EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex

Article

EED-Targeted PROTACs Degrade EED, EZH2, and

SUZ12 in the PRC2 Complex

Highlights

Authors

d

EED-targeted PROTACs bind to EED and promote ternary

complex formation with VHL

Jessie Hao-Ru Hsu,

Timothy Rasmusson,

James Robinson, ..., M. Paola Castaldi,

Stephen Fawell, Andrew Bloecher

EED, EZH2, and SUZ12 are selectively degraded by EED-

targeted PROTACs

Correspondence

andrew.bloecher@astrazeneca.com

EED-targeted PROTACs reduce proliferation of EZH2-

dependent cancer cells

In Brief

Hsu et al. describe the discovery of EED-

targeted PROTACs that lead to the

degradation of multiple components of

the PRC2 complex. Furthermore, they

demonstrate that these PROTACs inhibit

the proliferation of PRC2-dependent

cancer cells and may be considered an

attractive alternative to EZH2 catalytic

inhibitors currently in clinical

development.

Hsu et al., 2020, Cell Chemical Biology 27, 1–6

January 16, 2020 ª 2019 Elsevier Ltd.

https://doi.org/10.1016/j.chembiol.2019.11.004

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Cell Chemical Biology

Article

EED-Targeted PROTACs Degrade EED,

EZH2, and SUZ12 in the PRC2 Complex

Jessie Hao-Ru Hsu,¹Timothy Rasmusson,³James Robinson,⁴Fiona Pachl,³Jon Read,⁴Sameer Kawatkar,¹

Daniel H. O’ Donovan,²Sharan Bagal,²Erin Code,³Philip Rawlins,⁴Argyrides Argyrou,⁴Ronald Tomlinson,³Ning Gao,³

Xiahui Zhu,³Elisabetta Chiarparin,²Kelly Jacques,¹Minhui Shen,¹Haley Woods,¹Emma Bednarski,¹David M. Wilson,²

Lisa Drew,¹M. Paola Castaldi,³Stephen Fawell,¹and Andrew Bloecher^1,5,

¹Oncology R&D, AstraZeneca, Waltham, MA 02451, USA

*

²Oncology R&D, AstraZeneca, Cambridge CB4 0FZ, UK

³Discovery Sciences, BioPharmaceuticals R&D, AstraZeneca, Waltham, MA 02451, USA

⁴Discovery Sciences, BioPharmaceuticals R&D, AstraZeneca, Cambridge CB4 0FZ, UK

⁵Lead Contact

*Correspondence: andrew.bloecher@astrazeneca.com

https://doi.org/10.1016/j.chembiol.2019.11.004

SUMMARY

The role of PRC2 in disease has been the focus of intense

research. Supporting a role for PRC2 in oncogenesis, activating

Deregulation of the PRC2 complex, comprised of the mutations in EZH2 are frequently found in both GCB-DLBCL

and follicular lymphoma (Morin et al., 2010). These activating

mutations lead to increased H3K27me3 which represses

core subunits EZH2, SUZ12, and EED, drives aber-

rant hypermethylation of H3K27 and tumorigenicity

of many cancers. Although inhibitors of EZH2 have

shown promising clinical activity, preclinical data

suggest that resistance can be acquired through

secondary mutations in EZH2 that abrogate drug

target engagement. To address these limitations,

we have designed several hetero-bifunctional

expression of key tumor suppressor genes as well as genes

involved in B cell differentiation (Kim and Roberts, 2016). In sup-

port of a critical role for EZH2 in oncogenesis, EZH2 SAM-

competitive inhibitors have been identiﬁed that demonstrate

preclinical and clinical activity in B cell lymphomas (Garapaty-

Rao et al., 2013; McCabe et al., 2012; Knutson et al., 2014; Ital-

iano et al., 2018).

Recently, a novel mechanism of inhibition of the PRC2

complex has been reported using inhibitors that bind to an

PROTACs (proteolysis-targeting chimera) to efﬁ-

ciently target EED for elimination. Our PROTACs

bind to EED (pK_Dꢀ 9.0) and promote ternary complex allosteric site on EED thereby preventing recruitment to histones

and activation of EZH2 in the PRC2 complex (Qi et al., 2017; He

et al., 2017). These inhibitors have demonstrated anti-tumor

formation with the E3 ubiquitin ligase. The PROTACs

potently inhibit PRC2 enzyme activity (pIC₅₀ꢀ 8.1)

activity in EZH2 mutant DLBCL models and also in the EZH2 in-

hibitor-resistant setting. Given the likelihood of resistance

and induce rapid degradation of not only EED but

also EZH2 and SUZ12 within the PRC2 complex.

emerging against EZH2 small-molecule inhibitors, new thera-

Furthermore, the PROTACs selectively inhibit prolif-

peutic approaches against the activity of the PRC2 complex

eration of PRC2-dependent cancer cells (half

are much needed.

maximal growth inhibition [GI₅₀] = 49–58 nM). In sum-

mary, our data demonstrate a therapeutic modality to

Proteolysis-targeting chimaeras (PROTACs) represent an

emerging pharmacological mode of action that may offer advan-

target PRC2-dependent cancer through a PROTAC-

mediated degradation mechanism.

tages over traditional small-molecule inhibitors (Lai and Crews,

2017). PROTACs combine two small molecules via a linker motif,

one of which binds to an E3 ligase while the other engages a

target protein. In this manner, PROTACs recruit the cellular

protein degradation machinery to the proximity of the protein

of interest to ubiquitinate and ultimately degrade the target

protein. To determine whether PROTACs could serve as an alter-

native strategy to inhibit PRC2 activity, we have designed, devel-

oped, and characterized novel EED-targeted PROTACs that

can potently and selectively degrade both EED and associated

proteins within the PRC2 complex.

INTRODUCTION

The polycomb repressive complex 2 (PRC2) is an epigenetic

modulator of transcription and plays a role in a variety of biolog-

ical processes, such as stem cell maintenance, DNA repair, and

epithelial to mesenchymal transition (Dimou et al., 2017). EED,

SUZ12, and EZH2 form the core essential subunits of the

PRC2 complex that regulate gene expression through the

methylation of H3K27. EZH2, the catalytic subunit of PRC2, is RESULTS

regulated by the EED subunit through two mechanisms; recruit-

ment of the PRC2 complex to chromatin, and allosteric activa- Design of EED-Targeted PROTACs

tion of EZH2 methyltransferase activity (Qi et al., 2017; He We sought to develop a novel mechanism of inhibiting PRC2

et al., 2017).

activity by synthesizing PROTACs targeting the EED subunit

Cell Chemical Biology 27, 1–6, January 16, 2020 ª 2019 Elsevier Ltd.

1

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Figure 1. Chemical Structures of the Cap-

ped VHL Ligand, EED Targeting PROTACs,

and Capped EED Ligand

EED-Targeted PROTACs Bind and

Inhibit the PRC2 Complex

We proﬁled the two EED-targeted PRO-

TACs (compounds 2 and 3) alongside

the parent EED ligand (compound 4), the

VHL ligand (compound 1), and the two

inactive PROTACs (compounds 5 and 6)

(Figure 1). PROTACs are bifunctional molecules which comprise using surface plasmon resonance to assess EED binding afﬁnity

a ligand binding to the target protein (EED) covalently linked to and a biochemical MTase Glo assay (Promega) to measure

a second ligand binding to an E3 ubiquitin ligase, thereby recruit- effects on PRC2 catalytic activity (Figures 2A and 2B). Com-

ing the E3 ligase to induce ubiquitination and subsequent pound 4 bound to EED with a pK_Dof 9.61 0.11 and inhibited

degradation of the target protein (Bondeson et al., 2015). After PRC2 function with a pIC₅₀of 8.43

0.04. Attachment of

evaluating several reported EED ligands (Huang et al., 2017) the linkers and VHL warheads resulted in a slight decrease in

(patent US20160176882A1), we generated an X-ray structure EED afﬁnity and potency of PRC2 inhibition for compounds 2,

of an EED ligand compound bound to EED, which highlighted 3, 5, and 6; however, all compounds nonetheless inhibit PRC2

a solvent-exposed vector suitable to append a PROTAC catalytic activity with low nM potency and bind to EED with

linker (Read et al., unpublished). For the E3 ligase component, low nM or pM afﬁnity. As expected, capped VHL ligand com-

we used a well-studied ligand of von Hippel-Lindau (VHL) pound 1 does not bind to EED or inhibit PRC2 function. Next,

(Buckley et al., 2012) and evaluated various linker motifs to ulti- we used a TR-FRET in vitro ternary complex formation assay

mately generate the compounds 2 and 3 (PROTACs 1 and 2). to measure EED:PROTAC:VHL ternary complex formation (Fig-

As tools for comparison, we also synthesized the inactive ure 2C). Ternary complex formation occurs only in the presence

PROTACs analogs, compounds 5 and 6, which modify the ste- of compounds 2 and 3, which incorporate the appropriate ste-

reochemistry of the VHL binding motif to ablate afﬁnity for VHL reochemistry on the VHL warhead. We observed a bell-shaped

while maintaining similar EED binding and physicochemical curve in agreement with previously published observations

properties.

for PROTAC-induced ternary complex formation (Smith et al.,

Figure 2. EED-Targeted PROTACs Bind to EED and Promote Ternary Complex Formation with VHL

(A) MTase Glo biochemical assay measuring PRC2 catalytic activity shows that the capped EED ligand, the two active PROTACs, and the two inactive PROTACs

all inhibit PRC2 function.

(B) Table shows mean MTase Glo pIC₅₀values and mean EED surface plasmon resonance (SPR) pK_Dvalues for all six compounds tested.

(C) TR-FRET in vitro ternary complex formation assay shows that EED:PROTAC:VHL ternary complex formation only occurs in the presence of the two active

PROTACs, which have the correct stereochemistry on the VHL warhead.

2

Cell Chemical Biology 27, 1–6, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Figure 3. EED-Targeted PROTACs Promote

A

PROTAC PROTAC Inactive

EED

Degradation of EED, EZH2, and SUZ12

(A) Karpas422 cells were treated with 0.1, 1, and

3 mM of the compounds for 48 h and the protein

levels of EZH2, EED, and H3K27me3 were

analyzed by western blot.

#2

#1

PROTAC Ligand

#1

EED

(B) Karpas422 cells were treated with 1 mM of the

PROTACs for 1–24 h or 0.3 (+) or 1 (++) mM of the

inactive PROTACs for 24 h. Protein levels of EZH2,

EED, SUZ12, EZH1, H3, GAPDH, and H3K27me3

in the treated cells were analyzed by western blot.

(C) Quantiﬁcation of the western blot shown in (B)

by densitometry analysis. Protein band intensity

was normalized to GAPDH and DMSO control

samples.

EZH2

GAPDH

H3K27me3

0.12, 1, 3 PM

B

PROTAC #2

(hrs)

PROTAC #1 Inactive PROTACs

(hrs) (24hr )

#1 #2

1 2 4 6 8

not signiﬁcantly modulate EED and

EZH2 protein expression (Figures 3A,

3B, and S2). To understand the kinetics

of degradation, a time course experiment

was performed with PROTACs 1 and 2.

EED protein levels decreased within 1–

2 h of PROTAC treatment. EZH2 and

SUZ12 also decreased but with slightly

slower kinetics (Figures 3B and 3C).

Finally, the inactive PROTACs 1 and 2

did not lead a decrease in EED, EZH2,

or SUZ12 within 24 h (Figure 3B).

EZH2

EED

SUZ12

EZH1

GAPDH

H3K27me3

Total H3

PROTAC-Mediated Degradation of

EED Is Dependent on the Ubiquitin

C

150

100

50

EZH1 level in PROTAC #1 treated cells

Proteasome Pathway

To further corroborate that EED-targeted

EZH1 level in PROTAC #2 treated cells

SUZ12 level in PROTAC #1 treated cells

SUZ12 level in PROTAC #2 treated cells

EZH2 level in PROTAC #1 treated cells

EZH2 level in PROTAC #2 treated cells

EED level in PROTAC #1 treated cells

EED level in PROTAC #2 treated cells

PROTACs lead to degradation of EED via

an E3 ubiquitin ligase-dependent mecha-

nism, we treated cells with EED-targeted

PROTAC 1 and PROTAC 2 in the pres-

ence of excess VHL or MLN4924, which

outcompetes PROTAC-mediated binding

0

to E3 ubiquitin ligase or inhibits E3 ligase

activity, respectively. We observed that

0

2

4

6

8

1015

20

25

Hours

excess VHL or MLN4924 treatment alone

did not affect EED or EZH2 expression

2019). As expected, compounds 1, 4, 5, and 6 fail to induce

ternary complex formation (Figure 2C).

