Design of a fluorescent ligand targeting the S-adenosylmethionine binding site of the histone methyltransferase MLL1

Article abstract of DOI:10.1039/c5ob01794g

The histone methyltransferase MLL1 has been linked to translocation-associated gene fusion in childhood leukemias and is an attractive drug target. High-throughput biochemical analysis of MLL1 methyltransferase activity requires the production of at least a trimeric complex of MLL1, RbBP5 and WDR5 to elicit robust activity. Production of trimeric and higher order MLL1 complexes in the quantities and reproducibility required for high-throughput screening presents a significant impediment to MLL1 drug discovery efforts. We present here a small molecule fluorescent ligand (FL-NAH, 6) that is able to bind to the S-adenosylmethionine (SAM) binding site of MLL1 in a manner independent of the associated complex members. We have used FL-NAH to develop a fluorescence polarization-based SAM displacement assay in a 384-well format targeting the MLL1 SET domain in the absence of associated complex members. FL-NAH competes with SAM and is displaced from the MLL1 SET domain by other SAM-binding site ligands with K_disp values similar to the higher-order complexes, but is unaffected by the H3 peptide substrate. This assay enables screening for SAM-competitive MLL1 inhibitors without requiring the use of trimeric or higher order MLL1 complexes, significantly reducing screening time and cost.

Full text of DOI:10.1039/c5ob01794g

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Design of a fluorescent ligand targeting the
S-adenosylmethionine binding site of the histone
methyltransferase MLL1†
Cite this: DOI: 10.1039/c5ob01794g
Yepeng Luan,‡a Levi L. Blazer,‡b Hao Hu,a Taraneh Hajian,b Jing Zhang,a Hong Wu,b
Scott Houliston,b Cheryl H. Arrowsmith,b,c Masoud Vedadi*b,d and
Yujun George Zheng*a
The histone methyltransferase MLL1 has been linked to translocation-associated gene fusion in childhood
leukemias and is an attractive drug target. High-throughput biochemical analysis of MLL1 methyl-
transferase activity requires the production of at least a trimeric complex of MLL1, RbBP5 and WDR5 to
elicit robust activity. Production of trimeric and higher order MLL1 complexes in the quantities and re-
producibility required for high-throughput screening presents a significant impediment to MLL1 drug dis-
covery efforts. We present here a small molecule fluorescent ligand (FL-NAH, 6) that is able to bind to the
S-adenosylmethionine (SAM) binding site of MLL1 in a manner independent of the associated complex
members. We have used FL-NAH to develop a fluorescence polarization-based SAM displacement assay
in a 384-well format targeting the MLL1 SET domain in the absence of associated complex members.
FL-NAH competes with SAM and is displaced from the MLL1 SET domain by other SAM-binding site
ligands with Kdisp values similar to the higher-order complexes, but is unaffected by the H3 peptide
substrate. This assay enables screening for SAM-competitive MLL1 inhibitors without requiring the use of
trimeric or higher order MLL1 complexes, significantly reducing screening time and cost.
Received 26th August 2015,
Accepted 29th October 2015
DOI: 10.1039/c5ob01794g
www.rsc.org/obc
members of the SET1 family. MLL1 facilitates transcription by
increasing H3K4 methylation at the promoters, transcriptional
Introduction
The SET1 family of lysine methyltransferases is comprised of start sites, and 5′ transcribed regions of actively transcribed
six members in mammals (SET1A, SET1B, and MLL1-4) that genes.8 This stimulatory activity is essential for development
perform the majority of H3K4 mono-, di-, and trimethylation and expansion of the hematopoietic system by regulating the
in vivo.1–3 The SET1 family members require at least two expression of Meis and Hox cluster genes.1,6,8
accessory proteins (WDR5 and RbBP5) for efficient methyl-
MLL1 dysregulation has been strongly linked to a variety of
transferase activity.4–7 Formation of higher-order complexes by human malignancies; indeed, MLL1-rearrangements have
the sequential addition of ASH2L and DPY-30 further enhance been detected in approximately 5% of acute lymphoblastic
the enzymatic competency of these proteins.6 Due to its pro- leukemia (ALL) and 5–10% of acute myeloid leukemia (AML)
nounced oncogenic potential, MLL1 (mixed-lineage leukemia cases in adults as well as in more than 70% of infant ALL and
1) has become one of the most thoroughly characterized 35–50% of infant AML patients (reviewed by Chen and
Armstrong).9 These translocations generate a fusion protein of
the N-terminal region of MLL1 with one of the several tran-
scriptional elongators, including AF4, AF9, ENL, and ELL.9,10
The oncogenic potential of these fusion proteins stems from
the recruitment of the fusion protein to the super elongation
complex leading to the dysregulated expression of normal MLL
targets, including MEIS1 and the Hox cluster genes.8–10
aDepartment of Pharmaceutical and Biomedical Sciences, University of Georgia,
Athens, Georgia 30602, USA. E-mail: yzheng@uga.edu; Tel: +1 (706) 542-0277;
Fax: +1 (706) 542-5358
bStructural Genomics Consortium, University of Toronto, Toronto, Ontario M5G 1L7,
Canada
cPrincess Margaret Cancer Centre and Department of Medical Biophysics,
University of Toronto, Toronto, ON M5G 2M9, Canada
dDepartment of Pharmacology and Toxicology, University of Toronto, Toronto,
Several attempts to inhibit the function of MLL fusion
proteins with small molecules have been reported in the
literature.11–15 However, a wild-type allele of MLL1 is required
for the transformative capacity of the oncogenic fusion pro-
teins,16 leading us and others to target the catalytic activity of
ON M5S 1A8, Canada. E-mail: m.vedadi@utoronto.ca; Tel: +(416) 432-1980
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5ob01794g
‡These authors contributed equally.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
wild-type MLL1.17,18 Significant progress has been made in
both of these areas, including targeting the MLL-menin and
MLL-WDR5 interactions.11–15,17–19 Recently OICR-9429,
a
WDR5 antagonist has been reported to dissociate the MLL1–
WDR5–RbBP5 complex, selectively inhibit proliferation, and
induce differentiation in p30-expressing human AML cells.20
To date, no compounds that directly target the SAM- or
substrate-binding sites of the MLL1 SET domain have been
reported.
