Combined Omics Approach Identifies Gambogic Acid and Related Xanthones as Covalent Inhibitors of the Serine Palmitoyltransferase Complex

Article

Combined Omics Approach Identiﬁes Gambogic Acid

and Related Xanthones as Covalent Inhibitors of the

Serine Palmitoyltransferase Complex

Graphical Abstract

Authors

Dominic G. Hoch, Daniel Abegg,

J. Thomas Hannich, ..., Qidong You,

Howard Riezman, Alexander Adibekian

Correspondence

aadibeki@scripps.edu

In Brief

In this study, Hoch et al. discovered that

the natural product gambogic acid and

related synthetic xanthones bind to the

regulatory small subunit B of the serine

palmitoyltransferase complex, leading to

a decrease in sphingolipids, including

sphingosine-1-phosphate, in cells

and mice.

Highlights

d

Gambogic acid causes reduction of S1P levels through

binding to SPTSSB

Structurally related xanthone 18 is a more selective SPTSSB

binder

SPTSSB engagement by xanthone 18 leads to a decrease in

S1P levels in cells and mice

Xanthone 18 induces cellular responses characteristic for

diminished S1P signaling

Hoch et al., 2020, Cell Chemical Biology 27, 1–12

May 21, 2020 ª 2020 Elsevier Ltd.

https://doi.org/10.1016/j.chembiol.2020.03.008

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Cell Chemical Biology

Article

Combined Omics Approach Identiﬁes Gambogic

Acid and Related Xanthones as Covalent

Inhibitors of the Serine Palmitoyltransferase Complex

Dominic G. Hoch,^1,2,7Daniel Abegg,^1,7J. Thomas Hannich,²Dany Pechalrieu,¹Anton Shuster,¹Brendan G. Dwyer,¹

Chao Wang,^1,2,3Xiaojin Zhang,^4,5Qidong You,⁴Howard Riezman,²and Alexander Adibekian^1,6,

*

¹Department of Chemistry, The Scripps Research Institute, 130 Scripps Way, Jupiter, FL 33458, USA

²School of Chemistry and Biochemistry, National Centre of Competence in Research (NCCR) Chemical Biology, University of Geneva,

CH-1211 Geneva, Switzerland

³Department of Molecular Medicine, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA

⁴Jiangsu Key Laboratory of Drug Design and Optimization, China Pharmaceutical University, Nanjing 210009, China

⁵Department of Chemistry, School of Science, China Pharmaceutical University, Nanjing 211198, China

⁶Lead Contact

⁷These authors contributed equally

*Correspondence: aadibeki@scripps.edu

https://doi.org/10.1016/j.chembiol.2020.03.008

SUMMARY

Milstien, 2003; Maceyka et al., 2012). Binding to these receptors

leads to an activation of the ERK (Van Brocklyn et al., 2002; Ruth-

In this study, we identify the natural product gambo- erford et al., 2013), Akt (Means et al., 2007; Lee et al., 2001), and

mTOR (Liu et al., 2009; Maeurer et al., 2009) pathways, thus re-

sulting in an increase in procancerous cellular processes such as

gic acid as well as structurally related synthetic xan-

thones as ﬁrst-in-class covalent inhibitors of the de

novo sphingolipid biosynthesis. We apply chemo-

proteomics to determine that gambogic acid binds

to the regulatory small subunit B of the serine palmi-

toyltransferase complex (SPTSSB). We then test

structurally related synthetic xanthones to identify

18 as an equally potent but more selective binder of

apoptosis resistance, proliferation, inﬂammation, angiogenesis,

invasion, and metastasis (Maceyka et al., 2012). A monoclonal

antibody that depletes S1P, sonepcizumab, is being studied

for the treatment of renal cell carcinoma, advanced solid tumors,

and age-related macular degeneration (Pal et al., 2017; Gordon

et al., 2010; Lukowski et al., 2013). S1P signaling plays a key

role not only in cancer but also in the pathogenesis of various

other diseases. For example, S1PR functional antagonist ﬁngo-

limod that is derived from myriocin, a natural metabolite inhibitor

SPTSSB and show that 18 reduces sphingolipid

levels in situ and in vivo. Finally, using various biolog-

ical methods, we demonstrate that 18 induces of serine palmitoyltransferase (Brinkmann et al., 2010; Pelletier

and Haﬂer, 2012), is approved by the Food and Drug Administra-

tion for the treatment of relapsing-remitting multiple sclerosis.

cellular responses characteristic for diminished

sphingosine-1-phosphate (S1P) signaling. This study

demonstrates that SPTSSB may become a viable

therapeutic target in various diseases with patholog-

ical S1P signaling. Furthermore, we believe that our

compound will become a valuable tool for studying

the sphingolipid metabolism and serve as a blueprint

for the development of a new generation of sphingo-

Targeting S1P signaling has also been proposed as a possible

therapeutic option in inﬂammatory bowel disease (Danese

et al., 2018) various skin disorders (Holleran et al., 2006), sickle

cell disease, and vascular and cardiac diseases (Proia and Hla,

2015; Zhang et al., 2014; Yatomi, 2006).

The ﬁrst step in S1P de novo synthesis is catalyzed by serine

palmitoyltransferase (SPT), which condenses a fatty acyl-coen-

lipid biosynthesis inhibitors.

zyme A (CoA), typically palmitoyl-CoA, and serine (Merrill,

2002). Mammalian SPTs are heterotrimeric proteins consisting

of a large subunit dimer, SPTLC1 and SPTLC2 or SPTLC3, and

one of the two small activating subunits SPTSSA and SPTSSB

(Han et al., 2009). SPTs form 3-ketosphinganine, which is

INTRODUCTION

Ceramides and sphingosine-1-phosphate (S1P) are essential reduced to sphinganine and then N-acylated to produce various

bioactive signaling molecules that form a so-called sphingolipid dihydroceramides, which are then desaturated to ceramides.

rheostat, which controls the fate of mammalian cells through a Ceramides can be deacylated to form sphingosine, which in

rapid and dynamic interconversion of these species (Van Brock- turn can be phosphorylated by sphingosine kinases 1 and 2 to

lyn and Williams, 2012; Newton et al., 2015). Elevated ceramide form S1P (Berdyshev et al., 2006).

levels induce cell growth arrest and apoptosis, whereas S1P

Gambogic acid (GA) is a polyprenylated xanthone natural

suppresses ceramide-induced apoptosis and stimulates various product that is derived from the resin of Garcinia hanburyi, an

progrowth cellular processes (Hannun and Obeid, 1995; Cuvillier evergreen tree native to Southeast Asia (Banik et al., 2018).

et al., 1996). S1P is generated intracellularly and is then secreted GA-containing resin has been used as a traditional Chinese med-

to bind to the dedicated receptors S1PR1–S1PR5 (Spiegel and icine for detoxiﬁcation, homeostasis, anti-inﬂammatory, and

Cell Chemical Biology 27, 1–12, May 21, 2020 ª 2020 Elsevier Ltd.

1

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Figure 1. Gambogic Acid Reduces S1P

Levels in MCF-7 Cells

(A) Chemical structures of Michael acceptor-con-

taining natural products: gambogic acid, goya-

zensolide, helenalin, and deoxyelephantopin.

(B) Normalized S1P levels following 3 mM treatment

of MCF-7 cells with the indicated natural product

(relative values SD; n = 3). *p < 0.05 by unpaired

Student’s t test.

cancer activity in many different animal

models of cancer (Wu et al., 2004; Guo

et al., 2004; Liu et al., 2005; Qiang et al.,

2008), but it has not, to date, been

described as an inhibitor of sphingolipid

biosynthesis. The structure of GA bears

parasiticide purposes for thousands of years (Choudhury et al., two electrophilic a,b-unsaturated carbonyls, and numerous co-

2016; Tao et al., 2010). GA has recently been identiﬁed as a prom- valent protein targets of GA have been reported in Jurkat (Kasib-

ising anticancer agent against a broad variety of different cancers hatla et al., 2005), K-562 (Zhou et al., 2016), and HeLa (Zhang

(Li et al., 2012; Zhu et al., 2009; Wang et al., 2013). Several in vivo et al., 2010; Zhou et al., 2016) cells. However, the targets of

studies have shown that GA has an acceptable safety proﬁle (Wu GA in MCF-7 cells are not known, and we were unable to plau-

et al., 2004; Pandey et al., 2007; Yu et al., 2007a; Guo et al., 2004). sibly link the targets reported in other cell lines to the identiﬁed

Results of a phase IIa trial in China similarly showed a favorable phenotype of decrease in S1P levels. Thus, we decided to

safety proﬁle with promising disease control rates in 47 patients perform a deep proﬁling of the cellular targets of GA in MCF-7

with advanced malignant tumors (Chi et al., 2013).

cells using a classical chemoproteomics approach. For this pur-

Here, we apply a combined proteomics and metabolomics pose, we ﬁrst prepared an alkyne-tagged derivative of GA in a

approach to identify GA as well as structurally related synthetic single chemical step (Figure 2A). To examine whether the probe

xanthones as ﬁrst-in-class covalent inhibitors of mammalian GA alkyne 1 shares the same covalent protein targets as GA, we

sphingolipid de novo biosynthesis with an unprecedented mech- treated MCF-7 cell lysate with 30 mM 1; the probe-labeled pro-

anism of action. We ﬁrst use various chemoproteomic techniques teins were then conjugated to TAMRA (tetramethylrhodamine)

to determine that GA binds to the (so far undruggable) regulatory under standard biorthogonal CuAAC conditions (Speers and

small subunit B of the serine palmitoyltransferase complex Cravatt, 2004), separated by SDS-PAGE, and visualized in gel

(SPTSSB). We then screen a small collection of structurally related by ﬂuorescence scanning. The gel scan showed intense labeling

synthetic xanthones to identify 18 as an equally potent but more of the proteome (Figures 2B and S1A), indicating that, similarly to

selective binder of SPTSSB. Using metabolomics, we show that GA, probe 1 reacts with the proteome quite promiscuously. This

18 reduces key sphingolipid levels, including S1P, in situ and ﬂuorescence labeling was competed by GA in a concentration-

in vivo. We use various proteomic, biochemical, and cell biological dependent manner. Moreover, pretreatment of the lysate with

methods to demonstrate that 18 induces cellular and metabolic the broad-proﬁle cysteine-reactive reagent iodoacetamide

responses characteristic of diminished S1P signaling. We show (IAA) abolished the labeling by 1, indicating that GA speciﬁcally

that 18 decreases phosphorylation of the oncogenic kinase AKT reacts with proteomic cysteines (Figure S1B). In contrast, post-

and reduces the neosynthesis of the oncogenes transforming treatment with IAA did not signiﬁcantly diminish the labeling,

growth factor b (TGF-b), c-Fos, and jun-B.

an indication that 1 binds to cysteines irreversibly (Figure S1C).

RESULTS

Gambogic Acid Binds to SPTSSB

To identify the targets of GA in MCF-7 cells, we treated live cells

with 3 mM GA or DMSO for 4 h, then harvested and lysed the

Gambogic Acid Lowers S1P Levels in MCF-7 Cells

As part of our ongoing endeavor to discover novel covalent inhib- cells. Knowing that GA speciﬁcally binds cysteines, we treated

itors of human sphingolipid biosynthesis, we tested several cell lysates with 10 mM IAA alkyne (Weerapana et al., 2010).

Michael acceptor-containing natural products using a commer- IAA alkyne-labeled proteins were then ‘‘clicked’’ to a photocleav-

cial ELISA-based assay for quantitation of total S1P levels in able biotin azide linker, enriched with streptavidin beads, and di-

MCF-7 cells, a breast cancer cell line known to exhibit S1P- gested. The bead-bound peptides were then subjected to

dependent tumorigenesis (Nava et al., 2002). While goyazenso- dimethyl labeling (DML) (Boersema et al., 2009), released by irra-

lide, helenalin, and deoxyelephantopin (all 3 mM) did not show diation, and analyzed by liquid chromatography and tandem

any signiﬁcant effect on S1P levels in MCF-7 cells after 6 h of in- mass spectrometry (LC-MS/MS) (Figure 2C). This experiment

cubation, we observed a ꢀ50% decrease in S1P after treatment is based on the cysteine modiﬁcation identiﬁcation strategy pre-

with GA (Figures 1A and 1B).

viously reported by our group (Abegg et al., 2015) and allows se-

GA is an extensively studied xanthone natural product derived lective enrichment and quantiﬁcation of cysteine-containing and

from the resin of the Garcinia hanburyi tree with proven anti- IAA-modiﬁed peptides rather than whole proteins. We identiﬁed

2

Cell Chemical Biology 27, 1–12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Figure 2. Identiﬁcation of SPTSSB as

Cellular Target of Gambogic Acid

a

(A) Chemical structure of the GA alkyne probe 1.

(B) Fluorescence SDS-PAGE gel image of MCF-7

lysate labeled with 30 mM 1 and pretreated with

indicated concentrations of GA. See also Fig-

ure S1A.

(C) Catch-and-release proteomics workﬂow for the

in situ competitive proﬁling of GA targets using

iodoacetamide (IAA) alkyne.

(D) Log₂DML ratio of peptides identiﬁed in the GA-

and DMSO-treated samples (n

= 6, three bi-

ologicals 3 two technicals). Competed SPTSSB

peptide is highlighted and the ratio of GA versus

DMSO SD is shown.

(E) Labeling of mock and overexpressed WT,

C68A, and H56L SPTSSB-GFP mutants in

HEK293T lysate with 30 mM 1 and competition with

10 mM GA. Shown are the ﬂuorescence gel on top

and western blot membrane probed for GFP

below. See also Figure S2A.

(F) IC₅₀values of GA and other natural products

binding to SPTSSB measured via a competitive

ﬂuorescence gel assay using 30 mM 1 as probe and

various concentrations of compounds as

competitor.

See also Figure S3A.

product-like xanthones that still feature the a,b-unsaturated ke- cells from 18 (Figures 5C and 5D). This result shows that SPTSSB

tone group but lack the dihydropyran ring and the carboxylate inhibition is at least partially responsible for the 18-mediated

side chain of GA (Figure 3A). These compounds have previously toxicity, although we cannot exclude that binding to the other

been reported to have potent in vivo antitumor activity and solu- eight proteomic targets could also have contributed to the

bility superior to GA (Wang et al., 2011; Zhang et al., 2013; Wu toxicity of 18.

et al., 2015; Xu et al., 2016). Overall, we tested 17 xanthones

as potential binders to SPTSSB using the competitive gel-based 18 Reduces S1P Levels In Situ and In Vivo

assay. Among these compounds, 18 emerged as the most To investigate the precise effect of 18 on sphingolipid levels in

potent binder of the desired target (Figures 3B and S3A). We cells, we performed MS-based lipidomic analysis of sphingoli-

re-synthesized 18 for subsequent experiments using a previ- pids after treatment of MCF-7 cells with 18 or DMSO as con-

ously reported synthetic route (Wang et al., 2011; Wu et al., trol. Cells were treated with 3 mM 18 for 5 h and the lipids

2015) but with modiﬁed conditions for the last three steps to were extracted and subjected to multiple reaction monitoring

enable larger-scale synthesis (Figure S3B).

(MRM) analysis, as reported previously (Figure 6A) (Hannich

4

Cell Chemical Biology 27, 1–12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Figure 4. 18 Exhibits Similar Afﬁnity and

Greater Selectivity toward SPTSSB in Com-

parison with GA

(A) In situ competitive ﬂuorescence SDS-PAGE gel

and western blot membrane probed for GFP in

HEK293T cells overexpressing SPTSSB-GFP with

18 and 30 mM 1. See also Figure S4A.

(B) Kinetic parameters K_iand k_inactfor GA and 18

on SPTSSB. See also Figures S2D and S4C.

(C) Log₂DML ratio of peptides identiﬁed from the in

situ catch-and-release competitive pull-down

experiment with 3 mM 18 or DMSO and with 10 mM

IAA alkyne (n = 6, three biologicals and two tech-

nicals each).

