Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription

Article abstract of DOI:10.1016/j.chembiol.2019.02.009

The transcription factor Max is a basic-helix-loop-helix leucine zipper (bHLHLZ) protein that forms homodimers or interacts with other bHLHLZ proteins, including Myc and Mxd proteins. Among this dynamic network of interactions, the Myc/Max heterodimer has

Full text of DOI:10.1016/j.chembiol.2019.02.009

Article
Stabilization of the Max Homodimer with a Small
Molecule Attenuates Myc-Driven Transcription
Graphical Abstract
Authors
Nicholas B. Struntz, Andrew Chen,
Anja Deutzmann, ..., Charles Y. Lin,
Dean W. Felsher, Angela N. Koehler
Correspondence
koehler@mit.edu
In Brief
Myc/Max-mediated transcription is
deregulated in most of human cancers.
Struntz et al. discovered a small molecule
that stabilizes the Max homodimer and
attenuates Myc-driven transcription with
efficacy in cellular and murine cancer
models. This discovery reinforces an
alternative Myc-targeting strategy and
could inform development of compounds
to treat Myc-dependent cancers.
Highlights
d
KI-MS2-008 is a Max-binding small molecule that attenuates
Myc-driven transcription
d
d
The compound stabilizes the Max homodimer
Effects on DNA occupancy and the transcriptome resemble
loss of Myc
d
Treatment with KI-MS2-008 exhibits efficacy in cellular and
murine cancer models
Struntz et al., 2019, Cell Chemical Biology 26, 1–13
May 16, 2019 ª 2019 Elsevier Ltd.
https://doi.org/10.1016/j.chembiol.2019.02.009

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Cell Chemical Biology
Article
Stabilization of the Max Homodimer
with a Small Molecule Attenuates
Myc-Driven Transcription
Nicholas B. Struntz,1,2,3,13 Andrew Chen,1,2,3,13 Anja Deutzmann,4 Robert M. Wilson,1,2,3 Eric Stefan,1,2 Helen L. Evans,1,2,3
Maricela A. Ramirez,5 Tong Liang,5 Francisco Caballero,1,2 Mattheus H.E. Wildschut,1,2 Dylan V. Neel,1,2
David B. Freeman,1,2,6 Marius S. Pop,1,2,6 Marie McConkey,7,8,9 Sandrine Muller,9 Brice H. Curtin,1,2,3 Hanna Tseng,1,2
Kristen R. Frombach,1,2 Vincent L. Butty,1,10 Stuart S. Levine,1,10 Clementine Feau,9 Sarah Elmiligy,1 Jiyoung A. Hong,1,2,11
Timothy A. Lewis,9 Amedeo Vetere,9 Paul A. Clemons,9 Scott E. Malstrom,1 Benjamin L. Ebert,7,8,9 Charles Y. Lin,5,12
Dean W. Felsher,4 and Angela N. Koehler1,2,3,9,14,
*
1David H. Koch Institute for Integrative Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02142, USA
2Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
3MIT Center for Precision Cancer Medicine, Massachusetts Institute of Technology, Cambridge, MA 02142, USA
4Division of Oncology, Departments of Medicine and Pathology Stanford School of Medicine, Stanford, CA 94305, USA
5Department of Molecular and Human Genetics and Verna & Marrs McLean Department of Biochemistry and Molecular Biology, Baylor
College of Medicine, Houston, TX 77030, USA
6Kronos Bio, Inc., Cambridge, MA 02139, USA
7Division of Hematology, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115, USA
8Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA
9Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
10BioMicro Center, Massachusetts Institute of Technology, Cambridge, MA 02142, USA
11Division of Pediatric Hematology/Oncology, Boston Children’s Hospital, Boston, MA 02115, USA
12Therapeutic Innovation Center, Baylor College of Medicine, Houston, TX 77030, USA
13These authors contributed equally
14Lead Contact
*Correspondence: koehler@mit.edu
https://doi.org/10.1016/j.chembiol.2019.02.009
SUMMARY
INTRODUCTION
The transcription factor Max is a basic-helix-loop- Max is a transcription factor that acts as a critical hub in a
bHLHLZ transcription factor network capable of binding different
interaction partners in response to stimuli (Conacci-Sorrell et al.,
helix leucine zipper (bHLHLZ) protein that forms
homodimers or interacts with other bHLHLZ pro-
teins, including Myc and Mxd proteins. Among
this dynamic network of interactions, the Myc/
Max heterodimer has crucial roles in regulating
normal cellular processes, but its transcriptional
activity is deregulated in a majority of human
cancers. Despite this significance, the arsenal of
2014). Most notably among these is Myc, a master regulator of
growth and proliferation. Myc binds DNA as an obligate hetero-
dimer with Max, and this heterodimer recognizes canonical
E-box sequences (CACGTG) (Blackwell et al., 1990) abundant
throughout the genome and enriched at promoters of growth
and proliferation genes. Although Max preferentially heterodi-
merizes with Myc, Max can also form homodimers that effec-
tively block Myc/Max heterodimers from binding to DNA, thus
high-quality chemical probes to interrogate these
proteins remains limited. We used small molecule
´
attenuating transcriptional activity (Figure 1A) (Cascon and
microarrays to identify compounds that bind Max Robledo, 2012). In addition, Max can form heterodimers with
the Mxd family of proteins (Nair and Burley, 2003), which recruit
additional transcriptional repressors to actively shut down
in a mechanistically unbiased manner. We discov-
ered the asymmetric polycyclic lactam, KI-MS2-
008, which stabilizes the Max homodimer while
reducing Myc protein and Myc-regulated transcript
levels. KI-MS2-008 also decreases viable cancer
cell growth in a Myc-dependent manner and
suppresses tumor growth in vivo. This approach
demonstrates the feasibility of modulating Max
transcription (Laherty et al., 1997). Deregulation of Myc/Max-
mediated transcription can lead to tumorigenesis through activa-
tion of cardinal hallmark features of cancer including cellular
growth, proliferation, self-renewal, and evasion of apoptosis
(Dang, 2012; Gabay et al., 2014). As such, a longstanding goal
of the cancer research community has been to directly target
Myc. Inactivation of Myc in model systems with engineered
with small molecules and supports altering Max
dimerization as an alternative approach to target-
ing Myc.
gene regulation (Felsher and Bishop, 1999) or expression of
the dominant negative peptide Omomyc appears to elicit onco-
gene addiction and tumor regression (Soucek et al., 2002), and,
Cell Chemical Biology 26, 1–13, May 16, 2019 ª 2019 Elsevier Ltd.
1

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 1. Discovery of a Putative Max Binder
(A) Cartoon diagram of the Max (blue) dimerization network displaying transcriptional activation by the Myc (green)/Max heterodimer, transcriptional attenuation
by the Max/Max homodimer, and transcriptional repression by Max/Mxd (purple) heterodimer.
(B) Schematic of a small molecule microarray (SMM) showing SMM positives (red spots), which were detected by Alexa Fluor 647-labeled antibodies against the
His-tag on the purified Max protein.
(C) SMM screening results for DIV06 compound library, which is comprised mostly of products of diversity-oriented synthesis (n = 3 technical replicates).
Histogram of composite Z scores and 3D scatterplot showing Z scores of feature to background ratios in triplicate for compounds on the SMMs.
(D) Prioritization scheme for Max binders that modulate Myc-driven transcription leading to hit stock solution BRD-K19261677 (resynthesized as KI-MS2-001)
and KI-MS2-008 as a more potent and synthetically accessible probe.
See also Figures S1–S4 and Table S1.
in fact, Myc is one of the few oncogenes for which true oncogene Max interaction directly or by inducing either Max/Max homo-
addiction has been demonstrated in vivo (Felsher, 2010; Soucek dimers or Max/Mxd heterodimers (Conacci-Sorrell et al., 2014).
et al., 2008). Myc is highly regulated and degrades rapidly (Farrell The vast majority of small molecule inhibitors developed to
and Sears, 2014), so a dynamic equilibrium shift away from Myc/ inhibit Myc transcriptional activity have focused on disrupting
Max transcriptional activation can occur by inhibiting the Myc/ the Myc/Max interaction (ex. 10058-F4, 10074G5, KJ-Pyr-9,
2
Cell Chemical Biology 26, 1–13, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
and sAJM589) (Choi et al., 2017; Hart et al., 2014; Yin et al., pathway (8 versus 13 steps), which improved overall yield from
2003), but strategies to stabilize other Max dimers—such as 4% to 51%, while slightly improving the reporter IC50 value to
stabilization of the Max/Max homodimer with NSC13728 1.28 mM (Figures 2A and 2E). The enantiomers of KI-MS2-001
(Jiang et al., 2009)—remain relatively unexplored. Utilizing a and KI-MS2-008 demonstrated comparable activity with the
chemical probe to directly target Max will clarify its role in this racemic mixture when tested separately (Figure S3; Table S1).
dynamic equilibrium and improve our understanding of the tran- As seen by the CellTiter-Glo assay to measure ATP as an indica-
scriptional consequences of perturbing various Max protein tor for metabolically active cell levels (Riss et al., 2013), both
interactions.
compounds decreased levels of viable P493-6 cells, a well-
While modulating Myc/Max-mediated transcription via a characterized B cell line engineered with exogenous c-Myc
Max-directed small molecule is an appealing alternative under a doxycycline-repressible promoter (Schuhmacher et al.,
approach to targeting Myc directly, these transcription factors 1999), with KI-MS2-008 displaying an IC50 value of 2.15 mM after
have proven difficult to target due to their intrinsically disor- 3 days of treatment (Figures 2B and 2E). The effect of KI-MS2-
dered nature that lack traditional binding pockets and system- 008 was lost with co-treatment of doxycycline, which shuts
atic design strategies to perturb protein-protein or protein-DNA down expression of c-Myc (Figure 2D), as well as in EBNA2-
interactions (Darnell, 2002; Koehler, 2010; McKeown and driven cells that proliferate without exogenous c-Myc with
Bradner, 2014). In recent years, the small-molecule microarray co-treatment of both doxycycline and b-estradiol (Pajic et al.,
(SMM) platform has helped bridge this gap by enabling robust 2000) (Figure S4A), suggesting that c-Myc inhibition is the pri-
and unbiased binding assays capable of identifying small mary mechanism underlying diminished growth of viable cells.
molecules with multiple modes of binding regardless of how In contrast to levels of viable cells, P493-6 cell viability—as
disordered or intractable a target may be, thus making it an measured by propidium iodide exclusion—was unaffected by
attractive screening approach compared with other traditional compound treatment (Figure S4B). The PC12 rat pheochromo-
binding assays such as differential scanning fluorimetry (Hong cytoma cell line (Hopewell and Ziff, 1995), which lacks functional
et al., 2014). This high-throughput approach utilizes glass Max protein (Figure 2D), was insensitive to treatment with
slides functionalized with small molecules to permit fluorescent both compounds (Figures 2C and 2E), suggesting the possibility
detection of protein-small molecule interactions (Figure 1B) of on-target effects in cells. In addition, we evaluated the viable
(Vegas et al., 2008). This screening approach has been previ- cell levels of a variety of other cancer cell lines upon treatment
ously used to identify small molecule binders to proteins, with both compounds (Figures S1F, S1G, S4C, and S4D) in
DNA, and RNA, including binders to the ETV1 transcription CellTiter-Glo assays. As a preliminary assessment of speci-
factor, Hedgehog growth factors, and G-quadruplexes, among ficity, we profiled KI-MS2-008 (10 mM) for activity against a panel
others (Felsenstein et al., 2016; Pop et al., 2014; Stanton of 299 kinases followed by dose-response experiments for
et al., 2009).
selected kinases. KI-MS2-008 did not inhibit any of these
kinases with an IC50 value below 10 mM (Figures S4E and S4F).
These initial findings encouraged us to further characterize
KI-MS2-008 as a Max-directed small molecule that modulates
Myc transcriptional activity.
RESULTS
Discovery of Max Binders that Modulate Myc-Driven
Transcription
We used SMMs to screen purified recombinant Max protein KI-MS2-008 and Analogs Engage Max In Vitro and in
against 21,600 printed chemical array features, including com- Live Cells
mercial compounds, known bioactives, natural products, prod- Although the parent compound was discovered in a target-
ucts of diversity-oriented synthesis (DOS) (Clemons et al., directed primary screen, we used five approaches to validate
2010), and various negative controls, which yielded 117 putative the hypothesis that KI-MS2-008 directly engages Max. The first
Max binders (Figures 1C and S1A). After prioritizing assay posi- two strategies made use of previous SAR studies to develop
tives specific to Max as compared with more than 100 other pro- chemical probes for both affinity-enrichment experiments in
teins screened against the same set of molecules, we identified cell lysates and photo-crosslinking studies in live cells (Figures
six compounds that inhibited Myc transcriptional activity with a S2 and S3; Table S1). We conjugated KI-MS2-081, an analog
half maximal inhibitory concentration (IC50) less than 5 mM in a containing a primary amine (Figure 3A), to NHS-activated mag-
dual luciferase reporter assay containing tandem Myc response netic beads and incubated KI-MS2-081-loaded and unloaded
elements (Figures S1B–S1D). We prioritized the DOS polycyclic control beads with P493-6 cell lysates to pull down protein tar-
lactam BRD-K19261677 (KI-MS2-001) (Mitchell and Shaw, gets of the compound. We observed Max, as well as c-Myc,
2006) due to validated potency on resynthesis and reasonable bound to KI-MS2-008-loaded beads as determined by immuno-
physicochemical characteristics (Figures 1D and S1E).
blots (Figure 3B). Pull down of Max with these compound-loaded
To establish cursory structure-activity relationships (SARs), beads was similarly observed when doxycycline was used to
we synthesized analogs of KI-MS2-001 (Figure S2). Using the repress c-Myc expression. We also synthesized KI-MS2-085,
same Myc transcriptional dual luciferase reporter assay as an a bifunctional probe containing a diazirine for UV-activated
indicator of compound activity, we identified sites tolerant to covalent attachment to neighboring proteins (Mackinnon and
modification and determined that the propanediol side chain, Taunton, 2009), and an alkyne handle for copper-catalyzed
which is predicted to mediate attachment to the SMM surface, azide-alkyne cycloaddition to a fluorophore for visualization (Fig-
was not required for activity (Figure S3; Table S1). Removal of ure 3C). To determine the validity of this approach, purified
this group allowed KI-MS2-008 to be synthesized by a simplified full-length Max was covalently linked to KI-MS2-085 and
Cell Chemical Biology 26, 1–13, May 16, 2019
3

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 2. KI-MS2-001 and KI-MS2-008 Modulate Myc-Driven Transcription in Cells and Inhibit Viable Cell Levels in a Myc-Dependent Manner
(A) Dose-response curves for the Myc dual luciferase reporter assay in HEK293 cells in response to KI-MS2-001 or KI-MS2-008 treatment after 16 h (n = 3
technical replicates, error bars represent mean SD).
(B) Dose-response curves for P493-6 viable cell levels in response to KI-MS2-001 or KI-MS2-008 treatment with Myc expression left on or shut down with
doxycycline after 3 days (n = 3 technical replicates, error bars represent mean SD).
(C) Dose-response curves for PC12 viable cell levels in response to KI-MS2-001 or KI-MS2-008 treatment after 5 days (n = 3 technical replicates, error bars
represent mean SD).
(D) Immunoblots of c-Myc from P493-6 lysates, demonstrating doxycycline repressed expression of c-Myc and immunoblots of Max from P493-6 and PC12
lysates, showing the lack of functional Max in PC12 cells.
(E) Summary of estimated IC50 values for the Myc reporter assay and viable cell levels.
See also Figure S4.
conjugated to Alexa Fluor 647, which was visualized by SDS- stabilizing Max in a time-dependent manner, we observed
PAGE. Mass spectrometry was used to confirm the presence that the stabilization of Max occurs at about 30 min after
of a covalent protein-small molecule adduct (Figures S5A and initial dosing of KI-MS2-008, possibly leading to a subsequent
S5B). After validation of covalent reaction with pure protein, destabilization of c-Myc starting at 45 min (Figures 3E
P493-6 cells were treated with KI-MS2-085 (2.5, 5, and 10 mM) and 5A).
followed by irradiation for covalent linkage in cells. Following
In preliminary efforts to identify a possible binding site of
cell lysis, the same workflow allowed visualization of a pair of KI-MS2-008 on Max, we used NMR spectroscopy and mass
proteins between 20 and 25 kDa (Figure 3D). These proteins spectrometry methods. In the first approach, a chemically stabi-
matched the molecular weights of two Max isoforms and aligned lized form of Max (Jean-Franc¸ ois et al., 2003) in the absence of
with immunoblot-based detection of these Max isoforms (Fig- DNA was subjected to 15N-HSQC in the absence and presence
ure S5C). The intensities of these bands varied in a dose-depen- of KI-MS2-008, and the change in chemical shift perturbation
dent manner, were not observed without UV irradiation, and was quantified. After mapping the structure to the previously
were reduced by co-treatment with competitive KI-MS2-008 published NMR of Max (Sauve et al., 2004), it was revealed
(40 mM) (Figure 3D).
that residues L44 and I51 and the surrounding region were
In addition, we used a cellular thermal shift assay (Molina most affected by compound treatment, suggesting a possible
et al., 2013) to determine whether KI-MS2-008 produced binding region for KI-MS2-008 (Figures 3F and S5F). In the
ligand-induced thermal stabilization of Max in P493-6 cells. second approach, we digested Max samples, one containing
We demonstrated that KI-MS2-008 stabilized cellular Max the KI-MS2-085 covalent adduct and one without, and sub-
in a dose-responsive manner, further suggesting engagement jected them to liquid chromatography-mass spectrometry anal-
to Max (Figures 3E and S5D). Looking at the effects of ysis (Figures S5A and S5B). Comparative analysis of the two
4
Cell Chemical Biology 26, 1–13, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 3. KI-MS2-008 and Analogs Engage Max In Vitro and in Live Cells
(A) Chemical structure of KI-MS2-081 analog containing a primary amine for covalent coupling to NHS-activated magnetic beads.
(B) Immunoblots of Max and c-Myc from samples containing proteins bound to beads with or without KI-MS2-081 incubated with lysates from P493-6 with and
without doxycycline.
(C) Workflow schematic for KI-MS2-085 analog containing both diazirine and alkyne moieties, enabling crosslinking to targets through the diazirine (in blue) and
detection using the alkyne (in red) after copper-catalyzed azide-alkyne cycloaddition to a fluorophore.
(D) Fluorescent gel of proteins labeled by KI-MS2-085. P493-6 cells were incubated with 2.5, 5, or 10 mM KI-MS2-085. Irradiation of UV light to covalently modify
nearby proteins followed by installation of Alexa Fluor 647 allowed visualization of a pair of proteins between 20 and 25 kDa (molecular weight of Max isoforms) on
the subsequent gel loaded with labeled lysates.
(E) Immunoblot of Max in soluble fraction using cellular thermal shift assay (CETSA) in intact P493-6 cells following 1 h KI-MS2-008 dosing treatment (top). Time
course of c-Myc and Max protein levels in soluble fraction using CETSA in intact P493-6 cells at 10 mM KI-MS2-008 visualized by immunoblots (bottom).
(F) 15N-HSQC in the absence (blue) and presence (red) of 150 mM KI-MS2-008 (left) and the quantification of the change in chemical shift (right).
See also Figure S5.
samples showed a loss of the peptides which contained residues KI-MS2-008 Induces Max/Max Homodimerization
L44 and I51 with KI-MS2-085 treatment, suggesting formation To understand whether KI-MS2-008 affected the DNA binding
of the KI-MS2-085 covalent adduct in this region, thus functions of Max, we performed electrophoretic mobility
corroborating the NMR data (Figure S5G). Additional detailed shift assays (EMSAs). While a known c-Myc/Max disrupter,
structural biology studies are currently underway to precisely 10058-F4 (F4) (Yin et al., 2003), decreased the level of c-Myc/
map contacts.
Max heterodimers bound to DNA at active concentrations
Cell Chemical Biology 26, 1–13, May 16, 2019
5