While methods for generating SET1 family complexes have
been established in our lab, it is rather time consuming and
costly to produce the quantities of high-quality protein com-
plexes needed to perform high-throughput screening (HTS).
Keeping the MLL1 complex fully stable and active for duration
of the screening experiments may also be challenging. To this
end, we have developed a simple, HTS-amenable assay that
does not require generation of high-order MLL1 complexes.
This assay is based upon a small molecule fluorescent ligand
Fig. 1 Chemical structure of the fluorescent probe FL-NAH.
that specifically binds to the SAM-binding site of the MLL1 ligands with a fluorophore tethered to the junction between
SET domain. This ligand binds in a similar manner to the iso- the adenosyl and the 2-aminobutanoate moieties of aza-
lated MLL1 SET domain as it does to the higher order MLL1 adenosylhomocysteine.
complexes and is displaced by SAM analogs with similar Kdisp
The synthetic route is shown in Scheme 1. Compound 1
values in the isolated MLL1 domain and the higher-order was obtained in high yield by condensation between the com-
complexes. This compound represents a valuable tool for the mercially available 5-FAM and the 2-azido-ethanamine with
development of SAM-binding site-specific inhibitors of MLL1.
the coupling reagents HOBt and EDCI in anhydrous DCM. The
synthesis of compounds 2 to 5 started from 2′,3′-O-isopropyl-
ideneadenosine and followed the modified procedures
reported in the literature.24,25 The sequential azidation and
reduction of 2′,3′-O-isopropylideneadenosine yielded com-
pound 3. Then compound 3 was condensed with tert-butyl
Results and discussion
Design and synthesis of FL-NAH
The histone methyltransferase MLL1 is involved in trans- (S)-2-((tert-butoxycarbonyl)amino)-4-oxobutanoate in a reduc-
location-associated gene fusion in childhood leukemia and is tive environment to give the intermediate compound 4 which
a promising drug target. Biochemical screening for MLL1 was then alkylated with propargyl bromide to afford com-
inhibitors typically requires the production of at least a tri- pound 5. The final click reaction was performed with the
meric complex of MLL1, RbBP5 and WDR5 to elicit robust regular catalyst of copper(II) sulfate and sodium ascorbate in
activity. Production of trimeric and higher order MLL1 the mixture of methanol and ddH2O to yield the final product
complexes in the quantities and reproducibility required for FL-NAH (6).
high-throughput screening presents a significant challenge to
Measurement of MLL1-FL-NAH binding affinity. FL-NAH is
MLL1 drug discovery efforts. Therefore, we have developed a bound in a saturable manner to the SET domain of MLL1 in
fluorescent ligand that binds to the SAM-binding site of MLL1. the context of the enzymatically active trimeric, tetrameric,
We have used this ligand to develop a miniaturized biochemical and pentameric complexes as well as in the absence of these
assay to identify and characterize MLL1 SET-domain inhibitors.