(Figures 4 and S3). Treatment with PROTACs in the presence

of excess VHL or MLN4924 prevented PROTAC-mediated

decrease in EED protein levels (Figures 4 and S3). We found

that PROTAC 2 requires higher amounts of VHL to fully inhibit

EED degradation (Figure S3). Furthermore, treatment with a pro-

EED-Targeted PROTACs Promote Degradation of EED,

EZH2, and SUZ12

Having established that our EED-targeted PROTACs can bind teasome inhibitor MG132 in PROTAC-treated cells also stabi-

EED and VHL as well as inhibit PRC2 enzymatic activity, lized PROTAC-mediated loss of EED protein expression.

we determined whether they lead to degradation of EED in Together, these observations suggest that our PROTACs pro-

cells. As expected, H3K27me3 was reduced upon PROTAC mote degradation of EED via the ubiquitin proteasome pathway.

treatment (Figure 3A, S1, and S2). The inactive PROTACs and

the EED ligand inhibited tri-methylation of H3 at lysine 27, EED-Targeted PROTAC Is Selective against EED, EZH2,

suggesting that all our compounds bind to EED and inhibit and SUZ12

PRC2-mediated methylation of histone 3 (Figures 3A, S1, and S2). To help establish the selectivity proﬁle of our EED-targeted

Interestingly, PROTAC 1 and PROTAC 2 treatment led to PROTAC, we performed a global proteomics analysis to quantify

marked decrease in not only EED, but also EZH2. We also the relative abundance of proteins in PROTAC 1-treated cells as

demonstrated that the inactive PROTAC or the EED ligand do compared with DMSO-treated cells. We found that EED, EZH2,

Cell Chemical Biology 27, 1–6, January 16, 2020

3

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

5PM VHL

5PM MLN4924

5PM MG132

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

ꢀ

ꢁ

0.3PM PROTAC #2

0.3PM PROTAC #1

ꢁ

ꢀ

DMSO

EED

EZH2

GAPDH

H3K27me3

Figure 4. PROTAC-Mediated Degradation of EED Is Dependent on

the Ubiquitin Proteasome Pathway

log2 FC PROTAC#1/DMSO

Karpas422 cells were pre-treated with 5 mM of MG132, VHL warhead, and

MLN4924 for 2 h, followed by treatment with of 0.3 mM of PROTACs for an

additional 4 h. Protein levels of EZH2, EED, and H3K27me3 were analyzed by

western blot.

Figure 5. EED-Targeted PROTAC

EZH2, and SUZ12

1 Selectively Degrades EED,

Global proteomic analysis of Karpas422 cells treated with 1mM of PROTAC 1

for 24 h. EED, EZH2, and SUZ12 protein/peptides were signiﬁcantly decreased

in the PROTAC-treated cells as compared with DMSO-treated cells. The log2

fold change for EED, EZH2, and SUZ12 are À1.2, À0.6, and À0.4, respectively

(adjusted p < 0.05).

and SUZ12 were the only proteins that were signiﬁcantly

decreased 24 h after treatment (Figure 5; Table S3). As expected,

the abundance of EED was the most signiﬁcantly decreased, fol-

lowed by EZH2 and SUZ12. This suggests that PROTAC 1 is

highly selective against EED, EZH2, and SUZ12.

EED destabilizes of the PRC2 complex. Studies in the literature

show that knockout or knockdown of individual components of

the PRC2 complex leads to loss of the other members of the

complex (Leeb et al., 2010; Xu et al., 2015). Consistent with the

published reports, we have also found that knockout of EED pro-

tein in Karpas422 using CRISPR-based technology leads to

concomitant downregulation of EZH2 and SUZ12 (Figure S1),

comparable with PROTAC-mediated degradation pf EED.

Thus, PROTAC-induced degradation of EED likely triggers sub-

sequent degradation of EZH2 and SUZ12. Alternatively, it is

possible that once the ternary complex is formed, exposed

lysine residues in EZH2 or SUZ12, which are in proximity to the

EED binding site, could be sites of PROTAC-mediated ubiquitin

attachment. Intriguingly, we have observed rapid poly-

ubiquitination of EZH2 within 1 h of PROTAC treatment (Figures

S4B–S4D). Further studies are needed to determine the mecha-

EED-Targeted PROTACs Reduce Proliferation of EZH2

Mutant Karpas422 Cells

To assess whether the EED-targeted PROTACs would affect the

proliferation of EZH2 mutant cells, the DLBCL cell line Karpas422

was treated with PROTACs 1 and 2 as well as the EED targeting

ligand and the VHL warhead. PROTAC 1 and PROTAC 2 treat-

ment impaired cell proliferation similar to treatment with the

EED ligand alone (Figures 6A–6D). The VHL warhead did not

affect cell growth (Figure 6D). These data suggest that inhibition

of EED can lead to a profound decrease in cell proliferation in

EZH2-dependent cells.

DISCUSSION

Herein we have demonstrated that EED-targeted PROTACs nism(s) of ubiquitination of EZH2 with EED-targeted PROTACs.

can degrade EED and its associated proteins EZH2 and We have found that PROTAC 1 and PROTAC 2 can potently

SUZ12 in the PRC2 complex. In-cell concentration experi- inhibit proliferation of an EZH2 mutant DLBCL cell line, as well

ments also demonstrate that the PROTACs are found in both as an EZH2 WT rhabdoid cancer cell line (Figure 6; Table S2).

the nuclear and cytoplasmic fractions (Table S1, see the The PROTACs are ꢀ3- to 4-fold less potent than the EED

STAR Methods). Both PROTACs show comparable binding af- inhibitor as antiproliferative agents. This suggests that the

ﬁnity for EED and promote ternary complex formation with binding of PROTACs to EED and subsequent inhibition of

VHL in vitro. Co-immunoprecipation experiments have corrob- EZH2 catalytic activity is the primary driver of the antiproliferative

orated that EED, EZH2, and VHL are found together in the effects. However, PROTACs could provide extended durability

same complex in the presence of PROTACs (Figures S4A of inhibition compared with the EED ligand. EED ligand and

and S4B). In cells, both PROTACs induce rapid degradation PROTAC washout experiments are necessary to conﬁrm this

of EED, whereas the inactive PROTACs and the EED ligand hypothesis.

do not. Interestingly, PROTAC 2 seems to show faster kinetics

Previous publications have reported on non-catalytic func-

and more profound degradation of EED than PROTAC 1. tions for EZH2 in prostate and breast cancer (Xu et al., 2012;

Together, these data suggest that the linker structure is a Lawrence and Baldwin, 2016; Lee et al., 2011). Hypothetically,

key contributing factor to efﬁcient VHL/E3 recruitment and it is conceivable that our EED-targeted PROTACs may have

subsequent EED degradation.

activity in this setting. However, this hypothesis relies on the

The observation that these PROTACs induce EED degradation assumption that the PROTAC would efﬁciently lead to degrada-

more rapidly than EZH2 and SUZ12 implies that rapid loss of tion of all cellular EZH2. We have tested our PROTACs for

4

Cell Chemical Biology 27, 1–6, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

A

B

PROTAC #1

PROTAC #2

Day 4

Day 8

100

50

0

100

50

0

Day 12

Day 14

Day 12

Day 14

-50

0.01

0.1

1

10

100

0.01

0.1

1

10

100

cmpd,PM

C

D

EED Ligand

Day 4

Compound

EED ligand

PROTAC #1

PROTAC #2

VHL warhead

GI50, PM (Day 14)

100

Day 8

0.014

0.057

0.045

>31

Day 12

Day 14

50

0

-50

0.01

0.1

1

10

100

cmpd,PM

Figure 6. EED-Targeted PROTACs and EED Ligand Reduce Proliferation of Karpas422 Cells

(A–C) Dose-response curves for Karpas422 cells treated with PROTACs (A and B) or the EED ligand (C) from day 4 to 14. Percent cell growth relative to DMSO was

measured from day 4 to 14 after treatment. The error bars represent standard deviation.

(D) Summary of Karpas422 half maximal growth inhibition (GI₅₀) values measured after 14 days of treatment with indicated compounds (cmpds).

anti-proliferative effects in a limited number of reported EZH2- pathway. Importantly, the PROTACs inhibit cell proliferation

dependent (catalytically independent) cell lines and have not in PRC2-dependent cancer cells, comparable with EED

observed growth inhibition (Table S2). At present, it is unclear small-molecule inhibitor. This drug modality may potentially

whether the lack of growth inhibition is due to a cellular fraction help combat resistance to catalytic inhibitors or eliminate

of EZH2 that is immune to PROTAC-mediated degradation. cancer cells that are dependent on the non-catalytic func-

Further studies are warranted to further understand the non-cat- tion of the PRC2 complex.

alytic functions of EZH2.

In summary, we have discovered a novel PROTAC-based

approach to target the PRC2 complex. The PROTACs possess

STAR+METHODS

a unique and selective degradation proﬁle that may serve as

promising tools to help unravel PRC2 biology. In addition, they

Detailed methods are provided in the online version of this paper

and include the following:

may represent a novel therapeutic modality to treat PRC2-

dependent cancers.

d KEY RESOURCES TABLE

d LEAD CONTACT AND MATERIALS AVAILABILITY

SIGNIFICANCE

d EXPERIMENTAL MODEL AND SUBJECT DETAILS

d METHOD DETAILS

We have discovered a highly potent and selective EED-tar-

geted PROTAC that can inhibit PRC2 activity. The PROTACs

target EED and its associated proteins including EZH2

and SUZ12 for elimination via the ubiquitin proteasome

B Expression and Puriﬁcation of Human PRC2

B Preparation of Biotinylated EED

B Surface Plasmon Resonance (SPR) Assay for EED

Binding

Cell Chemical Biology 27, 1–6, January 16, 2020

5

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

B Methyltransferase Assay

B Ternary Complex Formation

B Proliferation Assay

Garapaty-Rao, S., Nasveschuk, C., Gagnon, A., Chan, E.Y., Sandy, P., Busby,

J., Balasubramanian, S., Campbell, R., Zhao, F., Bergeron, L., et al. (2013).

Identiﬁcation of EZH2 and EZH1 small molecule inhibitors with selective

impact on diffuse large

1329–1339.

B cell lymphoma cell growth. Chem. Biol. 20 ,

B Western Blot

B Immunoprecipitation (IP)

He, Y., Selvaraju, S., Curtin, M.L., Jakob, C.G., Zhu, H., Comess, K.M., Shaw,

B., The, J., Lima-Fernandes, E., Szewczyk, M.M., et al. (2017). The EED pro-

tein-protein interaction inhibitor A-395 inactivates the PRC2 complex. Nat.

Chem. Biol. 13 , 389–395.

B Cas9/RNP Nuceofection

B Reagents

B Sample Prep for Global Proteomics Analysis

B LC-MS/MS Analysis

Huang, Y., Zhang, J., Yu, Z., Zhang, H., Wang, Y., Lingel, A., Qi, W., Gu, J.,

Zhao, K., Shultz, M.D., et al. (2017). Discovery of ﬁrst-in-class, potent, and

orally bioavailable embryonic ectoderm development (EED) inhibitor with

robust anticancer efﬁcacy. J. Med. Chem. 60 , 2215–2226.

B Peptide and Protein Identiﬁcation and Quantiﬁcation

B Mass Spectrometry Analysis of Compound Concentra-

tion in Cells

Italiano, A., Soria, J.C., Toulmonde, M., Michot, J.M., Lucchesi, C., Varga, A.,

Coindre, J.M., Blakemore, S.J., Clawson, A., Suttle, B., and McDonald, A.A.

(2018). Tazemetostat, an EZH2 inhibitor, in relapsed or refractory B-cell non-

Hodgkin lymphoma and advanced solid tumours: a ﬁrst-in-human, open-label,

phase 1 study. Lancet Oncol. 19 , 649–659.

