Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and tumor growth

Article

Gene editing and synthetically accessible inhibitors

reveal role for TPC2 in HCC cell proliferation and

tumor growth

Graphical Abstract

Authors

€

Martin Muller, Susanne Gerndt,

Yu-Kai Chao, ..., Franz Bracher,

Angelika M. Vollmar, Karin Bartel

Correspondence

christian.grimm@med.uni-muenchen.de

(C.G.),

franz.bracher@cup.uni-muenchen.de

(F.B.),

angelika.vollmar@cup.uni-muenchen.de

(A.M.V.),

karin.bartel@cup.uni-muenchen.de (K.B.)

In Brief

€

Muller et al. describe that genetic TPC2

knockout reduces cell proliferation

in vitro, alters cellular energy metabolism,

and effectively suppresses tumor growth

in vivo, suggesting that this lysosomal ion

channel represents a suitable target for

cancer therapy. Concurrently, a potent

TPC2 inhibitor with antitumor efﬁcacy

was developed.

Highlights

d

TPC2 knockout reduces HCC cell proliferation and prevents

tumor growth

d

TPC2 deﬁciency alters the metabolic phenotype

Easily accessible, potent inhibitors with favorable drug-

likeness were developed

€

Muller et al., 2021, Cell Chemical Biology 28, 1–13

June 17, 2021 ª 2021 Elsevier Ltd.

https://doi.org/10.1016/j.chembiol.2021.01.023

ll

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Gene editing and synthetically accessible

inhibitors reveal role for TPC2 in HCC

cell proliferation and tumor growth

€

Martin Muller, Susanne Gerndt, Yu-Kai Chao, Themistoklis Zisis, Ong Nam Phuong Nguyen, Aaron Gerwien,

€

Nicole Urban, Christoph Muller, Florian A. Gegenfurtner, Franz Geisslinger, Carina Ortler, Cheng-Chang Chen,

Stefan Zahler,¹Martin Biel,⁷Michael Schaefer,⁵Christian Grimm,^3,Franz Bracher, Angelika M. Vollmar,

2 ,

1 ,

*

and Karin Bartel^1,8,

*

¹Department of Pharmacy, Pharmaceutical Biology, Ludwig-Maximilians-University Munich, 81377 Munich, Germany

²Department of Pharmacy – Center for Drug Research, Ludwig-Maximilians-University Munich, 81377 Munich, Germany

³Walther-Straub-Institute of Pharmacology and Toxicology, Ludwig-Maximilians-University Munich, 80336 Munich, Germany

⁴Department of Chemistry and Munich Center for Integrated Protein Science (CIPSM), Ludwig-Maximilians-University Munich, 81377 Munich,

Germany

⁵Rudolf-Boehm-Institute of Pharmacology and Toxicology, Leipzig University, 04107 Leipzig, Germany

⁶Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, 100 Taipei, Taiwan

⁷Department of Pharmacy, Pharmacology, Ludwig-Maximilians-University Munich, 81377 Munich, Germany

⁸Lead contact

*Correspondence: christian.grimm@med.uni-muenchen.de (C.G.), franz.bracher@cup.uni-muenchen.de (F.B.), angelika.vollmar@

cup.uni-muenchen.de (A.M.V.), karin.bartel@cup.uni-muenchen.de (K.B.)

https://doi.org/10.1016/j.chembiol.2021.01.023

SUMMARY

The role of two-pore channel 2 (TPC2), one of the few cation channels localized on endolysosomal mem-

branes, in cancer remains poorly understood. Here, we report that TPC2 knockout reduces proliferation of

cancer cells in vitro, affects their energy metabolism, and successfully abrogates tumor growth in vivo.

Concurrently, we have developed simpliﬁed analogs of the alkaloid tetrandrine as potent TPC2 inhibitors

by screening a library of synthesized benzyltetrahydroisoquinoline derivatives. Removal of dispensable sub-

structures of the lead molecule tetrandrine increases antiproliferative properties against cancer cells and im-

pairs proangiogenic signaling of endothelial cells to a greater extent than tetrandrine. Simultaneously, toxic

effects on non-cancerous cells are reduced, allowing in vivo administration and revealing a TPC2 inhibitor

with antitumor efﬁcacy in mice. Hence, our study unveils TPC2 as valid target for cancer therapy and provides

easily accessible tetrandrine analogs as a promising option for effective pharmacological interference.