accessory proteins (Fig. 2, ESI Fig. 1,† Table 1). In all cases the
The fluorescent ligand containing fluorescein linked to aza- maximal FP shift observed with saturating concentrations of
adenosylhomocysteine, namely fluorescein N-adenosylhomo- proteins was similar. FL-NAH bound the isolated MLL1 SET
cysteine (FL-NAH), is built based on the backbone structure of domain with an affinity similar to that of the enzymatically
the AdoMet cofactor (Fig. 1). It contains a fluorescein fluoro- competent complexes, suggesting that the SET domain of
phore linked to the cofactor mimetic aza-adenosylhomo- MLL1 possesses a structured SAM-binding site even in the
cysteine (NAH) via a triazole linker. NAH possesses all the absence of associated complex members (Fig. 2, Table 1). This
essential binding elements (e.g. electrostatic, H-bonds, and is consistent with the crystal structure of the isolated MLL1
π–π stacking) of the native cofactor required for strong inter- SET domain26 (PDB ID: 2W5Y), where the SAM-binding site is
actions with the active site of MLL1. We envisaged that the well-ordered in the absence of the associated complex
rigid triazole linker could also contribute to the ligand inter- members. Furthermore, increased catalytic efficiency observed
action with the enzyme. Previously, fluorescently labeled in the presence of higher-order MLL1 complexes can be
AdoMet analogs have been made with a fluorophore attached explained by a rearrangement of the I-SET domain that orients
to the N6 position of the adenyl ring or the ribose the substrate lysine side chain in a more stable conformation,
hydroxyls.21–23 However, there are no reports that describe suggesting that increased affinity for SAM is not necessarily a
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Synthesis of FL-NAH. (a) HOBt, EDCI, TEA, DMF; (b) DPPA, DBU, NaN3, 15-crown-5, under nitrogen, 1,4-dioxane; (c) hydrogen, Pd/C in
MeOH; (d) Na(AcO)3BH, DCE; (e) propargyl bromide, K2CO3, DMF, r.t.; (f) CuSO4, sodium ascorbate, H2O and MeOH; (g) TFA and water.
Table 1 Binding affinities and ΔFPmax of FL-NAH binding to the MLL1
SET domain. The isolated MLL1 SET domain (SET only) and in the
context of various complexes were allowed to bind to FL-NAH as
described in Materials and methods. Saturation data were fitted using
non-linear least squares regression analysis. Data are presented as the
mean SD from at least three independent experiments
Complex
Kd (µM)
ΔFPmax
240
260 40
300 30
270 20
SET only
Trimeric
Tetrameric
Pentameric
1
7
0.1
3
3
1.4 0.4
0.2 0.03
Fig. 2 Saturation of the isolated MLL1 SET domain with FL-NAH. The
isolated SET domain of MLL1 binds to FL-NAH with an affinity of 1 0.1
µM. FL-NAH (50 nM) was incubated with increasing concentrations of
the MLL1 SET domain. Fluorescence polarization was measured after
30 minutes incubation. Data are presented as the mean SD from three
independent experiments.
binding site ligands to displace FL-NAH. MLL1 complexes and
the isolated SET domain were diluted to their respective Kd
concentrations and incubated with 50 nM FL-NAH in the pres-
ence of increasing concentrations of SAM, the reaction product
SAH, or the fungal SAM-binding site inhibitor sinefungin
requisite component of the higher-order MLL1 complexes.26
Binding of FL-NAH to the MLL1 SET domain alone was also (Fig. 3A–C, Table 2) and the Kdisp was determined by fitting the
confirmed by saturation-transfer difference (STD) NMR (ESI data to eqn (2) using non-linear least squares regression in
GraphPad Prism v.6.05. A peptide corresponding to residues
Fig. 2†) and differential static light scattering (ESI Fig. 3†).
FL-NAH displacement by SAM-binding site ligands. To 1 to 25 of histone H3 (MLL1 substrate) was used as a control
confirm that FL-NAH binds specifically to the SAM-binding which did not displace FL-NAH (Fig. 3D). The reaction product
SAH is a particularly well-suited inhibitor for in vitro analysis
site of MLL1, we tested the ability of several known SAM-
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 Displacement of FL-NAH by various compounds. FL-NAH is displaced from the isolated SET domain by (A) SAM with a Kdisp of 14 2 µM, (B)
SAH with a Kdisp of 4 0.5 µM, (C) sinefungin with a Kdisp of 9 0.2 µM, but not by (D) the substrate peptide H3 (1–25). Data are presented as the
mean SD from three independent experiments.
Table 2 Displacement of FL-NAH from the MLL1 SAM-binding site. Displacement of FL-NAH from the SAM-binding site of the isolated MLL1 SET
domain or in various MLL1 complexes by SAH, SAM or sinefungin was measured as described in Materials and methods. Data are presented as the
mean SD from at least three independent experiments
SAH
SAM
Sinefungin
Kdisp
(µM)
Hill
slope
Kdisp
(µM)
Hill
slope
Kdisp
(µM)
Hill
slope
Complex
SET only
Trimeric
Tetrameric
Pentameric
4
2
0.5
0.2
−0.9 0.2
−1.5 0.3
−1.9 0.1
−1.4 0.3
14
7
1
2
1
0.1
0.2
−1.2 0.1
−1.6 0.4
−1.7 0.2
−1.6 0.3
9
0.2
10 2.5
0.2
0.5 0.1
−0.9 0.1
−1.1 0.3
−1.2 0.1
−1.1 0.3
0.3 0.1
0.2 0.02
2
1
of SAM-dependent methyltransferases. Indeed, in this assay
Amenability of the assay to screening in a 384-well format.