B Chemical Synthesis of Compounds

d QUANTIFICATION AND STATISTICAL ANALYSIS

d DATA AND CODE AVAILABILITY

SUPPLEMENTAL INFORMATION

Kim, K.H., and Roberts, C.W.M. (2016). Targeting EZH2 in cancer. Nat. Med.

22 , 128–134.

Supplemental Information can be found online at https://doi.org/10.1016/j.

chembiol.2019.11.004.

Knutson, S.K., Kawano, S., Minoshima, Y., Warholic, N.M., Huang, K.C., Xiao,

Y., Kadowaki, T., Uesugi, M., Kuznetsov, G., Kumar, N., et al. (2014). Selective

inhibition of EZH2 by EPZ-6438 leads to potent antitumor activity in EZH2-

mutant non-Hodgkin lymphoma. Mol. Cancer Ther. 13 , 842–854.

ACKNOWLEDGMENTS

The authors thank Sylvie Guichard, James Brownell, Alan Rosen, Nin Guan,

Jingwen Zhang, Scott Boiko, Kurt Pike, Beth Williamson, and members of

the AstraZeneca drug discovery team for various technical support and helpful

discussions throughout the project.

Lai, A.C., and Crews, C.M. (2017). Induced protein degradation: an emerging

drug discovery paradigm. Nat. Rev. Drug Discov. 16 , 101–114.

Lawrence, C.L., and Baldwin, A.S. (2016). Non-canonical EZH2 transcription-

ally activates RelB in triple negative breast cancer. PLoS One 11 , e0165005.

Lee, S.T., Li, Z., Wu, Z., Aau, M., Guan, P., Karuturi, R.K., Liou, Y.C., and Yu, Q.

(2011). Context-speciﬁc regulation of NF-k B target gene expression by EZH2

in breast cancers. Mol. Cell 43 , 798–810.

AUTHOR CONTRIBUTIONS

Conceptualization, J.H.-R.H., T.R., and A.B.; Methodology, J.H.-R.H., T.R.,

S.B., J.R., F.P., J.R., S.K., E.C., P.R., and A.A.; Investigation, J.H.-R.H., T.R.,

J.R., F.P., S.K., E.Code, P.R., D.H.O’D., A.A., M.S., H.W., E.B., and E.Chiar-

Leeb, M., Pasini, D., Novatchkova, M., Jaritz, M., Helin, K., and Wutz, A. (2010).

Polycomb complexes act redundantly to repress genomic repeats and genes.

Genes Dev. 24 , 265–276.

parin; Resources, R.T., N.G., X.Z., and K.J.; Writing

– Original Draft,

J.H.-R.H., T.R., J.R., F.P., S.K., E.C., N.G., X.Z., E.Chiarparin, D.H.O., and

A.B.; Writing – Review & Editing, J.H.-R.H., D.H.O’D., and A.B.; Supervision,

D.M.W., L.D., M.P.C., S.F., and A.B.

McCabe, M.T., Ott, H.M., Ganji, G., Korenchuk, S., Thompson, C., Van Aller,

G.S., Liu, Y., Graves, A.P., Della Pietra, A., 3rd, and Diaz, E. (2012). EZH2 inhi-

bition as a therapeutic strategy for lymphoma with EZH2-activating mutations.

Nature 492 , 108–112.

DECLARATION OF INTERESTS

Morin, R.D., Johnson, N.A., Severson, T.M., Mungall, A.J., An, J., Goya, R.,

Paul, J.E., Boyle, M., Woolcock, B.W., Kuchenbauer, F., et al. (2010).

Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large

B-cell lymphomas of germinal-center origin. Nat. Genet. 42 , 181–185.

The authors are current or former employees of AstraZeneca.

Received: May 31, 2019

Revised: September 3, 2019

Accepted: November 8, 2019

Published: November 27, 2019

Qi, W., Zhao, K., Gu, J., Huang, Y., Wang, Y., Zhang, H., Zhang, M., Zhang, J.,

Yu, Z., Li, L., et al. (2017). An allosteric PRC2 inhibitor targeting the H3K27me3

binding pocket of EED. Nat. Chem. Biol. 13 , 381–388.

Seki, A., and Rutz, S. (2018). Optimized RNP transfection for highly efﬁcient

CRI SPR/Cas9-mediated gene knockout in primary T cells. J. Exp. Med.

215 , 985–997.

REFERENCES

Bondeson, D.P., Mares, A., Smith, I.E.D., Ko, E., Campos, S., Miah, A.H.,

Mulholland, K.E., Routly, N., Buckley, D.L., Gustafson, J.L., et al. (2015).

Catalytic in vivo protein knockdown by small-molecule PROTACs. Nat.

Chem. Biol. 11 , 611–617.

Smith, B.E., Wang, S.L., Jaime-Figueroa, S., Harbin, A., Wang, J., Hamman,

B.D., and Crews, C.M. (2019). Differential PROTAC substrate speciﬁcity

dictated by orientation of recruited E3 ligase. Nat. Commun. 10 , 131.

Xu, K., Wu, Z.J., Groner, A.C., He, H.H., Cai, C., Lis, R.T., Wu, X., Stack, E.C.,

Loda, M., Liu, T., et al. (2012). EZH2 oncogenic activity in castration-resistant

prostate cancer cells is polycomb-independent. Science 338 , 1465–1469.

Buckley, D.L., Van Molle, I., Gareiss, P.C., Tae, H.S., Michel, J., Noblin, D.J.,

Jorgensen, W.L., Ciulli, A., and Crews, C.M. (2012). Targeting the von

Hippel-Lindau E3 ubiquitin ligase using small molecules to disrupt the VHL/

HIF-1 a interaction. J. Am. Chem. Soc. 134 , 4465–4468.

Xu, J., Shao, Z., Li, D., Xie, H., Kim, W., Huang, J., Taylor, J.E., Pinello, L.,

Glass, K., Jaffe, J.D., et al. (2015). Developmental control of polycomb subunit

composition by GATA factors mediates a switch to non-canonical functions.

Mol. Cell 57 , 304–316.

Dimou, A., Dincman, T., Evanno, E., Gemmill, R.M., Roche, J., and Drabkin,

H.A. (2017). Epigenetics during EMT in lung cancer: EZH2 as a potential ther-

apeutic target. Cancer Treat. Res. Commun. 12 , 40–48.

6

Cell Chemical Biology 27, 1–6, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE

SOURCE

IDENTIFIER

Antibodies

FlagꢀM2

Sigma

F3165; RRID: AB_259529

5246; RRID: AB_10694683

AF5827; RRID: AB_2246350

2118; RRID: AB_561053

3737; RRID: AB_561053

PA1-41114; RRID: AB_2102295

9733; RRID: AB_2616029

14269; RRID: AB_2756816

68547; RRID: AB_2716279

3900; RRID: AB_1550038

04-263; RRID: AB_612093

EZH2

Cell Signaling Technology

R&D Systems

EED

GAPDH

Cell Signaling Technology

Invitrogen

SUZ12

EZH1

H3K27me3

Cell Signaling Technology

Millipore

H3

VHL

IgG Isotype Control

Ubiquitin, clone FK2

Bacterial and Virus Strains

Sf21 Baculovirus expression system

E. coli expression strain BL21(DE3)

Chemicals, Peptides, and Recombinant Proteins

MLN4924

Invitrogen

10359016

C2527H

New England BioLabs

Selleckchem

S7109

S2619

9809

MG132

Selleckchem

Cell Lysis Buffer

Cell Signaling Technology

ThermoFisher

Lifesensors

DynabeadsꢁProtein G

Anti-Ub Agarose TUBE2

TrueCut Cas9 protein v2

Critical Commercial Assays

CellTiter Gloꢀ Luminescent Cell Viablity Assay

Mtase Glo

10009D

UM402

A36496

ThermoFisher

Promega

CisBio

G7570

N/A

Streptavidin-terbium

Streptavidin-XL665

Trypsin/Lys-C mix

Deposited Data

N/A

CisBio

N/A

Promega

V5073

Global proteomic proﬁling

Experimental Models: Cell Lines

Karpas422

This study

ProteomeXchange PRIDE: PXD016107

DSMZ

ATCC

ECACC

ATCC

ACC 32

G401

CRL-1441

CRL-5803

HTB-26

NCI-H1299

MDA-MB-231

CHP134

06122002

CCL-221

CRL-2343

CCL-127

CRL-1740

HTB-22

DLD1

HCC2218

IMR32

LNCAP

MCF7

MDA-MB453

HTB-131

Oligonucleotides

Alt-RꢀCRISPR-Cas9 EED crRNA #1

Alt-RꢀCRISPR-Cas9 EED crRNA #2

Integrated DNA Technologies (IDT)

Hs.Cas9.EED.1.AA

Hs.Cas9.EED.1.AB

(Continued on next page)

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e1

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Continued

REAGENT or RESOURCE

SOURCE

IDENTIFIER

Alt-RꢀCRISPR-Cas9 EED crRNA #3

Alt-RꢀCRISPR-Cas9 tracrRNA

Integrated DNA Technologies (IDT)

Hs.Cas9.EED.1.AC

1072534

None-targeting control crRNA sequence:

GTAGCGAACGTGTCCGGCGT

CTRL0001 (Custom-made)

Recombinant DNA

pFastBac EZH2

This study

pAZB0549

pAZB0547

pAZB0543

pAZB0542

pAZB0541

pFastBac FLAG EED

pFastBac SUZ12

pFastBac AEBP2

pFastBac RBBP4

Software and Algorithms

Proteome Discoverer 2.1

GraphPad Prism 8

Perseus

Thermo Fisher

N/A

GraphPad Software

https://www.graphpad.com/

https://maxquant.net/perseus/

Max Planck Institute of Biochemistry

LEAD CONTACT AND MATERIALS AVAILABILITY

Further information and requests for resources and reagents should be directed to and will be fulﬁlled by the Lead Contact, Andrew

Bloecher. There are restrictions to the availability of chemical reagents (Compounds 1 - 6) due to proprietary reasons, however, we

have provided full experimental details for the preparation of these reagents from commercial sources in this paper.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Karpas422 (female) cell line was obtained from DSMZ. All the other cell lines including G401 (male), NCI-H1299 (male), MDA-MB-231

(female), DLD1 (male), HCC2218 (female), IMR32 (male), LNCAP (male), MCF7 (female), and MDA-MB-453 (female) were obtained

from ATCC. CHP134 (male) cell line was obtained from ECACC. Cancer cell lines were grown in recommended growth media sup-

plemented with 10% FBS and 1% L-Glutamine at 37^ꢁC and 5% CO₂. All cell lines were authenticated and tested negative for my-

coplasma contamination. The Sf21 baculovirus expression system was purchased from Invitrogen. The E.Coli expression strain

used was BL21(DE3).

METHOD DETAILS

Expression and Puriﬁcation of Human PRC2

PRC2 is a 5-member complex, which was formed with the following members: human EZH2 (GenBank: NM_004456) untagged(MW=

86 kDa), human EED (GenBank: NM_003797) with N-terminal Flag-tag(MW= 51 kDa), human SUZ12 (GenBank: NM_015355) untag

(MW = 87 kDa), Human AEBP2 (GenBank: NM_153207) (amino acids 1-295, MW= 33 kDa), and human RbAp48 (GenBank:

NM_005610, MW = 48 kDa). The members of the complex were individually cloned into pFastBac1. All the subunits were co-ex-

pressed in baculovirus expression system. Sf21 cells were co-infected with ﬁve viruses at a density of 2 x10⁶cells/ml in Sf-900ꢁ

III SFM (Invitrogen). Sf21 cells were harvested by centrifugation 48 hr post infection. Cell pastes were stored at -20^ꢁC.