(D) Venn diagram showing the number of identiﬁed

in situ proteomic cysteine targets of GA or 18 with a

DML ratio %0.5 versus DMSO.

(E) Chemical scheme showing the reaction to

reduce the Michael acceptor in 18.

(F) Competitive ﬂuorescent SDS-PAGE gel and

western blot membrane probed for GFP in

HEK293T cells overexpressing SPTSSB-GFP with

18, 24, and 30 mM 1. See also Figure S5A.

likely compensated by the salvage

pathway

sphingomyelin.

Having conﬁrmed that 18 lowers sphingolipid levels in cells,

and

the

hydrolysis

of

et al., 2017). We observed robust reduction of C18/C20 sphin-

ganines and sphingosines as well as S1P (Figure 6B and Table we sought to investigate whether 18 can also decrease sphingo-

S2). The latter result was additionally conﬁrmed using an ELISA lipid levels in vivo. In brief, mice were intravenously injected with

assay (Figure S6A). Importantly, a decrease in S1P levels in 18 (50 mg/kg) or vehicle. After 8 h, the animals were sacriﬁced

MCF-7 cells caused by treatment with 18 was rescued by over- and tissues known to natively express high levels of SPTSSB

expression of SPTSSB C68A (Figure S6B). Furthermore, treat- (Uhlen et al., 2015), stomach and intestine, were subjected to

ment of MCF-7 cells with 3 mM 24 did not cause any changes lipid extraction and analyzed as described above. Once again,

in S1P levels (Figure S6C). Interestingly, small interfering RNA we observed a signiﬁcant reduction in C18 sphinganine, sphin-

(siRNA) knockdown of SPTSSB in MCF-7 cells did not alter gosine, and S1P levels in both tissues (Figures 6C and 6D; Table

S1P levels (Figure S6D). Assuming that SPTSSB knockdown S3). We also performed an MS-based pull-down experiment us-

led to a compensatory cellular response, for example through ing probe 1 to conﬁrm engagement of SPTSSB in mice. Intestine

upregulation of key S1P biosynthetic enzymes, we quantiﬁed and stomach tissue lysates from vehicle or 18-treated mice were

the levels of sphingosine kinase 1 (SPHK1) after 18 treatment treated with 30 mM probe 1, followed by click chemistry with

or SPTSSB knockdown. Strikingly, SPHK1 was strongly upre- biotin azide, subsequent enrichment with streptavidin, and

gulated in siRNA but not in compound-treated MCF-7 cells, SPTSSB-targeted LC-MS/MS analysis. Indeed, we were able

thus explaining the unchanged S1P levels and further under- to observe ꢀ90% competition of SPTSSB in both analyzed tis-

scoring the physiological importance of SPTSSB in this cell sues of 18-treated mice (Figures S7B and S7C; Table S4). These

line (Figures S6E and S6F). Importantly, we conﬁrmed that 18 preliminary results show that it is possible to decrease the S1P

treatment does not affect the S1P levels in two cell lines that levels in vivo via small-molecule inactivation of SPTSSB,

do not express SPTSSB (Uhlen et al., 2015), HeLa and A-431 although clearly more extensive studies in animal disease

cells (Figures S6G and S6H).

models with pathogenic upregulation of S1P signaling will be

Collectively, these results conﬁrm that the binding of xan- required to investigate the effect and the therapeutic potential

thones to SPTSSB is productive and leads to the loss of function of 18 binding to SPTSSB.

of this activating subunit and decrease in downstream sphingo-

lipid levels. Furthermore, the observed reduction of C18 and C20 18 Induces S1P-Dependent Changes in Cellular

species is in good accordance with the previously described Signaling

preference of SPTSSB toward C16 and C18 acyl-CoA substrates S1P is a lipid mediator that induces a multitude of signaling and

(Han et al., 2009). In contrast, we did not observe a signiﬁcant phosphorylation changes via its action on the S1P receptors. We

decrease in cellular levels of the proapoptotic ceramides upon performed a series of western blotting experiments to conﬁrm

treatment with 18 (Figure S7A and Table S2). A likely explanation that 18 causes critical changes in hallmark S1P-linked protein

is that there are three major pathways for ceramide synthesis in phosphorylation events in MCF-7 cells.

mammals (Kitatani et al., 2008). While SPTSSB inactivation can

First, we conﬁrmed that 18 treatment leads to a strongly

affect the de novo pathway, this loss in ceramide synthesis is decreased phosphorylation of Ser⁴⁷³on AKT serine/threonine

Cell Chemical Biology 27, 1–12, May 21, 2020

5

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Figure 5. SPTSSB Overexpression Protects Cells from 18-Induced Toxicity

(A) Cytotoxicity of GA, 18, and 24 in MCF-7 cells (relative values SD and IC₅₀s; n = 3).

(B) Flow-cytometry plots of annexin V versus propidium iodide of MCF-7 cells treated with DMSO, 1 mM staurosporine, 10 mM GA, or 50 mM 18.

(C) Cytotoxicity curves of mock, SPTSSB WT, and SPTSSB C68A overexpressing MCF-7 cells treated with indicated concentrations of 18 for 24 h. Quantiﬁcation

was performed using the WST-1 assay (relative values SD and IC₅₀s; n = 3). Calculated unpaired Student’s t test p values: ***p = 0.0004 for mock compared with

SPTSSB-V5 WT, **p = 0.0011 for mock compared with SPTSSB-V5 C68A, and *p = 0.0057 for SPTSSB-V5 WT compared with SPTSSB-V5 C68A; 95% con-

ﬁdence intervals: mock, 12.5–24.9 mM; SPTSSB-V5, WT 23.0–29.9 mM; SPTSSB-V5 C68A, 30.8–45.4 mM.

(D) qPCR analysis of expression levels of SPTSSB WT and C68A in MCF-7 cells. TBP (TATA-box-binding protein) was used as endogenous control. Shown are

normalized values SD, n = 4. ***p < 0.0005 by unpaired Student’s t test.

kinase 1 (AKT; Figures 7A, S7D, and S7E). AKT is an oncogene of fect of 18 on protein neosynthesis. Newly synthesized proteins in

major clinical relevance, and AKT phosphorylation (activation) is 10 mM 18 or DMSO-treated MCF-7 cells were identiﬁed using

a key predictor of poor clinical outcome in many cancers (Yu metabolic incorporation of homopropargylglycine (HPG) as a

et al., 2007b; Kreisberg et al., 2004; Hong et al., 2014, 2015). clickable methionine analog (van Hest et al., 2000; Kiick et al.,

S1P is known to increase the phosphorylation level of AKT (Mat- 2001) and subsequent conjugation to biotin azide, enrichment

suzaki et al., 2013). Second, we observed an increased phos- with streptavidin, and LC-MS/MS analysis using LFQ (label-

phorylation and, thus, activation of the p38 mitogen-activated free quantiﬁcation) intensities (Cox et al., 2014) (Figures 7B–

protein kinase (p38 MAPK), as previously reported for GA (Fig- 7D). Overall, we detected 149 R50% upregulated and

ures 7A, S7F, and S7G) (Chen et al., 2008). This result is in 106 R50% downregulated proteins, including three downregu-

good accordance with the reported activation of p38 MAPK in lated proteins with a previously reported biochemical link to

S1P lyase overexpressing cells, where the S1P levels are much S1P (Figure 7D and Table S5). For instance, it is known that there

lower than in normal cells (Min et al., 2005).

is a functional crosstalk between the TGF-b and the S1P

Knowing that 18 treatment induces phosphorylation changes signaling pathways (Takuwa et al., 2012; Radeke et al., 2005).

in critical signaling pathways, we also investigated the global ef- Our proteomics data show that the neosynthesis of TGFB1, a

6

Cell Chemical Biology 27, 1–12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Figure 6. In Situ and In Vivo Treatment of 18

Results in Decreased Sphingolipid and S1P

Levels

(A) Metabolomics workﬂow to analyze sphingolipid

levels in 18-treated cells or mice. AQC, 6-amino-

quinolyl-hydroxysuccinimidyl carbamate.

(B–D) Quantiﬁcation of sphingosines (So), sphin-

ganines (Sa), and S1P (So-1P) levels in (B) MCF-7

cells treated for 5 h with 3 mM 18, and in (C) in-

testines and (D) stomach of mice treated with

50 mg/kg 18 for 8 h (relative values SD; n = 6

biological replicates for MCF-7 cells and n = 4

biological replicates for mouse tissues). *p < 0.05,

**p < 0.005, ***p < 0.0005 by unpaired Student’s

t test.

with an unprecedented mechanism of ac-

tion. Using chemical proteomics, we

determined that GA covalently binds to

SPTSSB, the regulatory small subunit B

of the serine palmitoyltransferase com-

plex. We then screened a small collection

of structurally related synthetic xanthones

and identiﬁed 18 as an equally potent but

more selective binder of SPTSSB. We

showed by metabolomics that 18 reduces

key sphingolipid levels, including S1P, in

situ and in vivo. Finally, we used various

proteomic, biochemical, and cell biolog-

ical methods to demonstrate that 18 in-

duces cellular and metabolic responses

characteristic

of

diminished

S1P

signaling. We showed that 18 decreases

phosphorylation of the oncogenic kinase

AKT and reduces the neosynthesis of

the oncogenes TGF-b, c-Fos, and jun-B.

Myriocin, a widely applied inhibitor of

serine palmitoyltransferase and perhaps

secreted ligand that binds to various TGF-b receptors, is signif- the closest comparison with the mode of action of compound

icantly downregulated in 18-treated cells. S1P has been 18, binds to the main catalytic subunits SPTLC1 and SPTLC2

reported to induce the expression of the transcription factor and completely blocks the SPT activity (Wadsworth et al.,

c-Fos (Su et al., 1994; Segura et al., 2005), and our results 2013). Consequently, myriocin is highly toxic in animals (Johnson

show that a decrease in cellular S1P results in decreased neo- et al., 2004; Adachi and Chiba, 2007; Adachi et al., 1995; Kiuchi

synthesis of this important proto-oncogenic transcription factor. et al., 2000) and is thus of limited use as a research tool to study

Remarkably, our data show that also jun-B (Wang et al., 2007), a the mammalian sphingolipids. In contrast, we did not observe

protein that together with c-Fos forms the heterodimeric tran- any toxicity in non-cancerous MDCK1 cells or mice treated

scription factor AP-1 (activator protein 1) (Shaulian, 2010; Su with 18. This can be explained by the fact that 18 binds to a reg-

et al., 1994; Hsieh et al., 2008), is also downregulated upon 18 ulatory subunit of SPT and does not inhibit the SPT activity and

treatment (Figure 7D).

sphingolipid de novo biosynthesis completely. Furthermore,

Collectively, our results demonstrate that the 18-induced SPTSSB is signiﬁcantly expressed only in a few speciﬁc organs

decrease in S1P levels is physiologically relevant and leads to and tissues such as the gastrointestinal tract, skin, kidney, and

signiﬁcant changes in major signaling pathways.

urinary bladder (Uhlen et al., 2015). Therefore, inhibition of

SPTSSB in vivo should, in principle, cause fewer systemic side

effects. Thus, we believe that 18 represents a valuable research

tool to study the sphingolipid metabolism and S1P signaling in

DISCUSSION

We have here applied a combined omics approach to identify the situ and in vivo. The discrepancy in the effect of 18 and SPTSSB

cytotoxic xanthone natural product GA as well as structurally siRNA treatment in MCF-7 cells underscores the importance of

related synthetic xanthones as novel, ﬁrst-in-class covalent in- discovery of small-molecule binders as crucial tools for studying

hibitors of the mammalian sphingolipid de novo biosynthesis the physiological function of scarcely characterized proteins.

Cell Chemical Biology 27, 1–12, May 21, 2020

7

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Figure 7. 18 Induces Changes in Cellular S1P Signaling

(A) Western blots of cellular markers linked to S1P signaling after 24 h of 10 mM 18 treatment in MCF-7 cells. See also Figures S7D–S7G.

(B) Workﬂow for the global proteomics assay to measure protein neosynthesis using a clickable methionine derivative (HPG).

(C) Volcano plot of proteins enriched via HPG incorporation in 10 mM 18 versus DMSO for 24 h in MCF-7 cells. Dotted lines were calculated using a false discovery

rate of 1% and an S₀value of 0.1, indicating an adjusted p value of 0.01 (n = 6, three biologicals 3 two technicals).

(D) LFQ ratios of proteins linked to S1P signaling.

The unique in vivo expression pattern of SPTSSB also pro- eases will be required to explore the therapeutic potential of

vides an opportunity for selectively targeting the S1P biosyn- SPTSSB inhibition by compound 18.

thesis in these tissues for therapeutic purposes. For example,

Finally, we believe that 18 will serve as a blueprint for the

¨

Sjogren-Larsson syndrome, a rare ichthyosis-type skin disorder, development of smaller, structurally less complex, and more

is caused by mutations in the enzyme ALDH3A2 that is respon- potent covalent inhibitors of SPTSSB with improved pharmaco-

sible for metabolizing hexadecenal, the product of S1P degrada- kinetic properties. These studies are currently under way in our

tion by S1P lyase (Nakahara et al., 2012). Inhibition of SPTSSB in laboratories and the results will be reported in due course.

skin may decrease production and accumulation of this toxic

fatty acid aldehyde. It is also known that lowering S1P levels in SIGNIFICANCE

Crohn’s disease and ulcerative colitis reduces the disease

severity (Snider et al., 2009). Taking advantage of the high Our work shows that a small-molecule inhibitor of SPTSSB,

SPTSSB expression in the gastrointestinal tract, compounds the regulatory small subunit B of the serine palmitoyltrans-

like 18 may be applicable to treat these inﬂammatory bowel dis- ferase complex, inhibits the mammalian de novo sphingoli-

eases. Furthermore, SPTSSB is highly overexpressed in bladder pid biosynthesis pathway and reduces sphingolipid,

cancer (Uhlen et al., 2015; Ramakrishnan et al., 2018), which is including sphingosine-1-phosphate, levels in situ and in vivo.

known to be highly dependent on the crosstalk between vascular We also demonstrate that the SPTSSB inhibitor induces

endothelial growth factor and S1P signaling (Azuma et al., 2003). cellular responses characteristic for diminished S1P

Thus, SPTSSB inhibitors may one day ﬁnd a clinical application signaling. This study demonstrates that SPTSSB may

in the treatment of this particular type of cancer. Extensive and become a viable therapeutic target in various diseases

specialized studies in animal models of the aforementioned dis- with pathological S1P signaling. Because of the unique

8

Cell Chemical Biology 27, 1–12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Q.Y. provided the synthetic xanthones and performed the in vivo experiments.

native expression pattern of SPTSSB, our compound pro-

A.A. wrote the manuscript, and all authors reviewed and edited the

vides an opportunity to pharmacologically target S1P

manuscript.

biosynthesis in selective tissues, such as gastrointestinal

tract, kidney, urinary bladder, and skin, rather than globally.

DECLARATION OF INTERESTS

Furthermore, we believe that our compound will become a

valuable tool for studying the sphingolipid metabolism and

The authors declare no competing interests.

serve as a blueprint for the development of a new generation

of sphingolipid biosynthesis inhibitors.

Received: September 24, 2019

Revised: February 19, 2020

Accepted: March 9, 2020

Published: April 23, 2020

STAR+METHODS

Detailed methods are provided in the online version of this paper

and include the following:

REFERENCES

Abegg, D., Frei, R., Cerato, L., Prasad Hari, D., Wang, C., Waser, J., and

Adibekian, A. (2015). Proteome-wide proﬁling of targets of cysteine reactive

small molecules by using ethynyl benziodoxolone reagents. Angew. Chem.