format, SAH displaces FL-NAH from the isolated SET domain The isolated SET domain was used to develop a miniaturized
of MLL1 with a Kdisp of 4 0.5 µM (Fig. 3B, Table 2). The FP binding assay. The Kd concentration of the MLL1 isolated
potency of FL-NAH displacement by SAH increases as more SET domain (1 µM) was chosen to balance the assay signal-to-
complex members are added to the MLL SET domain, result- noise ratio with the sensitivity of the assay to detect inhibitors.
ing in a 20-fold increase in potency for the pentameric The assay was miniaturized to a 384-well format in a total
complex relative to the isolated SET domain. Similar increases volume of 10 µl and formatted in a mix-and-read manner com-
in potency were observed for SAM and sinefungin, suggesting prising two additions to the well; the first addition being a
that this is a general effect for this protein complex. This solution containing MLL1 and FL-NAH, and the second being
increase in potency presumably stems from a better packing of the addition of the test compound or appropriate control. In
SAM-binding sites in the higher order complexes leading this format, the assay is stable for several hours at ambient
towards higher affinity for SAM and its analogues.
temperature (∼23 °C, data not shown) and is insensitive to the
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 FL-NAH displacement assay in a 384-well format. (A) FL-NAH binding to MLL1 is stable in the presence of at least 10% DMSO. Data are pre-
sented as the mean SD from three independent experiments. (B) The optimized 384-well format provides a Z’ of 0.7 when 250 µM SAH is used as
■
a positive control. Data are presented from 192 wells each of positive (Δ; 250 µM SAH) and negative ( ; vehicle) controls. Dashed lines represent
3
standard deviations from the mean value of each control.
concentrations of DMSO (Fig. 4A) that are essential for com- gradient elution (B% from 5% to 60%, 0 min to 51 min,
pound screening. The assay is robust and amenable to HTS, 4 mL min−1). The structures of all the intermediates and final
routinely providing Z′ values of ≥0.7 when a saturating concen- product were confirmed by the mass spectrum and NMR. The
tration of SAH is used as a positive control (Fig. 4B).
purity of the final product was analyzed by analytical HPLC
(Waters Nova-Park C18, 60 Å, 4 μm 150 × 3.9 mm column,
1 mL min−1, under 214 nm, B% = 30% for 33 min). The spec-
tral data of compounds 1–5 in this manuscript are consistent
with the literature reports.24,25,27
Conclusion
We have described the synthesis and characterization of
FL-NAH, a SAM-binding site probe for MLL1 that does not
require accessory complex members. We have used FL-NAH to
develop an FP-based screening assay that will facilitate
efficient high-throughput screening for the discovery of MLL1
SAM-binding site ligands without the need for the production
of multi-protein complexes. This assay is robust and is inhib-
ited by known SAM-binding site ligands with Kdisp values that
are similar to the higher-order complexes currently being used
for chemical biology studies. While compounds discovered
using this assay format will require follow-up studies using
more physiologically relevant MLL1 complexes, it provides a
simple method by which one can quickly screen large com-
pound libraries for SAM-binding site ligands of the MLL1 SET
domain.
Synthesis procedures and compound characterization
5-(2′-Azidoethylamino)carbonylfluorescein (1).27 A mixture
of 5-FAM (51.6 mg, 0.137 mmol, 1.0 eq.), hydroxybenzotriazole
(HOBt, 26 mg, 0.192 mmol, 1.5 eq.) and 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide (EDCI, 36.8 mg, 0.192 mmol,