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 4. KI-MS2-008 Induces Max/Max Homodimerization
(A) Electrophoretic mobility shift assay (EMSA) utilizing Alexa Fluor 684-labeled E-box DNA incubated with P493-6 lysate treated with 10058-F4 (F4) or
KI-MS2-008.
(B) EMSA involving pure recombinant Max protein binding to Alexa Fluor 684-labeled E-box DNA incubated with NSC13728 or KI-MS2-008.
(C) Plot of the distribution of sedimentation coefficients (c(s)) versus S, calculated from analytical ultracentrifugation sedimentation velocity experiments carried
out with untreated Max protein or Max protein treated with 10 and 20 mM KI-MS2-008.
(D) Size-exclusion chromatography of pure recombinant Max protein, Max protein with high KCl concentration, and Max protein with high KCl concentration
treated with 10 mM KI-MS2-008.
See also Figure S5.
reported previously, KI-MS2-008 (up to 20 mM) had no effect on perturbed, we observed a time-dependent decrease in total
c-Myc/Max heterodimerization (Figure 4A), consistent with the c-Myc protein levels, in addition to concomitant decreases in
bead-based pull down from cell lysates. In contrast, KI-MS2- c-Myc protein levels phosphorylated at S62 and T58, on treat-
008 stabilized Max/Max homodimers bound to DNA in a solution ment with 10 mM KI-MS2-008 (Figures 5A and S6A). Decreased
containing pure recombinant Max, similarly to the Max/Max c-Myc protein levels were partially rescued when cells were
homodimer stabilizer NSC13728 (NSC) (Jiang et al., 2009), as co-treated with 10 mM MG132 (Tsubuki et al., 1996), suggesting
visualized and quantified by EMSA (Figures 4B and S5E). The sta- a role for proteasome-mediated degradation as one mechanism
bilization of the Max/Max homodimer by KI-MS2-008 was further for alterations in c-Myc protein levels in this cell line (Figure 5B).
supported by analytical ultracentrifugation experiments involving Similar observations were recently reported for two Myc/Max
pure Max (Figure 4C). Untreated Max protein exhibited two heterodimer antagonists (Choi et al., 2017; Wang et al., 2015),
peaks representing the monomer and homodimer, respectively, but increases of ubiquitin-conjugated c-Myc levels were not
at a 1.44:1 ratio. Incubation of the protein with 10 or 20 mM observed with KI-MS2-008 treatment (Figure S6B). We did not
KI-MS2-008 shifted the distribution to predominantly Max/Max detect changes in the protein levels of other Max network mem-
homodimers. When fit to a monomer-dimer self-association bers such as Mxd1, Mxd4, or Mlx (Ayer et al., 1993; Billin et al.,
model (Brautigam, 2011), the data demonstrated a strong 1999; Conacci-Sorrell et al., 2014) with 10 mM KI-MS2-008 treat-
enhancement of the Max dimerization affinity in the absence of ment over 24 h (Figure S6A). Overall, these results suggest that
DNA with KI-MS2-008 treatment to signify Max/Max homodimer KI-MS2-008 affects Myc protein levels in P493-6 cells to further
stabilization, which is similar to results observed with NSC13728 promote a shift in equilibrium away from the Myc/Max hetero-
(Jiang et al., 2009). Specifically, the data yielded a dimerization dimer within the dynamic network, although the precise molecu-
affinity Kd value of 8 mM in absence of the compound, which lar mechanism is still unclear and is the subject of ongoing study.
was strongly enhanced to nanomolar affinity (12 nM with 10 mM
To understand the impact of KI-MS2-008 treatment on the
KI-MS2-008 and 10 nM with 20 mM KI-MS2-008). In addition, global transcriptome and Myc-driven gene expression, we per-
size-exclusion chromatography demonstrated an increased con- formed RNA sequencing (RNA-seq) experiments in P493-6 cells.
centration of Max/Max homodimer relative to Max monomer after A strong correlation was observed between 10 mM KI-MS2-008
10 mM KI-MS2-008 treatment (Figure 4D), further supporting the treatment (8 h) and doxycycline addition to turn off c-Myc
hypothesis that KI-MS2-008 stabilizes the Max/Max homodimer. expression (r = 0.87) (Figure 5C). Furthermore, a group of genes
regulated by Myc (Liberzon et al., 2015) appeared enriched in the
In P493-6 Cells, KI-MS2-008 Treatment Decreases
c-Myc Protein Levels and Causes Global Changes to the
Transcriptome that Mimic Myc Inactivation
genes downregulated in response to 10 mM KI-MS2-008 treat-
ment (Figures 5C and S6C). This observation was confirmed
using gene set enrichment analysis and extended to other Myc
We next sought to monitor the effects of KI-MS2-008 at the target gene sets (Subramanian et al., 2005) at 1 and 10 mM
protein level in P493-6 cells. While Max protein levels were not KI-MS2-008 (Figures 5D and S6D). In addition, to observe
6
Cell Chemical Biology 26, 1–13, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 5. KI-MS2-008 Decreases Myc Protein Levels and Affects the Global Myc Transcriptional Program
(A) Time course of c-Myc and Max protein levels in P493-6 treated with 10 mM KI-MS2-008 visualized by immunoblots.
(B) c-Myc protein levels with 4 h of 10 mM KI-MS2-008 and/or proteasome inhibitor MG132 treatment in P493-6 visualized by immunoblots demonstrating partial
rescue of c-Myc protein levels.
(C) Scatterplot of RNA-seq data comparing Myc inactivation via doxycycline (8 h) and KI-MS2-008 treatment in P493-6 cells at 8 h of 10 mM KI-MS2-008. Green
dots correspond to genes from the HALLMARK_MYC_TARGETS _V2 (HALLMARK MYC) gene set.
(D) Gene set enrichment analysis (GSEA) plots of enrichment scores (ES) corresponding to RNA-seq datasets for P493-6 cells treated with 10 mM KI-MS2-008 for
8 h with normalized enrichment scores (NES) and nominal p values (n = 4 technical replicates). Four Myc target gene sets are shown.
(E) GSEA plots of ES corresponding to RNA-seq datasets for P3HR1 cells and ST486 cells treated with 10 mM KI-MS2-008 for 8 h with NES and nominal p values
(n = 4 technical replicates). HALLMARK_MYC_TARGETS _V2 (HALLMARK MYC) gene set is shown.
(F) Heatmap showing fold changes in Myc-driven genes in response to 10 mM KI-MS2-008, 10 mM MG132, or the combination at 4 h as determined by qPCR (n = 3
technical replicates).
See also Figure S6.
whether the transcriptional effects of KI-MS2-008 occurred in genes after 4 and 24 h (Figures 6A and 6B). These perturbations
non-engineered cell lines, we chose to monitor global transcrip- to the TSSs of NR1D1 and DDX31 corresponded with decreases
tional programs in ST486 and P3HR1, which were sensitive and in their mRNA levels at 4 and 8 h of compound treatment, with
insensitive, respectively, toward both KI-MS2-001 and KI-MS2- partially recovered levels at 24 h (Figures 6C and 6D). We then
008 treatment (Figures S1F, S1G, S4C, and S4D). KI-MS2-008 extended our analysis to 14,792 active genes (those with
downregulated c-Myc-regulated genes (Liberzon et al., 2015) H3K27ac at the promoter) and observed similar decreases in
in ST486 cells, but this enrichment was not observed in the H3K27ac signal and c-Myc binding and increases in Max binding
P3HR1 cells (Figures 5E and S6E). To determine the contribu- at these TSSs (Figures 6E–6G). A comparison of the top 1,000
tion of c-Myc protein level decreases toward transcriptional genes with the strongest signals for c-Myc occupancy, as
repression of Myc-driven genes, we monitored transcript levels measured by the area under the curve of c-Myc at the proximal
of MYC and a panel of canonical Myc reporter genes using promoter ( 1 kb TSS region) and proximal enhancers ( 50 kb of
qPCR. We observed repression of these genes in response to TSS) with other active genes, revealed significant differences be-
10 mM KI-MS2-008 by 2 h and a general reversal of this effect tween changes in expression of Myc-regulated genes and other
with co-treatment of 10 mM MG132 at 2 and 4 h (Figures 5F active genes over the 24-h KI-MS2-008 treatment, suggesting
and S6F).
that KI-MS2-008 can specifically perturb Myc-regulated genes
(Figure 6H). Taken together, these results demonstrate a global
loss of c-Myc and H3K27ac, and a global stabilization of Max
at promoter sites with KI-MS2-008 treatment, along with down-
regulation of genes with Myc-occupied TSSs. These results are
consistent with previous observations that KI-MS2-008 stabi-
KI-MS2-008 Treatment Decreases c-Myc Protein
Binding at Promoters of Active Myc-Regulated Genes
and Increases Max Protein Binding at the Same
Promoters
To determine the effect of KI-MS2-008 treatment on the global lizes the Max/Max homodimer and decreases Myc protein
chromatin landscape of Myc and Max, we used chromatin levels.
immunoprecipitation sequencing (ChIP-seq) (Lin et al., 2012).
We first focused on the 10-kb region surrounding the transcrip- KI-MS2-008 Exhibits Efficacy in Cellular and Murine
tional start sites (TSSs) of NR1D1 and DDX31 as representative Cancer Models
Myc-regulated genes (Altman et al., 2015). Treatment of To further test the applicability of KI-MS2-001 and KI-MS2-008
P493-6 cells with 1 mM KI-MS2-008 led to decreases in signals in a broader spectrum of cancer models, we first evaluated
of active enhancer marker H3K27ac and c-Myc, as well as sub- whether these compounds modulated viable cell levels of Myc-
stantial increases in Max signal at the core promoters of these driven cancer cell lines derived from tumors of transgenic mouse
Cell Chemical Biology 26, 1–13, May 16, 2019
7

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 6. KI-MS2-008 Treatment Decreases c-Myc
Protein Binding and Increases Max Protein Binding
at Promoters of Myc-Occupied Genes
(A and B) Gene tracks of H3K27ac (gold), c-Myc (red), and
Max (blue) binding at 0 h (top), 4 h (middle), and 24 h
(bottom) after 1 mM KI-MS2-008 treatment (n = 1 replicate)
at the (A) NR1D1 gene or (B) DDX31 gene.
(C and D) Bar graphs of the mean log2 fold changes (versus
0 h) of the amount of (C) NR1D1 or (D) DDX31 mRNA at 1, 4,
8, and 24 h of 1 mM KI-MS2-008 treatment. Error bars
represent the standard deviation of the mean (n = 3
replicates).
(E–G) Heatmap of (E) H3K27ac, (F) c-Myc, and (G) Max
levels at transcription start site (TSS) regions at 0, 4, and
24 h of 1 mM KI-MS2-008 treatment. TSSs for all active
promoters in P493-6 cells are shown and ranked by
average Myc occupancy at 0 h. Color scaled intensities are
in units of rpm/bp (n = 1 replicate).
(H) Boxplots of log2 fold changes (versus 0 h) of expression
of Myc-occupied genes at 1, 4, 8, and 24 h of 1 mM
KI-MS2-008 treatment (n = 3 replicates). The top 1,000
MYC genes based on Myc occupancy at enhancers and
promoters are shown in red and other active genes (n =
6,835) in gray. Significance is denoted (Welch’s two-tailed
t test): ***p < 1 3 10À9, **p < 1 3 10À6
.
8
Cell Chemical Biology 26, 1–13, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Figure 7. KI-MS2-008 Exhibits Efficacy in Cellular and Murine Cancer Models
(A) Dose-response curves for Myc-driven T-ALL cancer cell line (4188) viable cell levels in response to KI-MS2-001 or KI-MS2-008 treatment (n = 3, error bars
represent mean SEM).
(B) Dose-response curves for Myc-driven murine cancer cell lines (EC4: HCC, 4188: T-ALL), and normal human h-TERT immortalized fibroblasts (BJ5-tA: FIB)
viable cell levels in response to KI-MS2-008 treatment (n = 3, error bars represent mean SEM).
(C) Immunofluorescence images of a Myc-driven OS cell line treated with 10 mM KI-MS2-001, 10 mM KI-MS2-008, or with doxycycline to inactivate Myc
expression. a-tubulin and Apex1 were visualized using specific antibodies.
(D) Representative bioluminescence imaging (BLI) of day 12 (left) and data quantitation of BLI time course (right) for NSG mice bearing Myc-driven luciferase-
labeled T-ALL treated with vehicle or 0.06 mg/kg KI-MS2-008 intravenously on days 4–12 (vehicle: n = 6 mice, KI-MS2-008: n = 8 mice, error bars represent
mean SEM).
(E) BLI of day 21 (left) and data quantitation of BLI time course (right) for NSG mice with subcutaneous transplants of Myc-driven luciferase-labeled HCC treated
with vehicle or 0.24 mg/kg KI-MS2-008 intraperitoneally (on days 4–21, 5-on/2-off schedule) (vehicle: n = 4 mice, KI-MS2-008: n = 3 mice, error bars represent
mean SEM). **p < 0.01, ***p < 0.001; two-way ANOVA with p values of Bonferroni post test for treatment endpoints (D and E).
See also Figure S7.
models of Myc-induced T cell acute lymphoblastic leukemia with intraperitoneal injections of either vehicle or KI-MS2-008
(T-ALL), hepatocellular carcinoma (HCC), and osteosarcoma (0.24 mg/kg) starting after tumor cell engraftment (day 4, 5-on/
(OS) (Felsher and Bishop, 1999; Jain et al., 2002; Shachaf 2-off schedule). Significantly smaller tumors were observed in
et al., 2004). KI-MS2-008 showed a stronger effect on viable KI-MS2-008-treated mice compared with vehicle-treated con-
cell levels of Myc-driven T-ALL (4188) compared with KI-MS2- trol mice (Figure 7E). Following this, the in vivo pharmacokinetic
001 (Figure 7A). KI-MS2-008 inhibited viable cell levels of lines properties of KI-MS2-008 were analyzed by a single high dose
derived from HCC (EC4) and T-ALL (4188), but KI-MS2-008 (0.731 mg/kg) study in healthy KP53 mice (DuPage et al., 2009)
had no effect on normal human fibroblast (BJ5-tA) viable cell (n = 3), which demonstrated complete clearance from plasma
levels (Figure 7B). A Myc-driven OS cell line was used to assess in ꢀ24 h with 50% of the probe being cleared in 30 min, similarly
the potential of KI-MS2-001 and KI-MS2-008 to induce differen- to its metabolic stability in mouse microsomes (Figures S7A–
tiation via Myc inactivation. Morphological changes associated S7C). Repeated doses of KI-MS2-008 resulted in slightly higher
with differentiation were assessed by visualizing a-tubulin, while levels in the plasma 24 h after the last dose compared with
the expression of a known Myc target, Apex1 (Zeller et al., 2003), vehicle control-treated mice, and had no toxic effects on healthy
was used as a readout for Myc activity. Treatment with 10 mM mice over the course of a week, as observed by body weight
KI-MS2-008 resulted in differentiation and downregulation of (Figure S7D), behavior, and H&E stains of the liver and kidney.
Apex1 expression, which resulted in indistinguishable immuno- These results suggest that higher doses would be tolerable for
fluorescence images of cells treated with KI-MS2-008 from cells future in vivo studies exploring the efficacy of KI-MS2-008 in a
in which Myc expression was shut off by doxycycline (Figure 7C). broad swath of cancer models.
The in vivo efficacy of KI-MS2-008 was then tested in allograft
models of Myc-driven T-ALL and HCC. Luciferase-labeled DISCUSSION
T-ALL cells were injected into NOD/SCID/IL-2Rgnull (NSG) mice
intravenously and mice were treated intravenously with vehicle As Myc requires Max for DNA binding and subsequent tran-
or KI-MS2-008 (0.06 mg/kg) daily, starting after tumor cell scriptional activation of target genes, a major area of research
engraftment (day 4). Mice treated with KI-MS2-008 showed a has focused on disrupting the Myc/Max interaction. These
significant reduction in tumor burden compared with vehicle efforts have yielded a number of small molecules to mediate
control-treated mice (Figure 7D). In addition, NSG mice with sub- Myc-driven transcription via disruption of this heterodimer,
cutaneous transplants of Myc-driven HCC cells were treated but few have shown efficacy in mouse models (Soodgupta
Cell Chemical Biology 26, 1–13, May 16, 2019
9