Int. Ed. Engl. 54 , 10852–10857.

d KEY RESOURCES TABLE

d LEAD CONTACT AND MATERIALS AVAILABILITY

d EXPERIMENTAL MODEL AND SUBJECT DETAILS

B Cell Culture and Preparation of Lysates

B Animals

Adachi, K., and Chiba, K. (2007). FTY720 story. Its discovery and the following

accelerated development of sphingosine 1-phosphate receptor agonists as

immunomodulators based on reverse pharmacology. Perspect. Med. Chem.

1 , 11–23.

d METHOD DETAILS

B Materials and General Methods

B Synthesis Procedures

Adachi, K., Kohara, T., Nakao, N., Arita, M., Chiba, K., Mishina, T., Sasaki, S.,

and Fujita, T. (1995). Design, synthesis, and structure-activity-relationships of

2-substituted-2-amino-1,3-propanediols - discovery of a novel immunosup-

pressant, Fty720. Bioorg. Med. Chem. Lett. 5 , 853–856.

B Sample Preparation for SDS-PAGE Experiments

B Gel-Based In Vitro Competitive Experiments

B Cloning and Site-Directed Mutagenesis

B Transfection and Western Blot

Azuma, H., Takahara, S., Horie, S., Muto, S., Otsuki, Y., and Katsuoka, Y.

(2003). Induction of apoptosis in human bladder cancer cells in vitro and in vivo

caused by FTY720 treatment. J. Urol. 169 , 2372–2377.

B Determination of Kinetic Parameters

B Gel-Based In-Situ Labeling Experiments

B Cell Toxicity Assays

Banik, K., Harsha, C., Bordoloi, D., Sailo, B.L., Sethi, G., Leong, H.C., Arfuso,

F., Mishra, S., Wang, L.Z., Kumar, A.P., and Kunnumakkara, A.B. (2018).

Therapeutic potential of gambogic acid, a caged xanthone, to target cancer.

Cancer Lett. 416 , 75–86.

B ELISA for Detection of S1P

B Extraction of RNA and RT-PCR

B Western Blot for PARP-1, p-p38 MAPK and p-AKT

B Annexin V and PI Staining

Berdyshev, E.V., Gorshkova, I.A., Usatyuk, P., Zhao, Y., Saatian, B., Hubbard,

W., and Natarajan, V. (2006). De novo biosynthesis of dihydrosphingosine-1-

phosphate by sphingosine kinase 1 in mammalian cells. Cell Signal. 18 ,

1779–1792.

B IAA-Competitive DML Experiment for MS

B HPG Incorporation and Enrichment for MS

B LC-MS/MS Analysis

Boersema, P.J., Raijmakers, R., Lemeer, S., Mohammed, S., and Heck, A.J.

(2009). Multiplex peptide stable isotope dimethyl labeling for quantitative pro-

teomics. Nat. Protoc. 4 , 484–494.

B MaxQuant Analysis

B MCF-7 Treatment for Sphingoid Base Analysis

B In Vivo Experiment Using 18

Brinkmann, V., Billich, A., Baumruker, T., Heining, P., Schmouder, R., Francis,

G., Aradhye, S., and Burtin, P. (2010). Fingolimod (FTY720): discovery and

development of an oral drug to treat multiple sclerosis. Nat. Rev. Drug

Discov. 9 , 883–897.

B Lipidomics Sample Preparation for MS

B In Vivo SPTSSB Competitive Experiment for MS

d QUANTIFICATION AND STATISTICAL ANALYSIS

d DATA AND CODE AVAILABILITY

Chen, J., Gu, H.Y., Lu, N., Yang, Y., Liu, W., Qi, Q., Rong, J.J., Wang, X.T., You,

Q.D., and Guo, Q.L. (2008). Microtubule depolymerization and phosphoryla-

tion of c-Jun N-terminal kinase-1 and p38 were involved in gambogic acid

induced cell cycle arrest and apoptosis in human breast carcinoma MCF-7

cells. Life Sci. 83 , 103–109.

SUPPLEMENTAL INFORMATION

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

chembiol.2020.03.008.

Chi, Y.H.B.L., Zhan, X.K., Yu, H., Xie, G.R., Wang, Z.Z., Xiao, W., Wang, Y.G.,

Xiong, F.X., Hu, J.F., Yang, L., et al. (2013). An open-labeled, randomized,

multicenter phase IIa study of gambogic acid injection for advanced malignant

tumors. Chin. Med. J. 126 , 1642–1646.

ACKNOWLEDGMENTS

We thank The Scripps Research Institute, the National Centre of Competence

in Research Chemical Biology, the Swiss National Science Foundation, and

the Jiangsu Province Funds for Excellent Young Scientists (China)

(BK20170088) for ﬁnancial support.

Choudhury, B., Kandimalla, R., Bharali, R., Monisha, J., Kunnumakara, A.B.,

Kalita, K., and Kotoky, J. (2016). Anticancer activity of Garcinia morella on T-

cell murine lymphoma via apoptotic induction. Front. Pharmacol. 7 , 3.

Cox, J., Hein, M.Y., Luber, C.A., Paron, I., Nagaraj, N., and Mann, M. (2014).

Accurate proteome-wide label-free quantiﬁcation by delayed normalization

and maximal peptide ratio extraction, termed MaxLFQ. Mol. Cell Proteomics

13 , 2513–2526.

AUTHOR CONTRIBUTIONS

A.A. conceived the study and supervised the work. C.W., D.P., and A.A. were

involved in the design and execution of the synthesis. D.G.H., D.A., B.G.D.,

A.S., and A.A. designed and performed the biological studies. J.T.H., H.R.,

and A.A. designed and performed the metabolomics experiments. X.Z. and

Cuvillier, O., Pirianov, G., Kleuser, B., Vanek, P.G., Coso, O.A., Gutkind, S.,

and Spiegel, S. (1996). Suppression of ceramide-mediated programmed cell

death by sphingosine-1-phosphate. Nature 381 , 800–803.

Cell Chemical Biology 27, 1–12, May 21, 2020

9

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Danese, S., Furfaro, F., and Vetrano, S. (2018). Targeting S1P in inﬂammatory

bowel disease: new avenues for modulating intestinal leukocyte migration.

J. Crohns Colitis 12 , S678–S686.

an excellent predictor of poor clinical outcome in prostate cancer. Cancer

Res. 64 , 5232–5236.

Lee, M.J., Thangada, S., Paik, J.H., Sapkota, G.P., Ancellin, N., Chae, S.S.,

Wu, M.T., Morales-Ruiz, M., Sessa, W.C., Alessi, D.R., and Hla, T. (2001).

Akt-mediated phosphorylation of the G protein-coupled receptor EDG-1 is

required for endothelial cell chemotaxis. Mol. Cell 8 , 693–704.

Fujita, T., Yoneta, M., Hirose, R., Sasaki, S., Inoue, K., Kiuchi, M., Hirase, S.,

Adachi, K., Arita, M., and Chiba, K. (1995). Simple compounds, 2-alkyl-2-

amino-1,3-propanediols have potent immunosuppressive activity. Bioorg.

Med. Chem. Lett. 5 , 847–852.

Li, C.L., Qi, Q., Lu, N., Dai, Q.S., Li, F.N., Wang, X.T., You, Q.D., and Guo, Q.L.

(2012). Gambogic acid promotes apoptosis and resistance to metastatic po-

tential in MDA-MB-231 human breast carcinoma cells. Biochem. Cell Biol.

90 , 718–730.

Gordon, M.S., Just, R., Rosen, L.S., and Dorr, A. (2010). A phase I study of so-

nepcizumab (S), a humanized monoclonal antibody to sphingosine-1-phos-

phate (S1P), in patients with advanced solid tumors. J. Clin. Oncol. 28,

https://doi.org/10.1200/jco.2010.28.15_suppl.2560.

Liu, G., Burns, S., Huang, G., Boyd, K., Proia, R.L., Flavell, R.A., and Chi, H.

(2009). The receptor S1P1 overrides regulatory T cell-mediated immune sup-

pression through Akt-mTOR. Nat. Immunol. 10 , 769–777.

Guo, Q., Qi, Q., You, Q., Gu, H., Zhao, L., and Wu, Z. (2006). Toxicological

studies of gambogic acid and its potential targets in experimental animals.

Basic Clin. Pharmacol. Toxicol. 99 , 178–184.

Liu, W., Guo, Q.L., You, Q.D., Zhao, L., Gu, H.Y., and Yuan, S.T. (2005).

Anticancer effect and apoptosis induction of gambogic acid in human gastric

cancer line BGC-823. World J. Gastroenterol. 11 , 3655–3659.

Guo, Q.L., You, Q.D., Wu, Z.Q., Yuan, S.T., and zhao, L. (2004). General gam-

bogic acids inhibited growth of human hepatoma SMMC-7721 cells in vitro

and in nude mice. Acta Pharmacol. Sin. 25 , 769–774.

Lukowski, Z.L., Min, J., Beattie, A.R., Meyers, C.A., Levine, M.A., Stoller, G.,

Schultz, G.S., Samuelson, D.A., and Sherwood, M.B. (2013). Prevention of

ocular scarring after glaucoma ﬁltering surgery using the monoclonal antibody

LT1009 (Sonepcizumab) in a rabbit model. J. Glaucoma 22 , 145–151.

Han, G., Gupta, S.D., Gable, K., Niranjanakumari, S., Moitra, P., Eichler, F.,

Brown, R.H., Harmon, J.M., and Dunn, T.M. (2009). Identiﬁcation of small sub-

units of mammalian serine palmitoyltransferase that confer distinct acyl-CoA

substrate speciﬁcities. Proc. Natl. Acad. Sci. U S A 106 , 8186–8191.

Maceyka, M., Harikumar, K.B., Milstien, S., and Spiegel, S. (2012).

Sphingosine-1-phosphate signaling and its role in disease. Trends Cell Biol.

22 , 50–60.

Hannich, J.T., Mellal, D., Feng, S.H., Zumbuehl, A., and Riezman, H. (2017).

Structure and conserved function of iso-branched sphingoid bases from the

nematode Caenorhabditis elegans . Chem. Sci. 8 , 3676–3686.

Maceyka, M., Payne, S.G., Milstien, S., and Spiegel, S. (2002). Sphingosine ki-

nase, sphingosine-1-phosphate, and apoptosis. Biochim. Biophys. Acta 1585 ,

193–201.

Hannun, Y.A., and Obeid, L.M. (1995). Ceramide: an intracellular signal for

apoptosis. Trends Biochem. Sci. 20 , 73–77.

Holleran, W.M., Takagi, Y., and Uchida, Y. (2006). Epidermal sphingolipids:

Maeurer, C., Holland, S., Pierre, S., Potstada, W., and Scholich, K. (2009).

Sphingosine-1-phosphate induced mTOR-activation is mediated by the E3-

ubiquitin ligase PAM. Cell Signal. 21 , 293–300.

metabolism, function, and roles in skin disorders. FEBS Lett. 580 , 5456–5466.

Hong, J.Y., Hong, M.E., Choi, M.K., Chang, W., Do, I.G., Jo, J.S., Jung, S.H.,

Park, S., Kim, S.J., Ko, Y.H., and Kim, W.S. (2015). The clinical signiﬁcance of

activated p-AKT expression in peripheral T-cell lymphoma. Anticancer Res.

35 , 2465–2474.

Matsuzaki, E., Hiratsuka, S., Hamachi, T., Takahashi-Yanaga, F., Hashimoto,

Y., Higashi, K., Kobayashi, M., Hirofuji, T., Hirata, M., and Maeda, K. (2013).

Sphingosine-1-phosphate promotes the nuclear translocation of beta-catenin

and thereby induces osteoprotegerin gene expression in osteoblast-like cell

lines. Bone 55 , 315–324.

Hong, J.Y., Hong, M.E., Choi, M.K., Kim, Y.S., Chang, W., Maeng, C.H., Park,

S., Lee, S.J., Do, I.G., Jo, J.S., et al. (2014). The impact of activated p-AKT

expression on clinical outcomes in diffuse large B-cell lymphoma: a clinico-

pathological study of 262 cases. Ann. Oncol. 25 , 182–188.

Means, C.K., Xiao, C.Y., Li, Z., Zhang, T., Omens, J.H., Ishii, I., Chun, J., and

Brown, J.H. (2007). Sphingosine 1-phosphate S1P2 and S1P3 receptor-medi-

ated Akt activation protects against in vivo myocardial ischemia-reperfusion

injury. Am. J. Physiol. Heart Circ. Physiol. 292 , H2944–H2951.

Hsieh, H.L., Sun, C.C., Wu, C.B., Wu, C.Y., Tung, W.H., Wang, H.H., and Yang,

C.M. (2008). Sphingosine 1-phosphate induces EGFR expression via Akt/NF-

kappaB and ERK/AP-1 pathways in rat vascular smooth muscle cells. J. Cell

Biochem. 103 , 1732–1746.

Merrill, A.H., JR. (2002). De novo sphingolipid biosynthesis: a necessary, but

dangerous, pathway. J. Biol. Chem. 277 , 25843–25846.

Johnson, V.J., He, Q., Osuchowski, M.F., and Sharma, R.P. (2004). Disruption

of sphingolipid homeostasis by myriocin, a mycotoxin, reduces thymic and

splenic T-lymphocyte populations. Toxicology 201 , 67–75.

Min, J., Van Veldhoven, P.P., Zhang, L., Hanigan, M.H., Alexander, H., and

Alexander, S. (2005). Sphingosine-1-phosphate lyase regulates sensitivity of

human cells to select chemotherapy drugs in a p38-dependent manner. Mol.

Cancer Res. 3 , 287–296.

Kanehisa, M., and Goto, S. (2000). KEGG: Kyoto Encyclopedia of Genes and

Genomes. Nucleic Acids Res. 28 , 27–30.

Miyake, Y., Kozutsumi, Y., Nakamura, S., Fujita, T., and Kawasaki, T. (1995).

Serine palmitoyltransferase is the primary target of a sphingosine-like immu-

nosuppressant, ISP-1/myriocin. Biochem. Biophys. Res. Commun. 211 ,

396–403.

Kasibhatla, S., Jessen, K.A., Maliartchouk, S., Wang, J.Y., English, N.M.,

Drewe, J., Qiu, L., Archer, S.P., Ponce, A.E., Sirisoma, N., et al. (2005). A

role for transferrin receptor in triggering apoptosis when targeted with gambo-

gic acid. Proc. Natl. Acad. Sci. U S A 102 , 12095–12100.

Kiick, K.L., Weberskirch, R., and Tirrell, D.A. (2001). Identiﬁcation of an

expanded set of translationally active methionine analogues in Escherichia

coli . FEBS Lett. 502 , 25–30.

Nakahara, K., Ohkuni, A., Kitamura, T., Abe, K., Naganuma, T., Ohno, Y.,

Zoeller, R.A., and Kihara, A. (2012). The Sj o ¨ gren-Larsson syndrome gene en-

codes a hexadecenal dehydrogenase of the sphingosine 1-phosphate degra-

dation pathway. Mol. Cell 46 , 461–471.

Kitatani, K., Idkowiak-Baldys, J., and Hannun, Y.A. (2008). The sphingolipid

salvage pathway in ceramide metabolism and signaling. Cell Signal. 20 ,

1010–1018.

Nakamura, H., Oda, T., Hamada, K., Hirano, T., Shimizu, N., and Utiyama, H.

(1998). Survival by Mac-1-mediated adherence and anoikis in phorbol ester-

treated HL-60 cells. J. Biol. Chem. 273 , 15345–15351.

Kitz, R., and Wilson, I.B. (1962). Esters of methanesulfonic acid as irreversible

inhibitors of acetylcholinesterase. J. Biol. Chem. 237 , 3245.

Nakamura, S., Kozutsumi, Y., Sun, Y., Miyake, Y., Fujita, T., and Kawasaki, T.

(1996). Dual roles of sphingolipids in signaling of the escape from and onset of

apoptosis in a mouse cytotoxic T-cell line, CTLL-2. J. Biol. Chem. 271 ,

1255–1257.