1.5 eq.) was dissolved in 5 mL anhydrous DMF under nitrogen
protection. DIPEA (133 μL, 0.768 mmol, 3.0 eq.) was injected
into the reaction solution. The reaction mixture was stirred at
room temperature for 1 h. Then, the 2-azido-ethylamine
(32.3 mg, 0.376 mmol, 1.5 eq.) dissolved in 500 μL anhydrous
DMF was injected into the mixture. The reaction was contin-
ued for 24 h at room temperature in the dark. The DMF was
evaporated and the crude product was purified by column
chromatography (DCM : MeOH = 10 : 1) to give pure compound
1 as a yellow powder (18 mg, 35%). H+ESI = 445.1. 1H-NMR
(400 MHz, methanol-D4) δ = 3.55 (t, 2H, J = 5.76 Hz), 3.64
(t, 2H, J = 5.76 Hz), 6.69 (d, 2H, J = 8.40 Hz), 6.79–6.84 (m, 4H),
Materials and methods
Reagents and general procedures
7.38 (d, 1H, J = 8.00 Hz), 8.23 (dd, 1H, J1 = 8.04 Hz, J2
1.04 Hz), 8.53 (s, 1H).
=
5-Carboxyfluorescein (5-FAM) of high purity was purchased
from ChemPep Inc. All other starting materials were purchased furo[3,4-d]-1,3-dioxol-4-yl)-9H-purin-6-ylamine
9-((3aR,4R,6R,6aR)-6-Azidomethyl-2,2-dimethyl-tetrahydro-
(2).24,25 To
from Sigma-Aldrich. The solvents for chemical synthesis and 3.25 mmol of 2′,3′-O-isopropylideneadenosine (1.00 g, 1.0 eq.,
column chromatography purification were all analytical grade. 3.25 mmol) suspended in anhydrous 1,4-dioxane (10 mL)
The intermediates (compounds 1 to 5) were purified by silica were added 6.51 mmol of diphenylphosphoryl azide (DPPA,
gel (200–300 mesh) column chromatography. The final 1.40 mL, 2.00 eq., 6.51 mmol) and 9.77 mmol of DBU
product (compound 6) was purified by preparative HPLC (YMC (1.46 mL, 3.00 eq., 9.77 mmol) at room temperature under N2.
Pack Pro C18 AS12S05-1520WT, 120 Å, 5 µm, 150 × 20 mm) The solution was stirred for 16 hours after which 32.5 mmol of
with water (HPLC grade, +0.1% TFA) (A) and acetonitrile NaN3 (2.11 g, 10 eq.) and 65 µmol of 15-crown-5 (0.1 eq.) were
(HPLC grade, +0.1% TFA) (B) under 214 nm and 260 nm using added and the reaction mixture was heated to reflux. After
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
4 hours the solid was removed by filtration. The solvent was
FL-NAH (fluorescein linked with N-adenosylhomocysteine,
evaporated and the crude product was purified by column 6). Compound 5 (10 mg, 0.01664 mmol, 1.0 eq.) was dissolved
chromatography (ethyl acetate/petroleum ether 1 : 1, to in 3 mL methanol. Compound 1 (15 mg, 0.025 mmol, 1.5 eq.)
ethanol/ethyl acetate 1 : 5, silica). After drying under high was added to the solution. Under stirring, an aqueous solution
vacuum, a colorless solid (compound 2) was obtained (1.08 g, of sodium ascorbate (0.1664 mmol, 10.0 eq.) was added
1
100%). H+ESI = 333.1. H-NMR (400 MHz, CDCl3) δ = 1.38 (s, followed by an aqueous solution of copper(II) sulfate
3H), 1.60 (s, 3H), 3.57 (m, 2H), 4.37 (m, 1H), 5.05 (dd, 1H, J1 = (0.01664 mmol, 1.0 eq.). The mixture was stirred in the dark at
5.08 Hz, J2 = 2.72 Hz), 5.45 (dd, 1H, J1 = 5.12 Hz, J2 = 1.72 Hz), room temperature for 24 h. All liquid was removed by evapo-
6.10 (d, 1H, J = 1.76 Hz), 6.28 (s, 2H), 7.91 (s, 1H), 8.33 (s, 1H).
ration under strong vacuum. The residue was dissolved in
9-((3aR,4R,6R,6aR)-6-Aminomethyl-2,2-dimethyl-tetrahydro- trifluoroacetic acid (TFA, 5 mL) and was stirred in an ice bath
furo[3,4-d]-1,3-dioxol-4-yl)-9H-purin-6-ylamine (3).24,25 74 mg for 1 h. Then double deionized (dd) water (1 mL) was added to
of compound 2 was dissolved in 10 mL methanol, then 25 mg the solution. Stirring continued overnight at room tempera-
Pd/C was added. The mixture was stirred under a hydrogen ture. TFA and water were removed by evaporation and the
atmosphere at room temperature for 2.5 h. After filtering off residue was purified by preparative HPLC to give a yellow color
the Pd/C, the solvent was removed by evaporation and the powder (5 mg, 35%). Analytical HPLC purity was 97%. H+ESI =
1
residue was dried under vacuum to give a white solid (60 mg, 850.3. H-NMR (400 MHz, D2O) δ = 2.16 (m, 1H), 2.38 (m, 1H),
1
88%). H+ESI = 307.2. H-NMR (400 MHz, CD3OD) δ = 1.36 (s, 3.35–3.55 (m, 8H), 3.89 (m, 2H), 4.24 (t, 1H, J = 4.0 Hz),
3H), 1.58 (s, 3H), 2.86–2.90 (m, 2H), 4.24–4.18 (m, 1H), 5.00 4.48–4.60 (m, 3H), 5.97 (s, 1H), 6.54 (d, 2H, J = 8.4 Hz), 6.73
(dd, 1H, J1 = 6.80 Hz, J2 = 3.60 Hz), 5.46 (dd, 1H, J1 = 6.40 Hz, (m, 4H), 7.08 (d, 1H, J = 8.0 Hz), 7.81 (dd, 1H, J1 = 8.0 Hz, J2 =
J2 = 3.20 Hz), 6.13 (d, 1H, J = 2.80 Hz), 8.20 (s, 1H), 8.26 (s, 1.04 Hz), 8.12 (s, 1H), 8.18 (s, 1H), 8.26 (s, 2H). 13C-NMR
1H).