The frozen cells were suspended in Buffer A consisting of 25 mM Tris HCl, pH 8.0, 150 mM NaCl, 2 mM MgCl₂, 1mM DTT, 0.4 mM

EDTA, 0.005% Triton X-100, and 10% glycerol, supplemented with protease inhibitor cocktail tablets (Roche Molecular Biochem-

ical). Cells were lysed by sonication and clariﬁed by centrifugation at 150,000 g, for 40 min at 4^ꢁC. The supernatant was then

incubated with M2-agarose beads (Sigma) preequilibrated with Buffer B with the composition of 25 mM Tris HCl, pH 8.0, 150 mM

NaCl, 2 mM MgCl₂, 1 mM DTT, 0.4 mM EDTA, 0.005% Triton X-100, and 10% glycerol for 1 hr, and then extensively washed with

20 CV of buffer B. Proteins were eluted with 15 min incubations with 150 mg/ml 3X-FLAG peptide (Sigma). The fractions containing

PRC2 complex were pooled and concentrated using Amicon ultra 100 kDa MWCO concentrators (Millipore). Concentrated sample

was then applied onto a size exclusion chromatography Superdex200 Hiload 16/60 (GE healthcare) preequilibrated with buffer B at a

ﬂow rate of 1.0 mL/min. The PRC2 fractions were identiﬁed by SDS-PAGE, and pooled and concentrated. Protein concentration

was determined by absorbance at 280 nm. Protein was snap frozen with liquid nitrogen and store at -80^ꢁC.

Preparation of Biotinylated EED

The EED gene from H. sapiens was custom-synthesized with N-terminal His₆tag, thrombin protease cleavage site and AVI tag

(Life Technology). The synthetic gene comprising H. sapiens EED (Uniprot: O75530) was cloned into pET28a(+) as His₆-Thrombin-

TEV-EED using NdeI and XhoI restriction sites to create plasmid pAZB0739. The EED was transformed into E. coli expression strain

e2 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

BL21(DE3) and expressed with 0.25 mM IPTG at 16^ꢁC overnight. The frozen cell paste from 2 L of cell culture was suspended in 50 ml

of Buffer A consisting of 25 mM Tris-HCl (pH 8.0), 0.5 M NaCl, 5% (v/v) glycerol, supplemented with 1 EDTA-free protease inhibitor

cocktail tablet (Roche Molecular Biochemical). Cells were disrupted by French pressure at 18,000 psi twice at 4^ꢁC, and the crude

extract was centrifuged at 150,000 g (45Ti rotor, Bechman-Coulter) for 30 min at 4^ꢁC. The supernatant was applied at a ﬂow rate

of 2.0 ml/min onto a 5-ml HiTrap Ni²⁺chelating column (GE Healthcare Life Sciences) pre-equilibrated with Buffer A. The column

was washed with Buffer A, and EED was eluted by a linear gradient from 0 M to 0.5 M imidazole in Buffer A. Fractions containing

EED were pooled. And EED was biotinylated by using BirA biotin-protein ligase standard reaction kit (Avidity) and the biotinylation

was conﬁrmed by analytical LC-MS. To remove free biotin, the EED biotinylation reaction mixture was concentrated to 5 ml by

Amiconꢀ Ultracel-10K (Millipore, Billerica, MA). The 5 ml sample was applied at a ﬂow rate of 1.5 ml/min to a 120-ml Superdex

200 (HR 16/60) (GE Healthcare Life Sciences) pre-equilibrated with Buffer B consisting of 50 mM Tris-HCl, pH 8.0, 0.15 M NaCl,

1 mM dithiothreitol, 5% (v/v) glycerol. The fractions containing biotinylated EED were pooled and concentrated by Amiconꢀ Ultra-

cel-10K (Millipore). The protein concentration was determined by Bradford method and characterized by SDS-PAGE analysis. The

ﬁnal yield of puriﬁed biotinylated EED was 12 mg from 2 L of cell culture. The protein was stored at –80^ꢁC.

Surface Plasmon Resonance (SPR) Assay for EED Binding

A Biacore T200 or 8K instrument (GE Healthcare) was used to monitor binding interactions via SPR. 10 mg/mL 6His-AVI-EED (1À441)

was injected for 180 seconds on a streptavidin Biacore chip (GE Healthcare) at a ﬂow rate of 10 mL/min in buffer containing 20 mM

HEPES, 150 mM NaCl, 1 mM TCEP, 0.005% tween-20 (ﬁltered), 1% DMSO (v/v), pH 7.5 at 25^ꢁC. Typical immobilization density was

6000-8000 RU. The surface was blocked with 1 min. injection of 50 mM PEG-amine-biotin. Reference ﬂow cells were prepared

without protein. All binding measurements were performed in the buffer used for immobilisation, at 25^ꢁC. Prior to analysis, solvent

calibration and double referencing subtractions were made to eliminate bulk refractive index changes, injection noise, and data

drift. Afﬁnity and binding kinetic parameters were determined by global ﬁtting to a 1:1 binding model within the Biacore Evaluation

software (GE Healthcare).

Methyltransferase Assay

PRC2 methyltransferase activity is determined using a luminescent coupled assay, Promega Mtase-Glo, to detect transfer of methyl

groups from S-adenosylmethionine (SAM, Promega) to Lysine 27 of Histone H3 peptide 21-44 (Anaspec) by following production of

the product S-adenosylhomocysteine (SAH). The methyltransferase portion of the assay is performed in a reaction volume of 4 ul

containing 10 uM SAM, 20 uM peptide, and 20 nM PRC2 enzyme complex (Human EZH2, Flag-EED, SUZ12, AEBP2 and

RbAp48, AstraZeneca) in 20 mM Bicine pH 8.5, 0.01% triton x 100, 0.01% BSA, 0.5 mM DTT. The Mtase-Glo assay is run in 5 steps.

In step 1, SAM and peptide are added to 384 well compound plates (Greiner 784075) for a ﬁnal DMSO of 1%. In step 2, the reaction is

initiated by addition of PRC2 enzyme and incubated at RT for 3 hours. In step 3, methyltransferase activity is stopped by addition of

triﬂuoracetic acid (TFA, Sigma) for a ﬁnal of 0.2%. In step 4, 6X Mtase-Glo reagent (in reaction buffer supplemented for a ﬁnal con-

centration of 3 mM MgCl₂and 50 mM NaCl) is added to convert SAH into ADP for 30 minutes at RT. In step 5 (30 minutes RT), ADP

is converted to ATP and then to light by addition of Mtase–Glo detection solution (contains luciferase). The ﬁnal coupled reaction

volume is 12 ul. Luminescence is read on a PHERAStar reader and quantitated by comparison to a SAH (Promega) standard curve.

Ternary Complex Formation

Formation of ternary complexes composed of EED, PROTAC and VHL was monitored by time-resolved ﬂuorescence energy transfer

from a terbium donor to an XL665 acceptor. Assays were carried out in black 384-well low volume microplates (Greiner Bio-One,

catalog no. 784900) in a total volume of 10 mL. Compounds were dispensed in the assay plate using a Hewlett Packard HP D300

digital dispenser (TECAN). Mix 1 was prepared by preincubating 2 nM biotin-EED with 2 nM of streptavidin-terbium (Cis-Bio) for

15 min in Buffer C [50 mM Tris pH 7.5, 100 mM NaCl, 1 mM TCEP, 0.01% (w/v) pluronic F127, 1% (v/v) DMSO]. Mix 2 was prepared

by preincubating 40 nM biotin-VHL with 40 nM streptavidin-XL665 (Cis-Bio) for 15 min also in Buffer C. Mix1 and Mix2 were then

mixed at a ratio of 1:1 to give the following ﬁnal concentrations: 1 nM biotin-EED:1 nM streptavidin-terbium complex and 20 nM biotin

VHL:20 nM streptavidin-XL665 complex. This solution was then added to the assay plate containing the compounds and incubated

for 3 h prior to ﬂuorescence measurements [lex = 337 nm (laser), terbium emission (l = 620 nm), sensitized emission (l = 665 nm)]

using a PHERAstar FSX multimode reader (BMG LABTECH). Fluorescence data were normalized by taking the ratio of the sensitized

emission to the terbium emission (TR-FRET ratio) and plotted versus the compound concentration as shown in Figure 2C.

Proliferation Assay

Tumor cell lines were seeded onto 96-well tissue culture treated plates containing compound dilution from 30mM to 0.005mm (10

point, 3-fold dilution). Karpas422 cells were split every 4 days and cell viability was determined by CellTiter-Glo Luminescent Cell

Viability Assay (Promega) from day 4-14. For adherent cell lines, cell viability was measured 7 or 10 days after seeding. Luminescence

readings were normalized to day 0 and the DMSO control readings. Dose-response curves were analyzed using Prism software

(GraphPad) or Genedata Screenerꢀ.

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e3

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Western Blot

Cells were lysed in lysis buffer (100mM Tris-HCl, pH7.4, 10% glycerol, and 1%SDS). The lysates were boiled brieﬂy for 5 minutes

and then centrifuged at maximum speed at 4^ꢁC for 15 minutes. Protein quantiﬁcation was done using the Bio-Rad DC^TMProtein

Assay kit according to the manufacturer’s protocol. 5-20mg of protein were loaded onto 4-12% NuPAGE Bis-Tris protein gels, trans-

ferred to PVDF membranes, and probed with antibodies of interest. Antibodies against the following proteins were used: Flagꢀ M2

(Sigma: F3165), EZH2 (CST:#5246), EED (R&D Systems: AF5827), GAPDH(CST:#2118), SUZ12(CST:#3737) EZH1 (Invitrogen: PA1-

41114), H3K27me3 (CST:#9733), VHL(CST:#68547) and Histone H3(CST: #14269). CST, Cell Signaling Technology.

Western blot images were quantiﬁed using ImageQuant TL (General Electric Company).

Immunoprecipitation (IP)

Cells were lysed in 1x cell lysis buffer (CST: #9803) containing protease inhibitor cocktail (Roche). Protein lysates were pre-cleared

and quantiﬁed. Cell lysates and Dynabeadsꢁ Protein G (ThermoFisher) were incubated with anti- EZH2, anti- EED, anti-ubiquitiny-

lated protein (Millipore: 04-263) or IgG isotype control antibody (CST:#3900) overnight at 4^ꢁC. The beads were washed three times

with 1x cell lysis buffer the following day. The immunoprecipitated proteins were eluted by heating the samples in 2x SDS sample

to 95^ꢁC.

TUBE (tandem Ubiquitin Binding Entity)-based IP was conducted by incubating cell lysates with agarose-TUBE2 (Lifesensors:

UM402) overnight at 4^ꢁC. The beads were washed three times with 1x cell lysis buffer ,and then the proteins were eluted as described

above.

Cas9/RNP Nuceofection

Preparation of crRNA-tracrRNA duplex, pre-complexing of Cas9/RNA, and nucleofection using a 4D-nucleofector (Lonza) were pre-

viously described (Seki and Rutz, 2018). Brieﬂy, we mixed three EED speciﬁc crRNA-tracrRNA (IDT:Hs.Cas9.EED.1.AA, AB; Alt-Rꢀ

CRISPR-Cas9 tracrRNA) duplexes ( 3mL of each crRNA-tracrRNA equal to 150pmol, total of 9mL) and 6mL(180pmol) TrueCut Cas9

protein v2 (Thermo Fisher) at room temperature for 10 minutes. 20mL of cells(200,000 cells) in SF Cell Line Nucleofector ꢁ Solution

was combined with the Cas9/RNP mixture for 2 minutes at room temperature, followed by electroporation using a 4D nucleofec-

tor (Lonza).

Reagents

MLN4924 and MG132 were purchased from Selleckchem.

Sample Prep for Global Proteomics Analysis

2 million Karpas422 cells were treated with DMSO or 1uM PROTAC in biological triplicates for 24h, and cells were harvested

by trypsination. Lysis buffer (8 M urea, 50 mM NaCl, 50mM Tris-HCl, pH 8.5, 1% SDS, 1x Pierce HALT Protease and Phospha-

tase Inhibitor Cocktail) was added to the cell pellets, cells were homogenized by sonication and protein concentration

was determined by DC Protein Assay (Bio-Rad). 50mg protein for each sample were reduced by TCEP (tris(2-carboxyethyl)phos-

phine hydrochloride) and alkylated using MMTS (methyl methanethiosulfonate) followed by protein precipitation using methanol/

chloroform. Precipitated protein was resuspended in 4 M urea, 50 mM Tris HCl, pH 8.5 for digestion with Lys-C (1:50) for 12h at

room temperature, diluted to 0.5 M urea, 50mM Tris-HCl, pH 8.5, and digested with trypsin (1:50) for another 6h at room

temperature.