INTRODUCTION

nicotinic acid adenine dinucleotide phosphate (NAADP) or the

phosphoinositide PI(3,5)P₂(Gerndt et al., 2020). High gene

With an estimated number of 9.6 million deaths worldwide in expression of TPC2 was correlated with poor survival probability

2018 (Bray et al., 2018), cancer-related mortality represents a of prostate cancer patients (Li et al., 2019b), and there is growing

major health problem to date. Huge efforts have been made to evidence that TPC2 is involved in cancer-related processes,

identify suitable targets for cancer therapy, and 109 protein tar- namely neoangiogenesis (Favia et al., 2014) and migration of

gets were addressed by 85 protein-targeted oncology drugs in invasive tumor cells (Nguyen et al., 2017). Therefore, TPC2 might

2017. On the contrary, around 600 cancer drivers have been serve as an innovative target for antitumor drugs. Whether TPC2

identiﬁed, but a majority thereof lacks detailed biological investi- is involved in other cancer hallmarks, such as sustained prolifer-

gation or remains to be targeted by antitumor agents (Santos ation and deregulated cellular energetics (Hanahan and Wein-

et al., 2017; Workman and Al-Lazikani, 2013). This highlights berg, 2011), however, has remained largely unknown and was

the need for the identiﬁcation and validation of innovative, drug- therefore investigated in the context of this study. In the past

gable targets to provide additional options for cancer therapy.

years, several efforts have been made in the search for pharma-

Two-pore channels (TPCs) are cation channels that span cological inhibitors of TPCs, illustrating the increasing impor-

membranes of acidic vesicles of the endolysosomal (EL) system. tance of the ion channels as potential drug targets. The NAADP

TPC2 is one of two subtypes present in primates. It is predomi- antagonist Ned-19 reduces NAADP-induced Ca²⁺release (Favia

nantly found on lysosomal membranes (Ruas et al., 2010) and et al., 2014) and cancer cell migration (Nguyen et al., 2017), but

conducts different ions depending on its activation with either requires high-micromolar doses (R100 mM). Similarly, high

Cell Chemical Biology 28, 1–13, June 17, 2021 ª 2021 Elsevier Ltd.

1

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

concentrations (R500 mM) of naringenin are necessary to affect neic mice (Figures 1E and 1F). Similar bioluminescence signals

vascular endothelial growth factor (VEGF)-evoked tube forma- of wild-type (WT) and TPC2 KO tumors on day 2 ensured equiv-

tion (Pafumi et al., 2017; Penny et al., 2019) recently extended alent amounts of viable engrafted cells (Figure 1E), while average

the limited number of TPC2 inhibitors with ﬂuphenazine and ral- signal intensity of TPC2 KO tumors was reduced by 99% on day

oxifene. Fluphenazine, however, can cause serious extrapyra- 10 (Figures 1E and 1F). Thus, this study reveals that interfering

midal side effects (Divac et al., 2014), and raloxifene is approved with TPC2 as a single target is sufﬁcient to suppress tumor

for postmenopausal women only (Rossi et al., 2018). Conse- growth in vivo.

quently, there is an emerging need for the development of

more speciﬁc inhibitors to further unravel the involvement of TPC2 KO affects cellular energy metabolism

TPC2 in cancer and its potential as drug target. The alkaloid tet- To support sustained proliferation, cancers often reprogram their

randrine (1) was recently shown to reduce cancer cell migration energy metabolism. As a result of the Warburg effect, listed as a

(Nguyen et al., 2017) and infections with Ebola (Sakurai et al., cancer hallmark (Hanahan and Weinberg, 2011), cancer cells

2015), MERS-CoV (Gunaratne et al., 2018), and SARS-CoV-2 derive their cellular adenosine 5⁰-triphosphate (ATP) predomi-

(Ou et al., 2020) viruses via directly acting on TPCs, and thereby nantly from glycolysis to fuel their high energy demand, regard-

represents a promising lead structure for the development of less of oxygen availability (Gatenby and Gillies, 2004). Alkaloid

TPC2 blockers. However, its toxicity in animal models (Jin dauricine (4), which was identiﬁed as a TPC2 inhibitor in this

et al., 2011) is a serious drawback of this compound. In addition, study (Figure 4C), is known to suppress glycolysis in HCC cells

tetrandrine can only be obtained by extraction of plants or by (Li et al., 2018). Hence, we investigated whether TPC2 is impli-

complex chemical synthesis that involves 20 steps to gain enan- cated in energy metabolism. Metabolic proﬁling on a Seahorse

tiomerically pure tetrandrine (1) (Inubushi et al., 1968) or 12 steps Extracellular Flux Analyzer revealed that, following glucose star-

