

11 cells with EPZ5676 or MA at indicated concentrations. (D) The qRT-PCR analysis was

3

performed to examine the mRNA levels of HOXA9 and MEIS1 in MV4-11 cells over 7-day

4

treatment with EPZ5676 (10 μM) or MA (15 μM). Relative mRNA expression levels are

5

normalized to DMSO-treated control cells. *P

Computer-Aided Discovery of Massonianoside B as a Novel Selective

Recommend Documents