Kiuchi, M., Adachi, K., Kohara, T., Minoguchi, M., Hanano, T., Aoki, Y.,

Mishina, T., Arita, M., Nakao, N., Ohtsuki, M., et al. (2000). Synthesis and

immunosuppressive activity of 2-substituted 2-aminopropane-1,3-diols and

2-aminoethanols. J. Med. Chem. 43 , 2946–2961.

Nava, V.E., Hobson, J.P., Murthy, S., Milstien, S., and Spiegel, S. (2002).

Sphingosine kinase type 1 promotes estrogen-dependent tumorigenesis of

breast cancer MCF-7 cells. Exp. Cell Res. 281 , 115–127.

Kreisberg, J.I., Malik, S.N., Prihoda, T.J., Bedolla, R.G., Troyer, D.A.,

Kreisberg, S., and Ghosh, P.M. (2004). Phosphorylation of Akt (Ser(473)) is

10 Cell Chemical Biology 27, 1–12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Newton, J., Lima, S., Maceyka, M., and Spiegel, S. (2015). Revisiting the sphin-

golipid rheostat: evolving concepts in cancer therapy. Exp. Cell Res. 333 ,

195–200.

G(i)-coupled receptors: role of ERK MAP kinase and phosphatidylinositol 3-ki-

nase beta. Cancer Lett. 181 , 195–204.

Van Brocklyn, J.R., and Williams, J.B. (2012). The control of the balance be-

tween ceramide and sphingosine-1-phosphate by sphingosine kinase: oxida-

tive stress and the seesaw of cell survival and death. Comp. Biochem. Physiol.

B Biochem. Mol. Biol. 163 , 26–36.

Pal, S.K., Drabkin, H.A., Reeves, J.A., Hainsworth, J.D., Hazel, S.E.,

Paggiarino, D.A., Wojciak, J., Woodnutt, G., and Bhatt, R.S. (2017). A phase

2 study of the sphingosine-1-phosphate antibody sonepcizumab in patients

with metastatic renal cell carcinoma. Cancer 123 , 576–582.

van Hest, J.C.M., Kiick, K.L., and Tirrell, D.A. (2000). Efﬁcient incorporation of

unsaturated methionine analogues into proteins in vivo. J. Am. Chem. Soc.

122 , 1282–1288.

Pandey, M.K., Sung, B., Ahn, K.S., Kunnumakkara, A.B., Chaturvedi, M.M.,

and Aggarwal, B.B. (2007). Gambogic acid, a novel ligand for transferrin recep-

tor, potentiates TNF-induced apoptosis through modulation of the nuclear fac-

tor-kappa B signaling pathway. Blood 110 , 3517–3525.

Wadsworth, J.M., Clarke, D.J., Mcmahon, S.A., Lowther, J.P., Beattie, A.E.,

Langridge-Smith, P.R., Broughton, H.B., Dunn, T.M., Naismith, J.H., and

Campopiano, D.J. (2013). The chemical basis of serine palmitoyltransferase in-

hibition by myriocin. J. Am. Chem. Soc. 135 , 14276–14285.

Pelletier, D., and Haﬂer, D.A. (2012). Fingolimod for multiple sclerosis. N. Engl.

J. Med. 366 , 339–347.

Wang, W., Huang, M.C., and Goetzl, E.J. (2007). Type 1 sphingosine 1-phos-

phate G protein-coupled receptor (S1P1) mediation of enhanced IL-4 genera-

tion by CD4 T cells from S1P1 transgenic mice. J. Immunol. 178 , 4885–4890.

Proia, R.L., and Hla, T. (2015). Emerging biology of sphingosine-1-phosphate:

its role in pathogenesis and therapy. J. Clin. Invest. 125 , 1379–1387.

Qi, Q., You, Q., Gu, H., Zhao, L., Liu, W., Lu, N., and Guo, Q. (2008). Studies on

Wang, X., Deng, R.X., Lu, Y., Xu, Q., Yan, M., Ye, D.X., and Chen, W.T. (2013).

Gambogic acid as a non-competitive inhibitor of ATP-binding cassette trans-

porter B1 reverses the multidrug resistance of human epithelial cancers by

promoting ATP-binding cassette transporter B1 protein degradation. Basic

Clin. Pharmacol. Toxicol. 112 , 25–33.

the toxicity of gambogic acid in rats. J. Ethnopharmacol. 117 , 433–438.

Qiang, L., Yang, Y., You, Q.D., Ma, Y.J., Yang, L., Nie, F.F., Gu, H.Y., Zhao, L.,

Lu, N., Qi, Q., et al. (2008). Inhibition of glioblastoma growth and angiogenesis

by gambogic acid: an in vitro and in vivo study. Biochem. Pharmacol. 75 ,

1083–1092.

Wang, X., Lu, N., Yang, Q., Gong, D., Lin, C., Zhang, S., Xi, M., Gao, Y., Wei, L.,

Guo, Q., and You, Q. (2011). Studies on chemical modiﬁcation and biology of a

natural product, gambogic acid (III): determination of the essential pharmaco-

phore for biological activity. Eur. J. Med. Chem. 46 , 1280–1290.

Radeke, H.H., Von Wenckstern, H., Stoidtner, K., Sauer, B., Hammer, S., and

Kleuser, B. (2005). Overlapping signaling pathways of sphingosine 1-phos-

phate and TGF-beta in the murine Langerhans cell line XS52. J. Immunol.

174 , 2778–2786.

Weerapana, E., Wang, C., Simon, G.M., Richter, F., Khare, S., Dillon, M.B.D.,

Bachovchin, D.A., Mowen, K., Baker, D., and Cravatt, B.F. (2010). Quantitative

reactivity proﬁling predicts functional cysteines in proteomes. Nature 468 ,

790–U79.

Ramakrishnan, S., Huss, W., Foster, B., Ohm, J., Wang, J., Azabdaftari, G.,

Eng, K.H., and Woloszynska-Read, A. (2018). Transcriptional changes associ-

ated with in vivo growth of muscle-invasive bladder cancer cell lines in nude

mice. Am. J. Clin. Exp. Urol. 6 , 138–148.

Wu, Y., Hu, M.Y., Yang, L., Li, X., Bian, J.L., Jiang, F., Sun, H.P., You, Q.D., and

Zhang, X.J. (2015). Novel natural-product-like caged xanthones with improved

druglike properties and in vivo antitumor potency. Bioorg. Med. Chem. Lett.

25 , 2584–2588.

Rutherford, C., Childs, S., Ohotski, J., Mcglynn, L., Riddick, M., Macfarlane, S.,

Tasker, D., Pyne, S., Pyne, N.J., Edwards, J., and Palmer, T.M. (2013).

Regulation of cell survival by sphingosine-1-phosphate receptor S1P1 via

reciprocal ERK-dependent suppression of Bim and PI-3-kinase/protein kinase

C-mediated upregulation of Mcl-1. Cell Death Dis. 4 , e927.

Wu, Z.Q., Guo, Q.L., You, Q.D., Zhao, L., and Gu, H.Y. (2004). Gambogic acid

inhibits proliferation of human lung carcinoma SPC-A1 cells in vivo and in vitro

and represses telomerase activity and telomerase reverse transcriptase

mRNA expression in the cells. Biol. Pharm. Bull. 27 , 1769–1774.

Segura, B.J., Zhang, W., Xiao, L., Logsdon, C.D., and Mulholland, M.W. (2005).

Sphingosine-1-phosphate induces early response gene expression in C6 gli-

oma cells. Brain Res. Mol. Brain Res. 133 , 325–328.

Xu, X.L., Wu, Y., Hu, M.Y., Li, X., Bao, Q.C., Bian, J.L., You, Q.D., and Zhang,

X.J. (2016). Novel natural product-like caged xanthones bearing a carbamate

moiety exhibit antitumor potency and anti-angiogenesis activity in vivo. Sci.

Rep. 6 , 35771.

Shaulian, E. (2010). AP-1—the Jun proteins: oncogenes or tumor suppressors

in disguise? Cell Signal. 22 , 894–899.

Snider, A.J., Kawamori, T., Bradshaw, S.G., Orr, K.A., Gilkeson, G.S., Hannun,

Y.A., and Obeid, L.M. (2009). A role for sphingosine kinase 1 in dextran sulfate

sodium-induced colitis. FASEB J. 23 , 143–152.

Yatomi, Y. (2006). Sphingosine 1-phosphate in vascular biology: possible ther-

apeutic strategies to control vascular diseases. Curr. Pharm. Des. 12 ,

575–587.

Speers, A.E., and Cravatt, B.F. (2004). Proﬁling enzyme activities in vivo using

click chemistry methods. Chem. Biol. 11 , 535–546.

Yu, J., Guo, Q.L., You, Q.D., Zhao, L., Gu, H.Y., Yang, Y., Zhang, H.W., Tan, Z.,

and Wang, X.T. (2007a). Gambogic acid-induced G(2)/M phase cell-cycle ar-

rest via disturbing CDK7-mediated phosphorylation of CDC2/p34 in human

gastric carcinoma BGC-823 cells. Carcinogenesis 28 , 632–638.

Spiegel, S., and Milstien, S. (2003). Sphingosine-1-phosphate: an enigmatic

signalling lipid. Nat. Rev. Mol. Cell Biol. 4 , 397–407.

Su, Y., Rosenthal, D., Smulson, M., and Spiegel, S. (1994). Sphingosine 1-

phosphate, a novel signaling molecule, stimulates DNA binding activity of

AP-1 in quiescent Swiss 3T3 ﬁbroblasts. J. Biol. Chem. 269 , 16512–16517.

Yu, Z., Weinberger, P.M., Sasaki, C., Egleston, B.L., Speier, W.F.T., Haffty, B.,

Kowalski, D., Camp, R., Rimm, D., Vairaktaris, E., et al. (2007b).

Phosphorylation of Akt (Ser473) predicts poor clinical outcome in oropharyn-

geal squamous cell cancer. Cancer Epidemiol. Biomarkers Prev. 16 , 553–558.

Takuwa, Y., Okamoto, Y., Yoshioka, K., and Takuwa, N. (2012). Sphingosine-

1-phosphate signaling in physiology and diseases. Biofactors 38 , 329–337.

Tao, S.J., Guan, S.H., Li, X.N., and Gun, D.A. (2010). A highly rearranged pen-

taprenylxanthonoid from the resin of Garcinia hanburyi . Helvetica Chim. Acta

93 , 1395–1400.

Zhang, H.Z., Kasibhatla, S., Wang, Y., Herich, J., Guastella, J., Tseng, B.,

Drewe, J., and Cai, S.X. (2004). Discovery, characterization and SAR of gam-

bogic acid as a potent apoptosis inducer by a HTS assay. Bioorg. Med. Chem.

12 , 309–317.

Tyanova, S., Temu, T., Sinitcyn, P., Carlson, A., Hein, M.Y., Geiger, T., Mann,

M., and Cox, J. (2016). The Perseus computational platform for comprehen-

sive analysis of (prote)omics data. Nat. Methods 13 , 731–740.

Zhang, L., Yi, Y., Chen, J., Sun, Y., Guo, Q., Zheng, Z., and Song, S. (2010).

Gambogic acid inhibits Hsp90 and deregulates TNF-alpha/NF-kappaB in

HeLa cells. Biochem. Biophys. Res. Commun. 403 , 282–287.

Uhlen, M., Fagerberg, L., Hallstrom, B.M., Lindskog, C., Oksvold, P.,

Mardinoglu, A., Sivertsson, A., Kampf, C., Sjostedt, E., Asplund, A., et al.

(2015). Proteomics. Tissue-based map of the human proteome. Science

347, 1260419. www.proteinatlas.org.

Zhang, X., Li, X., Sun, H., Wang, X., Zhao, L., Gao, Y., Liu, X., Zhang, S., Wang,

Y., Yang, Y., et al. (2013). Garcinia xanthones as orally active antitumor agents.

J. Med. Chem. 56, 276–292.

Van Brocklyn, J.R., Letterle, C.A., Snyder, P.J., and Prior, T.W. (2002).

Sphingosine-1-phosphate stimulates human glioma cell proliferation through

Zhang, Y., Berka, V., Song, A., Sun, K., Wang, W., Zhang, W., Ning, C., Li, C.,

Zhang, Q., Bogdanov, M., et al. (2014). Elevated sphingosine-1-phosphate

Cell Chemical Biology 27, 1–12, May 21, 2020 11

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

promotes sickling and sickle cell disease progression. J. Clin. Invest. 124 ,

Zhou, Y., Li, W., Zhang, X., Zhang, H., and Xiao, Y. (2016). Global proﬁling of

cellular targets of gambogic acid by quantitative chemical proteomics.

Chem. Commun. (Camb) 52 , 14035–14038.

2750–2761.

Zhao, L.H., Spassieva, S., Gable, K., Gupta, S.D., Shi, L.Y., Wang, J.P.,

Bielawski, J., Hicks, W.L., Krebs, M.P., Naggert, J., et al. (2015). Elevation of

20-carbon long chain bases due to a mutation in serine palmitoyltransferase

small subunit b results in neurodegeneration. Proc. Natl. Acad. Sci. US A

112 , 12962–12967.

Zhu, X.L., Zhang, H.M., Lin, Y., Chen, P.S., Min, J., Wang, Z.Z., Xiao, W., and

Chen, B.A. (2009). Mechanisms of gambogic acid-induced apoptosis in non-

small cell lung cancer cells in relation to transferrin receptors. J. Chemother.

21 , 666–672.

12 Cell Chemical Biology 27, 1–12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE

Antibodies

SOURCE

IDENTIFIER

Mouse-monoclonal anti-GFP

Rabbit-polyclonal anti-SPHK1

Rabbit-monoclonal anti-PARP-1

Roche/Sigma-Aldrich

Abcam

Cat#11814460001

Cat#Ab71700

Cat#9532

Cell Signaling Technology

Rabbit-monoclonal anti-phospho-p38 MAPK (Thr180/

Tyr182)

Cat#4511

Rabbit-monoclonal anti-phospho-Akt (Ser473)

Rabbit-polyclonal anti-a-Tubulin

Anti-mouse IgG HRP

Cell Signaling Technology

GE Healthcare

Cat#4060T

Cat#2144S

Cat#NA931-1ML

Cat#711-605-152

Donkey anti-mouse Alexa Fluor 647-AfﬁniPure

Chemicals, Peptides, and Recombinant Proteins

Insulin

Jackson ImmunoResearch Inc

Sigma-Aldrich

Invitrogen

Cat#I6634

Cat#11668027

Cat# 11789020

Cat# 11791020

Cat# M0491S

Cat#P50700-1

Cat#P5726

Cat#90058

Cat#15596026

Cat#10437028

Cat#P4333

Cat#M7145

Cat#10-013-CV

Cat#P36935

Cat#14761

N/A

Lipofectamine 2000

Gateway BP Clonase Enzyme mix

Gateway LR Clonase Enzyme mix

Q5 High-Fidelity DNA Polymerase

Protease inhibitors

Thermo Scientiﬁc

New England BioLabs Inc

RPI Corp.