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
et al., 2015; Stellas et al., 2014). As an alternative approach, SIGNIFICANCE
stabilization of the Max/Max homodimer to indirectly target
the Myc/Max heterodimer was first explored using virtual ligand The bHLHLZ protein network—including Myc, Max, and
screening to predict candidate binders of the Myc/Max and Mxd—has crucial roles in normal cellular processes such
Max/Max dimers, followed by fluorescence resonance energy as cell proliferation, metabolism, apoptosis, and differentia-
transfer assays to prioritize Max/Max homodimer stabilizers. tion, but its activity is deregulated in a majority of human
This work led to the discovery of NSC13728, a symmetric small cancers. Despite this importance, there is still limited avail-
molecule that inhibits c-Myc-regulated oncogenic transforma- ability of high-quality chemical tools to investigate this
tion, cell growth, and target gene transcription via Max/Max network. Both the Myc/Max and Max/Max dimers bind to
homodimer stabilization (Jiang et al., 2009). Despite this suc- the canonical E-box (CACGTC) with the heterodimer acti-
cess, this strategy to indirectly modulate Myc-driven transcrip- vating transcription, while the homodimer competes for
tion via Max/Max homodimer stabilization has not been as the same genomic target sites to attenuate Myc-driven pro-
widely explored.
grams. An emerging hypothesis is that the tumor suppressor
Here we report the discovery of a small molecule with a activity of Max relies on its ability to modulate Myc-driven
unique asymmetric scaffold to perturb the balance of Myc/ programs via the homodimer. As such, we have explored
Max and Max/Max dimers. We used mechanistically unbiased alternative modes of Myc perturbation by modulating Max
binding assays involving SMMs to identify KI-MS2-001, a poly- with small molecules, utilizing the mechanistically unbiased
cyclic lactam that emerged from a library of DOS with a high SMM to help discover KI-MS2-008. We determined that
level of structural diversity (Mitchell and Shaw, 2006). SAR KI-MS2-008 stabilizes the Max/Max homodimer while also
studies led to the synthesis of the analog KI-MS2-008, and decreasing Myc protein levels and effectively perturbing
we demonstrated modulation of Max function by stabilizing Myc-driven transcriptional programs. To our knowledge,
the Max/Max homodimer using this compound. KI-MS2-008 KI-MS2-008 is the first Max-directed probe to demonstrate
also decreased Myc protein levels, perturbed Myc-driven tran- in vivo efficacy, and the unoptimized probe may serve as a
scriptional programs, induced Myc-dependent cellular differ- starting point for the development of improved compounds
entiation, and affected viable cell growth of engineered and to enable translational studies and to further clarify the
non-engineered cell lines. We also observed a global loss of relevance of Max as a therapeutic target.
c-Myc and a global stabilization of Max at promoter sites with
KI-MS2-008 treatment. Although unoptimized for potency or
STAR+METHODS
physicochemical properties, KI-MS2-008 exhibited preliminary
in vivo efficacy at low doses by significantly reducing tumor
Detailed methods are provided in the online version of this paper
and include the following:
volume in T-ALL and HCC mouse models, although we
acknowledge that this study does not assess the effect of
KI-MS2-008 on anti-tumor immunity. With the introduction of
KI-MS2-008, we provide further support that, by stabilizing
the Max homodimer, chemical probes can effectively perturb
this dynamic bHLHLZ protein network with overall effects
observed in cancer cell lines and mouse models, consistent
with Myc modulation.
d KEY RESOURCES TABLE
d CONTACT FOR REAGENT AND RESOURCE SHARING
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Animal Studies
B Cell Lines
d METHOD DETAILS
Several questions related to the molecular and cellular
mechanisms of KI-MS2-008 remain unanswered to date. For
example, the precise binding mode of the compound to Max
has not been defined, although preliminary NMR and proteomics
studies suggest binding to the L44 and I51 residues and the sur-
rounding region. Detailed investigations involving multiple struc-
tural biology approaches are currently underway in our labs. We
have not yet precisely characterized the mechanisms by which
Myc protein levels and Myc-regulated transcript levels are
decreased when the dynamic equilibrium of Myc/Max hetero-
dimer or Max/Max homodimer is perturbed (e.g., possibly selec-
tive degradation of Myc in the nucleolus and/or differential gene
B Chemical Synthesis
B Small Molecule Microarray
B Transcriptional Reporter Assay
B Viable Cell Assay
B Flow Cytometry
B Bead-Based Lysate Pull-down with KI-MS2-081
B In Vitro Labeling of Max with KI-MS2-085
B Live-Cell Pull-down with KI-MS2-085
B Electrophoretic Mobility Shift Assay
B Analytical Ultracentrifugation
B Size Exclusion Chromatography
B Cellular Thermal Shift Assay (Molina et al., 2013)
B
15N-Heteronuclear Single Quantum Coherence
´
regulation by Myc) (Amati and Sanchez-Arevalo Lobo, 2007;
Kress et al., 2015). We also have yet to understand how KI-
MS2-008 perturbs the wider Max interactome. Despite these
outstanding questions, KI-MS2-008 can serve as a useful tool
for interrogating the roles of Max in transcription and cell biology,
as well as understanding the importance of Max dimerization
status in cancer and other disease settings. Moving forward,
the compound may also serve as a scaffold for a next generation
of Max-directed chemical probes.
B Time Course of Protein Levels
B RNA-seq
B qPCR
B Primers for qPCR
B Ubiquitination Assay
B Protein Binding Assay
B Solubility of KI-MS2-001 in Aqueous Buffers
B Bidirectional Permeability of KI-MS2-001
10 Cell Chemical Biology 26, 1–13, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
B Liver Microsome Stability of KI-MS2-001
B Chromatin Immunoprecipitation Sequencing
B Myc-Addicted Tumor Cell Lines in Mice
B Immunofluorescence Staining and Imaging
B Metabolic Stability Studies
B.L.E., D.W.F., and A.N.K. supervised the study. N.B.S., A.C., and A.N.K.
wrote the manuscript with input from all authors.
DECLARATION OF INTERESTS
A.N.K. is a founder of Kronos Bio and a member of its scientific advisory board.
N.B.S., A.C., E.S., H.L.E., F.C., D.V.N., and A.N.K. have a patent related to this
work that is licensed by Kronos Bio. D.W.F. and C.Y.L. are consultants to
Kronos Bio.
B Pharmacokinetics
B Repeat Dose Toxicity Studies
B LC/MS Analysis of Plasma Samples
d QUANTIFICATION AND STATISTICAL ANALYSIS
d DATA AND SOFTWARE AVAILABILITY
Received: June 15, 2018
Revised: November 27, 2018
Accepted: February 7, 2019
Published: March 14, 2019
SUPPLEMENTAL INFORMATION
Supplemental Information includes seven figures and one table and can
be found with this article online at https://doi.org/10.1016/j.chembiol.2019.
02.009.
REFERENCES
Altman, Brian J., Hsieh, Annie L., Sengupta, A., Krishnanaiah, Saikumari Y.,
Stine, Zachary E., Walton, Zandra E., Gouw, Arvin M., Venkataraman, A., Li,
B., Goraksha-Hicks, P., et al. (2015). MYC disrupts the circadian clock and
metabolism in cancer cells. Cell Metab. 22, 1009–1019.
ACKNOWLEDGMENTS
The authors thank S. Doyle for critical reading of the manuscript. We also
thank J. Bradner, R. Eisenman, T. Golub, S. Levine, M. Schenone, R. Sears,
F. White, O. Witte, and R. Young for helpful discussions. We thank D. Eick for
P493-6 cells, O. McPherson for SMM technical support, and E. Richards for
qPCR support. We apologize to colleagues not cited due to limited space.
We thank the Koch Institute Swanson Biotechnology Center for technical
support, specifically the Animal Imaging and Preclinical Testing Facility, the
Flow Cytometry Facility, the Genomics Core Facility, the Biopolymers & Pro-
teomics Core facility, and the High-Throughput Screening Facility. The proj-
ect benefitted from the Biophysical Instrumentation and Department of
Chemistry Instrumentation facilities at MIT as well as the Center for Science
of Therapeutics at the Broad Institute. Bioluminescence imaging was per-
formed in the SCi3 Stanford Small Animal Imaging Service Center. This
work was financially supported by the National Cancer Institute through
R01-CA160860 (to A.N.K.), P30-CA14051 (Koch Institute Cancer Center
Support Core grant), U01-CA176152 (Broad Institute Cancer Target Discov-
ery and Development Center), and NIH/NCI CA170378PQ2 (to D.W.F.), and
1U01CA188383-01/S1 (to D.W.F.). The work was also supported in part by
the Leukemia & Lymphoma Society, the Ono Pharma Foundation, the MIT
Deshpande Center for Technological Innovation, the MIT Center for Precision
Cancer Medicine, the AACR-Bayer Innovation and Discovery Grant, and the
Merkin Institute Fellows Program at the Broad Institute (to A.N.K.). C.Y.L is
supported by the Cancer Prevention Research Institute of Texas
(RR150093), the NIH, and NCI (1R01CA215452-01), and is a Pew-Stewart
Scholar for Cancer Research (Alexander and Margaret Stewart Trust).
N.B.S. is supported by the GSK-MIT Gertrude B. Elion Research Fellowship
Program for Drug Discovery and Disease and the Ludwig Center Fund at
MIT. A.C. is supported by a Koch Institute Graduate Fellowship in Cancer
Research. A.D. is supported by a Lymphoma Research Foundation Postdoc-
toral Fellowship. H.L.E is supported by a Koch Institute Quinquennial Cancer
Research Fellowship. R.M.W. is supported by a Ludwig Center for Molecular
Oncology Graduate Fellowship.
´
Amati, B., and Sanchez-Arevalo Lobo, V.J. (2007). MYC degradation: deubi-
quitinating enzymes enter the dance. Nat. Cell Biol. 9, 729–731.
Anders, S., and Huber, W. (2010). Differential expression analysis for sequence
count data. Genome Biol. 11, R106.
Ayer, D.E., Kretzner, L., and Eisenman, R.N. (1993). Mad: a heterodimeric part-
ner for Max that antagonizes Myc transcriptional activity. Cell 72, 211–222.
Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate - a
practical and powerful approach to multiple testing. J. R. Stat. Soc. Series B
Stat. Methodol. 57, 289–300.
Billin, A.N., Eilers, A.L., Queva, C., and Ayer, D.E. (1999). Mlx, a novel max-like
BHLHZip protein that interacts with the max network of transcription factors.
J. Biol. Chem. 274, 36344–36350.
Blackwell, T.K., Kretzner, L.F., Blackwood, E.M., Eisenman, R.N., and
Weintraub, H. (1990). Sequence-specific DNA binding by the c-Myc protein.
Science 23, 1149–1151.
Blencowe, A., and Hayes, W. (2005). Development and application of diazir-
ines in biological and synthetic macromolecular systems. Soft Matter. 1, 178.
Brautigam, C.A. (2011). Using Lamm-equation modeling of sedimentation
velocity data to determine the kinetic and thermodynamic properties of macro-
molecular interactions. Methods 54, 4–15.
Calderon, M.R., Verway, M., An, B.S., DiFeo, A., Bismar, T.A., Ann, D.K.,
Martignetti, J.A., Shalom-Barak, T., and White J, H. (2012). Ligand-dependent
corepressor (LCoR) recruitment by Kruppel-like factor 6 (KLF6) regulates
expression of the cyclin-dependent kinase inhibitor CDKN1A gene. J. Biol.
Chem. 287, 8662–8674.
Casco´ n, A., and Robledo, M. (2012). MAX and MYC: a heritable breakup.
Cancer Res. 72, 3119–3124.
Champely, S. (2018). pwr: basic functions for power analysis. R package
version 1.2-2, https://CRAN.R-project.org/package=pwr.
Choi, S.H., Mahankali, M., Lee, S.J., Hull, M., Petrassi, H.M., Chatterjee, A.K.,
Schultz, P.G., Jones, K.A., and Shen, W. (2017). Targeted disruption of
Myc-Max oncoprotein complex by a small molecule. ACS Chem. Biol. 17,
2715–2719.
AUTHOR CONTRIBUTIONS
N.B.S., A.C., A.D., R.M.W., M.A.R., T.L., F.C., M.H.E.W., A.V., C.Y.L., D.W.F.,
and A.N.K. designed the experiments and analyzed the data. N.B.S., A.C.,
R.M.W., F.C., M.S.P., M.McM., S.M., H.T., K.R.F., D.B.F., P.A.C., B.L.E.,
and A.N.K. performed and analyzed screening assays. A.D. and D.W.F. per-
formed cell culture and mouse experiments. E.S., H.L.E., B.H.C., and T.A.L.
synthesized small molecules. M.S.P., D.B.F., and B.H.C. analyzed NMR
studies. M.A.R. and T.L. performed ChIP-seq and RNA-seq experiments.
A.C. and N.B.S. performed target engagement studies. N.B.S. performed
mechanistic studies. A.C. and R.M.W. performed RNA-seq sample process-
ing. R.M.W. performed RNA-seq analysis. V.L.B. and S.S.L. sequenced RNA
transcripts. S.E. and S.E.M. performed pharmacokinetic studies. D.V.N.,
C.F., and J.A.H. performed preliminary experiments that initiated the study.
Clemons, P.A., Bodycombe, N.E., Carrinski, H.A., Wilson, J.A., Shamji, A.F.,
Wagner, B.K., Koehler, A.N., and Schreiber, S.L. (2010). Small molecules
of different origins have distinct distributions of structural complexity that
correlate with protein-binding profiles. Proc. Natl. Acad. Sci. U S A 107,
18787–18792.
Conacci-Sorrell, M., McFerrin, L., and Eisenman, R.N. (2014). An overview of
MYC and its interactome. Cold Spring Harb. Perspect. Med. 4, a014357.
Cook, R.D. (1977). Detection of influential observation in linear regression.
Technometrics 19, 15–18.
Dang, C.V. (2012). MYC on the path to cancer. Cell 149, 22–35.
Cell Chemical Biology 26, 1–13, May 16, 2019 11