€

to obtain racemic tetrandrine (Schutz et al., 2020). We aimed to vation, TPC2 KO cells displayed a less glycolytic phenotype, as

overcome these drawbacks by chemical simpliﬁcation, substitu- indicated by an increased OCR/ECAR ratio (OCR is oxygen con-

tion, and derivatization of tetrandrine (1).

sumption rate, a parameter for oxidative phosphorylation

Here, we used genetic knockout to investigate the role of [OXPHOS]; ECAR is extracellular acidiﬁcation rate, a parameter

TPC2 in cell proliferation in vitro and tumor growth in vivo. By for glycolytic activity) (Figure 2A). This effect was abolished

combining chemical synthesis and state-of-the-art functional when saturating concentrations of D-glucose were injected.

testing systems, we succeeded in generating tetrandrine- Despite being glucose dependent, glucose deprivation is a com-

derived TPC2 inhibitors with improved potency, reduced mon feature of solid tumors due to poor vascularization (Bergers

toxicity, and antitumor efﬁcacy in vivo.

and Benjamin, 2003; Wu et al., 2012), and the OCR/ECAR ratio

can be used to decipher the reliance of cells on glycolysis or

OXPHOS (Zhang et al., 2012). To further assess the role of

RESULTS

¨

TPC2 in energy metabolism, we applied a Forster resonance en-

TPC2 knockout inhibits cancer cell proliferation and

abrogates tumor growth

ergy transfer (FRET) biosensor, which consists of a subunit of the

bacterial ATP synthase and which can be used for real-time visu-

To test whether TPC2 is involved in cancer cell proliferation alization of ATP levels inside single live cells (Nakano et al., 2011)

in vitro and tumor growth in vivo, we knocked out the channel us- (Figure 2B). Stressing the glycolytic pathway using the glycolysis

ing CRISPR/Cas9 in murine RIL175 hepatocellular carcinoma inhibitor 2-deoxy-D-glucose (2-DG) resulted in a stronger drop of

(HCC) cells. PI(3,5)P₂-induced currents of lysosomes obtained cellular ATP levels in WT cells, whereas TPC2 KO cells remained

from TPC2 knockout (KO) cells were strongly reduced, conﬁrm- unaffected (Figures 2C and 2D). Taken together, loss of TPC2

ing loss of channel function (Figures 1A and 1B). Deletion of exon function promotes a switch toward OXPHOS in vitro. Increased

2 and absence of off-target effects were conﬁrmed by PCR glycolysis, but also decreased mitochondrial function and respi-

methods (Figures S1A–S1F and Table S1). TPC2 KO cells in ration, are hypothesized to promote tumor growth (Denko, 2008).

fact showed an impaired ability to migrate in vitro and to dissem- As TPC2 KO favors the opposite effect, this study suggests that

inate into the lungs of mice (Figures S1G and S1H), both conﬁrm- the metabolic implication of TPC2 contributes to the reduced

ing as well as extending our knowledge on the effect of TPC2 on proliferation rate in vitro and tumor growth in vivo.

tumor cell migration (Nguyen et al., 2017).

Of note, TPC2 also affects cell proliferation and tumor growth. Development of cytostatic TPC2 inhibitors

Cell proliferation in vitro was monitored using impedance mea- After having shown that genetic ablation of the channel is

surements (Figure 1C) and staining with the ﬂuorescent dye sufﬁcient to induce antitumor effects (Figure 1), we aimed at

CFSE (Figure 1D), which can be used to monitor cell divisions. identifying pharmacological TPC2 inhibitors, derived from the

As evident from a reduced slope over a period of 4 days as known blocker tetrandrine (1) (Nguyen et al., 2017; Sakurai

well as from the presence of two CFSE-positive populations after et al., 2015), with potent anticancer properties but reduced

3 days, proliferation of TPC2 KO cells was considerably reduced toxicity. Hence, we screened tetrandrine-related compounds,