Phosphatase inhibitors

Sigma-Aldrich

Thermo Scientiﬁc

Invitrogen

Pierceꢀ Trypsin Protease, MS Grade

TRIzolꢀ Reagent

Fetal Bovine Serum

Life Technologies

Sigma-Aldrich

Mediatech Inc

Life Technologies

Cayman Chemical

This paper

Penicillin-Streptomycin

MEM Non-essential Amino Acid

DMEM

ProLong Gold antifade mountant with DAPI

Gambogic acid

Compound 1

Compound 2

(Wang et al., 2011), This paper

(Zhang et al., 2013)

(Xu et al., 2016)

(Wu et al., 2015)

This paper

N/A

Compounds 3, 4, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17

Compound 5

N/A

Compounds 6, 7

N/A

2,6-Dimethoxybenzoyl chloride

1,5,6-Trimethoxy-9H-xanthen-9-one

Compound 19

N/A

This paper

N/A

This paper

N/A

7-Hydroxy-2,2-diphenyl-6H-[1,3]dioxolo[4,5-c]

This paper

N/A

xanthen-6-one

7-(Methoxymethoxy)-2,2-diphenyl-6H-[1,3]dioxolo

This paper

N/A

[4,5-c]xanthen-6-one

Compound 20

Compound 21

Compound 22

Compound 23

Compound 18

Compound 24

DAPI

This paper

Invitrogen

N/A

Cat#62248

Cat#7285

Iodoacetamide alkyne

Setareh Biotech

(Continued on next page)

Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020 e1

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Continued

REAGENT or RESOURCE

Streptavidin agarose

Tris(2-carboxyethyl)phosphine hydrochloride

Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine

TAMRA azide

SOURCE

IDENTIFIER

Cat#G4108

Cat#C4706

Cat#1061

ProteoChem

Sigma-Aldrich

Click Chemistry Tools

Sigma-Aldrich

Cat#AZ109

Cat#AZ118

Cat#762024

Cat#1119

Cy5 azide

Biotin azide

Photocleavable (PC) biotin azide

Formaldehyde

Click Chemistry Tools

Sigma-Aldrich

Cat#F8775

Cat#492620

Cat#156159

Cat#T7408

Cat#r124.aq

Formaldehyde-d2

Sigma-Aldrich

Sodium cyanoborohydride

Triethylammonium bicarbonate buffer

ReproSil-Pur 120 C18-AQ 2.4 mm beads

Critical Commercial Assays

S1P ELISA kit

Sigma-Aldrich

Dr. Maisch GmbH

Amsbio

Cat#EH4398

Cat#11828681001

Cat#CELLPRO-RO

Cat#27016

Annexin V-FLUOS

Roche

WST-1

Roche

QIAprep Spin Miniprep Kit

QIAquick Gel Extraction Kit

iTaq Universal SYBR Green One-Step Kit

Bradford Assay

Qiagen

Cat#28704

Bio-Rad

Cat#1725150

Cat#5000006

Cat#23225

Bio-Rad

BCA protein assay

Thermo Fisher Scientiﬁc

Experimental Models: Cell Lines

MCF-7

ATCC

Cat#HTB-22

Cat#CCL-2

HeLa

HEK293T

Cat#ACS-4500

Cat#CRL-1555

Cat#CRL-2935

A-431

MDCK1

Experimental Models: Organisms/Strains

Mouse: ICR mice

Jiangsu Key Laboratory

of Drug Design and

N/A

Optimization, China

Pharmaceutical University

Oligonucleotides

For primers see Table S6

Microsynth/ Genewiz

Dharmacon

N/A

siRNA: ON-TARGETplus Human SPTSSB

Recombinant DNA

Cat#L-016481-02-0005

Gatewayꢀ pDONRꢀ221 Vector

Vivid Colorsꢀ pcDNAꢀ6.2/C-EmGFP-DEST Vector

pcDNAꢀ6.2-DEST Mammalian Expression Vector

Software and Algorithms

Invitrogen

Cat#12536017

Cat#V35520

Cat#12489027

GraphPad Prism (V7.00)

GraphPad

https://www.graphpad.com/

scientiﬁc-software/prism/

ImageJ

https://imagej.nih.gov/ij/

MaxQuant (V1.6.1.0)

Perseus (V1.6.0.2)

StepOneꢀ

MaxQuant

Perseus

https://www.maxquant.org/

https://maxquant.net/perseus/

Applied Biosystems

https://www.thermoﬁsher.com/

us/en/home/technical-resources/

software-downloads/StepOne-

and-StepOnePlus-Real-Time-

PCR-System.html

e2 Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

LEAD CONTACT AND MATERIALS AVAILABILITY

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

Adibekian email: aadibeki@scripps.edu. All unique/stable reagents generated in this study are available from the Lead Contact with a

completed Materials Transfer Agreement.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Cell Culture and Preparation of Lysates

MCF-7 (female), HeLa (female), A-431 (female), HEK293T (female) and MDCK1 (female) cells were maintained in DMEM media. All

media were supplemented with 10% fetal calf serum (FCS), non-essential amino acids and penicillin/streptomycin. Cells were

grown at 37^ꢁC under 5% CO₂atmosphere. Cells were allowed to grow to conﬂuence and were harvested by scraping, centrifuged

at 1’500 x g for ﬁve minutes at 4^ꢁC and resuspended in PBS. Cells were lysed by sonication to form cell lysates and protein con-

centration was determined using the Bradford assay. Alternatively for the screening experiment, cell lysates were fractionated into

a soluble and pellet fraction using a MTX150 ultracentrifuge (100’000 x g for 1 hour at 4^ꢁC, Thermo Scientiﬁc) and the pellet frac-

tion was used.

Animals

Animal studies were performed according to the experimental protocols approved by the Institutional Animal Care and Use Com-

mittee of China Pharmaceutical University (No. 20161031-1). Mice were housed in a controlled environment (12-h dark-light cycle)

with access to water and fed a standard diet. Five-week-old male ICR mice with a body weight of 18-22 g were used for

experiments.

METHOD DETAILS

Materials and General Methods

All commercial reagents were purchased from Sigma-Aldrich or Acros of the highest purity grade unless otherwise noted. All re-

agents were weighed and handled in air and used without further puriﬁcation, unless otherwise noted. Liquid reagents were trans-

ferred with stainless steel syringes or cannula. Thin layer chromatography (TLC) was performed on plates of silica precoated with

0.25 mm Kieselgel 60 F254. Flash chromatography was performed using silica gel 40–63 mm (230–400 mesh) from SiliCycle. Gam-

bogic Acid was bought from Cayman Chemical (Item No. 14761).

¹H and ¹³C NMR spectra were recorded on a Bruker Avance 400 MHz spectrometer. Chemical shifts (d) are expressed in ppm

relative with the solvent resonance used as internal reference, and coupling constants (J) are reported in hertz (Hz). Multiplicities

were deﬁned as s (singlet), d (doublet), dd (doublet of doublets), t (triplet), dt (doublet of triplets), m (multiplet) or br (broad). High res-

olution mass spectra data (HRMS) were recorded on a Q Exactive HF-X mass spectrometer (Thermo Scientiﬁc). MW reactions were

performed using a Biotage Initiator instrument.

DMEM/High glucose media with GlutaMAX (with sodium pyruvate), phosphate buffered saline (PBS), MEM Non-Essential

Amino Acids, Penicillin Streptomycin (Pen/Strep) and Trypsin-EDTA were obtained from Life Technologies. Protein concentra-

tion was determined using the Bradford assay (Bio-Rad). Cloning of SPTSSB-GFP was done using the Gateway technology

(Thermo Scientiﬁc). Plasmids were prepared using the QIAprep Spin Miniprep Kit (Qiagen). For all PCR reactions, Q5 High-Fi-

delity DNA Polymerase (New England BioLabs Inc.) was used as recommended by manufacturer. PCR cleanup was performed

using agarose gel electrophoresis with subsequent cutting of ampliﬁed bands using QIAquick Gel Extraction Kit (Qiagen)

following the protocol of the producer. For the BP and the LR reactions the Gateway BP Clonase Enzyme mix and the Gateway

LR Clonase Enzyme mix (Thermo Scientiﬁc) were used according to the protocol of the manufacturer. SPTSSB siRNA was pur-

chased from Dharmacon (ON-TARGETplus Human SPTSSB). S1P ELISA kit was purchased from Amsbio LLC. Primary anti-

bodies were purchased from Cell Signaling Technology Inc. and secondary antibodies were acquired from Jackson ImmunoR-

esearch Laboratories Inc.

Synthesis Procedures

Synthetic procedures of the following compounds can be found in previously published articles:

DDO-6101 (2) (Wang et al., 2011)

DDO-6106 (3), DDO-6114 (4), DDO-6202 (8), DDO-6214 (9), DDO-6215 (10), DDO-6235 (11), DDO-6239 (12), DDO-6243 (13),

DDO-6257 (14), DDO-6276 (15), DDO-6283 (16) and DDO-6303 (17) (Zhang et al., 2013)

DDO-6122 (5) (Xu et al., 2016)

DDO-6127 (6) and DDO-6128 (7) (Wu et al., 2015)

Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020 e3

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Synthesis of Gambogic Acid Alkyne 1

To a stirred solution of Gambogic Acid (mixture of epimers bought from Cayman Chemical with 95 % purity, 6.3 mg, 9.52 mmol, 1.0

eq.) and HATU (5.4 mg, 14.20 mmol, 1.5 eq.) in dry DCM (1.0 mL) at r.t., 1-amino-3-butyne (95 % purity bought from Sigma-Aldrich,

1.0 mL, 11.60 mmol, 1.2 eq.) and DIPEA (3.0 mL, 17.22 mmol, 1.8 eq.) were added. The resulting reaction mixture was stirred at r.t. for 23

hours. After that, the reaction mixture was directly applied to silica column eluting with 20 % EtOAc in DCM to afford the epimeric

mixture of products (5.8 mg, 89 % yield) as a yellow solid. ¹H NMR (400 MHz, CDCl₃) d 12.86/12.85 (s, 1H), 7.57 – 7.55 (m, 1H),

6.77 – 6.66 (m, 2H), 5.50 – 5.44 (m, 1H), 5.36 – 5.32 (m, 1H), 5.11 – 5.03 (m, 2H), 3.49 – 3.46 (m, 1H), 3.45 – 3.28 (m, 4H), 3.22 –

3.17 (m, 1H), 2.70 – 2.61 (m, 1H), 2.55 – 2.44 (m, 2H), 2.41 – 2.37 (m, 2H), 2.34 – 2.30 (m, 1H), 2.13 – 2.01 (m, 2H), 1.97 – 1.95 (m,

1H), 1.79 (s, 3H), 1.74 (s, 3H), 1.70 (s, 3H), 1.66 – 1.55 (m, 11H), 1.45 – 1.37 (m, 3H), 1.29 (s, 3H); ¹³C NMR (100 MHz, CDCl₃)

d 203.80/203.75, 178.89/178.86, 169.02/168.93, 161.76/161.66, 157.75/157.69, 157.34/157.26, 135.70/135.63, 135.11/135.08,

133.20/133.15, 132.14/131.92, 131.83/131.73, 125.12, 125.02/124.85, 123.71/123.58, 122.01/121.98, 115.84/115.76, 107.98/

107.77, 102.96/102.77, 100.44/100.39, 91.07/91.03, 84.02, 83.67, 81.90, 81.54/81.40, 69.70/69.67, 48.98/48.96, 46.96/46.94,

42.10/41.88, 38.05, 30.00/29.98, 29.17/29.13, 28.89, 27.94, 27.19, 25.73/25.69, 25.65, 25.26, 22.91/22.69, 21.64/21.17, 19.25/

19.23, 18.17/18.13, 17.65/17.63. HRMS (ESI): Calculated for C₄₂H₅₀NO₇[M+H]⁺680.3582, found: 680.3577.

Synthesis of 18

2,6-dimethoxybenzoyl chloride. A suspension of 2,6-dimethoxybenzoic acid (10 g, 54.9 mmol) in thionyl chloride (20 mL ,

274.2 mmol) was heated at 50^ꢁC for 2 hours. The solvent was removed in vacuo to afford the title compound (10.9 g, 99 % yield)

as a yellowish solid. ¹H NMR (400 MHz, CDCl₃) d 7.26 (t, J = 8.4 Hz, 1H), 6.49 (d, J = 8.6 Hz, 2H), 3.78 (s, 6H); ¹³C NMR (100

MHz, CDCl₃) d 166.16, 156.22, 132.73, 117.71, 104.05, 56.27.

1,5,6-trimethoxy-9H-xanthen-9-one. Aluminum trichloride (11 g, 82.5 mmol) was added at 0^ꢁC to a solution of 2,6-dimethoxyben-

zoyl chloride (5.5 g, 27.5 mmol) and 1,2,3-trimethoxybenzene (5.1 g, 30.3 mmol) in dry ether (40 mL) and then stirred for six hours at

room temperature under inert atmosphere. After that, a mixture of 15 % HCl and ethyl acetate (180 mL, 1:1) was added. The organic

phase was then washed with brine, dried over Na₂SO₄, ﬁltered and concentrated in vacuo. The residue was suspended in a solution

containing methanol (47 mL) and water (31 mL) and NaOH (10.3 g, 259 mmol) was added. The reaction mixture was heated at 110^ꢁC

for 40 minutes in a sealed reaction vessel under microwave irradiation. After cooled to 0^ꢁC, the reaction mixture was acidiﬁed with 2 M

HCl until pH = 2 – 3. The formed precipitate was ﬁltered and washed with cold water to afford 1,5,6-trimethoxy-9H-xanthen-9-one

(6.2 g, 78 %) as a pale yellow solid. ¹H NMR (400 MHz, CDCl₃) d 8.04 (d, J = 8.9 Hz, 1H), 7.58 (t, J = 8.2 Hz, 1H), 7.12 (dd, J = 0.8, 8.5 Hz,

1H), 6.96 (d, J = 9.1 Hz, 1H), 6.79 (brd, J = 8.1 Hz, 1H), 4.01 (s, 3H), 4.01 (s, 3H), 3.98 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) d 176.12,

160.86, 158.29, 157.12, 149.54, 135.99, 134.71, 122.51, 118.11, 112.27, 110.30, 108.58, 105.74, 61.69, 56.55, 56.49. HRMS (ESI):

e4 Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Calculated for C₁₆H₁₅O₅[M+H]⁺287.0914, found: 287.0911.

1,5,6-trihydroxy-9H-xanthen-9-one (19). 1,5,6-trimethoxy-9H-xanthen-9-one (10 g, 35 mmol) was added to a stirred mixture of

48% hydrobromic acid (229 mL) and acetic acid (540 mL). The reaction mixture was heated at 120^ꢁC for 12 hours. The solution

was then cooled to 0^ꢁC and 20% NaOH solution was added until pH = 3 – 4. The formed precipitate was ﬁltered and washed

with cold water to afford 1,5,6-trihydroxy-9H-xanthen-9-one (19) (6.1 g, 71%) as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆)

d 12.92 (s, 1H), 10.70 (brs, 1H), 9.55 (brs, 1H), 7.62 (t, J = 8.3 Hz, 1H), 7.57 (d, J = 8.9 Hz, 1H), 7.05 (dd, J = 0.7, 8.4 Hz, 1H), 6.97

(d, J = 8.8 Hz, 1H), 6.77 (dd, J = 0.8, 8.2 Hz, 1H); ¹³C NMR (100 MHz, DMSO-d₆) d 181.24, 161.13, 155.79, 152.59, 146.45,

136.75, 132.56, 116.25, 113.52, 113.17, 109.90, 107.57, 107.07. HRMS (ESI): Calculated for C₁₃H₉O₅[M+H]⁺245.0444, found:

245.0443.

7-hydroxy-2,2-diphenyl-6H-[1,3]dioxolo[4,5-c]xanthen-6-one. a,a-dichlorodiphenylmethane (5.6 mL, 29 mmol) was added to

1,5,6-trihydroxy-9H-xanthen-9-one (19) (5.9 g, 24.2 mmol) and stirred for 20 minutes at 180^ꢁC. The solution was then cooled to

room temperature and diluted with dichloromethane. The organic phase was then washed with brine, dried over Na₂SO₄, ﬁltered

and concentrated in vacuo. The residue was puriﬁed by column chromatography (silica gel, pentane-ethyl acetate 8:1) to afford

7-hydroxy-2,2-diphenyl-6H-[1,3]dioxolo[4,5-c]xanthen-6-one (8 g, 81 %) as a yellow solid. ¹H NMR (400 MHz, CDCl₃) d 12.76

(s, 1H), 7.90 (d, J = 8.4 Hz, 1H), 7.67 – 7.61 (m, 4H), 7.56 (t, J = 8.3 Hz, 1H), 7.46 – 7.39 (m, 6H), 7.00 (d, J = 8.5 Hz, 1H), 6.97 (dd,

J = 0.9, 8.4 Hz, 1H), 6.79 (dd, J = 0.9, 8.4 Hz, 1H); ¹³C NMR (100 MHz, CDCl₃) d 181.45, 162.29, 156.08, 153.39, 141.09, 139.18,

136.75, 133.94, 129.81, 128.60, 126.48, 121.32, 116.85, 111.08, 108.56, 107.09, 106.53. HRMS (ESI): Calculated for C₂₆H₁₇O₅

[M+H]⁺409.1071, found: 409.1067.