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Darnell, J.E., Jr. (2002). Transcription factors as targets for cancer therapy.
Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009). Ultrafast and
memory-efficient alignment of short DNA sequences to the human genome.
Genome Biol. 10, R25.
Nat. Rev. Cancer 2, 740–749.
Derr, A., Yang, C., Zilionis, R., Sergushichev, A., Blodgett, D.M., Redick, S.,
Bortell, R., Luban, J., Harlan, D.M., Kadener, S., et al. (2016). End Sequence
Analysis Toolkit (ESAT) expands the extractable information from single-cell
RNA-seq data. Genome Res. 26, 1397–1410.
Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G.,
Abecasis, G., and Durbin, R. (2009). The sequence alignment/map format and
SAMtools. Bioinformatics 25, 2078–2079.
Devgan, V., Harris, A., Jiang, Q., and Pattison, S. (2012). Cignal reporter assay
kit: a high-performance tool for assessing the functions of genes, biologics,
and small molecule compounds. https://www.qiagen.com/fi/resources/
download.aspx?id=6fc8144b-010b-43d8-9125-bf541a5d9c9c&lang=en.
Liberzon, A., Birger, C., Thorvaldsdottir, H., Ghandi, M., Mesirov, J.P., and
Tamayo, P. (2015). The Molecular Signatures Database (MSigDB) hallmark
gene set collection. Cell Syst. 1, 417–425.
Lin, C.Y., Loven, J., Rahl, P.B., Paranal, R.M., Burge, C.B., Bradner, J.E., Lee,
T.I., and Young, R.A. (2012). Transcriptional amplification in tumor cells with
elevated c-Myc. Cell 151, 56–67.
DuPage, M., Dooley, A.L., and Jacks, T. (2009). Conditional mouse lung cancer
models using adenoviral or lentiviral delivery of Cre recombinase. Nat. Protoc.
4, 1064–1072.
Livak, K.J., and Schmittgen, T.D. (2001). Analysis of relative gene expression
data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method.
Methods 25, 402–408.
Farrell, A.S., and Sears, R.C. (2014). MYC degradation. Cold Spring Harb.
Perspect. Med. 4, a014365.
Felsenstein, K.M., Saunders, L.B., Simmons, J.K., Leon, E., Calabrese, D.R.,
Zhang, S., Michalowski, A., Gareiss, P., Mock, B.A., and Schneekloth, J.S.
(2016). Small molecule microarrays enable the identification of a selective,
Quadruplex-binding inhibitor of MYC expression. ACS Chem. Biol. 11,
139–148.
Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of
fold change and dispersion for RNA-seq data with DESeq2. Genome Biol.
15, 550.
Lu, T.Y., Lu, R.M., Liao, M.Y., Yu, J., Chung, C.H., Kao, C.F., and Wu, H.C.
(2010). Epithelial cell adhesion molecule regulation is associated with the
maintenance of the undifferentiated phenotype of human embryonic stem
cells. J. Biol. Chem. 285, 8719–8732.
Felsher, D.W. (2010). MYC inactivation elicits oncogene addiction through
both tumor cell-intrinsic and host-dependent mechanisms. Genes Cancer 1,
597–604.
Mackinnon, A.L., and Taunton, J. (2009). Target identification by diazirine
Felsher, D.W., and Bishop, J.M. (1999). Reversible tumorigenesis by MYC in
photo-cross-linking and click chemistry. Curr. Protoc. Chem. Biol. 1, 55–73.
hematopoietic lineages. Mol. Cell 4, 199–207.
McKeown, M.R., and Bradner, J.E. (2014). Therapeutic strategies to
inhibit MYC. Cold Spring Harb. Perspect. Med. 4, https://doi.org/10.1101/
cshperspect.a014266.
Gabay, M., Li, Y., and Felsher, D.W. (2014). MYC activation is a hallmark of
cancer initiation and maintenance. Cold Spring Harb. Perspect. Med. 4,
a014241.
Mitchell, J.M., and Shaw, J.T. (2006). A structurally diverse library of polycyclic
lactams resulting from systematic placement of proximal functional groups.
Angew. Chem. Int. Ed. 45, 1722–1726.
Grzesiek, S., and Bax, A. (1993). The importance of not saturating water in pro-
tein NMR. Application to sensitivity enhancement and NOE measurements.
J. Am. Chem. Soc. 115, 12593–12594.
Molina, D.M., Jafari, R., Ignatushchenko, M., Seki, T., Larsson, E.A., Dan, C.,
Sreekumar, L., Cao, Y., and Nordlund, P. (2013). Monitoring drug target
engagement in cells and tissues using the cellular thermal shift assay.
Science 341, 84–87.
Hart, J.R., Garner, A.L., Yu, J., Ito, Y., Sun, M., Ueno, L., Rhee, J.-K., Baksh,
M.M., Stefan, E., Hartl, M., et al. (2014). Inhibitor of MYC identified in a
Kro¨ hnke pyridine library. Proc. Natl. Acad. Sci. U S A 111, 12556–12561.
Hong, J.A., Neel, D.V., Wassaf, D., Caballero, F., and Koehler, A.N. (2014).
Recent discoveries and applications involving small-molecule microarrays.
Curr. Opin. Chem. Biol. 18, 21–28.
Nair, S.K., and Burley, S.K. (2003). X-ray structures of Myc-Max and Mad-Max
recognizing DNA. Cell 112, 193–205.
Pajic, A., Spitkovsky, D., Christoph, B., Kempkes, B., Schuhmacher, M.,
Staege, M.S., Brielmeier, M., Ellwart, J., Kohlhuber, F., Bornkamm, G.W.,
et al. (2000). Cell cycle activation by c-myc in a Burkitt lymphoma model cell
line. Int. J. Cancer 87, 787–793.
Hopewell, R., and Ziff, E.B. (1995). The nerve growth factor-responsive PC12
cell line does not express the Myc dimerization partner Max. Mol. Cell. Biol. 15,
3470–3478.
Jafari, R., Almqvist, H., Axelsson, H., Ignatushchenko, M., Lundback, T.,
Nordlund, P., and Martinez Molina, D. (2014). The cellular thermal shift assay
for evaluating drug target interactions in cells. Nat. Protoc. 9, 2100–2122.
Palmer, A.G., Cavanagh, J., Wright, P.E., and Rance, M. (1991). Sensitivity
improvement in proton-detected two-dimensional heteronuclear correlation
NMR spectroscopy. J. Magn. Reson. 93, 151–170.
Jain, M., Arvanitis, C., Chu, K., Dewey, W., Leonhardt, E., Trinh, M., Sundberg,
C.D., Bishop, J.M., and Felsher, D.W. (2002). Sustained loss of a neoplastic
phenotype by brief inactivation of MYC. Science 297, 102–104.
Pop, M.S., Stransky, N., Garvie, C.W., Theurillat, J.P., Hartman, E.C., Lewis,
T.A., Zhong, C., Culyba, E.K., Lin, F., Daniels, D.S., et al. (2014). A small mole-
cule that binds and inhibits the ETV1 transcription factor oncoprotein. Mol.
Cancer Ther. 13, 1492–1502.
Jean-Franc¸ ois, N., Fre´ de´ ric, G., Raymund, W., Benoit, C., and Lavigne, P.
(2003). Improving the thermodynamic stability of the leucine zipper of max in-
creases the stability of its b-HLH-LZ: E-box complex. J. Mol. Biol. 326,
1577–1595.
Riss, T.L., Moravec, R.A., Niles, A.L., Duellman, S., Benink, H.A., Worzella,
T.J., and Minor, L. (2013). Cell Viability Assays (Eli Lilly & Company and the
National Center for Advancing Translational Sciences).
Jiang, H., Bower, K.E., Beuscher, A.E.t., Zhou, B., Bobkov, A.A., Olson, A.J.,
and Vogt, P.K. (2009). Stabilizers of the Max homodimer identified in virtual
ligand screening inhibit Myc function. Mol. Pharmacol. 76, 491–502.
Sauve, S., Tremblay, L., and Lavigne, P. (2004). The NMR solution structure of
a mutant of the Max b/HLH/LZ free of DNA: insights into the specific and
reversible DNA binding mechanism of dimeric transcription factors. J. Mol.
Biol. 342, 813–832.
Kay, L., Keifer, P., and Saarinen, T. (1992). Pure absorption gradient enhanced
heteronuclear single quantum correlation spectroscopy with improved sensi-
tivity. J. Am. Chem. Soc. 114, 10663–10665.
Schleucher, J., Schwendinger, M., Sattler, M., Schmidt, P., Schedletzky, O.,
Glaser, S.J., Sørensen, O.W., and Griesinger, C. (1994). A general enhance-
ment scheme in heteronuclear multidimensional NMR employing pulsed field
gradients. J. Biomol. NMR 4, 301–306.
Koehler, A.N. (2010). A complex task? Direct modulation of transcription
factors with small molecules. Curr. Opin. Chem. Biol. 14, 331–340.
`
Kress, T.R., Sabo, A., and Amati, B. (2015). MYC: connecting selective
Schneider, C.A., Rasband, W.S., and Eliceiri, K.W. (2012). NIH image to
transcriptional control to global RNA production. Nat. Rev. Cancer 15,
imageJ: 25 years of image analysis. Nat. Methods 9, 671–675.
593–607.
Laherty, C.D., Yang, W.-M., Sun, J.-M., Davie, J.R., Seto, E., and Eisenman,
R.N. (1997). Histone deacetylases associated with the mSin3 corepressor
mediate mad transcriptional repression. Cell 89, 349–356.
Schuhmacher, M., Staege, M.S., Pajic, A., Polack, A., Weidle, U.H.,
Bornkamm, G.W., Eick, D., and Kohlhuber, F. (1999). Control of cell growth
by c-Myc in the absence of cell division. Curr. Biol. 9, 1255–1258.
12 Cell Chemical Biology 26, 1–13, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
˚
Shachaf, C.M., Kopelman, A.M., Arvanitis, C., Karlsson, A., Beer, S., Mandl, S.,
Bachmann, M.H., Borowsky, A.D., Ruebner, B., Cardiff, R.D., et al. (2004).
MYC inactivation uncovers pluripotent differentiation and tumour dormancy
in hepatocellular cancer. Nature 431, 1112.
Stellas, D., Szabolcs, M., Koul, S., Li, Z., Polyzos, A., Anagnostopoulos, C.,
Cournia, Z., Tamvakopoulos, C., Klinakis, A., and Efstratiadis, A. (2014).
Therapeutic effects of an anti-Myc drug on mouse pancreatic cancer.
J. Natl. Cancer Inst. 106, https://doi.org/10.1093/jnci/dju320.
Smith, E., and Collins, I. (2015). Photoaffinity labeling in target- and binding-
Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L.,
Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al.
(2005). Gene set enrichment analysis: a knowledge-based approach for inter-
preting genome-wide expression profiles. Proc. Nat. Acad. Sci. U S A 102,
15545–15550.
site identification. Future Med. Chem. 7, 159–183.
Soodgupta, D., Pan, D., Cui, G., Senpan, A., Yang, X., Lu, L., Weilbaecher,
K.N., Prochownik, E.V., Lanza, G.M., and Tomasson, M.H. (2015). Small
molecule MYC inhibitor conjugated to integrin-targeted nanoparticles extends
survival in a mouse model of disseminated multiple myeloma. Mol. Cancer
Ther. 14, 1286–1294.
Tsubuki, S., Saito, Y., Tomioka, M., Ito, H., and Kawashima, S. (1996).
Differential inhibition of calpain and proteasome activities by peptidyl alde-
hydes of Di-leucine and Tri-leucine. J. Biochem. 119, 572–576.
`
Soucek, L., Jucker, R., Panacchia, L., Ricordy, R., Tato, F., and Nasi, S. (2002).
Omomyc, a potential Myc dominant negative, enhances Myc-induced
Vegas, A.J., Fuller, J.H., and Koehler, A.N. (2008). Small-molecule microarrays
apoptosis. Cancer Res. 62, 3507–3510.
as tools in ligand discovery. Chem. Soc. Rev. 37, 1385–1394.
Soucek, L., Whitfield, J., Martins, C.P., Finch, A.J., Murphy, D.J., Sodir, N.M.,
Karnezis, A.N., Swigart, L.B., Nasi, S., and Evan, G.I. (2008). Modelling Myc
inhibition as a cancer therapy. Nature 455, 679–683.
Wang, H., Teriete, P., Hu, A., Raveendra-Panickar, D., Pendelton, K., Lazo,
J.S., Eiseman, J., Holien, T., Misund, K., Oliynyk, G., et al. (2015). Direct
inhibition of c-Myc-Max heterodimers by celastrol and celastrol-inspired
triterpenoids. Oncotarget 6, 32380–32395.
Soumillon, M., Cacchiarelli, D., Semrau, S., van Oudenaarden, A., and
Mikkelsen, T.S. (2014). Characterization of directed differentiation by high-
throughput single-cell RNA-Seq. bioRxiv. https://doi.org/10.1101/003236.
Yin, X., Giap, C., Lazo, J.S., and Prochownik, E.V. (2003). Low molecular
weight inhibitors of Myc-Max interaction and function. Oncogene 22,
6151–6159.
Speers, A.E., and Cravatt, B.F. (2009). Activity-based protein profiling (ABPP)
and click chemistry (CC)-ABPP by MudPIT mass spectrometry. Curr. Protoc.
Chem. Biol. 1, 29–41.
Zeller, K.I., Jegga, A.G., Aronow, B.J., O’Donnell, K.A., and Dang, C.V. (2003).
An integrated database of genes responsive to the Myc oncogenic transcrip-
tion factor: identification of direct genomic targets. Genome Biol. 4, R69.
Stanton, B.Z., Peng, L.F., Maloof, N., Nakai, K., Wang, X., Duffner, J.L.,
Taveras, K.M., Hyman, J.M., Lee, S.W., Koehler, A.N., et al. (2009). A small
molecule that binds Hedgehog and blocks its signaling in human cells. Nat.
Chem. Biol. 5, 154–156.
Zhang, Y., Liu, T., Meyer, C.A., Eeckhoute, J., Johnson, D.S., Bernstein, B.E.,
Nusbaum, C., Myers, R.M., Brown, M., Li, W., et al. (2008). Model-based anal-
ysis of ChIP-seq (MACS). Genome Biol. 9, R137.
Cell Chemical Biology 26, 1–13, May 16, 2019 13

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
STAR+METHODS
KEY RESOURCES TABLE
REAGENT or RESOURCE
Antibodies
SOURCE
Qiagen
IDENTIFIER
Cat#35370
Mouse monoclonal Alexa Fluorꢀ
647-conjugated anti-His5
Mouse monoclonal anti-c-Myc (9E10)
Rabbit polyclonal anti-c-Myc (N-262)
Mouse monoclonal anti-c-Myc (9E11)
Rabbit IgG control
Santa Cruz Biotechnology
Santa Cruz Biotechnology
Abcam
Cat#sc-40; RRID: AB_627268
Cat#sc-764; RRID: AB_631276
Cat#ab56; RRID: AB_304976
Cat#sc-2027; RRID: AB_737197
Cat#sc-765; RRID: AB_2281780
Cat#ab9700; RRID: AB_2141674
Cat#ab199489
Santa Cruz Biotechnology
Santa Cruz Biotechnology
Abcam
Rabbit polyclonal anti-Max (C-124)
Goat polyclonal anti-Max
Rabbit monoclonal anti-Max
Abcam
Rabbit polyclonal anti-Max (C-17)
Horse anti-mouse IgG, HRP-linked
Goat anti-rabbit IgG, HRP-linked
Rabbit polyclonal anti-c-Myc (pT58)
Mouse monoclonal anti-c-Myc (pS62)
Mouse monoclonal anti-MLX (F-12)
Mouse monoclonal anti-Mxd1 (F-1)
Goat polyclonal anti-Mxd4 (N-19)
Mouse anti-ubiquitin (FK2)
Santa Cruz Biotechnology
Cell Signaling Technology
Cell Signaling Technology
Abcam
Cat#sc-197, lot C3016; RRID: AB_2281783
Cat#7076; RRID: AB_330924
Cat#7074; RRID: AB_2099233
Cat#ab28842; RRID: AB_731667
Cat#ab78318; RRID: AB_1566069
Cat#sc-393086
Abcam
Santa Cruz Biotechnology
Santa Cruz Biotechnology
Santa Cruz Biotechnology
Enzo Life Sciences
Sigma-Aldrich
Cat#sc-8012; RRID: AB_627900
Cat#sc-1221; RRID: AB_2235630
BML-PW8810; RRID: AB_10541840
Cat#V9131; RRID: AB_477629
Cat#T9026; RRID: AB_477593
Cat#4128S; RRID: AB_2057933
Cat#925-32211; RRID: AB_2651127
Cat#A-11035; RRID: AB_2534093
Cat#A-21236; RRID: AB_2535805
Cat#ab4729; RRID: AB_2118291
Mouse monoclonal anti-vinculin (hVIN-1)
Mouse monoclonal anti-tubulin (DM1A)
Rabbit polyclonal anti-Apex1
Sigma-Aldrich
Cell Signaling Technology
LICOR
Goat polyclonal IRDyeꢀ 800CW anti-Rabbit IgG
Goat polyclonal anti-rabbit Alexa Fluor-546
Goat polyclonal anti-mouse Alexa Fluor-647
Rabbit polyclonal to Histone H3 (acetyl K27)
Chemicals, Peptides, and Recombinant Proteins
His6-tagged recombinant Max protein
Recombinant human Max protein
NSC13728
Thermo Fisher Scientific
Thermo Fisher Scientific
Abcam
Panomics
Cat#RP0080
Cat#ab95309
N/A
Abcam
Resynthesized in the Koehler lab.
SigmaAldrich
10058-F4
Cat#F3680-5MG
Critical Commercial Assays
Cignal Myc Reporter Assay Kit (LUC)
Dual-Gloꢀ Luciferase Assay System
CellTiter-Gloꢀ Luminenscent Cell Viability Assay
AurumTM Total RNA Mini Kit
Qiagen
Cat#CCS-012L
Cat#E2940
Promega
Promega
BioRad
Cat#G7572
Cat#7326820
Deposited Data
Small molecule microarray datasets
Clemons et al., 2010,
http://chembank.broadinstitute.org/
protein/view-protein.htm?id=5000358
18787-18792.1125.0116
(DIV06) and 1125.0117 (NPC)
RNA-seq dataset
ChIP-seq dataset
This paper. NCBI GEO: GSE107222
This paper. NCBI GEO: GSE125863
https://www.ncbi.nlm.nih.gov/geo/
https://www.ncbi.nlm.nih.gov/geo/
(Continued on next page)
e1 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Continued
REAGENT or RESOURCE
Experimental Models: Cell Lines
Human: P493-6
SOURCE
IDENTIFIER
Laboratory of D. Eick
ATCC
N/A
Human: HEK293T (female)
Rat: PC12 (male)
Cat#CRL-3216; RRID: CVCL_0063
Cat#CRL-1721; RRID:CVCL_0481
Cat#CRL-1647; RRID:CVCL_1712
Cat#CCL-240; RRID:CVCL_0002
Cat#CRL-1648; RRID:CVCL_1101
Cat#HTB-62; RRID:CVCL_2676
Cat#HTB-10; RRID:CVCL_0530
Cat#HTB-14; RRID:CVCL_0022
Cat#CRL-1690; RRID:CVCL_0556
Cat#CRL-5982; RRID:CVCL_1510
Cat#HTB-171; RRID:CVCL_1562
Cat#HTB-175; RRID:CVCL_1591
Cat#CRL-2233; RRID:CVCL_0077
Cat#CRM-CRL-3211; RRID:CVCL_1644
Cat#CCL-220; RRID:CVCL_0219
Cat#CRL-2331; RRID:CVCL_1256
Cat#CRL-5974; RRID:CVCL_0076
Cat#CRL-9068; RRID:CVCL_1600
Cat#CRL-2081; RRID:CVCL_1430
Cat#HTB-81; RRID:CVCL_0105
Cat#CRL-1740; RRID:CVCL_1379
Cat#CRL-1435; RRID:CVCL_0035
ATCC
Human: ST-486 (female)
Human: HL-60 (female)
Human: CA46 (male)
ATCC
ATCC
ATCC
Human: P3HR1 (male)
Human: SKNMC (female)
Human: U87-MG (male)
Human: T98G (male)
ATCC
ATCC
ATCC
ATCC
Human: NCI-H1963 (male)
Human: NCI-H446 (male)
Human: NCI-H82 (male)
Human: SNU-398 (male)
Human: PSN-1 (male)
Human: COLO-320 (female)
Human: HCC-1599 (female)
Human: SNU-16 (female)
Human: NCI-H929 (female)
Human: MSTO-211H (male)
Human: DU145 (male)
Human: LNCaP (male)
Human: PC3 (male)
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
ATCC
Experimental Models: Organisms/Strains
Mouse: NOD-SCIDIL-2Rg-/-Sex: male
Laboratory of Felsher
N/A
N/A
Mouse: KP53
Sex: male
AIPT colony (Laboratory
of Malstrom)
Oligonucleotides
See qPCR section for list of Primers used
Software and Algorithms
R
R Core Team. (2016).
R: A Language and
https://www.r-project.org/
Environment for Statistical
Computing (R Foundation
for Statistical Computing)
(Champely, 2018).
GenePix Pro 6.0
Molecular Devices
http://mdc.custhelp.com/app/answers/detail/
a_id/18691/ꢀ/genepix%C2%AE-pro-6-microarray-
acquisition-%26-analysis-software-download-page
Image Lab 6.0
BioRad
http://www.bio-rad.com/en-us/product/image-
lab-software?ID=KRE6P5E8Z
ImageStudioLite
ChemDraw Professional 15.0
GraphPad Prism 7
SedFit
LI-COR Biosciences
PerkinElmer
Prism
https://www.licor.com/bio/products/software/
image_studio_lite/
http://www.perkinelmer.com/product/
chemdraw-professional-chemdrawpro
https://www.graphpad.com/scientific-
software/prism/
NIH
https://sedfitsedphat.nibib.nih.gov/software/
default.aspx
(Continued on next page)
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e2