(125 MHz, D2O) δ = 172.60, 169.86, 168.15, 162.99, 162.71,
5′-N-[4-{(2S)-2-(N-tert-Butoxycarbonyl)amino-butyric acid}]- 153.88, 150.05, 147.55, 144.88, 142.57, 135.79, 135.19, 132.69,
tert-butyl ester-5′-deoxy-2′,3′-O,O-(1-methyl ethylidene) adeno- 129.84, 127.98, 126.34, 125.57, 118.84, 117.36, 114.98, 111.19,
sine (4).24,25 Compound 3 (0.544 g, 1.78 mmol, 1.0 eq.) was 102.40, 90.00, 78.27, 73.10, 71.44, 61.61, 54.65, 52.56, 51.72,
dissolved in 10 mL anhydrous dichloroethene (DCE). Under 49.98, 39.67, 24.67.
stirring, the tert-butyl (S)-2-((tert-butoxycarbonyl)amino)-4-oxo-
butanoate (0.46 g, 1.7 mmol, 1.0 eq.) was added in one portion
Concentration determination of FL-NAH
followed by sodium triacetoxyborohydride (0.38 g, 1.78 mmol, The FL-NAH purified from HPLC was dissolved in dd-H2O. A
1 eq.). The mixture was stirred at room temperature for 3 h. small amount of solution was diluted and adjusted to pH 9 to
Saturated aqueous sodium bicarbonate (20 mL) was added to make sure that fluorescein existed in a dianion form. The UV
the mixture and then was extracted by dichloromethane (DCM, absorption was measured at 495 nm. The concentration was
5 mL × 3). The organic layer was collected, washed with brine calculated with an extinction coefficient of 76 000 M−1 cm−1
.
and dried over Na2SO4. After removal of the solvent, the crude
product was purified by column chromatography to give pure
Protein expression and purification
compound 4 as a white solid (0.762 g, 76%). H+ESI = 564.3. Detailed protocols for protein expression and purification can
1H-NMR (400 MHz, CDCl3) δ = 1.35 (s, 9H), 1.36 (s, 3H), 1.43 be found in the ESI.†
(s, 9H), 1.60 (s, 3H), 1.84 (m, 1H), 2.36 (m, 1H), 3.04 (m, 1H),
Fluorescence polarization assay
3.14 (m, 1H), 3.40 (m, 2H), 4.18 (s, 1H), 4.54 (d, 1H, J =
3.64 Hz), 5.08 (s, 1H), 5.25 (d, 1H, J = 5.20 Hz), 5.55 (d, 1H, J = All binding experiments were performed in a total volume of
5.20 Hz), 6.20 (s, 1H), 8.28 (s, 1H), 8.30 (s, 1H).
5′-N-Propynylamino-N′-[4-{(2S)-2-(N-tert-butoxycarbonyl)amino- 0.01% Triton X-100 and 5 mM DTT) in 384-well black poly-
butyric acid}]tert-butyl ester-5′-deoxy-2′,3′-O,O-(1-methyl- propylene PCR plates (Axygen, PCR-384-BK). Fluorescence
ethylidene)adenosine (78 mg, polarization (FP) was measured using a Biotek Synergy 4 after
(5).24,25 Compound
10 μl binding buffer (20 mM Tris pH 8.0 supplemented with
4
0.15 mmol, 1.0 eq.) was dissolved in anhydrous DMF (5 mL). 30 min incubation at ambient temperature. The excitation and
Under stirring, propargyl bromide (18.2 mg, 0.15 mmol, emission wavelengths were 485 nm and 528 nm, respectively.
1.2 eq.) was added followed by K2CO3 (21 mg, 0.15 mmol, Polarization values were expressed in millipolarization
1 eq.). The reaction mixture was stirred at room temperature units (mP).
for 24 h. Water was added and the mixture was extracted by
Binding of MLL1 to FL-NAH. Increasing concentrations
DCM (×3). The organic layer was collected and dried over of MLL1 or MLL1 complexes were incubated with 50 nM
Na2SO4. The crude product was purified by column chromato- FL-NAH-1. The binding affinity Kd was calculated by fitting the
graphy to give pure compound 5 (67 mg, 75%). H+ESI = 602.3. data to nonlinear least squares regression, single-site binding
1H-NMR (400 MHz, CDCl3) δ = 1.37 (s, 3H), 1.41 (s, 9H), 1.43 model using eqn (1) in GraphPad Prism v. 6.05.