7-(methoxymethoxy)-2,2-diphenyl-6H-[1,3]dioxolo[4,5-c]xanthen-6-one. 7-hydroxy-2,2-diphenyl-6H-[1,3]dioxolo[4,5-c]

xanthen-6-one (5.4 g, 13.2 mmol) was dissolved in dry DMF (100 mL) at 0^ꢁC and sodium hydride (0.8 g, 33.3 mmol) was added. The

mixture was stirred at this temperature for 30 minutes. Chloromethyl methyl ether (2 mL, 26.4 mmol) was then added dropwise and

the reaction was warmed to room temperature and left stirring for two hours. The mixture was then diluted with ethyl acetate and

poured into water. The organic phase was then washed with brine, dried over Na₂SO₄, ﬁltered and concentrated in vacuo. The res-

idue was puriﬁed by column chromatography (silica gel, pentane-ethyl acetate 6:4) to afford 7-(methoxymethoxy)-2,2-diphenyl-

6H-[1,3]dioxolo[4,5-c]xanthen-6-one (5.1 g, 85 %) as a yellow solid. ¹H NMR (400 MHz, CDCl₃) d 7.89 (d, J = 8.5 Hz, 1H), 7.68 –

7.61 (m, 4H), 7.56 (t, J = 8.3 Hz, 1H), 7.46 – 7.37 (m, 6H), 7.18 (dd, J = 0.9, 8.5 Hz, 1H), 7.05 (dd, J = 0.7, 8.3 Hz, 1H), 6.95 (d, J =

8.5 Hz, 1 H), 5.37 (s, 2H), 3.57 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) d 175.40, 158.28, 157.80, 152.33, 140.24, 139.44, 134.65,

129.69, 128.55, 126.53, 121.70, 119.39, 111.51, 110.86, 106.15, 95.59, 56.74, 29.85. HRMS (ESI): Calculated for C₂₈H₂₁O₆

Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020 e5

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

[M+H]⁺453.1333, found: 453.1329.

5,6-dihydroxy-1-(methoxymethoxy)-9H-xanthen-9-one (20). 7-(methoxymethoxy)-2,2-diphenyl-6H-[1,3]dioxolo[4,5-c]xanthen-

6-one (5 g, 11 mmol) was solubilized in a mixture of THF and methanol (100 mL, 1:1) and then 10 % Pd/C (500 mg) was added

and the reaction mixture was stirred for 18 hours at room temperature under hydrogen atmosphere. After that, the mixture was

ﬁltered through a plug of silica gel, concentrated in vacuo. The residue was washed with pentane to afford 5,6-dihydroxy-1-(me-

thoxymethoxy)-9H-xanthen-9-one (2.83 g, 89 %) as a yellow-brown solid and used for the next step without further puriﬁcation.

¹H NMR (400 MHz, DMSO-d₆) d 10.33 (s, 1H), 9.34 (s, 1H), 7.66 (t, J = 8.3 Hz, 1H), 7.48 (d, J = 8.7 Hz, 1H), 7.21 (dd, J = 0.8,

8.5 Hz, 1H), 7.01 (dd, J = 0.8, 8.3 Hz, 1H), 6.89 (d, J = 8.8 Hz, 1H), 5.30 (s, 2H), 3.45 (s, 3H); ¹³C NMR (100 MHz, DMSO-d₆)

d 174.47, 157.37, 157.26, 150.91, 145.31, 134.56, 132.17, 116.47, 115.95, 112.90, 112.37, 111.17, 110.65, 95.18, 56.08. HRMS

(ESI): Calculated for C₁₅H₁₃O₆[M+H]⁺289.0707, found: 289.0704.

1-hydroxy-5,6-bis((2-methylbut-3-yn-2-yl)oxy)-9H-xanthen-9-one (21). Potassium iodide (3.28 g, 19.8 mmol), K₂CO₃(3.1 g,

22.6 mmol) and cuprous iodide (171.4 mg, 0,9 mmol) were added to a solution of 5,6-dihydroxy-1-(methoxymethoxy)-9H-

xanthen-9-one (20) (2.6 g, 9 mmol) in acetone (60 mL). The reaction mixture was stirred at room temperature for 30 minutes and

then, 3-chloro-3-methylbut-1-yne (5 mL, 45 mmol) was added. The solution was heated to 45^ꢁC and stirred for four hours. Upon

completion, a mixture of water and ethyl acetate (300 mL, 2:1) was added and the resulted mixture was stirred for 30 minutes.

The organic phase was then washed with brine, dried over Na₂SO₄, ﬁltered and concentrated in vacuo. The residue was suspended

in a solution containing a mixture of 4M HCl and THF (40 mL, 0.2:1) was added and stirred at room temperature for six hours. Ethyl

acetate and water were then added and the organic phase was washed with brine, dried over Na₂SO₄, ﬁltered and concentrated in

vacuo. The residue was puriﬁed by column chromatography (silica gel, pentane-ethyl acetate 20:1) to afford 1-hydroxy-5,6-bis((2-

methylbut-3-yn-2-yl)oxy)-9H-xanthen-9-one (21) (1.76 g, 52 %) as a yellow oil. ¹H NMR (400 MHz, CDCl₃) d 12.73 (s, 1H), 7.99 (d,

J = 8.8 Hz, 1H), 7.66 (d, J = 9.1 Hz, 1H), 7.56 (t, J = 8.3 Hz, 1H), 6.95 (dd, J = 1.0, 8.5 Hz, 1H), 6.78 (dd, J = 0.9, 8.3 Hz, 1H), 2.67 (s, 1H),

2.30 (s, 1H), 1.82 (s, 6H), 1.77 (s, 6H); ¹³C NMR (100 MHz, CDCl₃) d 182.07, 162.04, 156.93, 156.22, 152.67, 136.51, 135.54, 121.27,

116.10, 115.93, 110.56, 108.74, 107.21, 85.24, 85.18, 75.20, 74.22, 73.91, 30.70, 29.78. HRMS (ESI): Calculated for C₂₃H₂₁O₅[M+H]⁺

377.1384, found: 377.1380.

1-hydroxy-5,6-bis((2-methylbut-3-en-2-yl)oxy)-9H-xanthen-9-one (22). To a solution of 1-hydroxy-5,6-bis((2-methylbut-3-yn-2-

yl)oxy)-9H-xanthen-9-one (21) (900 mg, 2.4 mmol) in ethanol (46 mL) was added 10 % Pd/BaSO₄(104 mg). The reaction mixture

was stirred for three hours at room temperature under hydrogen atmosphere. After that, the mixture was ﬁltered through a plug of

silica gel, concentrated under reduced pressure and puriﬁed by column chromatography (silica gel, pentane-ethyl acetate 20:1)

to afford 1-hydroxy-5,6-bis((2-methylbut-3-en-2-yl)oxy)-9H-xanthen-9-one (22) (727 mg, 80 %) as a yellow oil. ¹H NMR (400

MHz, CDCl₃) d 12.75 (s, 1H), 7.86 (d, J = 9.1 Hz, 1H), 7.56 (t, J = 8.3 Hz, 1H), 7.14 (d, J = 8.9 Hz, 1H), 6.93 (dd, J = 0.8, 8.4 Hz,

1H), 6.77 (dd, J = 0.9, 8.4 Hz, 1H), 6.32 – 6.12 (m, 2H), 5.27 – 5.13 (m, 3H), 5.03 (dd, J = 1.1, 10.8 Hz, 1H), 1.58 (s, 6H), 1.58 (s,

e6 Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

6H); ¹³C NMR (100 MHz, CDCl₃) d 182.10, 162.05, 157.84, 156.25, 152.70, 143.64, 143.44, 136.45, 120.54, 116.82, 115.79, 114.44,

113.38, 110.51, 108.65, 107.09, 83.89, 82.56, 27.40, 27.10. HRMS (ESI): Calculated for C₂₃H₂₅O₅[M+H]⁺381.1697, found: 381.1691.

8-hydroxy-1-(3-methylbut-2-en-yl)-3,3-dimethyl-3,3a,4,5-tetrahydro-1,5-methano-1H,7H-furo[3,4-d]xanthene-7,13-dione (2).

A

solution of 1-hydroxy-5,6-bis((2-methylbut-3-en-2-yl)oxy)-9H-xanthen-9-one 8-hydroxy-2,2-dimethyl-3a-(3-methylbut-2-en-

1-yl)-1,2-dihydro-1,5-methanofuro[2,3-d]xanthene-4,7(3aH,5H)-dione (22) (421 mg, 1.11 mmol) in 80 % methanol aqueous solu-

tion (15 mL) was heated at 110^ꢁC for ﬁve minutes in a sealed reaction vessel under microwave irradiation. The reaction mixture was

then cooled to room temperature and concentrated in vacuo. The residue was puriﬁed by column chromatography (silica gel,

pentane-ethyl acetate 6:1) to afford 8-hydroxy-1-(3-methylbut-2-en-yl)-3,3-dimethyl-3,3a,4,5-tetrahydro-1,5-methano-

1H,7H-furo[3,4-d]xanthene-7,13-dione (2) (295 mg, 70 %) as a yellow solid. ¹H NMR (400 MHz, CDCl₃) d 12.09 (s, 1H), 7.48 (dd,

J = 0.5, 7 Hz, 1H), 7.39 (t, J = 8.32 Hz, 1H), 6.53 (dd, J = 1.0, 8.4 Hz, 1H), 6.51 (dd, J = 0.9, 8.3 Hz, 1H), 4.45 – 4.36 (m, 1H), 3.53

(d, J = 4.2, 6.8 Hz, 1H), 2.69 – 2.56 (m , 2H), 2.46 (d, J = 9.6 Hz, 1H), 2.35 (dd, J = 4.7, 13.6 Hz, 1H), 1.70 (s, 3H), 1.37 (s, 3H), 1.33

– 1.27 (m, 4H), 1.01 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) d 202.88, 181.45, 162.99, 159.70, 138.89, 135.45, 135.01, 133.97, 118.77,

109.49, 107.52, 106.21, 90.11, 84.58, 83.68, 48.91, 47.09, 30.39, 29.27, 29.18, 25.63, 25.07, 16.80. HRMS (ESI): Calculated for

C₂₃H₂₅O₅[M+H]⁺381.1697, found: 381.1695.

(E)-8-hydroxy-3a-(4-hydroxy-3-methylbut-2-en-1-yl)-2,2-dimethyl-1,2-dihydro-1,5-methanofuro[2,3-d]xanthene-4,7(3aH,5H)-di-

one (23). 8-hydroxy-1-(3-methylbut-2-en-yl)-3,3-dimethyl-3,3a,4,5-tetrahydro-1,5-methano-1H,7H-furo[3,4-d]xanthene-7,13-

dione (2) (212 mg, 0.56 mmol) in DCM (1 mL) was added to a stirring solution of 70 % t-BuOOH aqueous (325 mL, 2.52 mmol), salicylic

acid (11.6 mg, 0.084 mmol) and SeO₂(1.9 mg, 0.017 mmol) in DCM (4 mL) and stirred at room temperature for 36 hours. Upon comple-

tion, the reaction mixture was diluted with DCM, washed with dilute Na₂SO₃and brine. The organic phase was then dried over Na₂SO₄,

ﬁltered and concentrated in vacuo. The residue was puriﬁed by column chromatography (silica gel, pentane-ethyl acetate 2:1) to afford

(E)-8-hydroxy-3a-(4-hydroxy-3-methylbut-2-en-1-yl)-2,2-dimethyl-1,2-dihydro-1,5-methanofuro[2,3-d]xanthene-4,7(3aH,5H)-

dione (23) (141 mg, 64 %; 81 % BORSM (44 mg)) as a yellow solid. ¹H NMR (400 MHz, CDCl₃) d 11.90 (s, 1H), 7.50 (d, J = 7.0 Hz, 1H),

7.42 (t, J = 8.3 Hz, 1H), 6.57 – 6.51 (m, 2H), 4.83 – 4.75 (m, 1H), 3.75 – 3.60 (m, 2H), 3.59 – 3.52 (m, 1H), 2.76 – 2.61 (m, 2H), 2.48 (d, J =

9.5 Hz, 1H), 2.37 (dd, J = 4.7, 13.5 Hz, 1H), 1.71 (s, 3H), 1.33 – 1.28 (m, 4H), 1.05 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) d 202.66, 182.34,

162.92, 159.64, 139.37, 138.24, 135.80, 134.11, 118.79, 109.76, 107.57, 106.35, 90.11, 84.35, 83.87, 67.96, 48.85, 47.18, 30.39, 29.11,

28.73, 24.97, 12.73. HRMS (ESI): Calculated for C₂₃H₂₅O₆[M+H]⁺397.1646, found: 397.1644.

(E)-4-(8-hydroxy-2,2-dimethyl-4,7-dioxo-1,2,5,7-tetrahydro-1,5-methanofuro[2,3-d]xanthen-3a(4H)-yl)-2-methylbut-2-en-1-yl nico-

tinate (18). To a solution of nicotinic acid (38 mg, 0.31 mmol) in DCM (5.2 mL), DMAP (51 mg, 0.42 mmol) was added and stirred for

15 minutes. The reaction was cooled to 0^ꢁC, DIC (65 mL, 0.42 mmol) and (E)-8-hydroxy-3a-(4-hydroxy-3-methylbut-2-en-1-yl)-2,2-

dimethyl-1,2-dihydro-1,5-methanofuro[2,3-d]xanthene-4,7(3aH,5H)-dione (23) (111 mg, 0.28 mmol) solubilized in DCM (4 mL)

Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020 e7

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

was then added and the mixture was stirred at room temperature for three more hours. Upon completion, DCM was added and the

organic phase was washed with saturated NH₄Cl and brine. Finally, it was dried over Na₂SO₄, ﬁltered and concentrated in vacuo. The

residue was puriﬁed by column chromatography (silica gel, pentane-ethyl acetate 1:1) to afford (E)-4-(8-hydroxy-2,2-dimethyl-4,7-

dioxo-1,2,5,7-tetrahydro-1,5-methanofuro[2,3-d]xanthen-3a(4H)-yl)-2-methylbut-2-en-1-yl nicotinate (18) (101 mg, 72 %) as a

yellow solid. ¹H NMR (400 MHz, CDCl₃) d 11.95 (s, 1H), 9.17 (brs, 1H), 8.78 (dd, J = 1.3, 4.7 Hz, 1H), 8.25 (dt, J = 2.0, 7.9 Hz, 1H), 7.52

(d, J = 6.8 Hz, 1H), 7.43 – 7.38 (m, 2H), 6.54 – 6.50 (m, 2H), 5.07 – 5.02 (m, 1H), 4.44 (s, 2H), 3.57 – 3.52 (m, 1H), 2.75 – 2.69 (m, 1H), 2.67

– 2.61 (m, 1H), 2.48 (d, J = 9.4 Hz, 1H), 2.36 (dd, J = 4.6, 13.7 Hz, 1H), 1.71 (s, 3H), 1.35 – 1.29 (m, 4H), 1.23 (s, 3H); ¹³C NMR (100 MHz,

CDCl₃) d 202.59, 181.29, 164.97, 162.99, 159.52, 153.54, 151.10, 139.22, 137.29, 136.27, 133.82, 133.00, 126.26, 123.40, 122.28,

109.79, 107.49, 106.20, 90.27, 83.88, 83.82, 69.94, 48.93, 47.12, 30.33, 29.15, 28.64, 25.14, 13.30. HRMS (ESI): Calculated for

C₂₉H₂₇NO₇[M+H]⁺502.1860, found: 502.1857.