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Continued
REAGENT or RESOURCE
SedPhat
SOURCE
NIH
IDENTIFIER
https://sedfitsedphat.nibib.nih.gov/software/
default.aspx
AmiView V1.7.06
BD FACSDIVATM Software
Spectral Instruments Imaging
BD Biosciences
https://specimg.com/instruments/ami/
http://www.bdbiosciences.com/us/instruments/
research/software/flow-cytometry-acquisition/
bd-facsdiva-software/m/111112/overview
Topsin 3.5 pl.7
Bruker
https://www.bruker.com/service/
support-upgrades/software-downloads/nmr.html
CONTACT FOR REAGENT AND RESOURCE SHARING
Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Angela N.
Koehler (koehler@mit.edu).
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Animal Studies
Mouse experiments were approved by the Institutional Animal Care and Use Committee (IUCAC) at the Massachusetts Institute
of Technology or the Administrative Panel on Laboratory Animal Care (APLAC) at Stanford University, and were carried out in
accordance with institutional and national guidelines. The efficacy studies used 4-6 week-old male NOD-SCIDIL-2Rg-/- (NSG)
mice. The cell lines used for transplantation experiments (4188 (T-ALL), and EC4 (HCC)) were isolated from tumors of male mice.
The PK experiments used 6-9 month-old male LSL-Kras G12D/+ ; Trp53 flox/flox mice.
Cell Lines
Cell lines were obtained from ATCC with exception of P493-6, which was a gift from D. Eick. Cell lines were not authenticated in
our hands. Cell lines were tested intermittently throughout studies using MycoAlert Mycoplasma Detection Kit (Lonza,
Cat#LT07-418), generally a few days after thawing and immediately before significant studies (e.g. RNA-seq experiment).
Cells were cultured in a humidified incubator at 37.0ꢁC and 5.0% CO2. P493-6 cells were cultured in RPMI Medium 1640 (gibco,
11875-093) supplemented with 10% Tet-approved FBS (Takara, Cat#631106) and penicillin-streptomycin (Corning, 30-002-CI).
HEK293T (female) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (gibco, Cat#11965-092) supplemented
with penicillin-streptomycin and 10% FBS (ATCC, Cat#30-2020). PC12 cells were cultured in RPMI-1640 supplemented with
penicillin-streptomycin, 5% FBS and 10% heat-inactivated horse serum. ST-486 (female), NCI-H1963 (male), NCI-H446
(male), NCI-H82 (male), SNU-398 (male), PSN-1 (male), COLO-320 DM (female), HCC-1599 (female), SNU-16 (female),
MSTO-211H (male), LNCaP (male), PC3 (male) cells were cultured in RPMI-1640 supplemented with penicillin-streptomycin
and 10% FBS. CA46 (male) and P3HR1 (male) cells were cultured in RPMI-1640 supplemented with penicillin-streptomycin
and 20% FBS. HL-60 (female) cells were cultured in Iscove’s Modified Dulbecco’s Medium (IMDM) (ATCC, Cat#30-2005) sup-
plemented with penicillin-streptomycin and 20% FBS. SKNMC (female), U87-MG (male), T98G (male), DU145 (male) cells were
cultured in Eagle’s Minimum Essential Medium (EMEM) (ATCC, Cat#30-2003) supplemented with penicillin-streptomycin and
10% FBS. NCI-H929 (female) cells were cultured in RPMI-1640 supplemented with penicillin-streptomycin, 0.05 mM 2-mercap-
toethanol and 10% FBS.
METHOD DETAILS
Chemical Synthesis
KI-MS2-001, KI-MS2-002 and 2-(azidomethyl)benzaldehyde were prepared according to published procedures (Mitchell and Shaw,
2006). For the SMM screen, the compounds were originally synthesized as part of a diversity-oriented chemical library on solid phase
involving 529 compounds involving a key catalytic enantioselective Suga-Ibata cycloaddition between an oxazole and aldehyde
so as to develop enantiomerically pure compounds. Upon resynthesis, the compounds were prepared using a modified procedure
carried out in solution that did not provide enantiomerically pure material. For KI-MS2-001 and KI-MS2-008, the racemic mixtures
(KI-MS2-001a and KI-MS2-008a) were separated into the respective enantiomers (KI-MS2-001b-c and KI-MS2-008b-c, respectively)
and tested in follow-up assays.
e3 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Synthesis of KI-MS2-008
KI-MS2-008 was obtained as a mixture of enantiomers according to the procedure outlined below.
5-Methoxy-2-phenyloxazole was synthesized using the method reported by Mitchell et al. with minor modifications:
Methyl Benzoylglycinate
To a solution of benzoic acid (4.35 g, 35.31 mmol) and glycine methyl ester hydrochloride (4.03 g, 32.10 mmol) in DMF (50 mL) was
added EDC hydrochloride (6.77 g, 35.31 mmol), HOBt (4.77 g, 35.31 mmol) and DIPEA (12.30 mL, 70.62 mmol) at room temperature.
The mixture was stirred overnight, prior to being quenched with sat. NaHCO3 (150 mL). The aqueous layer was extracted with EtOAc
(3 x 30 mL), and the combined organic layers were washed with sat. NaHCO3 (1 x 20 mL), brine (2 x 30 mL), dried over MgSO4, filtered
and concentrated to furnish methyl benzoylglycinate as an off-white solid (6.07 g, 98% yield). HRMS (m/z): [M]+ calcd. for
C10H11NO3, 176.0706; found 176.0698.
5-methoxy-2-phenyloxazole
A solution of methyl benzoylglycinate (4.25 g, 22.00 mmol) and Et3N (14.7 mL, 104.5 mmol) in CH2Cl2 (10 mL) was added
dropwise to a solution of iodine (12.56 g, 49.48 mmol) and triphenylphosphine (12.98 g, 49.48 mmol) in CH2Cl2 (150 mL) at
room temperature. After 5 hr, the mixture was concentrated and purified by short plug on silica gel (15 to 40% EtOAc in
hexanes) to provide 5-methoxy-2-phenyloxazole (3.8 g, 98% yield). HRMS (m/z): [M]+ calcd. for C10H9NO2, 176.0706; found
176.0698.
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e4

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Methyl (5S)-5-(2-(azidomethyl)phenyl)-2-phenyl-4,5-dihydrooxazole-4-carboxylate (KI-MS2-082)
An oven-dried flask was charged with 5-methoxy-2-phenyloxazole (2.54 g, 14.51 mmol), 2-(azidomethyl)benzaldehyde (2.58 g,
15.96 mmol) and anhydrous 1:1 CH2Cl2:CH3CN (100 mL). The flask was purged with Arg, cooled to 0ꢁC and SnCl4 (17.6 mL,
1.0 M in CH2Cl2) was added. The reaction was stirred overnight at room temperature then cooled to 0ꢁC. Saturated NaHCO3
(100 mL) was carefully added followed by CH2Cl2 (75 mL) and the solution was vigorously stirred for 3 hr. The aqueous layer was
extracted with EtOAc (4 x 30 mL). The combined organic layers were dried over MgSO4, filtered concentrated and purified by
column chromatography (10 to 45% EtOAc in hexanes) to afford KI-MS2-082 (4.49 g, 92% yield) and as a mixture of diastereomers,
1
as a yellow oil. KI-MS2-082cis (higher Rf ) H NMR (500 MHz, CDCl3): d 8.10 (d, J = 8.0 Hz, 2H), 7.55 (m, 1H), 7.45-7.50 (m, 3H),
7.31-7.39 (m, 3H), 6.18 (d, J = 10.5 Hz, 1H), 5.35 (d, J = 10.5 Hz, 1H), 4.45 (d, J = 13.5 Hz, 1H), 4.37 (d, J = 13.5 Hz, 1H), 3.18
(s, 3H). 13C NMR (125 MHz, CDCl3): d 169.8, 167.0, 135.1, 132.7, 132.4, 129.7, 129.1, 129.0, 128.8, 127.0, 126.9, 79.9, 73.9, 53.1,
52.1. KI-MS2-082Trans (lower Rf) 1H NMR (500 MHz, CDCl3): d 8.08 (d, J = 8.0 Hz. 1H), 7.55 (t, J = 8.0 Hz, 1H), 7.46 (t, J =
8.0 Hz, 2H), 7.34-7.40 (4H, m), 6.21 (d, J = 7.0 Hz, 1H), 4.79 (d, J = 7.0 Hz, 1H), 4.55 (d, J = 13.5 Hz, 1H), 4.46 (d, J = 13.5 Hz,
1H), 3.87 (s, 3H). 13C NMR (125 MHz, CDCl3): d 171.6, 165.8, 138.6, 132.5, 132.4, 130.5, 129.8, 129.1, 129.0, 128.7, 126.9, 126.4,
80.0, 77.3, 53.2, 52.6. HRMS (m/z): [M]+ calcd. for C18H16N4O3, 337.1295; found 337.1268.
( )-(4S,5S)-methyl 5-(2-(azidomethyl)phenyl)-4-benzyl-2-phenyl-4,5-dihydrooxazole-4-carboxylate (KI-MS2-006)
A cis/trans mixture of KI-MS2-082 (4.49 g, 13.35 mmol) was diluted in DMF (50 mL). BTPP (5.40 mL, 17.35 mmol) was then
added at room temperature followed by benzyl bromide (6.40 mL, 53.40 mmol), K2CO3 (3.69 g, 26.70 mmol) and TBAI
(2.46 g, 6.67 mmol). After 16 hr, the reacting mixture was concentrated and purified by column chromatography on
silica gel (5 to 40% EtOAc in hexanes) to afford KI-MS2-006 (5.28 g, 93% yield) as a 95:5 dr. 1H NMR (500 MHz, CDCl3):
d 8.04 (m, 2H), 7.54 (m, 1H), 7.46 (m 2H), 7.39 (m, 2H), 7.23-7.38 (m, 7H), 5.93 (s, 1H), 4.58 (d, J = 14.0 Hz, 1H), 4.45
(d, J = 14.0 Hz, 1H), 3.57 (d, J = 13.5 Hz, 1H), 3.45 (d, J = 13.5 Hz, 1H), 3.16 (s, 3H). 13C NMR (125 MHz, CDCl3): d 171.6,
165.0, 136.5, 135.2, 133.0, 132.2, 131.3, 130.0, 129.0, 128.8, 128.7, 128.4, 127.4, 127.2, 127.1, 84.7, 82.5, 53.0, 52.4,
45.0. HRMS (m/z): [M]+ calcd. for C25H22N4O3, 427.1652; found 427.1678. The other diastereomer (lower Rf) can also be
isolated. 1H NMR (500 MHz, CDCl3): d 8.18 (d, J = 7.5 Hz, 2H), 7.62 (t, J = 7.5 Hz, 1H), 7.53 (t, J = 7.5 Hz, 2H), 7.46 (t, J =
7.5 Hz, 1H), 7.31-7.41 (m, 7H), 6.04 (s, 1H), 5.01 (d, J = 11.0 Hz, 1H), 4.74 (d, J = 14.5 Hz, 1H), 4.73 (d, J = 11.0 Hz, 1H),
3.15 (s, 3H).
( )-(3aS,10bS)-3a-benzyl-2-phenyl-5,6-dihydro-3aH-benzo[c]oxazolo[5,4-e]azepin-4(10bH)-one (KI-MS2-007)
KI-MS2-006 (1.73 g, 4.06 mmol) was diluted in THF:H2O (10:1, 100 mL) followed by addition of DBU (0.64 mL, 4.26 mmol) and
PMe3 (4.50 mL, 1.0 M in THF) at room temperature. After 16 hr, water was added and the aqueous layer extracted with
CH2Cl2 (x 3). The combined organic layers were dried over MgSO4, filtered concentrated and purified by column
chromatography on silica gel (0 to 15% MeOH in CH2Cl2) providing KI-MS2-007 (1.21g, 81% yield). 1H NMR (500 MHz,
CDCl3): d 8.11 (d, J = 9.0 Hz, 2H), 7.56 (t, J = 7.5 Hz, 1H), 7.45 (t, J = 8.0 Hz, 2H), 7.36 (m, 2H), 7.16-7.32 (m, 7H), 5.84
(bs, 1H), 4.22 (m, 1H), 3.95 (dd, J = 15.5, 7.5 Hz, 2H), 3.51 (s, 2H). 13C NMR (125 MHz, CDCl3): d 180.1, 165.9, 138.3,
134.3, 133.5, 131.3 (2), 129.5 (4), 128.8 (2), 128.5 (2), 128.4 (2), 128.2 (2), 127.5, 83.1, 65.0, 44.1, 32.3. HRMS (m/z): [M]+ calcd.
for C24H20N2O2, 369.1598; found 369.1580.
e5 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
( )- (3aS,10bS)-3a-benzyl-2-phenyl-5-(2-(phenylsulfonyl)ethyl)-5,6-dihydro-3aH-benzo[c]oxazolo[5,4-e]azepin-
4(10bH)-one (KI-MS2-008)
To a solution of KI-MS2-007 (1.10 g, 2.99 mmol) in DMF (15 mL) was successively added BTPP (1.20 mL, 3.86 mmol) and phenyl vinyl
sulfone (1.00 g, 5.97 mmol) at room temperature. The reacting mixture was stirred overnight then diluted with water (50 mL). The
aqueous layer was extracted with EtOAc (3 x 15 mL). The combined organic layers were dried over MgSO4, filtered concentrated
and purified by column chromatography (20 to 60% EtOAc in hexanes) to afford KI-MS2-008 as a pale yellow solid (1.55 g, 97% yield).
1H NMR (400 MHz, CD3OD): d 7.98 (dd, J = 14.5, 7.5 Hz, 3H), 7.94 – 7.84 (m, 1H), 7.74 (d, J = 7.0 Hz, 1H), 7.68 (t, J = 7.5 Hz, 2H), 7.58
(t, J = 7.5 Hz, 1H), 7.48 (t, J = 7.5 Hz, 2H), 7.41 – 7.36 (m, 1H), 7.36 – 7.26 (m, 3H), 7.24 – 7.15 (m, 2H), 5.93 (s, 1H), 4.57 (s, 1H), 4.44 –
4.24 (m, 1H), 3.92 (s, 1H), 3.71 (s, 1H), 3.56 – 3.39 (m, 2H). 13C NMR (101 MHz, CD3OD): d 170.66, 146.59, 139.23, 136.87, 135.56,
134.38, 133.90, 133.86, 133.81, 133.57, 131.92, 130.90, 129.31, 129.29, 129.27, 128.31, 128.27, 128.16, 127.91, 127.85, 127.72,
127.69, 127.14, 126.85, 126.42, 125.71, 81.91, 74.25, 53.07, 49.97, 44.09, 43.11. HRMS (m/z): [M]+ calcd. for C32H28N2O4S,
537.1845; found 537.1866.
Synthesis of KI-MS2-081 (Probe for Displaying Compound on Beads)
( )-2-((3aS,10bS)-3a-benzyl-4-oxo-2-phenyl-3aH-benzo[c]oxazolo[5,4-e]azepin-5(4H,6H,10bH)-yl)acetic acid
(KI-MS2-084)
To a solution of KI-MS2-007 (300 mg, 0.81 mmol) in DMF (5 mL) were successively added BTPP (0.48 mL, 1.22 mmol) and tert-butyl
bromoacetate (0.48 mL, 3.26 mmol). The resultant mixture was stirred overnight at room temperature and diluted with water (30 mL).
The aqueous layer was then extracted with EtOAc (3 x 10 mL), and the combined organic layers were dried over MgSO4, filtered
concentrated and purified by column chromatography (5 to 60% EtOAc in hexanes) to afford ( )tert-butyl 2-((3aS,10bS)-3a-
benzyl-4-oxo-2-phenyl-3aH-benzo[c]oxazolo[5,4-e]azepin-5(4H,6H,10bH)-yl)acetate (300 mg, 76% yield). The resulting ester was
stirred overnight in 20% TFA/CH2Cl2 (15 mL). Upon completion, the mixture was concentrated azeotropically distilled with MeOH
(x 3), and purified by column chromatography (0 to 15% MeOH in CH2Cl2) to afford KI-MS2-084 as a white solid (210 mg,
0.49 mmol, 60% yield over 2 steps). 1H NMR (400 MHz, CD3OD): d 7.94 (d, J = 7.5 Hz, 2H), 7.45 (t, J = 7.5 Hz, 2H), 7.34 (t, J =
7.5 Hz, 2H), 7.24 – 7.18 (m, 2H), 7.18 – 7.04 (m, 6H), 5.73 (s, 1H), 4.44 – 4.15 (m, 2H), 4.11 (q, J = 7.0 Hz, 1H), 3.98 – 3.56 (m, 1H),
3.52 – 3.23 (m, 2H). 13C NMR (125 MHz, CDCl3): d 172.90, 171.64, 170.99, 132.44, 131.56 (2), 130.21, 129.02 (3), 128.85 (5),
128.62, 128.56 (2), 127.52, 127.44, 126.77, 71.97, 60.83, 50.94, 45.16, 30.09. HRMS (m/z): [M]+ calcd. for C26H22N2O4, 427.1652;
found 427.1678.
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e6