(s, 9H), 1.60 (s, 3H), 1.91 (m, 1H), 2.17 (m, 1H), 2.46 (s, 1H),
DMSO tolerance and competitor Kdisp determination. The
3.11 (m, 2H), 3.36–3.46 (m, 2H), 3.91 (s, 2H), 4.12 (s, 1H), 4.59 MLL1 SET domain or complexes were diluted to their
(d, 1H, J = 5.20 Hz), 5.00 (s, 1H), 5.25 (d, 1H, J = 5.20 Hz), 6.13 respective Kd (FL-NAH) concentrations and incubated with 50 nM
(s, 1H), 8.16 (s, 1H), 8.30 (s, 1H).
FL-NAH in the presence of increasing concentrations of DMSO
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
or competitor compounds for 30 minutes at ambient tempera-
ture. Fluorescence polarization was measured as described
above. The concentration of the competitor compound that
achieved 50% displacement (Kdisp) was calculated by fitting
the data to nonlinear least squares regression analysis using
eqn (2) in GraphPad Prism v.6.05.
Z′ factor analysis. The MLL1 SET domain (1 μM) was incu-
bated with 50 nM FL-NAH in the presence or absence of
250 μM SAH. The Z′ factor was calculated as previously
described using eqn (3).28
3 J. H. Lee and D. G. Skalnik, J. Biol. Chem., 2005, 280,
41725–41731.
4 J. Wysocka, T. Swigut, T. A. Milne, Y. Dou, X. Zhang,
A. L. Burlingame, R. G. Roeder, A. H. Brivanlou and
C. D. Allis, Cell, 2005, 121, 859–872.
5 R. van Nuland, A. H. Smits, P. Pallaki, P. W. Jansen,
M. Vermeulen and H. T. Timmers, Mol. Cell. Biol., 2013, 33,
2067–2077.
6 S. A. Shinsky, K. E. Monteith, S. Viggiano and
M. S. Cosgrove, J. Biol. Chem., 2015, 290, 6361–6375.
7 Y. Dou, T. A. Milne, A. J. Ruthenburg, S. Lee, J. W. Lee,
G. L. Verdine, C. D. Allis and R. G. Roeder, Nat. Struct. Mol.
Biol., 2006, 13, 713–719.
Equations used for binding evaluations
The binding affinity Kd was calculated as eqn (1), where ΔFP is
the change of fluorescence polarization, ΔFPmax is the
maximum ΔFP when MLL1 saturates all the FL-NAH in solu-
tion, and [MLL] is the total concentration of the MLL1 SET
domain.
8 M. G. Guenther, R. G. Jenner, B. Chevalier, T. Nakamura,
C. M. Croce, E. Canaani and R. A. Young, Proc. Natl. Acad.
Sci. U. S. A., 2005, 102, 8603–8608.
9 C. W. Chen and S. A. Armstrong, Exp. Hematol., 2015, 43,
673–684.
½MLL1ꢀ ꢁ ΔFPmax
10 R. L. Wright and A. T. Vaughan, Crit. Rev. Oncol./Hematol.,
2014, 91, 283–291.
ΔFP ¼
ð1Þ
½MLL1ꢀ þ Kd
11 A. Shi, M. J. Murai, S. He, G. Lund, T. Hartley, T. Purohit,
G. Reddy, M. Chruszcz, J. Grembecka and T. Cierpicki,
Blood, 2012, 120, 4461–4469.
12 J. Manka, R. N. Daniels, E. Dawson, J. S. Daniels,
N. Southall, A. Jadhav, W. Zheng, C. Austin, J. Grembecka,
T. Cierpicki, C. W. Lindsley and S. R. Stauffer, in Probe
Reports from the NIH Molecular Libraries Program,
Bethesda, MD, 2010.
13 L. Li, R. Zhou, H. Geng, L. Yue, F. Ye, Y. Xie, J. Liu,
X. Kong, H. Jiang, J. Huang and C. Luo, Bioorg. Med. Chem.
Lett., 2014, 24, 2090–2093.
14 S. He, T. J. Senter, J. Pollock, C. Han, S. K. Upadhyay,
T. Purohit, R. D. Gogliotti, C. W. Lindsley, T. Cierpicki,
S. R. Stauffer and J. Grembecka, J. Med. Chem., 2014, 57,
1543–1556.
15 J. Grembecka, S. He, A. Shi, T. Purohit, A. G. Muntean,
R. J. Sorenson, H. D. Showalter, M. J. Murai, A. M. Belcher,
T. Hartley, J. L. Hess and T. Cierpicki, Nat. Chem. Biol.,
2012, 8, 277–284.
16 A. T. Thiel, P. Blessington, T. Zou, D. Feather, X. Wu,
J. Yan, H. Zhang, Z. Liu, P. Ernst, G. A. Koretzky and
X. Hua, Cancer Cell, 2010, 17, 148–159.
17 G. Senisterra, H. Wu, A. Allali-Hassani, G. A. Wasney,
D. Barsyte-Lovejoy, L. Dombrovski, A. Dong, K. T. Nguyen,
D. Smil, Y. Bolshan, T. Hajian, H. He, A. Seitova, I. Chau,
F. Li, G. Poda, J. F. Couture, P. J. Brown, R. Al-Awar,
M. Schapira, C. H. Arrowsmith and M. Vedadi, Biochem. J.,
2013, 449, 151–159.