Peptide samples were labeled with TMT 6-plex (Thermo Fisher Scientiﬁc) at a ratio of 4:1 TMT reagent : sample according to man-

ufacturer’s instructions. Labeled peptides were combined into a single sample and desalted using C18 solid phase extraction car-

tridges (Sep-Pak, Waters). Samples were then subjected to ofﬂine fractionation by high pH reverse-phase HPLC (Agilent 1200) on a

˚

Gemini C18 column (5 uM, 110 A, 150 x 2 mm; Phenomenex) with 20mM ammonium hydroxide in water as mobile phase A and 20 mM

ammonium hydroxide in acetonitrile. The 96 resulting fractions were pooled in a non-continuous manner into 24 fractions and every

fraction was used for subsequent mass spectrometry analysis.

LC-MS/MS Analysis

Nanoﬂow LC-MS/MS was performed by coupling an easy-nLC 1200 (Thermo Fisher Scientiﬁc) to a Q Exactive HF

˚

(Thermo Fisher Scientiﬁc). Peptides were trapped on an Acclaim Pepmax column (2 cm x 75 um, 3um, 100 A; Themro Fisher)

˚

in water with 0.1% formic acid (FA)and separated on a C18 EASY-spray column (50 cm x 75 um, 2 um, 100 A, Thermo Fisher

Scientiﬁc) at 45^ꢁC. Gradient elution was performed from 2% acetonitrile to 30% acetonitrile in 0.1% formic acid over 2h at

250 nL/min ﬂow rate. The Q Exactive HF mass spectrometer was operated in data dependent top 12 mode. Full scan MS

spectra were acquired at 60,000 resolution after accumulation to a target value of 3e6 ions. Higher energy collisional dissoci-

ation (HCD) scans were performed with 34% NCE at 60,000 resolution and the ion target settings was set to 5e5 after accumu-

lation for max 140ms.

e4 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Peptide and Protein Identiﬁcation and Quantiﬁcation

RAW mass spectral data was processed using Proteome Discoverer 2.1 (Thermo Fisher Scientiﬁc) for peptide and protein identiﬁ-

cation and TMT quantiﬁcation, and MS/MS spectra were searched against a SwissProt human database (January 2017). Search

parameters were the following: full tryptic speciﬁcity with up to two missed cleavage sites, a precursor mass tolerance of 10

ppm, fragment ion mass tolerance of 0.02 Da, static methylthio alkylation of cysteine (45.988 Da), dynamic TMT labeling of lysine

residues and peptide N-termini (229.163 Da), and dynamic oxidation of methionine (15.995 Da) and acetylation of protein N-termini

(42.011 Da). Peak integration for TMT reporter ion intensities was done within a 20 ppm mass window and spectra with a co-isolation

value >50 were excluded from quantiﬁcation. Search results were ﬁltered to a maximum false discovery rate (FDR) of 0.01 for proteins

and peptides. Statistical analysis was performed in Perseus and Excel, and GraphPad Prism was used for visualization. Statistical

signiﬁcance between each pair of groups was calculated using Student’s T-test and p-values were adjusted for multiple testing to

control the false discover rate at 5%. Data are available via ProteomeXchange with identiﬁer PXD016107.

Mass Spectrometry Analysis of Compound Concentration in Cells

G401 cells were treated with 10mM of compounds 1, 2 and 3 for 2 hours. Cells were trypsinised, washed three times with ice-cold PBS

and counted, before lysing in RIPA buffer. For cellular fractionation experiments, cells were processed using the ThermoFisher sub-

cellular fractionation kit (78840). Samples were analysed by UPLC-MS utilising a Waters Xevo TQ-XS (WBA0259) and an Acquity

UPLC system from Waters consisting of Sample Manager (M16UFL953M), Acquity PDA (F17UPD457A), Column Oven

(E17CMP703G) and Binary Solvent Manager (E17BUR621G).

Chemical Synthesis of Compounds

General Information

All solvents and chemicals used were reagent grade. Anhydrous solvents THF, DCM and DMF were purchased from Aldrich. Flash

column chromatography was carried out using prepacked silica cartridges (from 4 g up to 330 g) from Redisep^TMor Silicycle and

eluted using an Isco Companion system. Purity and characterization of compounds were established by a combination of liquid chro-

matography-mass spectroscopy (LC-MS), gas chromatography-mass spectroscopy (GC-MS) and NMR analytical techniques and

was >95% for all test compounds. ¹H NMR were recorded on a Varian INOVA (600 MHz), Varian Gemini 2000 (300 MHz) or Bruker

Avance DPX400 (400 MHz) and were determined in CDCl₃, DMSO-d₆or MeOH-d₄. Chemical shifts are reported in ppm relative to

TMS (0.00 ppm) or solvent peaks as the internal reference. Splitting patterns are indicated as follows: s, singlet; d, doublet; t, triplet;

m, multiplet; br, broad peak. Elevated temperatures were used where necessary to sharpen broad NMR peaks due to rotamers and

the temperature used is noted for such compounds. Merck precoated TLC plates (silica gel 60 F₂₅₄, 0.25 mm, art. 5715) were used for

TLC analysis.

Abbreviations used: DIPEA, N,N-diisopropylethylamine; EtOAc, ethyl acetate; EtOH, ethanol; MeOH, methanol; HATU, 1-[bis(di-

methylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexaﬂuorophosphate; pTSA, 4-methylbenzenesulfonic acid;

DIAD, diisopropyl (E)-diazene-1,2-dicarboxylate; n-BuOH, n-butanol; ether, diethylether; TEA, triethylamine; DMF, dimethylforma-

mide; DCM, dichloromethane; THF, tetrahydrofuran; TFA, triﬂuoroacetic acid.

Scheme 1. Synthesis of cpd 1, Capped VHL Ligand.

Reagents and conditions: (a) Acetic anhydride, Et₃N, DMSO, rt, 81%.

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e5

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

(2S,4R)-1-((S)-2-acetamido-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (Cap-

ped VHL Ligand).

A vial was charged with acetic anhydride (0.017 mL, 0.18 mmol) and (2S,4R)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hydroxy-N-

(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (7, 69 mg, 0.16 mmol). Et₃N (0.068 mL, 0.48 mmol) was added and the re-

action was stirred for 4.5 h. Solvents were removed in vacuo and the mixture was puriﬁed by C18-ﬂash chromatography, using

gradient elution of MeCN in water (with 0.1 % formic acid). The puriﬁed fractions were combined and evaporated to afford the desired

product (2S,4R)-1-((S)-2-acetamido-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide

(Capped VHL ligand, 61 mg, 81%). ¹H NMR (400 MHz, DMSO-d₆) d 8.98 (s, 1H), 8.54 (t, J = 6.0 Hz, 1H), 7.92 (d, J = 9.4 Hz, 1H),

7.35 – 7.46 (m, 4H), 5.10 (d, J = 3.6 Hz, 1H), 4.54 (d, J = 9.4 Hz, 1H), 4.38 – 4.48 (m, 2H), 4.36 (s, 1H), 4.22 (dd, J = 15.8, 5.5 Hz,

1H), 3.66 (d, J = 4.0 Hz, 2H), 2.45 (s, 3H), 1.91 (s, 2H), 1.89 (s, 3H), 0.94 (s, 9H). 13C NMR (126 MHz, DMSO, 27^ꢁC) 172.41,

170.16, 169.57, 152.00, 148.06, 140.02, 131.71, 130.05, 129.11, 127.90, 69.35, 59.15, 56.88, 56.85, 42.12, 38.43, 35.66, 26.83,

22.80, 16.35. HRMS (ESI+): m/z found [M+ H]+ 473.2213, C₂₄H₃₂N₄O₄S requires 473.2223.

Scheme 2. Synthesis of cpd 2, PROTAC 1.

Reagents and conditions: (a) HATU, Et₃N, DMF, rt, 85%; (b) HCl, dioxane, rt, 23%; (c) HATU, DIPEA, DMF, 76%.

e6 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

tert-butyl N-[5-[[(1S)-1-[(2S,4R)-4-hydroxy-2-[[4-(4-methylthiazol-5-yl)phenyl]methylcarbamoyl]pyrrolidine-1-carbonyl]-2,2-dimethyl-

propyl]amino]-5-oxo-pentyl]carbamate (8).

Et₃N (2.59 mL, 18.58 mmol) and HATU (2.65 g, 6.97 mmol) were added to (2S,4R)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hy-

droxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (7, 2.0 g, 4.65 mmol) and 5-((tert-butoxycarbonyl)amino)penta-

noic acid (1.009 g, 4.65 mmol) in DMF (30 mL) at 25^ꢁC . The reaction mixture was stirred at 25^ꢁC for 1 h. The reaction mixture

was diluted with EtOAc (500 mL) and washed with saturated brine. The organic layer was dried over Na₂SO₄, ﬁltered and evaporated

to afford crude product. The crude product was puriﬁed by ﬂash silica chromatography, elution gradient 0 to 5% MeOH in DCM. Pure

fractions were evaporated to dryness to afford tert-butyl (5-(((S)-1-((2S,4R)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)benzyl)carbamoyl)

pyrrolidin-1-yl)-3,3-dimethyl-1-oxobutan-2-yl)amino)-5-oxopentyl)carbamate (8, 2.50 g, 85 %) as a yellow solid. ¹H NMR (300 MHz,

DMSO-d₆) d 8.99 (s, 1H), 8.57 (t, J = 6.0 Hz, 1H), 7.84 (d, J = 9.3 Hz, 1H), 7.48 – 7.33 (m, 4H), 6.78 (s, 1H), 5.13 (d, J = 3.5 Hz, 1H), 4.54

(d, J = 9.4 Hz, 1H), 4.50 – 4.32 (m, 3H), 4.22 (dd, J = 15.9, 5.4 Hz, 1H), 3.66 (d, J = 4.8 Hz, 2H), 2.89 (q, J = 6.3 Hz, 2H), 2.45 (s, 3H), 2.31 –

1.82 (m, 5H), 1.52 – 1.41 (m, 2H), 1.37 (s, 9H), 1.33 (d, J = 7.0 Hz, 1H), 0.94 (s, 9H). m/z: ES+ [M+H]+ = 630.

(2S,4R)-1-[(2S)-2-(5-aminopentanoylamino)-3,3-dimethyl-butanoyl]-4-hydroxy-N-[[4-(4-methylthiazol-5-yl)phenyl]methyl]pyrroli-

dine-2-carboxamide (9).

HCl (4.0 M in dioxane) (28.6 mL, 114.32 mmol) was added to tert-butyl (5-(((S)-1-((2S,4R)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)

benzyl)carbamoyl)pyrrolidin-1-yl)-3,3-dimethyl-1-oxobutan-2-yl)amino)-5-oxopentyl)carbamate (8, 2.40 g, 3.81 mmol) and the

mixture was stirred at 25^ꢁC for 2 h. Solvent was removed under reduced pressure. The crude product was puriﬁed by reverse phase

HPLC using a gradient of MeCN and water (containing 0.05% TFA w/w). Pure fractions were evaporated to dryness to afford (2S,4R)-

1-((S)-2-(5-aminopentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (9,

0.500 g, 23 %) as a white solid. ¹H NMR (300 MHz, DMSO-d₆) d 9.00 (s, 1H), 8.58 (t, J = 6.1 Hz, 1H), 7.92 (d, J = 9.3 Hz, 1H), 7.75

(s, 3H), 7.42 (p, J = 8.6, 8.2 Hz, 4H), 4.56 (d, J = 9.3 Hz, 1H), 4.43 (td, J = 15.0, 13.7, 9.6 Hz, 3H), 4.22 (dd, J = 16.0, 5.4 Hz, 1H),

3.85 - 3.39 (m, 2H), 2.79 (q, J = 6.2 Hz, 2H), 2.52 (s, 1H), 2.45 (s, 3H), 2.37 - 2.24 (m, 1H), 2.17 (dt, J = 12.5, 6.1 Hz, 1H), 2.11 -

1.99 (m, 1H), 1.91 (ddd, J = 13.0, 8.7, 4.6 Hz, 1H), 1.52 (dd, J = 8.6, 4.2 Hz, 4H), 0.95 (s, 9H). m/z (ES+), [M+H]+ = 530.

(2S,4R)-1-((S)-2-(5-(3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyr-

idin-2-yl)propanamido)pentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide

(PROTAC 1).