Synthesis of 24

(E)-4-(8-hydroxy-2,2-dimethyl-4,7-dioxo-1,2,5,6,6a,7-hexahydro-1,5-methanofuro[2,3-d]xanthen-3a(4H)-yl)-2-methylbut-2-en-

1-yl nicotinate (24). (E)-4-(8-hydroxy-2,2-dimethyl-4,7-dioxo-1,2,5,7-tetrahydro-1,5-methanofuro[2,3-d]xanthen-3a(4H)-yl)-

2-methylbut-2-en-1-yl nicotinate (18) (24.2 mg, 0.048 mmol) was solubilized in 4 mL of ethanol and then 10 % Pd/C (2.42 mg)

was added and the reaction mixture was stirred for 1 h at room temperature under hydrogen atmosphere. After that, the mixture

was ﬁltered through a plug of silica gel, concentrated in vacuo. The residue was puriﬁed by column chromatography (silica gel,

pentane-ethyl acetate 4:6) to afford (E)-4-(8-hydroxy-2,2-dimethyl-4,7-dioxo-1,2,5,6,6a,7-hexahydro-1,5-methanofuro[2,3-d]

xanthen-3a(4H)-yl)-2-methylbut-2-en-1-yl nicotinate (24) (15.3 mg, 63 %) as a colorless solid. ¹H NMR (400 MHz, CDCl₃)

d 11.38 (s, 1H), 9.14 (brs, 1H), 8.78 (brs, 1H), 8.16 (d, J = 7.8 Hz, 1H), 7.42 – 7.35 (m, 2H), 6.55 (dd, J = 0.9, 8.4 Hz, 1H), 6.49 (dd,

J = 0.9, 8.2 Hz, 1H), 5.61 – 5.57 (m, 1H), 4.75 – 4.65 (m, 2H), 3.33 (dd, J = 2.7, 11.4 Hz, 1H), 3.02 – 2.96 (m, 1H), 2.92 – 2.87 (m,

1H), 2.82 – 2.77 (m, 1H), 2.58 (d, J = 8.9 Hz, 1H), 2.47 – 2.43 (m, 1H), 2.06 – 2.01 (m, 1H), 1.79 (s, 3H), 1.73 – 1.63 (m, 2H), 1.35 (s,

3H), 1.13 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) d 209.88, 197.59, 161.79, 158.43, 138.79, 137.06, 134.33, 121.98, 110.07, 108.27,

108.15, 88.50, 86.51, 81.94, 70.52, 44.40, 40.84, 38.64, 29.85, 29.81. 27.51, 27.19, 23.70, 22.41, 14.53. HRMS (ESI): Calculated

for C₂₉H₃₀NO₇[M+H]⁺504.2017, found: 504.2014.

Sample Preparation for SDS-PAGE Experiments

Proteins (2 mg/mL, 25 mL) was treated with indicated compound (1 mL of 25x stock in DMSO) or DMSO for one hour at r.t. Sample was

then treated with 30 mM 1 (1 mL of 25x stock in DMSO) for one hour at r.t. Click chemistry was initiated by the addition of TAMRA azide

(Click Chemistry Tools, 50 mM, 25x stock in DMSO) or Cy5 azide (Click Chemistry Tools, 50 mM, 25x stock in DMSO), tris(2-carbox-

yethyl)phosphine hydrochloride (TCEP, Alfa Aesar, 1 mM, fresh 50x stock in water), tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine

(TBTA, Sigma-Aldrich, 100 mM, 16x stock in DMSO:tBuOH 1:4), and copper(II) sulfate (1 mM, 50x stock in water) to the lysate and

incubated in the dark for one hour at r.t. SDS-PAGE reducing loading buffer (4x) was added and proteins were separated using a

10% SDS-PAGE gel. Gels were visualized using a FMBIO II Multi-View ﬂuorescence scanner (Hitachi) or a Sapphire Biomolecular

Imager (Azure Biosystems), then stained using Coomassie. Images were quantiﬁed with ImageJ.

Gel-Based In Vitro Competitive Experiments

MCF-7 lysate (2 mg/mL, 25 mL) was treated with indicated concentrations of indicated compound (1 mL of 25x stock in DMSO) for one

hour at r.t. Subsequently, lysate was incubated with 30 mM 1 (1 mL of 25x stock in DMSO) for one hour at r.t. Click chemistry, reducing

SDS-PAGE and visualization were performed as described above.

Cloning and Site-Directed Mutagenesis

The SPTSSB gene was ampliﬁed from a cDNA library derived from MCF-7 cells. The gene was ﬂanked with Attb-sites using the

following primer:

Fwd-Attb1-SPTSSB 5’ GGG GAC AAG TTT GTA CAA AAA AGC AGG CTC CAT GGA TTT GAG GCG TGT G 3’

Rv-Attb2-SPTSSB 5’ GGG GAC CAC TTT GTA CAA GAA AGC TGG GTC ATT AGA AAT TGT ACT GTG ATA TCC A 3’

e8 Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

PCR reactions were cleaned up and the PCR product inserted into the pDONR 221. The resulting pENTR was used for PCR-based

site-directed mutagenesis to generate the mutated pENTR vector SPTSSB-C68A-pENTR (forward primer: 5’-ATA GCC GGA TAT

CAC AGT ACA-3’; reverse primer: 5’-TCC GGC TAT TTT TGA GAA AAA-3’) or SPTSSB-H56L (forward primer: 5’-ATC CTC ATT

CGC CTG GCT TG-3’; reverse primer: 5’-AAT GAG GAT TGG AAT AAA GAC ATA GGC-3’). Afterwards the PCR reaction was incu-

bated with DpnI (New England BioLabs Inc., 37^ꢁC, 12 hours followed by 80^ꢁC for 20 minutes) to degrade SPTSSB-WT-pENTR. All

entry vectors were used for recombination with the pcDNA 6.2 N-EmGFP-DEST or pcDNAꢀ6.2-DEST (V5-tagged) via the LR clonase

to obtain the pEXPs. The pENTR and pEXP plasmids were propagated via transformation into E. coli DH5a strain using heat shock.

Transfection and Western Blot

HEK293T cells were seeded (350’000 cells/mL) in a 3.5 cm Petri dish and left overnight to attach. Cells were transfected using Lip-

ofectamine 2000 (Invitrogen) according to protocol of the manufacturer. After 48 hours, lysates were prepared and treated as

described above. Click chemistry, reducing SDS-PAGE and visualization were performed as described above. Proteins were trans-

ferred to a nitrocellulose membrane using a Trans-Blot Turbo transfer system (Bio-Rad). Membrane was blocked with 5% milk in Tris-

Buffered Saline with 0.1 % Tween 20 (TBST) for one hour. Mouse monoclonal antibody against GFP (Roche, 1/1’000 in 5% milk in

TBST) and a-tubulin (CST, #2144S) were added and incubated overnight at 4^ꢁC. Membrane was washed three times for ﬁve minutes

with TBST and incubated with anti-mouse HRP antibody (GE Healthcare, 1/2’000 in 5 % milk in TBST) for one hour at r.t. Membrane

was washed four times for ﬁve minutes with TBST and visualized with a Fusion Solo (Vilber Lourmat) after ECL (Witec AG). In the case

of the detection of SPHK1, 7.5 mL of 20 mM of siRNA (SPTSSB) were mixed with 7.5 mL of RNAiMAX reagent in 500 mL OptiMEM and

incubated for 20 minutes. Subsequently, MCF-7 cells (500’000 cells/mL) were seeded on top and incubated for 96 hours at 37^ꢁC.

Cells were washed twice with PBS, scraped and centrifuged at 1’500 x g for 5 minutes to obtain cell pellets. Protein concentration

was measured using the Bradford assay. SDS-PAGE was performed as described above and a wet transfer to a PVDF membrane

was done at 350 mA for one hour. Membrane was blocked with 5 % milk in TBST for one hour. Primary antibody against SPHK1

(Abcam, ab71700, 1/1’000) and a-tubulin (CST, #2144S) were added and incubated overnight at 4^ꢁC. Membrane was washed three

times for ﬁve minutes with TBST and incubated with Alexa Fluor 647-AfﬁniPure Donkey Anti-Rabbit IgG (Jackson ImmunoResearch

Inc., 711-605-152, 1/500). Membrane was washed four times for ﬁve minutes with TBST and visualized with a Sapphire Biomolecular

Imager (Azure Biosystems).

Determination of Kinetic Parameters

GFP or SPTSSB-GFP was overexpressed in HEK293T cells using Lipofectamine 2000. Lysate (25 mg in PBS, 25 mL) was treated with

indicated concentrations of GA or 18 (1 mL of 25x stock in DMSO) for 10, 20, 30, 40, 50 or 60 min. at r.t. Samples were then treated with

30 mM 1 (1 mL of 25x stock in DMSO) for one hour at r.t. Click chemistry, reducing SDS-PAGE and visualization were performed as

described above. Images were quantiﬁed with ImageJ and the kinetics values were calculated using GraphPad Prism (V7.00).

Gel-Based In-Situ Labeling Experiments

HEK293T cells were transfected with SPTSSB-GFP as mentioned above in 6-well plates. After 24 hours the plates were washed one

time with PBS and DMSO, GA or 18 was added at indicated concentrations (1’000x stock in DMSO) for four hours at 37^ꢁC. Lysate was

prepared as described above and treated with 30 mM 1 (25x stock in DMSO) for one hour at r.t. Click chemistry, reducing SDS-PAGE

and visualization were performed as described above.

Cell Toxicity Assays

MCF-7 (30’000 cells/mL) and MDCK1 (30’000 cells/mL) cells were seeded in a 96-well plate and allowed to attach overnight. Cells

were incubated with indicated concentrations of either GA, 18 or 24 (1’000x stock in DMSO) for 24 hours in cell culture media. Toxicity

was determined using the WST-1 assay (Roche) according to the protocol of the manufacturer. IC₅₀values were calculated with

GraphPad Prism (V7.02). For the overexpression experiments, SPTSSB-V5 WT and C68A were overexpressed in MCF-7 cells for

24 hours. Cells were trypsinized and seeded in 96-well plates (30’000 cell/mL for MCF-7) and allowed to attach overnight, then cells

were incubated with indicated concentrations of 18 (1’000x stock in DMSO) for 24 hours in cell culture media. WST-1 assay (10%)

was performed as readout.

ELISA for Detection of S1P

Cells were grown to conﬂuence in a 6-well plate, washed twice with PBS and then serum starved overnight. Then, cells were treated

with DMSO, 3 mM GA, 3 mM goyazensolide, 3 mM helenalin, 3 mM deoxyelephantopin or 3 mM 18 (1000x stock in DMSO) in media

without serum for two hours at 37^ꢁC. Afterwards, cells were washed with PBS and 1 mL of fresh media without serum was put on

the cells for four hours at 37^ꢁC. The media was collected, centrifuged for 20 minutes at 1000 x g and lyophilized overnight. The residue

was solubilized with 80 mL of the sample buffer from the ELISA kit and S1P concentration was measured according to the manufac-

turer (Amsbio, Cat.EH4398).

Extraction of RNA and RT-PCR

RNA was extracted using TRIzol reagent (Thermo Scientiﬁc) according to the protocol of the manufacturer. The puriﬁed RNA was

used in a qPCR setup using the iTaq Universal SYBR Green One-Step Kit (Bio-Rad). The data was acquired using a StepOnePlus

Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020 e9

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Real-Time PCR System from Applied Biosystems. Double delta C_tanalysis using TBP (TATA-box-binding protein) as endogenous

control was used to determine the relative quantiﬁcation. Following primer were used for the experiments

Fwd-TBP 5‘ CCG AAT ATA ATC CCA AGC GGT 3‘

Rv-TBP 5‘ AAA TCA GTG CCG TGG TTC GT 3‘

Fwd-SPHK1 5’ GCT GCG AAG TTG AGC GAA AA 3’

Rv-SPHK1 5’ GGC TGG ACC CAG TCG G 3’

Fwd-SPTSSB 5’ GGG GAC AAG TTT GTA CAA AAA AGC AGG CTC CAT GGA TTT GAG GCG TGT G 3’

Rv-SPTSSB 5’ GGG GAC CAC TTT GTA CAA GAA AGC TGG GTC ATT AGA AAT TGT ACT GTG ATA TCC A 3’

Fwd-MAPK14 5’ GCC GAG CTG TTG ACT GGA AG 3’

Rv-MAPK14 5’ GGA GGT CCC TGC TTT CAA AGG 3’

Western Blot for PARP-1, p-p38 MAPK and p-AKT

MCF-7 cells were seeded in 6-well plates and grown to conﬂuency. Afterwards, they were treated with DMSO or 10 mM 18 for 24

hours (p-p38 MAPK and p-AKT) or 50 mM 18 for ﬁve hours (PARP-1). In case of the detection of p-AKT, cells were subsequently

treated with 100 nM Insulin for 15 minutes in media with 10 % FCS. Cells were lysed using RIPA buffer (20 mM Tris, 150 mM

NaCl, 2 mM EDTA, 0.1 % SDS, 1 % NP-40, 0.25 % deoxycholate, 1 mM PMSF, 1 mM NaF) plus protease (100x, RPI Corp., Cat.No.

P50700-1) and phosphatase inhibitor (100x, Sigma-Aldrich, Cat.No. P5726) directly in the dish for ﬁve minutes on ice. Cells were

quickly sonicated using a needle sonicator and protein concentration was measured using the BCA protein assay (Thermo Fisher,

Cat No. 23225). SDS-PAGE was performed as described above and wet transfer to a PVDF membrane was done at 350 mA for

one hour. Membrane was blocked with 5 % milk in TBST for one hour. Primary antibodies against respective proteins (PARP-1 anti-

body (CST, #9532), phospho-p38 MAPK (CST, #4511), phospho-AKT (CST, #4060T) and a-Tubulin (CST, #2144S) were added and

incubated overnight at 4^ꢁC. Membrane was washed three times for ﬁve minutes with TBST and incubated with Alexa Fluor 647-Afﬁ-

niPure Donkey Anti-Rabbit IgG (Jackson ImmunoResearch Inc., 711-605-152). Membrane was washed four times for ﬁve minutes

with TBST and visualized with a Sapphire Biomolecular Imager (Azure Biosystems).

Annexin V and PI Staining

For ﬂow cytometry, MCF-7 cells were grown to ꢀ70% conﬂuency in 6-well plates and incubated with 10 mM GA, 50 mM 18, 1 mM STS

or DMSO for 5 hours in cell culture media. Cells were then trypsinized, washed twice with PBS and resuspended in 400 mL PBS. An-

nexin V-FLUOS and propidium iodide (PI) were added to cell suspensions in 100 mL incubation buffer for ﬁnal concentrations of 4 mg/

mL each (Annexin V-FLUOS staining kit, Roche). Samples were then ﬁltered through 35 mm nylon mesh strainers (BD Falcon), incu-

bated for 20 minutes on ice and analyzed with a Gallios Flow Cytometer (Beckman Coulter) using standard laser parameters for FITC

and PI. Flow cytometry data and plots were analyzed using FlowJo v10.6.1 with at least 10000 events per analysis.For cell micro-

scopy, MCF-7 cells were seeded (150’000 cells/mL) in a 24-well plate containing a poly-lysine coated coverslip and left overnight

to attach and grow. Cells were washed one time with PBS and incubated with 1 mM staurosporine (for six hours), 10 mM GA (ﬁve

hours), 50 mM 18 (for ﬁve hours) or DMSO (for ﬁve hours) at 37^ꢁC. Cells were centrifuged at 200 x g for 1 minute, washed one

time with PBS⁺⁺(0.1 g/L CaCl₂and 0.1 g/L MgCl₂) and coverslips were transferred to a wet box. Cells were stained with Annexin

V-FLUOS staining kit (Roche) for 30 minutes, washed two times with PBS⁺⁺and ﬁxed with 4 % paraformaldehyde for 20 minutes

at r.t. Cells were washed three times with PBS⁺⁺, one time with PBS, one time with water and then mounted on a slide with ProLong

Gold antifade mountant with DAPI (Life Technologies). Cells were visualized with a BX53 ﬂuorescent microscope (Olympus).