18 F. Cao, E. C. Townsend, H. Karatas, J. Xu, L. Li, S. Lee,
L. Liu, Y. Chen, P. Ouillette, J. Zhu, J. L. Hess, P. Atadja,
M. Lei, Z. S. Qin, S. Malek, S. Wang and Y. Dou, Mol. Cell,
2014, 53, 247–261.
The concentration of the competitor compound that
achieved 50% displacement, Kdisp, was calculated by using eqn
(2), where ΔFPbottom is the lower bound of the concentration–
response curve, ΔFPtop is the upper bound of the concen-
tration–response curve, [Compound] is the log of the com-
pound concentration and HS is the Hill coefficient.
ðΔFPtop ꢂ ΔFPbottom
Þ
ΔFP ¼ ΔFPbottom
þ
ð2Þ
dispꢂ½CompoundꢀÞꢁHSÞ
1 þ 10ððLog K
The Z′ factor was calculated as eqn (3),28 where σN and σP
are standard deviations for samples without or with SAH,
respectively, and μN and μP are mean FP values for samples in
the absence or presence of SAH, respectively.
ð3σN þ 3σPÞ
Z′ ¼ 1 ꢂ
ð3Þ
jμN ꢂ μPj
Acknowledgements
The SGC is a registered charity (number 1097737) that receives
funds from AbbVie, Bayer Pharma AG, Boehringer Ingelheim,
Canada Foundation for Innovation, Eshelman Institute for
Innovation, Genome Canada, Innovative Medicines Initiative
(EU/EFPIA), Janssen, Merck & Co., Novartis Pharma AG,
Ontario Ministry of Economic Development and Innovation,
Pfizer, São Paulo Research Foundation-FAPESP, Takeda, and
Wellcome Trust. YGZ acknowledges the National Institutes of
Health and National Science Foundation for financial support.
References
19 T. Cierpicki and J. Grembecka, Future Med. Chem., 2014, 6,
447–462.
1 A. Shilatifard, Annu. Rev. Biochem., 2012, 81, 65–95.
2 J. H. Lee, C. M. Tate, J. S. You and D. G. Skalnik, J. Biol. 20 M. V. Florian Grebien, M. Getlik, R. Giambruno, A. Grover,
Chem., 2007, 282, 13419–13428.
R. Avellino, A. Skucha, S. Vittori, E. Kuznetsova, D. Smil,
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
D. Barsyte-Lovejoy, F. Li, G. Poda, M. Schapira, H. Wu, 23 S. D. Barrett, D. A. Bochar, L. L. Blazer, F. L. Ciske,
A. Dong, G. Senisterra, A. Stukalov, K. V. M. Huber,
A. Schönegger, R. Marcellus, M. Bilban, C. Bock,
P. J. Brown, J. Zuber, K. L. Bennett, R. Al-awar, R. Delwel,
G. W. Endres, J. K. Johnson, G. S. Keyes, R. S. Sidhu,
R. C. Trievel and M. L. Collins, USPTO, US 2012/0295283A1,
2012.
C. Nerlov, C. H. Arrowsmith and G. Superti-Furga, Nat. 24 J. Dowden, W. Hong, R. V. Parry, R. A. Pike and S. G. Ward,
Chem. Biol., 2015. Bioorg. Med. Chem. Lett., 2010, 20, 2103–2105.
21 L. J. Brown, M. Baranowski, Y. Wang, A. K. Schrey, T. Lenz, 25 G. Zhang, S. L. Richardson, Y. Mao and R. Huang, Org.
S. D. Taverna, P. A. Cole and M. Sefkow, Anal.l Biochem.,
2014, 467, 14–21.
22 J. S. Yi, A. J. Federation, J. Qi, S. Dhe-Paganon, M. Hadler,
Biomol. Chem., 2015, 13, 4149–4154.
26 S. M. Southall, P. S. Wong, Z. Odho, S. M. Roe and
J. R. Wilson, Mol. Cell, 2009, 33, 181–191.
X. Xu, R. St Pierre, A. C. Varca, L. Wu, J. J. Marineau, 27 P. Wanat, S. Walczak, B. A. Wojtczak, M. Nowakowska,
W. B. Smith, A. Souza, E. J. Chory, S. A. Armstrong and J. Jemielity and J. Kowalska, Org. Lett., 2015, 17, 3062–3065.
J. E. Bradner, ACS Chem. Biol., 2015, 10, 667– 28 J. H. Zhang, T. D. Chung and K. R. Oldenburg, J. Biomol.
674.
Screening, 1999, 4, 67–73.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015