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e7

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

(2S,4R)-1-((S)-2-(5-aminopentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-car-

boxamide (9, 262 mg, 0.50 mmol) was dissolved in DMF (0.8 mL). HATU (7.91 mg, 0.02 mmol) was added, followed by DIPEA

(0.236 mL, 1.35 mmol), then 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-

methylpyridin-2-yl)propanoic acid (10, 202 mg, 0.45 mmol) was added and the reaction stirred for 16 h at 25^ꢁC. Solvents were

removed in vacuo and the residue was puriﬁed by preparative HPLC (Waters XSelect CSH C18 ODB column, 5m silica, 30 mm diam-

eter, 100 mm length), using decreasingly polar mixtures of water (containing by volume 1% NH₄OH (28-30% in H2O)) and MeCN as

eluents. Fractions containing the desired compound were evaporated to dryness to afford 3-((5-amino-2-(1H-pyrazol-5-yl)pyrazolo

[1,5-a]pyrimidin-7-yl)amino)propan-1-ol (PROTAC 1, 367 mg, 76%) as a white solid. ¹H NMR (300 MHz, MeOD-d₄) d 9.64 (s, 1H), 9.26

(s, 1H), 8.73 (t, J = 7.5 Hz, 1H), 8.10 (dd, J = 15.5, 7.5 Hz, 2H), 7.66 – 7.73 (m, 4H), 7.06 – 7.16 (m, 1H), 6.92 (dd, J = 8.6, 3.8 Hz, 1H), 4.7 –

4.9 (m, 7H), 4.62 (d, J = 15.5 Hz, 1H), 4.14 (d, J = 11.0 Hz, 1H), 4.05 (dd, J = 11.0, 3.8 Hz, 1H), 3.64 (m, 5H), 3.46 (d, J = 6.6 Hz, 2H), 3.09

(t, J = 6.6 Hz, 2H), 2.94 (d, J = 10.2 Hz, 3H), 2.73 (s, 3H), 2.56 (h, J = 7.0 Hz, 2H), 2.43 – 2.49 (m, 1H), 2.33 (ddd, J = 13.2, 9.2, 4.5 Hz, 1H),

1.82 (m, 4H), 1.28 (s, 9H). 5 exchangeable protons not observed. ¹³C NMR (126 MHz, MeOD-d₄) d 175.6, 174.5, 173.5, 172.3, 160.5,

160.3, 158.4, 157.6, 157.1, 156.5, 154.4, 153.3, 148.8, 146.4, 146.3, 140.5, 132.4, 131.2, 130.5, 130.5, 130.4, 130.4, 129.4, 129.0,

125.5, 122.2, 122.0, 115.2, 115.1, 110.0, 109.9, 107.4, 73.0, 71.1, 61.1, 60.8, 59.1, 58.0, 49.9, 49.7, 49.7, 43.7, 40.1, 39.1, 39.0,

39.0, 37.5, 36.6, 36.0, 34.6, 30.0, 29.8, 29.8, 29.7, 27.0, 24.1, 19.1, 15.6. HRMS (ESI+): m/z found [M+ H]+ 960.4323, C₅₀H₅₈N₁₁O₆SF

requires 960.4355.

Scheme 3. Synthesis of Intermediate 10.

Reagents and conditions: (a) (3-ethoxy-3-oxopropyl)zinc(II) bromide, Pd(PPh₃)₄, THF, 85^ꢁC, 16 h, 60%; (b) Bis(pinacolato)diboron,

Pd(dppf)Cl₂.CH₂Cl₂, , dioxane, 85^ꢁC, 8 h; (c) Pd(dppf)Cl₂, Cs2CO₃, dioxane/H₂O, 90^ꢁC, 4 h; (d) LiOH, THF/MeOH/H₂O, 30% over

3 steps.

e8 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

ethyl 3-(5-bromo-6-methylpyridin-2-yl)propanoate (11).

A mixture of Pd(PPh₃)₄(0.593 g, 0.51 mmol) and 3,6-dibromo-2-methylpyridine (1.95 g, 7.77 mmol) under N₂was treated with

(3-ethoxy-3-oxopropyl)zinc(II) bromide 0.5 M in THF (23.3 mL, 11.66 mmol) at rt. The reaction was heated to 85^ꢁC overnight. Solvents

were removed in vacuo and the residue was puriﬁed by ﬂash silica chromatography, elution gradient 0 to 25% EtOAc in heptane. Pure

fractions were evaporated to dryness to afford ethyl 3-(5-bromo-6-methylpyridin-2-yl)propanoate (11, 1.27 g, 60%) as a colourless

oil. ¹H NMR (400 MHz, CDCl₃) d 7.70 (1H, d), 6.92 (1H, d), 4.14 (2H, q), 3.05 (2H, t), 2.78 (2H, t), 2.64 (3H, s), 1.25 (3H, t). m/z: ES+

[M+H]+ = 272.

ethyl 3-(6-methyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-2-yl)propanoate (12).

Ethyl 3-(5-bromo-6-methylpyridin-2-yl)propanoate (11, 1.3 g, 4.78 mmol) was dissolved in degassed dioxane (23.88 mL) to which

potassium acetate (1.172 g, 11.94 mmol), Pd(dppf)Cl₂.CH₂Cl₂(0.390 g, 0.48 mmol) and 4,4,4’,4’,5,5,5’,5’-octamethyl-2,2’-bis(1,3,2-

dioxaborolane) (1.820 g, 7.17 mmol) were added. The reaction was heated to 100^ꢁC for 5 h, then additional Pd(dppf)Cl₂.CH₂Cl₂(0.390

g, 0.48 mmol) was added and the reaction heated for an additional 3 h and cooled to rt. The mixture was diluted with EtOAc (50 mL)

and ﬁltered through a plug of celite. Solvents were stripped in vacuo to afford the crude product 12 as a black tar which was used

immediately without further analysis or isolation.

3-[5-[5-[(5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methylamino]-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl]-6-methyl-2-pyridyl]propanoic

acid (10).

A mixture of crude ethyl 3-(6-methyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-2-yl)propanoate (12, 1.543 g,

4.83 mmol), 8-bromo-N-((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)-[1,2,4]triazolo[4,3-c]pyrimidin-5-amine (13, 1.10 g,

3.02 mmol), cesium carbonate (1.968 g, 6.04 mmol) and Pd(dppf)Cl₂(0.247 g, 0.30 mmol) in dioxane (12.59 mL) and water

(2.52 mL) was degassed with N₂for 5 min, then heated to 90^ꢁC under N₂for 4 h. The reaction was cooled to rt, diluted with

EtOAc (100 mL) and ﬁltered through celite. Solvents were removed in vacuo and the residue was puriﬁed by ﬂash silica chromatog-

raphy, elution gradient 0 to 7% MeOH in DCM. Pure fractions were evaporated to dryness to afford the intermediate ester as a

black gum. This was redissolved in 5 mL THF to which 2.5 mL H₂O and 2.5 mL MeOH were added followed by lithium hydroxide

hydrate (0.127 g, 3.02 mmol) and the reaction was stirred at rt for 1.5 h. Solvents were stripped in vacuo and the residue was trans-

ferred to a seperating funnel with 100 mL EtOAc and 50 mL 1M NaOH. The organic phase was discarded and the aqueous phase

brought to pH 7.0 with dropwise HCl. The organic phase was extracted with 10% MeOH in DCM (4 x 100 mL) and solvents removed

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e9

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

in vacuo to afford 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-

2-yl)propanoic acid (10, 0.400 g, 29.5 %) as an off-white solid. ¹H NMR (500 MHz, MeOD-d₄) d 9.35 (1H, s), 7.90 (1H, s), 7.72 (1H, s),

7.43 (1H, s), 6.75 – 6.94 (1H, m), 6.65 (1H, dd), 4.58 (2H, t), 3.40 (2H, t), 3.15 – 3.22 (2H, m), 2.82 (2H, t), 2.50 (3H, s). 2Hs obscured by

solvent peak. m/z: ES+ [M+H]+ = 449.

Scheme 4. Synthesis of cpd 3, PROTAC 2.

Reagents and conditions: (a) methyl 2-bromoacetate, K₂CO₃, DMF, rt, 4 h, 85%; (b) Pd/C, H2, 50^ꢁC, MeOH, 97%; (c) Cs₂CO₃,

DMF, rt, 2 days, 79%; (d) NaOH aq., MeOH, rt, 16 h, 94%; (e) HATU, Et₃N, DMF, 2 h, 96%; (f) 4M HCl/dioxane, DCM, rt, 1 h,

86%; (g) HATU, Et₃N, DMF, rt, 16 h, 67%.

Methyl 2-(4-(benzyloxy)phenoxy)acetate (14).

e10 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

K₂CO₃(4.14 g, 29.96 mmol) was added to a mixture of 4-(benzyloxy)phenol (2.0 g, 9.99 mmol) and methyl 2-bromoacetate (1.681 g,

10.99 mmol) in DMF (30 mL). The resulting mixture was stirred at 25^ꢁC for 4 h. The reaction mixture was diluted with EtOAc (50 mL),

and washed sequentially with water (50 mL) then saturated brine (20 mL). The organic layer was dried over Na₂SO₄, ﬁltered and evap-

orated to afford the crude product which was puriﬁed by ﬂash silica chromatography, elution gradient 0 to 10% EtOAc in petroleum

ether. Pure fractions were evaporated to dryness to afford methyl 2-(4-(benzyloxy)phenoxy)acetate (14, 2.300 g, 85 %) as a white

solid. ¹H NMR (300 MHz, MeOD-d₄) d 7.72 – 7.09 (m, 5H), 7.02 – 6.64 (m, 4H), 5.02 (s, 2H), 4.64 (s, 2H), 3.77 (s, 3H). m/z: ES+

[M+H]+ = 295.

ethyl 2-(4-hydroxyphenoxy)acetate (15).

10% Pd/C (0.907 g, 8.52 mmol) was added to a solution of methyl 2-(4-(benzyloxy)phenoxy)acetate (14, 4.64 g, 17.04 mmol) in

MeOH (3 mL).The resulting mixture was stirred at 50^ꢁC for 16 h under a hydrogen atmosphere. The reaction mixture was ﬁltered

through silica gel and solvents evaporated to dryness to afford methyl 2-(4-hydroxyphenoxy)acetate (15, 3.00 g, 97 %) as a white

solid. ¹H NMR (300 MHz, MeOD-d₄) d 6.88 – 6.50 (m, 4H), 4.59 (s, 2H), 3.76 (s, 3H). m/z (ES+), [M+Na]+ = 205.

ethyl 2-[4-[2-(tert-butoxycarbonylamino)ethoxy]phenoxy]acetate (16).

To a solution of ethyl 2-(4-hydroxyphenoxy)acetate (15, 240 mg, 1.22 mmol) in DMF (3 mL) was added cesium carbonate (399 mg,

1.22 mmol). The mixture was stirred at rt for 30 min, then tert-butyl (2-bromoethyl)carbamate (356 mg, 1.59 mmol) was added neat

and the reaction was stirred at rt for 2 days. The reaction mixture was diluted with EtOAc and washed with 1M HCl then sat aq.

NaHCO3. The organic phase was dried over Na₂SO₄, ﬁltered and concentrated to afford ethyl 2-(4-(2-((tert-butoxycarbonyl)

amino)ethoxy)phenoxy)acetate (16, 329 mg, 79 %) as a yellow oil. ¹H NMR (500 MHz, CD₂Cl₂-d₂) d 6.83 (m, 2H), 6.77 (d, 2H),

4.90 - 5.08 (s, 1H), 4.55 (s, 2H), 4.22 (q, 2H), 3.95 (t, 2H), 3.41 - 3.53 (m, 2H) 1.40 - 1.50 (m, 9H), 1.27 (t, 3H). m/z (ES+), [M+Na]+ = 348.

2-[4-[2-(tert-butoxycarbonylamino)ethoxy]phenoxy]acetic acid (17).

To a solution of ethyl 2-(4-(2-((tert-butoxycarbonyl)amino)ethoxy)phenoxy)acetate (16, 320 mg, 0.94 mmol) in MeOH (5 mL) was

added 1M aq. NaOH (3 mL, 3.00 mmol) and the reaction was stirred overnight at rt. The reaction mixture was diluted with EtOAc

and washed with 1M HCl, then brine. The organic phase was dried over Na₂SO₄, ﬁltered and concentrated to afford 2-(4-(2-((tert-

butoxycarbonyl)amino)ethoxy)phenoxy)acetic acid (17, 275 mg, 94 %) which was used without further puriﬁcation. m/z: ES- [M-

H]- = 310.

tert-butyl N-[2-[4-[2-[[(1S)-1-[(2S,4R)-4-hydroxy-2-[[4-(4-methylthiazol-5-yl)phenyl]methylcarbamoyl]pyrrolidine-1-carbonyl]-

2,2-dimethyl-propyl]amino]-2-oxo-ethoxy]phenoxy]ethyl]carbamate (18).