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
( )-N-(2-(2-(2-aminoethoxy)ethoxy)ethyl)-2-((3aS,10bS)-3a-benzyl-4-oxo-2-phenyl-3aH-benzo[c]oxazolo[5,4-e]
azepin-5(4H,6H,10bH)-yl)acetamide 2,2,2-trifluoroacetate (KI-MS2-081)
To a solution of KI-MS2-084 (20 mg, 47 mmol) and tert-butyl (2-(2-(2-aminoethoxy)ethoxy)ethyl)carbamate (14 mg, 56 mmol) in DMF
(1 mL) was added HATU (23 mg, 61 mmol) and DIPEA (17 mL, 94 mmol). The mixture was stirred for 18 hr, at room temperature after
which time the mixture was concentrated and purified by column chromatography on silica gel (25 to 80% Acetone/hexanes) to afford
the boc-protected product. The material was re-dissolved in 10% TFA/DCM, and stirred at room temperature for 5 hr, after which
time the mixture was concentrated azeotropically distilled with MeOH (3 x), and purified by column chromatography (0 to 15%
1
MeOH in CH2Cl2) to afford KI-MS2-081 as a colorless oil (20 mg, 36 mmol, 77% yield over 2 steps). H NMR (400 MHz, CD3OD):
d 7.98 (d, J = 7.5 Hz, 2H), 7.60 (t, J = 7.5 Hz, 1H), 7.51 (t, J = 7.5 Hz, 2H), 7.44 (d, J = 7.5 Hz, 1H), 7.41 – 7.27 (m, 5H), 7.26 – 7.19
(m, 3H), 6.04 (s, 1H), 4.76 (bs, 1H), 4.32 (bs, 1H), 4.23 – 4.01 (m, 2H), 3.66 (s, 2H), 3.62 (t, J = 5.0 Hz, 2H), 3.59 – 3.55 (m, 2H),
3.55 – 3.47 (m, 4H), 3.41 – 3.35 (m, 2H), 3.21 (q, J = 7.4 Hz, 1H), 3.07 (t, J = 5.0 Hz, 2H), 1.31 (dd, J = 14.0, 6.5 Hz, 2H). 13C NMR
(125 MHz, CD3OD): d 171.49, 169.71, 148.64, 136.59, 134.41, 132.21, 130.98 (2), 128.50 (2), 128.40, 128.23 (4), 127.87 (3),
127.66, 127.02, 126.49, 87.94, 70.01, 69.95, 68.97, 66.80, 66.47, 51.91, 50.43, 44.59, 39.41, 38.89. HRMS (m/z): [M]+ calcd. for
C32H36N4O5, 557.2758; found 557.2757.
Synthesis of KI-MS2-085 (Diazirine/Alkyne Functionalized Probe)
Preparation of N-(2-((2-aminoethyl)amino)-2-oxoethyl)-3-(3-methyl-3H-diazirin-3-yl)-N-(prop-2-yn-1-yl)propanamide
2,2,2-trifluoroacetate
To a solution of 3-(3-methyl-3H-diazirin-3-yl)propanoic acid (0.87 g, 6.70 mmol, prepared following Org. Lett. 15(19), 5060-5063,
2013) and ethyl 2-(prop-2-yn-1-ylamino)acetate (0.80 g, 5.66 mmol, prepared following Synthesis 3, 488-494, 2009) in DMF
(10 mL) were added PyBop (2.95 g, 5.66 mmol) and DIPEA (1.97 mL, 11.32 mol). The mixture was stirred at room temperature
for 18 hr, concentrated and purified by column chromatography on silica gel (25% EtOAc in hexanes) affording ethyl 2-(3-(3-
methyl-3H-diazirin-3-yl)-N-(prop-2-yn-1-yl)propanamido)acetate (1.1 g). NaOH (0.39 g, 9.74 mmol) was added to the resulting
ester (1.10 g, 4.87 mmol) in MeOH:H2O (5:1, 20 mL) and the mixture was stirred overnight. The solvent was removed under
vacuum, and the residue partitioned between CH2Cl2 and H2O. 6N HCl was slowly added until pH 2 was reached and the product
was extracted with CH2Cl2. The combined organic layers were dried over MgSO4, filtered and concentrated affording the
carboxylic acid (2-(3-(3-methyl-3H-diazirin-3-yl)-N-(prop-2-yn-1-yl)propanamido)acetic acid). LCMS: M+1=224. To a solution of
the carboxylic acid (210 mg, 941 mmol) and tert-butyl (2-aminoethyl)carbamate (137 mg, 855 mmol) in DMF (2 mL) were added
EDC hydrochloride (0.21 g, 1.11 mmol), HOBt (0.14 g, 1.03 mmol) and DIPEA (0.30 mL, 1.71 mmol). The reacting mixture
was stirred at room temperature overnight, then diluted with sat. NaHCO3, and the aqueous layer extracted with EtOAc.
The combined organic layers were dried over MgSO4, filtered and concentrated to afford the boc-protected product (tert-butyl
e7 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
(2-(2-(3-(3-methyl-3H-diazirin-3-yl)-N-(prop-2-yn-1-yl)propanamido)acetamido)ethyl)carbamate) in good purity. This material
was stirred in 10% TFA/DCM for 3 hr which showed completed conversion by LCMS. The mixture was concentrated and
azeotropically distilled with MeOH (3x) to afford N-(2-((2-aminoethyl)amino)-2-oxoethyl)-3-(3-methyl-3H-diazirin-3-yl)-N-(prop-
2-yn-1-yl)propanamide 2,2,2-trifluoroacetate.
( )-N-(2-((2-(2-((3aS,10bS)-3a-benzyl-4-oxo-2-phenyl-3aH-benzo[c]oxazolo[5,4-e]azepin-5(4H,6H,10bH)-yl)
acetamido)ethyl)amino)-2-oxoethyl)-3-(3-methyl-3H-diazirin-3-yl)-N-(prop-2-yn-1-yl)propanamide (KI-MS2-085)
To a solution of KI-MS2-084 (3 mg, 7.0 mmol) and N-(2-((2-aminoethyl)amino)-2-oxoethyl)-3-(3-methyl-3H-diazirin-3-yl)-N-(prop-2-
yn-1-yl)propanamide 2,2,2-trifluoroacetate (3 mg, 8.5 mmol) in DMF (0.3 mL) were added HATU (4 mg, 9.1 mmol) and DIPEA (4 mL,
21 mmol). The mixture was stirred for 18 hr, concentrated and purified by preparative TLC (7% MeOH in CH2Cl2) to afford
KI-MS2-085 (4 mg, 5.9 mmol, 85% yield over 2 steps). 1H NMR (400 MHz, CDCl3): d 8.03 (d, J = 7.5 Hz, 2H), 7.59 (t, J =
7.5 Hz, 1H), 7.47 (t, J = 7.5 Hz, 2H), 7.35 (m, 6H), 7.27 (m, 4H), 6.90 (s, 1H), 5.87 (s, 1H), 4.24 (m, 2H), 4.07 (s, 2H), 3.87 (m, 1H),
3.51 (m, 2H), 3.26 (m, 3H), 2.36 (m, 2H), 1.74 (m, 2H), 1.43 (d, J = 6.5 Hz, 1H), 1.00 (s, 3H). 13C NMR (125 MHz, CDCl3): d 172.55,
171.93, 170.50, 170.02, 169.76, 169.17, 168.45, 167.93, 134.42, 133.08, 131.25, 128.98, 128.94, 128.85, 128.55, 128.47, 127.58,
125.37, 78.66, 77.79, 73.58, 72.93, 54.16, 53.41, 51.01, 50.15, 49.79, 44.88, 40.09, 40.00, 38.98, 38.79, 36.11, 29.45, 29.37,
27.34, 27.27, 25.55. HRMS (m/z): [M]+ calcd. for C38H39N7O5, 674.3085; found 674.3070.
Small Molecule Microarray
Small molecule microarrays were manufactured and analyzed against purified Max as previously described (Clemons et al., 2010)
Briefly, pure His6-tagged Max (Panomics, Cat # RP0080) was incubated on glass slides with printed compounds at 1 mg/mL in
TBST buffer supplemented with 5 mM MgCl2 and 10% (v/v) glycerol. Binding to the arrays was detected using an Alexa Fluorꢀ
647-conjugated monoclonal anti-His5 antibody (Qiagen, Cat #35370) at 1:1000 dilution in the same buffer. Incubated arrays were
scanned using an Axon 4000B scanner at 5 mm resolution using 635 nm and 532 nm lasers to detect assay positives and printed
sentinels. Screened arrays were analyzed using GenePix Pro 6.0 software from Molecular Devices and hit calls were determined
as outlined in the Clemons et al. manuscript.
Transcriptional Reporter Assay
HEK 293T cells were transfected with the Myc reporter construct (Devgan et al., 2012) (Cignal Myc Reporter (luc) Kit; #CCS-
012L, Qiagen) by adding attractene and Myc reporter to cells in OptiMEM. Specifically, 0.6 mL of attractene and 25 mL of
OptiMEM were mixed per well, in addition to 0.5 mL of Myc reporter and 25 mL of OptiMEM. After 5 min, these two solutions
were mixed together and incubated at room temperature for 30 min. 20,000 cells in OptiMEM with 5% (v/v) FBS and no anti-
biotics were seeded into each well and 50 mL of the attractene/Myc reporter solution was added to each well. Cells were
allowed to incubate at 37ꢁC for about 17 hr. Media was aspirated and replaced with 50 mL OptiMEM + 0.5% (v/v) FBS with
1x nonessential amino acids for each well. ꢀ7 hr later, media was aspirated and replaced with 50 mL of media containing
0.4% (v/v) DMSO and compound concentrations ranging from 0 to 40 mM. After 16 hr of compound treatment, cells were
allowed to reach room temperature. 50 mL of Dual-Glo Luciferase Reagent (#E2940, Promega) was added to each well and
plates were incubated for 15 min, followed by a luminescence reading for firefly luciferase activity (automatic attenuation,
500 ms integration time). 50 mL of Dual-Glo Stop & Glo Reagent (Promega) was added to each well and incubated for
15 min, followed by another luminescence reading for Renilla luciferase activity.
Viable Cell Assay
96-well plates were seeded with 2000 or 5000 cells in 100 mL media with 0.4% (v/v) DMSO and compound concentrations ranging
from 0 to 40 mM. For some conditions, 0.1 mg/ mL doxycycline (D9891, Sigma) and/or 1 mM b-estradiol (E2758, Sigma) was added to
the cell culture. After the appropriate amount of time (3 – 5 days), plates were allowed to reach room temperature for 30 min and
100 mL of CellTiter-Glo reagent (#G7572, Promega) was added to each well. The plates were placed on a shaker with medium
setting for 2 min and allowed to incubate at room temperature for 10 min. Plates were read for luminescence with automatic
attenuation and 500 ms integration time.
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e8