IAA-Competitive DML Experiment for MS

MCF-7 cells were seeded in 14 cm Petri dishes, grown to 90 % conﬂuence and were subsequently treated with DMSO, 3 mM GA or

3 mM 18 (1’000x stock in DMSO) for four hours at 37^ꢁC. The cells were washed two times with PBS, harvested by scraping, centri-

fuged at 1’500 x g for ﬁve minutes at 4^ꢁC and resuspended in PBS. Cells were lysed by sonication and the protein concentration of the

lysate was determined using the Bradford assay. Lysates (2 mg/mL, 5 mL) were treated with 10 mM iodoacetamide (IAA) alkyne (Se-

tareh Biotech, 100x stock in DMSO) for one hour. Photocleavable (PC) biotin azide (Click Chemistry Tools, 20 mM, 50x stock in

DMSO), tris(2-carboxyethyl)phosphine hydrochloride (TCEP) (1 mM, 50x fresh stock in water), tris[(1-benzyl-1H-1,2,3-triazol-4-yl)

methyl]amine (TBTA) (100 mM, 16x stock in DMSO:tBuOH 1:4), and copper(II) sulfate (1 mM, 50x stock in water) were added to

the lysates and left to react for one hour at r.t. Protein was precipitated by adding methanol/chloroform/water (4:1:3 v/v) to the re-

action mixture and the turbid mixture was centrifuged for 10 minutes at 20’000 x g at 4^ꢁC yielding a protein layer between the aqueous

and organic layers. The protein layer was isolated, dried and solubilized in 2 % SDS in PBS via sonication. The proteome was centri-

fuged at 4’700 x g for ﬁve minutes and the supernatant was transferred to a new tube. PBS was added to give a ﬁnal SDS concen-

tration of 0.2 %. Streptavidin agarose beads (500 mL) were added and the mixture was rotated for four hours at r.t. Beads were

washed with 1 % SDS in PBS (2x 10 mL), PBS (3x 10 mL), and water (3x 10 mL). Beads were resuspended in 6 M urea in PBS

(500 mL), reduced with 10 mM neutralized TCEP (20x fresh stock in water) for 30 minutes at r.t., and alkylated with 25 mM iodoace-

tamide (400 mM fresh stock in water) for 30 minutes at r.t. in the dark. Beads were pelleted by centrifugation (1’400 x g, two minutes)

and resuspended in 300 mL of 2 M urea in 50 mM NH₄HCO₃, 1 mM CaCl₂(100x stock in water) and trypsin (2 mL of 0.5 mg/mL, Thermo

e10 Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

Scientiﬁc). The digestion was performed for six hours at 37^ꢁC. Beads were washed with 1 % SDS in PBS (1x 10 mL), PBS (3x 10 mL),

and water (3x 10 mL). Beads were transferred to a microtube in 500 mL of 100 mM Triethylammonium bicarbonate buffer (TEAB)

buffer, centrifuged (1’400 x g, 2 minutes) and resuspended in 500 mL of 100 mM TEAB buffer. Dimethyl labeling was performed

as reported previously (Boersema et al., 2009). Brieﬂy, 20 mL of either 4 % formaldehyde or 4 % formaldehyde-d2 solution was added

together with 20 mL of 0.6 M NaBH₃CN (freshly made in water) and left to react for one hour at r.t. with rotation. Beads were washed

three times with PBS and mixed at a ratio 1:1. Beads were resuspended in 2 mL PBS and photocleavage was performed with irra-

diation at 365 nm in a CL-1000 UV crosslinker (UVP) for one hour at 4^ꢁC. Beads were centrifuged and supernatant was transferred to a

new tube. Beads were resuspended in 1 mL PBS and incubated for one hour at 37^ꢁC with shaking then centrifuged, and the super-

natants were combined. Beads were again resuspended in 1 mL PBS with 1 M NaCl and incubated for six hours with shaking then

centrifuged and the supernatants were combined. Peptides were desalted over a self-packed C18 spin column and dried. Samples

were analyzed by LC-MS/MS and the MS data was processed with MaxQuant.

HPG Incorporation and Enrichment for MS

MCF-7 cells were seeded (100’000 cells/mL) in 10 cm Petri dishes and left to attach overnight. Cells were washed one time with

PBS and treated with either DMSO or 10 mM 18 for 24 hours in DMEM without serum at 37^ꢁC. After that the cells were starved

for one hour at 37^ꢁC using the cell culture media without FCS, L-cystine and L-methionine. Subsequently the cells were incu-

bated with DMEM containing 0.2 mM L-cystine and 1 mM HPG for ﬁve hours at 37^ꢁC. Cells were washed two times with PBS,

harvested by scraping, centrifuged at 1’500 x g for ﬁve minutes at 4^ꢁC and resuspended in PBS. Cells were lysed by sonication

and the protein concentration of the lysate was determined using the Bradford assay. Lysates (2 mg/mL, 0.5 ml) were subjected

to click chemistry using biotin azide (100 mM, 50x stock in DMSO), tris(2-carboxyethyl)phosphine hydrochloride (TCEP) (1 mM,

50x fresh stock in water), tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) (100 mM, 16x stock in DMSO:tBuOH 1:4), and

copper(II) sulfate (1 mM, 50x stock in water) for one hour at r.t. Protein was precipitated by adding methanol/chloroform/water

(4:1:3 v/v) to the reaction mixture and the turbid mixture was centrifuged for ﬁve minutes at 20’000 x g at 4^ꢁC yielding a protein

layer between the aqueous and organic layers. The protein layer was isolated, dried and solubilized in 2 % SDS in PBS via son-

ication. The proteome was centrifuged at 4’700 x g for ﬁve minutes and the supernatant was transferred to a new tube. PBS was

added to give a ﬁnal SDS concentration of 0.2 %. Streptavidin agarose beads (160 mL, ProteoChem) were added and the

mixture was rotated overnight at r.t. Beads were washed with 1 % SDS in PBS (1x 10 mL), PBS (3x 10 mL), and water (3x

10 mL). Beads were resuspended in 6 M urea in PBS (500 mL), reduced with 10 mM neutralized TCEP (20x fresh stock in water)

for 30 minutes at r.t. and alkylated with 25 mM iodoacetamide (400 mM fresh stock in water) for 30 minutes at r.t. in the dark.

Beads were pelleted by centrifugation (1’400 x g, two minutes) and resuspended in 150 mL of 2 M urea in 50 mM NH₄HCO₃,

1 mM CaCl₂(100x stock in water) and trypsin (1.5 mL of 0.5 mg/mL, Thermo Scientiﬁc). The digestion was performed for six hours

at 37^ꢁC. Samples were acidiﬁed to a ﬁnal concentration of 5 % acetic acid, desalted over a self-packed C18 spin column and

dried. Samples were analyzed by LC-MS/MS and the MS data was processed with MaxQuant and Perseus (Tyanova

et al., 2016).

LC-MS/MS Analysis

Peptides were resuspended in water with 0.1 % formic acid (FA) and analyzed using EASY-nLC 1200 nano-UHPLC coupled to Q

Exactive HF-X Quadrupole-Orbitrap mass spectrometer (Thermo Scientiﬁc). The chromatography column consisted of a 40 cm

long, 75 mm i.d. microcapillary capped by a 5 mm tip and packed with ReproSil-Pur 120 C18-AQ 2.4 mm beads (Dr. Maisch

GmbH). LC solvents were 0.1 % FA in water (Buffer A) and 0.1 % FA in 90 % MeCN: 10 % water (Buffer B). Peptides were eluted

into the mass spectrometer at a ﬂow rate of 300 nL/minute over 240-minute linear gradient (5-35 % Buffer B) at 65^ꢁC. Data was

acquired in data-dependent mode (top-20, NCE 30, R = 17’500) after full MS scan (R = 60’000, m/z 400-1’300). Dynamic exclusion

was set to 10 seconds with peptide match set to prefer and isotope exclusion enabled. For targeted identiﬁcation of the mouse

SPTSSB protein, an inclusion list containing m/z and charge of the expected SPTSSB peptides and deactivated dynamic exclu-

sion were used.

MaxQuant Analysis

The MS data was analyzed with MaxQuant (Cox et al., 2014) and searched against the human or mouse proteome (Uniprot) and a

common list of contaminants (included in MaxQuant). The ﬁrst peptide search tolerance was set to 20 ppm and the main peptide

search was set to 10 ppm with fragment mass tolerance at 0.02 Da. The false discovery rate for peptides, proteins and sites iden-

tiﬁcation was set to 1 %. The minimum peptide length was set to 6 amino acids and peptide requantiﬁcation was enabled. The min-

imal number of peptides per protein was set to one for IAA-competitive and SPTSSB engagement in mice experiments or two for

HPG incorporation pull-downs. For IAA-competitive photocleavage experiments, PC iodoacetamide alkyne adduct (+195.1120

Da) and carbamidomethylation on cysteine and oxidation of methionine were set as variable modiﬁcations. Dimethyl labels were

searched as N-terminus and lysine modiﬁcation for light +28.0313 Da and heavy +32.0564 Da. Modiﬁed peptides with PEP value

%1% and present in at least 3 out of 6 injections were considered. For other experiments, carbamidomethylation on cysteine

was set as ﬁxed modiﬁcation and oxidation of methionine as variable modiﬁcation. For the targeted LC-MS/MS experiments pep-

tides were manually conﬁrmed.

Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020 e11

Please cite this article in press as: Hoch et al., Combined Omics Approach Identiﬁes Gambogic Acid and Related Xanthones as Covalent Inhibitors of

the Serine Palmitoyltransferase Complex, Cell Chemical Biology (2020), https://doi.org/10.1016/j.chembiol.2020.03.008

MCF-7 Treatment for Sphingoid Base Analysis

MCF-7 cells were grown to conﬂuence in 6-well plates, serum-starved overnight, washed twice with PBS and treated with either 3 mM

GA, 3 mM 18 or DMSO for ﬁve hours at 37^ꢁC. Cells were washed twice with PBS, scraped and centrifuged at 1’500 x g for ﬁve minutes

to obtain cell pellets. Pellets were either used directly for lipid extraction or stored at -80^ꢁC.

In Vivo Experiment Using 18

Five-week-old male ICR mice (n = 10) with a body weight of 18-22 g were randomly divided into two groups, the 18 treated group (5

mice) and vehicle control group (5 mice). 18 (50 mg/kg) was injected intravenously once while the control group received 0.9 %

normal saline solution. All animals were sacriﬁced 8 hours after the treatment. Tissues (stomach and intestine) of the animals

were ﬂash frozen, lyophilized and subjected to lipid extraction and sphingoid base analysis or tissues were solubilized with a dounce

homogenizer in four volumes of PBS and centrifuged at 1’400 x g at 4^ꢁC for ﬁve minutes for protein extraction. Animal studies were

performed according to the experimental protocols approved by the Institutional Animal Care and Use Committee of China Pharma-

ceutical University (No. 20161031-1).

Lipidomics Sample Preparation for MS

Extraction and LC-MS analysis was done according to Hannich et al. (Hannich et al., 2017) In short, crude extracts of metabolites

were obtained using pyridine extraction solvent (ethanol: water: diethylether: pyridine, containing 2.1 mM ammonium hydroxide).

In the case of tissue samples 1.4 mm zirconium oxide beads (Bertin Technologies, France) were added to lyophilized tissue powder

(5 mg) to support quantitative extraction. Dried extracts were resuspended in a borate buffer (200 mM boric acid pH 8.8, 10 mM tris(2-

carboxyethyl)phosphine, 10 mM ascorbic acid and 33.7 mM ¹³C¹⁵N-valine), derivatized with 6-aminoquinolyl-hydroxysuccinimidyl

carbamate (AQC, 2.85 mg/mL in acetonitrile) at 55^ꢁC for 15 minutes and left at 24^ꢁC overnight. Samples were analyzed using a

reverse-phase C18 column (HPLC EC 100/2 Nucleoshell RP-18 2.7 mm) on an Accela high-performance liquid chromatography sys-

tem (Thermo Scientiﬁc), coupled to a TSQ Vantage triple quadrupole mass spectrometer (Thermo Scientiﬁc) using multiple reaction

monitoring.

In Vivo SPTSSB Competitive Experiment for MS

Protein concentration from tissue lysates was determined using the Bradford assay. Lysates (2 mg/mL, 0.75 mL) were treated with

30 mM 1 (100x stock in DMSO) for one hour at r.t. Lysates were subjected to click chemistry. Biotin azide (60 mM, 50x stock in DMSO),

tris(2-carboxyethyl)phosphine hydrochloride (TCEP) (1 mM, 50x fresh stock in water), tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]

amine (TBTA) (100 mM, 16x stock in DMSO:tBuOH 1:4), and copper(II) sulfate (1 mM, 50x stock in water) were added to the lysates

and left to react for one hour at r.t. Protein was precipitated by adding methanol/chloroform/water (4:1:3, v/v) to the reaction mixture

and the turbid mixture was centrifuged for ﬁve minutes at 14’000 x g at 4^ꢁC yielding a protein layer between the aqueous and organic

layers. The protein layer was isolated, dried and solubilized in 2 % SDS in PBS via sonication. The proteome was centrifuged at 4’700

x g for ﬁve minutes and the supernatant was transferred to a new tube. PBS was added to give a ﬁnal SDS concentration of 0.2 %.

Streptavidin agarose beads (160 mL, ProteoChem) were added and the mixture was rotated overnight at r.t. Beads were washed with

1 % SDS in PBS (1x 10 mL), PBS (3x 10 mL), and water (3x 10 mL). Beads were resuspended in 6 M urea in PBS (500 mL), reduced with

10 mM neutralized TCEP (20x fresh stock in water) for 30 minutes at r.t., and alkylated with 25 mM iodoacetamide (400 mM fresh

stock in water) for 30 minutes at r.t. in the dark. Beads were pelleted by centrifugation (1’400 x g, two minutes) and resuspended

in 150 mL of 2 M urea in 50 mM NH₄HCO₃, 1 mM CaCl₂(100x stock in water) and trypsin (1 mL of 0.5 mg/mL, Thermo Scientiﬁc).

The digestion was performed for six hours at 37^ꢁC. Samples were acidiﬁed to a ﬁnal concentration of 5 % acetic acid, desalted

over a self-packed C18 spin column and dried. Samples were analyzed by LC-MS/MS and the MS data was processed with

MaxQuant.

QUANTIFICATION AND STATISTICAL ANALYSIS

Statistical analysis was performed with GraphPad Prism unless noted otherwise. Statistical values including the number of times the

data was replicated (‘‘n’’) and statistical signiﬁcance are reported in the Figure legends. Data are represented as mean SD or SEM

and p-values calculated using an unpaired Student’s t-test. *P< 0.05, **P< 0.005, ***P< 0.0005. LC-MS/MS proteomics experiments

were analyzed with MaxQuant using high conﬁdence parameters indicated in the corresponding section. Statistical analysis of the

HPG-based proteomics dataset was performed with Perseus using indicated parameters in the Figure 7C.

DATA AND CODE AVAILABILITY

The published article includes all datasets generated or analyzed during this study (Tables S1, S2, S3, S4, and S5).

e12 Cell Chemical Biology 27, 1–12.e1–e12, May 21, 2020

Article Doi

Combined Omics Approach Identifies Gambogic Acid and Related Xanthones as Covalent Inhibitors of the Serine Palmitoyltransferase Complex

DOI: 10.1016/j.chembiol.2020.03.008

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.chembiol.2020.03.008

Full text of DOI:10.1016/j.chembiol.2020.03.008

Products guided by the article

R&D Labs maybe for 1301159-81-8

Relevant to this article

Hot Product