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e11

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Et₃N (1.209 mL, 8.67 mmol) was added to a mixture of (2S,4R)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (7, 1369 mg, 3.18 mmol), 2-(4-(2-((tert-butoxycarbonyl)amino)ethoxy)phenoxy)acetic

acid (17, 900 mg, 2.89 mmol) and HATU (1649 mg, 4.34 mmol) in DMF (20 mL). The resulting mixture was stirred at rt for 2 h. The

crude product was puriﬁed by ﬂash silica chromatography, elution gradient 0 to 10% MeOH in DCM. Pure fractions were evaporated

to dryness to afford tert-butyl (2-(4-(2-(((S)-1-((2S,4R)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)benzyl)carbamoyl)pyrrolidin-1-yl)-3,3-

dimethyl-1-oxobutan-2-yl)amino)-2-oxoethoxy)phenoxy)ethyl)carbamate (18, 2000 mg, 96 %) as a yellow solid. ¹H NMR

(400 MHz, MeOD-d₄) d 8.90 (s, 1H), 7.56 – 7.34 (m, 4H), 7.03 – 6.77 (m, 4H), 4.67 – 4.31 (m, 5H), 4.06 – 3.75 (m, 4H), 3.41 (t, J =

5.7 Hz, 2H), 3.01 (s, 2H), 2.49 (s, 3H), 2.33 – 2.05 (m, 2H), 1.46 (s, 9H), 1.37 – 0.92 (m, 9H). m/z: ES+ [M+H]+ = 724.

(2S,4R)-1-((S)-2-(2-(4-(2-aminoethoxy)phenoxy)acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)

pyrrolidine-2-carboxamide (19).

4N HCl in dioxane (15 g, 4.84 mmol) was added to tert-butyl (2-(4-(2-(((S)-1-((2S,4R)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)benzyl)

carbamoyl)pyrrolidin-1-yl)-3,3-dimethyl-1-oxobutan-2-yl)amino)-2-oxoethoxy)phenoxy)ethyl)carbamate (18, 3.5 g, 4.84 mmol) in

DCM (30 mL). The resulting solution was stirred at rt for 1 h. Solvents were evaporated to afford crude product, which puriﬁed by

C18-ﬂash chromatography, elution gradient 0 to 40% MeCN in water. Pure fractions were evaporated to dryness to afford

(2S,4R)-1-((S)-2-(2-(4-(2-aminoethoxy)phenoxy)acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)

pyrrolidine-2-carboxamide (19, 2.60 g, 86 %) as a white solid. ¹H NMR (300 MHz, MeOD-d₄) d 9.34 (d, J = 2.0 Hz, 1H), 7.78 – 7.21 (m,

4H), 6.97 (s, 4H), 4.73 (s, 1H), 4.56 (ddd, J = 12.4, 6.3, 2.3 Hz, 5H), 4.38 (d, J = 15.7 Hz, 1H), 4.17 (dd, J = 5.5, 4.4 Hz, 2H), 3.91 (d, J =

11.2 Hz, 1H), 3.81 (dd, J = 11.2, 3.7 Hz, 1H), 3.35 (d, J = 5.0 Hz, 2H), 2.52 (s, 3H), 2.18 (dd, J = 36.4, 6.1 Hz, 1H), 2.06 (d, J = 14.6 Hz,

1H), 0.99 (s, 9H). m/z (ES+), [M+H]+ = 624.

(2S,4R)-1-[(2S)-2-[[2-[4-[2-[3-[5-[5-[(5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methylamino]-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl]-6-

methyl-2-pyridyl]propanoylamino]ethoxy]phenoxy]acetyl]amino]-3,3-dimethyl-butanoyl]-4-hydroxy-N-[[4-(4-methylthiazol-5-yl)

phenyl]methyl]pyrrolidine-2-carboxamide (PROTAC 2).

(2S,4R)-1-((S)-2-(2-(4-(2-aminoethoxy)phenoxy)acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)

pyrrolidine-2-carboxamide (19, 27.8 mg, 0.04 mmol) was dissolved in DMF (0.8 mL). HATU (7.91 mg, 0.02 mmol) was added, followed

by Et₃N (0.019 mL, 0.13 mmol), then 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-

yl)-6-methylpyridin-2-yl)propanoic acid (10, 20 mg, 0.04 mmol) and the reaction stirred for 16 h at rt. Solvents were removed in vacuo

and the mixture was puriﬁed by C18-ﬂash chromatography, elution gradient 20 to 80 % MeCN in water (0.1 % formic acid). The pu-

riﬁed fractions were combined and evaporated to afford the desired product (2S,4R)-1-[(2S)-2-[[2-[4-[2-[3-[5-[5-[(5-ﬂuoro-2,3-dihy-

drobenzofuran-4-yl)methylamino]-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl]-6-methyl-2-pyridyl]propanoylamino]ethoxy]phenoxy]acetyl]

amino]-3,3-dimethyl-butanoyl]-4-hydroxy-N-[[4-(4-methylthiazol-5-yl)phenyl]methyl]pyrrolidine-2-carboxamide (PROTAC 2, 35 mg,

67%) as an off-white solid. ¹H NMR (500 MHz, MeOD-d₄) d 9.37 (s, 1H), 8.97 (s, 1H), 8.28 (d, J = 8.2 Hz, 1H), 7.79 (t, J = 4.3 Hz, 2H),

7.4 – 7.49 (m, 4H), 6.8 – 6.87 (m, 5H), 6.65 (dd, J = 8.7, 3.8 Hz, 1H), 4.71 (s, 1H), 4.41 – 4.61 (m, 8H), 4.35 (d, J = 15.4 Hz, 1H), 3.94 (t, J =

5.2 Hz, 2H), 3.87 (s, 1H), 3.80 (dd, J = 11.0, 3.6 Hz, 1H), 3.55 (t, J = 5.1 Hz, 2H), 3.39 (m, 5H), 2.76 – 2.87 (m, 2H), 2.67 (s, 3H), 2.47 (s,

e12 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

3H), 2.17 – 2.27 (m, 1H), 2.01 - 2.10 (m, 1H), 0.98 (s, 9H). 5 exchangeable protons not observed. ¹³C NMR (125 MHz, MeOD-d₄)

d 216.5, 216.3, 216.0, 215.8, 174.3, 173.6, 171.8, 170.6, 160.8, 160.5, 158.4, 157.6, 156.9, 156.5, 155.2, 154.4, 153.2,153.2,

150.0, 148.7, 148.4, 146.4, 146.3, 140.5, 134.4, 133.8, 132.4, 131.2, 130.5, 130.5, 130.4, 129.5, 129.0, 125.4, 116.9, 116.6, 115.3,

115.1, 110.0, 109.9, 107.3, 73.0, 71.1, 68.7, 68.0, 60.9, 58.3, 58.2, 49.7, 48.3, 43.7, 40.4, 39.1, 39.1, 39.0, 37.1, 34.6, 30.1, 29.8,

26.9, 19.1, 15.5. HRMS (ESI+): m/z found [M+ H]+ 1054.4393, C₅₅H₆₀N₁₁O₈FS requires 1054.4409.

Scheme 5. Synthesis of cpd 4, Capped EED Ligand.

Reagents and conditions: (a) Methylamine, HATU, DIPEA, DMF, rt, 86%.

3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-2-yl)-N-methyl-

propanamide (Capped EED Ligand).

To a solution of 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-

2-yl)propanoic acid (10, 30 mg, 0.07 mmol), N-ethyl-N-isopropylpropan-2-amine (35.5 ml, 0.20 mmol) and methanamine 33% in

ethanol (9.99 ml, 0.08 mmol) in DMF (289 ml) was added HATU (33.1 mg, 0.09 mmol) in 3 portions over 5 min and the reaction was

stirred at rt overnight. The reaction mixture was diluted with 1 mL DMSO and puriﬁed by preparative HPLC using decreasingly polar

mixtures of water (containing 1% NH₃) and MeCN as eluents. Fractions containing the desired compound were evaporated to dry-

ness to afford 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-2-

yl)-N-methylpropanamide (Capped EED ligand, 26.5 mg, 86 %) as a white solid. ¹H NMR (500 MHz, DMSO-d₆) d 9.47 (s, 1H),

8.68 (s, 1H), 7.83 (d, J = 4.4 Hz, 1H), 7.54 – 7.69 (m, 2H), 7.17 (d, J = 7.8 Hz, 1H), 6.91 – 7.02 (m, 1H), 6.73 (dd, J = 8.6, 3.9 Hz,

1H), 4.72 (d, J = 3.1 Hz, 2H), 4.56 (t, J = 8.7 Hz, 2H), 3.33 – 3.36 (m, 2H), 2.94 – 3.01 (m, 2H), 2.59 (d, J = 4.6 Hz, 3H), 2.53 (s, 2H),

2.37 (s, 3H). ¹³C NMR (125 MHz, DMSO-d₆) d 172.24, 159.95, 156.71, 156.42, 156.24, 154.83, 149.18, 144.11, 142.43, 139.01,

133.37, 129.81, 129.78, 126.64, 121.91, 121.77, 120.22, 114.52, 114.33, 110.93, 108.96, 108.89, 72.06, 40.91, 37.78, 37.75,

35.28, 33.55, 29.01, 25.96, 23.74. HRMS (ESI+): m/z found [M+ H]+ 462.2066, C₂₄H₂₄N₇O₂F requires 462.2054.

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e13

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Scheme 6. Synthesis of cpd 5, Inactive PROTAC 1.

Reagents and conditions: (a) HATU, DIPEA, DMF, rt; (b) HCl, dioxane, rt, 43% over 2 steps; (c) HATU, Et₃N, DMF, 92%.

(2R,4S)-1-((S)-2-(5-aminopentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carbox-

amide (21).

5-((tert-butoxycarbonyl)amino)pentanoic acid (30 mg, 0.14 mmol) was dissolved in DMF (0.8 ml). HATU (7.91 mg, 0.02 mmol) was

added, followed by DIPEA (0.049 mL, 0.28 mmol), then (2R,4S)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthia-

zol-5-yl)benzyl)pyrrolidine-2-carboxamide (20, 60 mg, 0.14 mmol) and the reaction stirred for 16 h at rt. The reaction was partitioned

between EtOAc and water. The organic layer was extracted twice with EtOAc, washed with brine, dried over Na₂SO₄and solvents

removed in vacuo. The boc-protected intermediate was puriﬁed by ﬂash chromatography over silica gel using a gradient of 0 - 10%

MeOH in DCM. This material was dissolved in DCM (1 mL) to which 4N HCl in dioxane was added (0.5 mL). The reaction was stirred

for 2 h, evolving a white precipitate. The reaction was evaporated to dryness and puriﬁed by SCX ion-exchange chromatography to

afford the desired product (2R,4S)-1-((S)-2-(5-aminopentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)

benzyl)pyrrolidine-2-carboxamide (21, 32 mg, 43%) as an off-white solid. ¹H NMR (500 MHz, MeOD-d₄) d 9.31 (s, 1H), 8.83 (s,

1H), 7.63 (s, 2H), 7.38 – 7.42 (m, 2H), 7.36 (d, J = 8.5 Hz, 2H), 7.18 (d, J = 7.8 Hz, 1H), 6.76 – 6.93 (m, 1H), 6.65 (dd, J = 8.7,

3.9 Hz, 1H), 4.81 (s, 2H), 4.57 (q, J = 8.5 Hz, 3H), 4.43 – 4.51 (m, 3H), 4.34 (d, J = 15.6 Hz, 1H), 3.98 (dd, J = 10.8, 4.9 Hz, 1H),

3.70 (dd, J = 10.6, 3.0 Hz, 1H), 3.39 (t, J = 8.7 Hz, 2H), 3.07 (q, J = 6.9 Hz, 4H), 2.58 (t, J = 7.6 Hz, 2H), 2.45 (s, 3H), 2.40 (s, 3H),

2.17 – 2.31 (m, 2H), 2.09 – 2.18 (m, 1H), 2.01 – 2.1 (m, 1H), 1.34 – 1.49 (m, 4H), 1.06 (s, 9H). m/z (ES+), [M+H]+ = 530.