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Flow Cytometry
P493-6 cells were treated with 0 – 40 mM KI-MS2-008 for 3 – 5 days. 1 – 5 million cells were resuspended in PBS with or without
10 mg/mL propdium iodide (81845, Sigma-Aldrich). Cell mixtures were passed through a cell strainer (352235, Falcon), and counted
using a BD FACSCANTOTM II (BD BioSciences), and analyzed with BD FACSDIVATM software.
Bead-Based Lysate Pull-down with KI-MS2-081
50 mL of well-mixed NHS activated magnetic beads (#88826, ThermoFisher Scientific) were washed in DMSO three times. 100 mL of
DMSO, 0.5 mL of 10 mM KI-MS2-081 or DMSO and 0.75 mL of triethylamine were added to the beads. The mixture was protected from
light and allowed to rotate overnight at room temperature. The next day, aminoethanol (2.5 mL) was added to the mixture and allowed
to incubate at room temperature overnight. The third day, beads were washed once in DMSO, followed by three washes in wash
buffer (150 mM NaCl, 50 mM Tris, protease inhibitor cocktail, phosphatase inhibitor cocktail, pH 7.5). Whole cell lysate was prepared
by washing P493-6 cells in cold PBS twice and resuspending in lysis buffer (200 mM NaCl, 50 mM Tris, 1% (w/v) NP-40, 0.1 g/100 mL
sodium deoxycholate, protease inhibitor cocktail, phosphatase inhibitor, 0.4 mL benzonase/10 mL buffer) at a density of roughly
50 million cells/mL. Following a brief vortex, this mixture was iced for 20 min and centrifuged at 4ꢁC for 10 min at 14,000 g. 4 mg
of protein was incubated with the washed beads for 1 hr. The beads were washed three times in wash buffer and boiled in 50 mL
2x SDS loading buffer (#1610737, Bio-Rad) at 90ꢁC for 5 min. c-Myc and Max proteins were visualized on immunoblots using
anti-c-Myc antibody (9E10, Santa Cruz Biotechnology) and anti-Max antibody (C-124, Santa Cruz Biotechnology), with anti-mouse
IgG, HRP-linked Antibody (#7076, Cell Signaling Technology) and anti-rabbit IgG, HRP-linked antibody (#7074, Cell Signaling
Technology) as secondary antibodies.
In Vitro Labeling of Max with KI-MS2-085
In the case of non-covalent probes, photoaffinity tags can be added to create a covalent bond between the probe and protein target
in addition to a clickable handle to ensure pull-down of all targets including the lower affinity binders. (Blencowe and Hayes, 2005;
Mackinnon and Taunton, 2009; Smith and Collins, 2015) The alkyne handles can be clicked to a fluorophore to visually analyze the
protein targets. (Speers and Cravatt, 2009).
Recombinant human Max protein (1 mL, 50 mg/mL, Abcam, ab95309) was treated with DMSO or 5 mM KI-MS2-085. After treating for
1 hr, the samples were irradiated at 365 nm for 10 min (except for the –UV sample). The proteins were incubated with AlexaFluorꢁ 647
azide (1 mL, 10 mM, ThermoFisher Scientific, #A10277), TCEP (1 mL, 50 mM), and TBTA (3 mL, 1.7 mM). After a brief vortex, CuSO4
(1 mL, 50 mM) was added and the click reaction was incubated at room temp for 1 hr. The KI-MS2-085-labeled samples were
added to loading buffer (25 mL, 2X Laemmli Sample Buffer, Bio-Rad, #1610737) and the proteins were boiled at 90ꢁC for 5 min.
Each sample was loaded onto a 4-20% denaturing PAGE gel. Samples were electrophoresed at 200V until the loading dye
was at the bottom of the gel. The gels were visualized using the Odyssey classic infrared imaging system (LI-COR Biotech.). After
visualization, KI-MS2-085-labeled Max was submitted for MALDI analysis.
Live-Cell Pull-down with KI-MS2-085
P493-6 cell culture (5 mL) at 1-2 million cells/mL was treated with DMSO, 2.5, 5, or 10 mM KI-MS2-085, or 10 mM KI-MS2-085 with
40 mM KI-MS2-008. After treating for 1 hr, the cells were washed in PBS at 4ꢁC and irradiated at 365 nm for 10 min (except for the –UV
sample). The cells were pelleted and resuspended in lysis buffer (200 mM NaCl, 50 mM Tris, 1% (w/v) NP-40, 0.1 g/100 mL sodium
deoxycholate, protease inhibitor cocktail) and incubated on ice for 25 min. The supernatant was recovered by spinning at 14000 g for
10 min at 4ꢁC. The KI-MS2-085-labeled lysates were diluted with H2O (400 mL) and incubated with Albumin depletion resin (200 mL
slurry, ThermoFisher Scientific, #85160) for 10 min at room temp. The beads were filtered at 14000 g for 1 min. The beads were
washed once with PBS (100 mL) and filtered at 14000 g for 1 min.
The lysates (500 mL) were incubated with AlexaFluorꢁ 647 azide (1 mL, 10 mM, ThermoFisher Scientific, #A10277), TCEP (1 mL,
50 mM), and TBTA (3 mL, 1.7 mM). After a brief vortex, CuSO4 (1 mL, 50 mM) was added and the click reaction was incubated at
room temp for 1 hr.
Acetone (600 mL, chilled to -20ꢁC) was added to the HSA-free lysates. The precipates were pelleted at 21,000 g for 6 min. The pellet
was dispersed in acetone (1000 mL, chilled to -20ꢁC) using a sonicator. These steps were repeated twice. The resulting pelleted
lysates were resuspended in 1% (w/v) SDS in PBS (100 mL). The protein was quantified by Bradford assay. Protein lysates (50 mg)
was added to loading buffer (25 mL, 2X Laemmli Sample Buffer, Bio-Rad, #1610737) and the proteins were boiled at 90ꢁC for
5 min. Each sample was loaded onto a 4-20% denaturing PAGE gel. Samples were electrophoresed at 200V until the loading dye
was at the bottom of the gel. The gels were visualized using the Odyssey classic infrared imaging system (LI-COR Biotech.). Relative
Max protein levels were determined via immunoblotting, using anti-Max antibody (ab199489, Abcam) and IRDyeꢀ 800CW Goat
anti-Rabbit IgG (925-32211, LICOR).
Electrophoretic Mobility Shift Assay
For c-Myc/Max heterodimer disruption: Binding reactions containing binding buffer (2 mL of a 10X solution; 10X solution: 10X PBS,
1 mM EDTA, 1% (w/v) NP-40, 50% (v/v) glycerol), P493-6 lysate (0.5 mg, prepared fresh), KI-MS2-008 (or DMSO, 10058-F4), and
e9 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
DNase/RNase free H2O (to a final volume of 20 mL) were prepared in microcentrifuge tubes (0.65 mL) and incubated at 37ꢁC for
20 min. AlexaFluorꢁ 660-labeled E-Box DNA (1 mL, 50 fmol/mL, IDT) was added to the binding reaction and then incubated at
37ꢁC for 10 min.
For Max/Max homodimer induction: Binding reactions containing binding buffer (2 mL of a 10X solution; 10X solution: 10X PBS,
1 mM EDTA, 1% (w/v) NP-40, 50% (v/v) glycerol), KCl (6 mL, 1 M), recombinant Max (1 mL, 12.5 ng/mL, Abcam, ab95309),
KI-MS2-008 (or DMSO, NSC13728), and DNase/RNase free H2O (to a final volume of 20 mL) were prepared in microcentrifuge tubes
(0.65 mL) and incubated at room temp for 20 min. AlexaFluorꢁ 660-labeled E-Box DNA (1 mL, 50 fmol/mL, IDT) was added to the
binding reaction and then incubated at room temp for 10 min.
Loading dye (1 mL, 20X solution; 0.5X TBE, 40% (v/v) glycerol, 2 mg/mL Orange G dye, Sigma) was added to each reaction and
samples were loaded onto a 4-20% non-denaturing PAGE gel that was pre-run at 200 V for 1 hr in 0.5X TBE. Samples were electro-
3
phoresed at 200V until the loading dye was ꢀ = down the gel. The gels were visualized using the Odyssey classic infrared imaging
4
system (LI-COR Biotech.).
Analytical Ultracentrifugation
Recombinant human Max protein (Abcam, ab95309) was dialyzed (50 mM phosphate, pH 7.2, 100 mM NaCl, 1 mM EDTA) three times
to stabilize the chemical potential of the protein. Three samples were prepared for the experiment. Max protein (450 mL, 0.056 mg/ml)
with DMSO and Max protein (450 mL, 0.056 mg/ml) pre-incubated (30 min) with KI-MS2-008 (10 mM or 20 mM) were added to the
analytical ultracentrifugation (AUC) assembly cells. The assembly cells were positioned into a Beckman XL-I Analytical Ultracentri-
fuge and centrifuged at 42,000 rpm at 20ꢁC. Sedimentation velocity experiments were started scanning A280 every ꢀ1.2 min for 18 hr.
The data (both absorbance and interference) were analyzed with SedFit software utilizing the continuous c(s) distribution model
keeping resolution (200), partial specific volume (0.707), buffer density (1.00), buffer viscosity (0.01002), bottom (7.15), and one sigma
confidence level (0.68) constant. The data were fitted for the frictional ratio (1.68 for DMSO, 1.22 for 10 mM, and 1.53 for 20 mM),
meniscus (5.94 for DMSO and 10 mM, and 5.92 for 20 mM), baseline (0.00 for all samples), RI noise, and time independent noise.
The estimations for the molecular weights of the two species were 25 kDa and 51 kDa, respectively.
The data were exported from SedFit and imported into SedPhat. The data were analyzed in SedPhat utilizing the monomer-dimer
self-association model (Brautigam, 2011) keeping M1 (20,000), S1 (1.964 for DMSO, 1.770 for 10 mM, and 1.964 for 20 mM), S2 (2.683
for DMSO, 2.447 for 10 mM, and 2.685 for 20 mM), mass conservation, and log10 nonid sed field (-2.0458) constant. The data were
fitted for logKa (5.12153 for DMSO, 7.93937 for 10 mM, and 8.01024 for 20 mM), logk- (-1.02487 for DMSO, -1.21487 for 10 mM,
and -1.01234 for 20 mM), and incompetent fraction (0.00000 for DMSO, 0.03591 for 10 mM, and 0.04173 for 20 mM).
Size Exclusion Chromatography
Recombinant human Max protein (Abcam, ab95309) was dialyzed (50 mM phosphate, pH 7.2, 100 mM NaCl, 1 mM EDTA) three times
to stabilize the chemical potential of the protein. Two samples were prepared for the experiment. Max protein (429 mL, 0.50 mg/mL
and 0.056 mg/ml) with DMSO and Max protein (429 mL, 0.50 mg/mL and 0.056 mg/ml) pre-incubated (30 min) with KI-MS2-008
(10 mM) were injected onto a Superdex 200 GL size exclusion column using 100% buffer (50 mM NaHPO4, 150 mM NaCl, 1 mM
BME, 5% (v/v) glycerol, and 2% (v/v) DMSO) to separate the species in solution. The monomer vs. dimer distribution was analyzed
by integrating the A280
.
Cellular Thermal Shift Assay (Molina et al., 2013)
Protocol adapted from Jafari et al. (Jafari et al., 2014). For dose dependent CETSA, DMSO or KI-MS2-008 (0.078 mM, 0.156 mM,
0.313 mM, 0.625 mM, 1.25 mM, 2.50 mM, and 5.00 mM) were incubated with 4 x 106 cells for 1 hr at 37ꢁC. For time course CETSA,
10 mM KI-MS2-008 was incubated with 4 x 106 cells for 0, 0.25 hr, 0.5 hr, 0.75 hr, 1 hr, 4 hr, 8 hr, and 24 hr at 37ꢁC. Cells were
then washed and harvested with cold PBS. After transferring to falcon tubes these were spun down for 3 min at 500 g, and the
supernatant discarded. The cell pellet was resuspended in PBS containing protease inhibitor cocktail (100 mL) and transferred to
PCR tubes. Cells were heat shocked in a thermal cycler at 60.0ꢁC for 3 min. The cell suspensions were then subjected to three
freeze-thaw cycles with liquid nitrogen to lyse cells. For a uniform thawing this step was performed in a thermal cycler at 25ꢁC.
Lysates were spun at 21,000 g for 45 min at 4ꢁC to clarify and pellet aggregated protein. The tubes were carefully handled after
this step to avoid disturbing the pellet. The supernatant containing the soluble protein fraction was transferred to a new tube. Protein
lysates (7.5 mL) was added to loading buffer (7.5 mL, 2X Laemmli Sample Buffer, Bio-Rad, #1610737) and the proteins were boiled at
90ꢁC for 5 min. Each sample was loaded onto a 4-20% denaturing PAGE gel. Samples were electrophoresed at 200 V until the
loading dye was at the bottom of the gel. Relative c-Myc and Max protein levels were determined via immunoblotting, using
anti-c-Myc antibody (9E10, Santa Cruz Biotechnology) and anti-Max antibody (ab199489, Abcam).
15N-Heteronuclear Single Quantum Coherence
Experiments were conducted at NMX. HSQC was performed on 15N-Max*VL(tr) (50 mM in 50 mM NaPi pH 6.8, 250 mM KCl, 3.6%
DMSO d-6 and 10% D2O at 35ꢁC) in the absence and presence of KI-MS2-008 (150 mM). Protein NMR was recorded on a Bruker
Avance600 with the HSQCETFPF3GPSI sequence: 2D [1H,15N]-correlation via double inept transfer using sensitivity improvement;
15N-decoupling during acquisition using the F3 channel; phase sensitive using Echo/Antiecho-TPPI gradient selection; and a
flip-back water pulse (Grzesiek and Bax, 1993; Kay et al., 1992; Palmer et al., 1991; Schleucher et al., 1994).
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e10

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
Time Course of Protein Levels
P493-6 cell culture (5 mL) at 1-2 million cells/mL was treated with 10 mM KI-MS2-008 or 0.1% (v/v) DMSO. After the indicated time,
cells were washed in PBS at 4ꢁC, resuspended in lysis buffer (200 mM NaCl, 50 mM Tris, 1% (w/v) NP-40, 0.1 g/100 mL sodium
deoxycholate, protease inhibitor cocktail, phosphatase inhibitor, 0.4 mL benzonase/10 mL buffer) and incubated on ice for 20 min.
The supernatant was recovered by spinning at 14,000 g for 10 min at 4ꢁC. Relative protein levels were determined via immunoblot-
ting, using anti-c-Myc antibody (9E10, Santa Cruz Biotechnology), anti-c-Myc (pT58) antibody (ab28842, Abcam), anti-c-Myc (pS62)
antibody (ab78318), anti-Max antibody (ab199489, Abcam), anti-MLX antibody (sc393086, Santa Cruz Biotechnology), anti-Mxd1
antibody (sc8012, Santa Cruz Biotechnology), anti-Mxd4 antibody (sc1221, Santa Cruz Biotechnology) and anti-vinculin antibody
(V 9131, Sigma).
RNA-seq
Cells were plated in 6-well plates at a concentration of 3 million cells in 3 mL media and treated with 1 mM or 10 mM KI-MS2-008,
0.1 mg/mL doxycycline, or DMSO for 8 hr with 4 replicates per condition. RNA was isolated from each sample guided by the AurumTM
Total RNA Mini Kit (Bio-Rad). RNA samples were quantified and quality assessed using an Advanced Analytical Fragment Analyzer.
20 ng of total RNA was used for library preparation with ERCC Spike-in control Mix A (Ambion 10-6 final dilution). All steps were per-
formed on a Tecan EVO150. 3’DGE-custom primers 3V6NEXT-bmc#1-24 were added to a final concentration of 1.2 mM. (5’-/5Biosg/
ACACTCTTTCCCTACACGACGCTCTTCCGATCT[BC6]N10T30VN-3’ where 5Biosg = 5’ biotin, [BC6] = 6bp barcode specific to each
sample/well, N10 = Unique Molecular Identifiers, Integrated DNA technologies).
After addition of the oligonucleotides, samples were denatured at 72ꢁC for 2 min followed by addition of SMARTScribe RT per man-
ufacturer’s recommendations with Template-Switching oligo5V6NEXT (12mM, [5V6NEXT : 5’-iCiGiCACACTCTTTCCCTACACGA
CGCrGrGrG-3’ where iC: iso-dC, iG: iso-dG, rG: RNA G ]) and incubation at 42ꢁC for 90 min followed by inactivation at 72ꢁC for
10 min. Following the template switching reaction, cDNA from 24 wells containing unique well identifiers were pooled together
and cleaned using RNA Ampure beads at 1.0X. cDNA was eluted with 90 mL of water followed by digestion with Exonuclease I at
37ꢁC for 45 min, inactivation at 80ꢁC for 20 min. Single stranded cDNA was then cleaned using RNA Ampure beads at 1.0X and eluted
in 50 mL of water.
Second strand synthesis and PCR amplification was done using the Advantage 2 Polymerase Mix (Clontech) and the SINGV6
primer (10 pmol, Integrated DNA Technologies 5’-/5Biosg/ACACTCTTTCCCTACACGACGC-3’). 12 cycles of PCR were performed
followed by clean up using regular SPRI beads at 1.0X, and was eluted with 20 mL of EB. Successful amplification of cDNA was
confirmed using the Fragment Analyzer. Illumina libraries were then produced using standard Nextera tagmentation substituting
P5NEXTPT5-bmc primer (25mM, Integrated DNA Technologies, (5’-AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACA
CGACGCTCTTCCG*A*T*C*T*-3’ where * = phosphorothioate bonds.) in place of the normal N500 primer. Final libraries were cleaned
using SPRI beads at 0.8X and quantified using the Fragment Analyzer and qPCR before being loaded for paired-end sequencing us-
ing the Illumina NextSeq500. Post-sequencing, quality-control on each of the libraries was performed to assess coverage depth,
enrichment for messenger RNA (exon/intron and exon/intergenic density ratios), fraction of rRNA reads and number of detected
genes using bespoke scripts. Sequences were aligned against the human genome hg19 and ERCC reference sequences separately
using bwa [http://bio-bwa.sourceforge.net/]. Gene expression was estimated based on reads mapping near the 3’ end of transcripts
using ESAT (Derr et al., 2016) based on the hg19 Refseq annotation. Results were summarized as counts per million mapped reads
(CPMs), merged across samples, log-transformed and subjected to hierarchical clustering and visualization. For ERCC quantifica-
tion, bam files from bwa were sorted and indexed with samtools (Li et al., 2009) and counts were retrieved from the indices using
idxstats (Anders and Huber, 2010; Benjamini and Hochberg, 1995; Cook, 1977; Langmead et al., 2009; Li et al., 2009; Love et al.,
2014; Soumillon et al., 2014).
qPCR
Cells were plated in 6-well plates at a concentration of 3 million cells/mL in 3 mL media and treated with 10 mM KI-MS2-008, 10 mM
MG132, both KI-MS2-008 and MG132, or 0.4% (v/v) DMSO for 15 min, 30 min, 45 min, 1 hr , 2 hr or 4 hr. RNA was isolated from each
sample guided by the AurumTM Total RNA Mini Kit (Bio-Rad). RNA samples were quantified using a NanoDrop OneC (Thermo
Scientific). A High Capacity cDNA Reverse Transcription Kit (appliedbiosystems) was used to generate cDNA from the isolated
RNA. Primers (Eurofins) and PowerUpTM SYBRTM Green Master Mix (appliedbiosystems) were added to cDNA samples and DNA
was amplified using a CFX384 TouchTM Real-Time PCR Detection System (Bio-Rad). Data was normalized to DMSO controls and
beta-actin for a reference gene, using the DDCt method (Livak and Schmittgen, 2001).
Primers for qPCR
APEX1: forward primer – 5’-CAGATCTCGCGAGTAGGGCAACG-3’;
reverse primer – 5’-GTCTTACTCTTCTTGGCCTCTGG-3’.
DUSP2: forward primer – 5’-GGGCTCCTGTCTACGACCA-3’;
reverse primer – 5’-GCAGGTCTGACGAGTGACTG-3’.
HK2: forward primer – 5’-GGGACAATGGATGCCTAGATG-3’;
reverse primer – 5’-GTTACGGACAATCTCACCCAG-3’.
IPO4: forward primer – 5’-CTCGCAAGTTGTACGCAATGC-3’;
e11 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
reverse primer – 5’-GGCACCGACTTCAAGTAGGC-3’.
MAX: forward primer – 5’-ATAATGCACTGGAACGAAAACG-3’;
reverse primer – 5’-TGGCTTTGTCTAGGATTTGGG-3’.
MYC (Lu et al., 2010): forward primer – 5’-AAACACAAACTTGAACAGCTAC-3’;
reverse primer – 5’-ATTTGAGGCAGTTTACATTATGG-3’.
PES1: forward primer – 5’-GGGCATTTATCCCCATGAACC-3’;
reverse primer – 5’-CACCTTGTATTCACGGAACTTGT-3’.
PPRC1: forward primer – 5’-AAGACAGCTCTGCAGAGAAGC-3’;
reverse primer – 5’-CCACAGGTTTGCTTACAGGTG-3’.
RRP12: forward primer – 5’-GTGACCTGACAGTCGATGCTG-3’;
reverse primer – 5’-AAAGGCCACTCAGGAAGGTAC-3’.
SRM: forward primer – 5’-AGAGAGACGAGTTCTCCTACC-3’;
reverse primer – 5’-CCCGATGATCAGCACCTTTCG-3’.
TERT: forward primer – 5’-ATCCTGGCCAAGTTCCTGCAC-3’;
reverse primer – 5’-CAACTTGCTCCAGACACTCTTC-3’.
ACTB (Calderon et al., 2012): forward primer – 5’-CACCATTGGCAATGAGCGGTTC-3’;
reverse primer – 5’-AGGTCTTTGCGGATGTCCACGT-3’.
Ubiquitination Assay
15 million P493-6 cells were resuspended in 15 mL media and plated in 20 cm dishes. Cells were treated with 0.4% (v/v) DMSO,
10 mM KI-MS2-008, 10 mM MG132, or 10 mM KI-MS2-008 and 10 mM MG132 for 4 hr. Cells were lysed in 100 mL modified RIPA buffer
(200 mM NaCl, 50 mM Tris, 1% (w/v) NP-40, 0.1 g/100 mL sodium deoxycholate, protease inhibitor cocktail, phosphatase inhibitor,
0.4 mL benzonase/10 mL buffer) and lysate corresponding to 0.5 mg of total protein with 5 mL c-Myc antibody (N-262, Santa Cruz
Biotechnology) or 2.5 mL normal rabbit IgG (sc-2027, Santa Cruz Biotechnology) was incubated overnight at 4ꢁC. 25 mL Dynabeads
(10007D, invitrogen) were added to antibody-protein mixture and incubated for 1 hr at 4ꢁC. Beads were washed four times and
samples were boiled in 2x SDS loading buffer (#1610737, Bio-Rad) at 90ꢁC for 5 min. c-Myc (9E10, Santa Cruz Biotechnology)
and ubiquitin (BML-PW8810-0100, Enzo) were visualized via immunoblotting.
Protein Binding Assay
Experiments were conducted at Cerep. Protein binding assay that simulates human plasma was spiked with 10 mM of compound at
1% (v/v) DMSO. A dialysate compartment was loaded with PBS. After incubation for 4 hr at 37ꢁC, samples were taken from each
compartment, centrifuged and supernatants were submitted to HPLC-MS/MS analysis.
Solubility of KI-MS2-001 in Aqueous Buffers
Compounds were prepared at 200 mM from 10 mM stock solutions. They were mixed with PBS buffer, simulated intestinal fluid
(pH 6.5), or simulated gastric fluid (pH 1.2) with a final DMSO concentration at 2%. They were then incubated for 24 hr followed
by HPLC analysis.
Bidirectional Permeability of KI-MS2-001
Experiments were conducted at Cerep. KI-MS2-001 was prepared at 10 mM in HBSS-MES buffer, pH 6.5 and added to the apical
side. HBSS-HEPES buffer, pH 7.4 was added to the basolateral side. Incubation was for 60 min total. Samples were taken at time
zero from the donor side and at the end of the incubation from both the donor and the receiver sides. Samples were analyzed by
HPLC-MS/MS.
Liver Microsome Stability of KI-MS2-001
Experiments were conducted at Cerep. KI-MS2-001 was pre-incubated with pooled liver microsomes in phosphate buffer for 5 min
at 37ꢁC. The reaction was initiated by adding NADPH system and incubated for 0, 15, 30, 45 and 60 min. Concentration was
measured by HPLC-MS/MS analysis.
Chromatin Immunoprecipitation Sequencing
ChIP-seq for c-Myc, Max, and H3K27ac was performed as in (Lin et al., 2012) with minor changes. P493-6 cells were grown to a
density of 1 3 106 cells / mL prior to the start of time course. At each time point, pellets of 50 million cells were isolated and
cross-linked with 1% formaldehyde (10 min) followed by quenching (125 mM glycine). Cells were washed in cold PBS and harvested
by cell scraper in cold PBS with protease inhibitors (Roche). Cells were centrifuged at 1,650g for 5 min and flash frozen and stored at
–80ꢁC at 5 3 106 cells per pellet. Pellets were resuspended in cytosolic and then nuclear lysis buffer, and DNA was sheared at 4ꢁC
using a water-bath sonicator (Bioruptor, Diagenode) for 25 min at high output (30 s on, 30 s off) in 1 ml of sonication buffer supple-
mented with 0.5% SDS. Sonicated lysates were cleared by centrifuging at 20,000g for 10 min and incubated overnight end over end
at 4ꢁC with magnetic beads prebound with antibody. Beads were washed three times with sonication buffer, one time with sonication
buffer with 500 mM NaCl added, once with LiCl wash buffer (20 mM Tris pH 8.0, 1 mM EDTA, 250 mM LiCl, 0.5% NP-40, 0.5%
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e12

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
sodium deoxycholate), and once with TE. DNA was eluted in elution buffer (50 mM Tris-HCl pH 8, 10 mM EDTA, and 1% SDS). Cross-
links were reversed overnight at 65ꢁC. RNA and protein were digested with 0.2 mg/ml RNase A for 2 h followed by 0.2 mg/ml pro-
teinase K for 1 hour. DNA was purified with phenol chloroform extraction and ethanol precipitation.
The following antibodies were used: Max (sc-197 Max (C-17) lot C3016, Santa Cruz), Myc (ab56, Abcam), H3K27ac (ab4729,
Abcam). Libraries for sequencing were prepared using the Rubicon Thruplex DNA-seq/FD library preparation kit. An input of
50 ng of DNA or less was used, and following ligation, libraries were amplified per the manufacturer’s instructions. Amplified libraries
were then size selected using AMPure beads (Agencourt AMPure XP) according to the manufacturer’s instructions. Further size
selection was performed using a 2% gel cassette in the Pippin Prep (SAGE Sciences) set to capture fragments of 200–700 bp in
size. Libraries were multiplexed at equimolar ratios and run on a NextSeq 500 instrument (75-bp, single-end reads).
Reads were aligned to the HG19 NCBI Refseq (GRCh37) reference using bowtie2 with default parameters. Enriched regions were
determined using MACS1.4.2 (Zhang et al., 2008) using a p-value threshold of 1e-9. To quantify ChIP-seq signal, we utilized the
bamliquidator tool (https://github.com/BradnerLab/pipeline/wiki/bamliquidator) which reports ChIP-seq signal in units of reads
per million per basepair (rpm/bp). To generate heatmaps of c-Myc, Max, and H3K27ac occupancy at active gene promoters, we first
defined active genes as those with a H3K27ac enriched region in at least one sample at in the TSS +/- 1kb region (n=14,792 genes).
ChIP-seq signal in units of rpm/bp was calculated in 200 bins evenly distributed across the +/-5kb region flanking the TSS. Meta
representations are plotted below heatmaps and show the mean ChIP-seq signal in units of rpm/bp.
Myc-Addicted Tumor Cell Lines in Mice
Myc-addicted cell lines were derived from either Em-tTA/tet-O-MYC (T-ALL) or LAP-tTA/tet-O-MYC (HCC) mice and labeled with
firefly luciferase. The minimal number of mice in each group was calculated using the ‘pwr.t.test’ function in the R/pwr package
(R Development Core Team 2009; http://www.r-project.org; Champely, 2018). 2 x 106 luciferase-labeled T-ALL or HCC cells were
injected intravenously (T-ALL) or subcutaneously (HCC) into 4-6 week-old male NOD-SCIDIL-2Rg-/- (NSG) mice. Engraftment and
tumor progression were visualized and quantified by bioluminescence imaging (BLI) using an in vivo bioluminescence/optical imag-
ing system (Ami HT, Spectral Instruments Imaging). The luminescence signal was assessed 10 min after intraperitoneal injection of
3 mg D-Luciferin (Promega) dissolved in PBS. Imaging was performed under general anesthesia with 2% isoflurane. Image analysis
was performed using AmiView software (V1.7.06, Spectral Instruments Imaging). Upon detection of tumor cell proliferation by BLI
(day 4), mice were injected with KI-MS2-008: T-ALL bearing mice received daily intravenous injections of 100 mL of either vehicle
(saline) or 20 mM KI-MS2-008/saline; HCC bearing mice were injected intraperitoneally (5 days on, 2 days off) with 200 mL of either
vehicle (10%(2-Hydroxypropyl)-b-cyclodextrin (HPBCD)/saline) or 40 mM KI-MS2-008/10%HPBCD/saline. Mice were imaged at the
times indicated. Mouse experiments were approved by the Administrative Panel on Laboratory Animal Care (APLAC) at Stanford Uni-
versity, and were carried out in accordance with institutional and national guidelines.
Immunofluorescence Staining and Imaging
A conditional Myc-addicted osteosarcoma cell line was cultured on glass slides and treated for two days with either vehicle (DMSO
or PBS), doxycycline (20 ng/ml), or KI-MS2-001 analogues (10 mM). Cells were labeled with antibodies specific for a-tubulin
(Anti-a-tubulin antibody, 1:1000, clone DM1A, Sigma-Aldrich, T9026) and Apex1 (Apex1 antibody, 1:200, Cell Signaling Technology,
#4128). Goat anti-rabbit Alexa Fluor-546 and goat anti-mouse Alexa Fluor-647 (both from Invitrogen) were used as secondary anti-
bodies. DNA was counterstained using Hoechst33342 (Invitrogen). Image acquisition was performed using a DMI6000B epifluores-
cence microscope (Leica) equipped with an HC PL APO 40x/0.85 DRY objective. Image processing was performed using ImageJ
(Schneider et al., 2012).
Metabolic Stability Studies
Microsomes (pooled male and female) were prepared at Cyprotex or were purchased from a reputable commercial supplier. Micro-
somes (final protein concentration 0.5 mg/mL), 0.1 M phosphate buffer pH 7.4 and test compound (final substrate concentration
3 mM; final DMSO concentration 0.25 %) were pre-incubated at 37ꢁC prior to the addition of NADPH (final concentration 1 mM) to
initiate the reaction. The final incubation volume was 50 mL. A control incubation was included for each compound tested where
0.1 M phosphate buffer pH 7.4 is added instead of NADPH (minus NADPH). Two control compounds are included with each species.
Each compound was incubated for 0, 5, 15, 30 and 45 min. The control (minus NADPH) was incubated for 45 min only. The reactions
were stopped by transferring 20 mL of incubate to 60 mL methanol at the appropriate time points. The termination plates were
centrifuged at 2,500 rpm for 20 min at 4ꢁC to precipitate the protein. The samples were centrifuged (2,500 rpm at 4ꢁC for 30 min)
and the supernatants at each time point pooled for analysis by LC-MS/MS using Cyprotex generic methods. Following the micro-
somal stability assays, the time point at the 45 min sample was chosen and utilized for metabolite profiling. The 45 min sample
was compared against the 0 min control sample to establish which and how many metabolites were formed. The metabolites found
have been displayed as extracted ion chromatograms (XIC) and the representative mass spectra. The areas and percentages re-
ported for the parent and metabolites have been calculated using the XIC data; it has been assumed that each metabolite has
the same ionization efficiency and that the sensitivity of the metabolite has not been affected by the biotransformation. Any potential
matrix effects have also been assumed to be consistent with the parent molecule. Metabolites observed at greater than 1% of the
total of drug related material are reported. The parent compound and its associated metabolites produced both protonated and
sodiated molecular ions with the sodium adducts being at a higher intensity; for the tabulated information the areas and associated
e13 Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019

Please cite this article in press as: Struntz et al., Stabilization of the Max Homodimer with a Small Molecule Attenuates Myc-Driven Transcription, Cell
Chemical Biology (2019), https://doi.org/10.1016/j.chembiol.2019.02.009
PPM errors have been reported from the sodium adducts. The m/z found for each metabolite in each species and its associated
ppm error, have been displayed for each species from the time point showing the largest peak area. For accuracy a ppm error of
less than 5 is desirable; when the response for a metabolite observed is low the ppm error maybe greater than 5. Where a metabolite
has been referred to as a potential oxidation, it can refer to a potential hydroxylation, epoxidation or oxide formation. Representative
collision induced dissociation (CID) mass spectra were obtained for the metabolites found. For elucidation the CID mass spectra for
the protonated species provided more informative fragment ions for interpretation purposes; therefore these spectra are used.
Pharmacokinetics
KI-MS2-008 was administered orally at 0.731 mg/kg in 20% cyclodextrin + 1.4% DMSO into healthy 6-9 month-old male LSL-Kras
G12D/+ ; Trp53 flox/flox mice. Blood samples were collected from animals (n=3/group) for vehicle control, 30 min, 1, 2, 4, 6 and 24 hr
post-compound administration in EDTA containing tubes (VWR). Plasma was isolated from the blood samples and frozen.
Subsequent LC/MS analysis was used to determine compound levels. Mouse experiments were approved by the Institutional
Animal Care and Use Committee (IUCAC) at the Massachusetts Institute of Technology, and were carried out in accordance with
institutional and national guidelines.
Repeat Dose Toxicity Studies
Healthy 6-9 month-old male LSL-Kras G12D/+ ; Trp53 flox/flox mice. (n=5 per group) were orally dosed with vehicle, 0.364 mg/kg or
0.731 mg/kg KI-MS2-008 in 20% cyclodextrin + 1.4% DMSO daily for 7 days. Animal weights were acquired each day. On day 8,
blood samples were collected to look at plasma concentration of compound, mice were sacrificed by CO2 asphyxiation, liver and
kidney were harvested and fixed in formalin. 24 hr later, fixed samples were transferred to 70% EtOH, embedded in paraffin,
sectioned to 4 mm thick sections, stained with hematoxylin and eosin (H&E) and analyzed for microscopic signs of organ damage.
Mouse experiments were approved by the Institutional Animal Care and Use Committee (IUCAC) at the Massachusetts Institute
of Technology, and were carried out in accordance with institutional and national guidelines.
LC/MS Analysis of Plasma Samples
Plasma Extraction
Samples of 10 mL plasma were extracted in eppendorf tubes by quickly adding 500 mL of cold MeOH, followed by 300 mL cold water,
and 500 mL cold chloroform. Tubes were vortexed at 4ꢁC, centrifuged for 5 min at max speed at 4ꢁC, the chloroform layer was trans-
ferred to a new tube and evaporated in a SpeedVac. Tubes were frozen at À80ꢁC until ready for LCMS analysis.
LC/MS Conditions
Frozen samples were brought to room temperature then reconstituted in 20 mL methanol and 80 mL water. Samples were aliquoted
into LCMS tubes with preslit caps and were run on C18 column with 5% acetonitrile (ACN, Sigma) in water + 0.1% formic acid (FA,
Sigma) as mobile phase A and 95% ACN in water + 0.1% as mobile phase B. A linear gradient was used: 0-0.5 min 0% B; 0.5-10.5 min
0-100% B; 10.5-15.5 min 100% B; 15.5-17.5 min 100-0% B; run stopped at 23 min. Flow rate was 0.4 mL/min and injection volume
was 10 mL. The mass spectrometer (Sciex API4000) was operated under multiple reaction monitoring mode scanning for the 4 most
prominent fragments of the compound: m/z 105.1, 519.1, 416, 205.1. The parameters were: curtain gas 25 units; ion source gas 1, 30
units; ion source gas 2, 30 units; temperature 550ꢁC; ion spray voltage 5500 V; declustering potential 96 V, and collision energy 57 eV.
QUANTIFICATION AND STATISTICAL ANALYSIS
For relevant experiments, number of replicates, error bars, statistical test, and P values are noted in the respective figure legends.
Most experiments were repeated at least 3 times to establish reproducibility with the following exceptions: The RNA-seq data for
Figure 5 was collected from 4 replicates treated separately, but from the same cell culture. The experiment with OS cells was
performed twice. The MALDI, LC/MS, 15N HMQC, and ChIP-seq data were acquired once. The flow cytometry experiment was
completed once, but with 10,000 counts per data point. No data were excluded except for one data point that was determined to
be an outlier and excluded from the analysis in Figure 5F. Image quantification was not blinded, but was analyzed by the same inves-
tigator in an unbiased manner. For the mouse studies, the minimal number of mice in each group was calculated using the ’pwr.t.test’
function in the R/pwr package (R Development Core Team, 2009; http://www.r-project.org; Champely, 2018). Mouse images were
coded with cage and mouse ID numbers, and were analyzed without knowledge of treatment group (although by the same
investigator).
DATA AND SOFTWARE AVAILABILITY
R code for RNA-seq analysis is available upon reasonable request. The accession number for the RNA-seq data reported in this pa-
per is GEO: GSE107222. The accession number for the ChIP-seq data reported in this paper is GEO: GSE125863. Full small molecule
screening data sets, along with compound structures for the libraries, are available upon reasonable request.
Cell Chemical Biology 26, 1–13.e1–e14, May 16, 2019 e14