(2R,4S)-1-((S)-2-(5-(3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyr-

idin-2-yl)propanamido)pentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide

(Inactive PROTAC 1).

e14 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

(2R,4S)-1-((S)-2-(5-aminopentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-car-

boxamide (21, 5.9 mg, 0.01 mmol) was dissolved in DMF (0.8 mL). HATU (7.91 mg, 0.02 mmol) was added, followed by Et₃N (4.7 ml,

0.03 mmol) and then 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyr-

idin-2-yl)propanoic acid (10, 5 mg, 0.01 mmol) was added and the reaction was stirred for 16 h at rt. Solvents were removed in vacuo

and the mixture was puriﬁed by C18-ﬂash chromatography, elution gradient 20 to 80 % MeCN in water (0.1 % formic acid). The pu-

riﬁed fractions were combined and evaporated to afford the desired product (2R,4S)-1-((S)-2-(5-(3-(5-(5-(((5-ﬂuoro-2,3-dihydroben-

zofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-2-yl)propanamido)pentanamido)-3,3-dimethylbuta-

noyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (Inactive PROTAC 1, 11 mg, 92%) as a white powder.

¹H NMR (300 MHz, MeOD-d₄) d 9.31 (s, 1H), 8.83 (s, 1H), 7.63 (s, 2H), 7.34 – 7.44 (m, 4H), 7.18 (d, J = 7.8 Hz, 1H), 6.8 – 6.92 (m, 1H),

6.65 (dd, J = 8.6, 3.9 Hz, 1H), 4.81 (s, 2H), 4.57 (q, J = 8.6 Hz, 3H), 4.42 – 4.49 (m, 3H), 4.34 (d, J = 15.6 Hz, 1H), 3.98 (dd, J = 10.8,

4.9 Hz, 1H), 3.70 (dd, J = 10.6, 3.0 Hz, 1H), 3.39 (t, J = 8.7 Hz, 2H), 3.07 (t, J = 7.5 Hz, 4H), 2.58 (t, J = 7.5 Hz, 2H), 2.45 (s, 3H), 2.40 (s,

3H), 2.18 – 2.31 (m, 2H), 2.09 – 2.18 (m, 1H), 1.96 – 2.09 (m, 1H), 1.34 – 1.48 (m, 4H), 1.06 (s, 9H). 5 exchangeable protons not

13

observed. ¹³C NMR (125 MHz, DMSO-d₆) C NMR (126 MHz, DMSO, 27^ꢁC) 173.37, 172.12, 171.63, 170.23, 159.93, 156.71,

156.39, 156.23, 154.83, 151.91, 149.17, 148.23, 144.13, 142.44, 139.81, 138.97, 133.38, 131.59, 129.27, 127.84, 126.62, 121.95,

121.80, 120.17, 114.51, 114.32, 108.94, 108.87, 72.05, 68.93, 59.27, 57.49, 55.76, 42.04, 38.72, 38.36, 37.80, 35.40, 34.84,

34.73, 33.63, 29.20, 29.01, 26.95, 23.71, 23.31, 16.43. HRMS (ESI+): m/z found [M+ H]+ 960.4323, C₅₀H₅₈N₁₁O₆SF requires

960.4355.

Scheme 7. Synthesis of cpd 6, Inactive PROTAC 2.

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e15

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

Reagents and conditions: (a) HATU, DIPEA, DMF, rt, 2 h, 22%; (b) 4M HCl/dioxane, DCM, rt, 2 h, 95%; (c) HATU, DIPEA, DMF, rt, 16

h, 32%.

tert-butyl (2-(4-(2-(((S)-1-((2R,4S)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)benzyl)carbamoyl)pyrrolidin-1-yl)-3,3-dimethyl-1-oxobutan-

2-yl)amino)-2-oxoethoxy)phenoxy)ethyl)carbamate (22).

To 2-(4-(2-((tert-butoxycarbonyl)amino)ethoxy)phenoxy) acetic acid (17, 31.8 mg, 0.10 mmol) in DMF (415 ml) was added

HATU (42.4 mg, 0.11 mmol) and DIPEA (49.4 ml, 0.28 mmol) and the reaction was stirred for 30 min. To the solution was added

(2R,4S)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (20, 40 mg,

0.09 mmol) and the reaction was stirred overnight at rt. The reaction mixture was diluted with 1 mL DMSO and puriﬁed by preparative

HPLC using decreasingly polar mixtures of water (containing 1% NH3) and MeCN as eluents. Fractions containing the desired com-

pound were evaporated to dryness to afford tert-butyl (2-(4-(2-(((S)-1-((2R,4S)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)benzyl)carba-

moyl)pyrrolidin-1-yl)-3,3-dimethyl-1-oxobutan-2-yl)amino)-2-oxoethoxy)phenoxy)ethyl)carbamate (22, 15.0 mg, 22%) as a dry

1

ﬁlm. H NMR (500 MHz, CD₂Cl₂-d₂) d 8.91 (s, 1 H), 7.37 - 7.42 (m, 1 H), 7.31 - 7.35 (m, 4 H) 7.11 - 7.15 (m, 1 H), 6.61 - 6.82 (m, 4

H), 4.54 - 4.83 (m, 2 H), 4.32 - 4.42 (m, 1 H), 4.17 - 4.27 (m, 2 H), 4.09 - 4.15 (m, 1 H), 3.92 - 3.99 (m, 2 H), 3.64 - 3.84 (m, 2 H),

3.46 (t, J=5.3 Hz, 2 H), 2.46 (s, 3H), 2.37 - 2.42 (m, 1 H), 2.15 - 2.30 (m, 1 H), 1.46 (s, 9 H), 1.05 (s, 9 H). 3 exchangable protons

not observed. m/z: ES+ [M+H]+ = 724.

(2R,4S)-1-((S)-2-(2-(4-(2-aminoethoxy)phenoxy)acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)

pyrrolidine-2-carboxamide (23).

To tert-butyl (2-(4-(2-(((S)-1-((2R,4S)-4-hydroxy-2-((4-(4-methylthiazol-5-yl)benzyl)carbamoyl)pyrrolidin-1-yl)-3,3-dimethyl-1-oxo-

butan-2-yl)amino)-2-oxoethoxy)phenoxy)ethyl)carbamate (22, 15 mg, 0.02 mmol) was added HCl (4N in Dioxane) (3 mL,

12.00 mmol) and the reaction was stirred for 2 at rt. Solvents were removed in vacuo to afford the crude product. This was puriﬁed

by SCX ion-exchange chromatography, eluting with 1N Ammonia in MeOH to afford (2R,4S)-1-((S)-2-(2-(4-(2-aminoethoxy)phenoxy)

acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (23, 12.30 mg, 95 %) as

a dry ﬁlm. ES+ [M+H]+ = 624.

(2R,4S)-1-((S)-2-(2-(4-(2-(3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-

methylpyridin-2-yl)propanamido)ethoxy)phenoxy)acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)

pyrrolidine-2-carboxamide (Inactive PROTAC 2).

e16 Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020

Please cite this article in press as: Hsu et al., EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex, Cell Chemical Biology

(2019), https://doi.org/10.1016/j.chembiol.2019.11.004

To 3-(5-(5-(((5-ﬂuoro-2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-2-yl)propa-

noic acid (10, 10.4 mg, 0.02 mmol) in DMF (1.1 mL) was added O-(7-Azabenzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexaﬂuor-

ophosphate (10.58 mg, 0.03 mmol) and DIPEA (0.021 mL, 0.12 mmol). The reaction was stirred at rt for 40 min, then(2R,4S)-1-((S)-2-

(2-(4-(2-aminoethoxy)phenoxy)acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-car-

boxamide (23, 14.90 mg, 0.02 mmol) was added and the reaction was stirred at rt for 2 days. Solvents were removed in vacuo and the

crude product was puriﬁed by preparative HPLC using decreasingly polar mixtures of water (containing 1% NH3) and MeCN as el-

uents. Fractions containing the desired compound were evaporated to dryness to afford (2R,4S)-1-((S)-2-(2-(4-(2-(3-(5-(5-(((5-ﬂuoro-

2,3-dihydrobenzofuran-4-yl)methyl)amino)-[1,2,4]triazolo[4,3-c]pyrimidin-8-yl)-6-methylpyridin-2-yl)propanamido)ethoxy)phenoxy)

acetamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (Inactive PROTAC 2,

7.7 mg, 31.5%) as a white solid. ¹H NMR (500 MHz, DMSO-d₆) d 9.47 (s, 1H), 8.97 (s, 1H), 8.68 (t, J = 5.5 Hz, 1H), 8.33 (t, J =

5.5 Hz, 1H), 8.16 (t, J = 5.5 Hz, 1H), 7.81 (d, J = 9.0 Hz, 1H), 7.60 (m, 2H), 7.42 (d, J = 8.0 Hz, 2H), 7.34 (d, J = 8.0 Hz, 2H), 7.16

(d, J = 8.0 Hz, 1H), 6.97 (t, J = 9.0 Hz, 1H), 6.83 (m, 4H), 6.72 (dd, J = 8.6, 3.8 Hz, 1H), 5.13 (d, J = 3.8 Hz, 1H), 4.72 (d, J = 4.8 Hz,

2H), 4.51 – 4.59 (m, 3H), 4.48 (d, J = 2.6 Hz, 2H), 4.28 – 4.42 (m, 3H), 4.23 (dd, J = 15.8, 5.7 Hz, 1H), 3.89 (t, J = 5.7 Hz, 2H), 3.75

(dd, J = 10.4, 4.8 Hz, 1H), 3.57 (dd, J = 10.4, 2.6 Hz, 1H), 3.41 (q, J = 5.3 Hz, 2H), 3.34 (d, J = 8.7 Hz, 2H), 2.99 (t, J = 7.7 Hz, 2H),

2.57 (t, J = 7.8 Hz, 2H), 2.42 (s, 3H), 2.36 (s, 3H), 2.07 (ddd, J = 12.8, 8.2, 4.5 Hz, 1H), 1.94 (dt, J = 12.3, 5.7 Hz, 1H), 0.94 (s, 9H).

¹³C NMR (125 MHz, DMSO-d₆) d 203.54, 187.06, 171.7, 171.51, 168.99, 167.69, 159.36, 155.88, 155.72, 155.2, 154.32, 152.87,

151.74, 151.38, 148.65, 147.72, 143.59, 141.89, 139.33, 138.46, 131.05, 129.74, 128.74, 127.39, 126.11, 121.39, 121.25, 119.69,

115.47, 115.43, 115.37, 114.00, 113.8, 113.32, 111.20, 110.41, 108.44, 108.37, 71.53, 68.48, 66.99, 66.79, 58.77, 56.12, 55.38,

41.49, 38.24, 37.85, 37.26, 37.24, 34.97, 34.73, 32.99, 28.49, 26.31, 26.24, 23.2, 15.89, 1.06. HRMS (ESI+): m/z found [M+ H]+

1054.4393, C₅₅H₆₀N₁₁O₈FS requires 1054.4409.

QUANTIFICATION AND STATISTICAL ANALYSIS

Statistical analysis was performed using Prism 8 (Graphpad) or Genedata Screener software (Genedata AG). Values are expressed as

mean standard deviation from the mean or SEM of the replicates, as indicated in the ﬁgure legends.

DATA AND CODE AVAILABILITY

The accession number for the proteomic dataset generated for Figure 5 is ProteomeXchange PRIDE: PXD016107 (http://

proteomecentral.proteomexchange.org/cgi/GetDataset).

Cell Chemical Biology 27, 1–6.e1–e17, January 16, 2020 e17

Article Doi

EED-Targeted PROTACs Degrade EED, EZH2, and SUZ12 in the PRC2 Complex

DOI: 10.1016/j.chembiol.2019.11.004

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.chembiol.2019.11.004

Full text of DOI:10.1016/j.chembiol.2019.11.004

Products guided by the article

R&D Labs maybe for 81466-35-5

Relevant to this article

Hot Product