(Figures 1C and 1D). Furthermore, small interfering RNA (siRNA) including commercially available bisbenzyltetrahydroisoquino-

silencing of TPC2 impaired proliferation and colony formation of line (BBTHIQ) alkaloids, such as dauricine (4), oxyacanthine (5),

human HCC cells (Hep3B) (Figures S2A–S2C) and decreased and berbamine (6) (Figure 3A), and, additionally, a library of syn-

proliferation of non-malignant hepatocytes (HepaRG) (Figures thesized truncated variants (monomeric 1-benzyltetrahydroiso-

S2A–S2D). Remarkably, TPC2 KO completely abrogated tumor quinolines [BTHIQs], Figure 3B) for their potential to inhibit can-

growth in an ectopic in vivo model using RIL175 cells and synge- cer cell growth and TPC2 function. Some simpliﬁed tetrandrine

2

Cell Chemical Biology 28, 1–13, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 1. TPC2 knockout impairs cancer cell proliferation and tumor growth

(A and B) Loss of TPC2 function was conﬁrmed by measuring PI(3,5)P₂-evoked currents on isolated endolysosomes of RIL175 cells. The TPC2 inhibitor ATP and

the TRPML1 activator ML-SA1 served as controls. (A) Representative patch-clamp recordings of wild-type (WT) and knockout (KO) cells are shown. (B) Statistical

analysis of data obtained from (A) as mean SEM of three independent experiments (two-way ANOVA, interaction p value).

(C and D) TPC2 KO reduces proliferation of RIL175 cells, monitored as (C) cell index over time from a representative experiment. Statistical analysis of the slope is

displayed as mean SEM of three independent experiments (unpaired t test with Welch’s correction). (D) CFSE stainings of WT and TPC2 KO cells are presented

upon initial staining and after 72 h in comparison with the unstained (unst) control. Two visible populations indicate a reduced number of cell divisions of TPC2 KO

cells. One out of three independent experiments is shown.

(E and F) RIL175 WT and TPC2 KO cells were implanted into the ﬂanks of C57Bl/6-Tyr mice and TPC2 KO suppressed tumor growth. (E) The bar graph indicates

similar levels of viable cells among both groups on day 2 after implantation. The line chart displays bioluminescence signal intensities of tumors over time as

mean SEM (n = 10 for KO, n = 9 for WT, Grubbs’ outlier test [a = 0.05], unpaired t test with Welch’s correction). The cartoon visualizes the relation between tumor

size, bioluminescence, and total ﬂux/area. (F) Bioluminescence images from day 10 are shown. The outlier is marked with an x. One excised tumor is shown for

each group.

congeners were previously probed by others for their afﬁnities to benzenoid rings were found to have outstanding properties.

a₁-adrenergic receptors and L-type Ca²⁺channels (Iturriaga- When modifying the amino group, antiproliferative activity re-

´

Vasquez et al., 2003). The antiproliferative properties of our com- mained in a similar range (IC₅₀2.4–4.8 mM) when N-alkyl residues

pound library were evaluated using RIL175 cells. Thereby, of different lengths (NH [Z3 (7)], N-methyl [SG-005 (2), SG-159

numerous molecules with increased potency as compared (8)], N-ethyl [SG-158 (9)]) were used, provided that the nitrogen

with tetrandrine (1) (half-maximal inhibitory concentration [IC₅₀

]

atom retained its basicity. N-Acyl variants lost their antiprolifera-

9.1 mM) were identiﬁed, including truncated variants with IC₅₀tive properties (SG-089 (10), SG-145 (11), IC₅₀R33.0 mM). Vari-

values in the low-micromolar range (Figure 3 and Table S2). To ations at C-6 and C-7 of the isoquinoline unit had only slight ef-

analyze structure-activity relationships, we modiﬁed different fects. In most cases, the loss of aromatic substituents (aryl or

positions of the THIQ scaffold and examined the antiproliferative benzyl ether) in the C-7 position slightly decreased antiprolifera-

effects. To cut a long story short, monomeric BTHIQs bearing tive activity (SG-083 (12), SG-127 (13), SG-157 (14), IC₅₀7.6–

additional aromatic residues (phenyl or benzyl ethers) at both 11.0 mM). The same applied if the 1-benzyl group was changed

Cell Chemical Biology 28, 1–13, June 17, 2021

3

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 2. TPC2 KO alters the metabolic phenotype of cancer cells

(A) RIL175 WT and TPC2 KO cells were subjected to extracellular ﬂux analysis on a Seahorse device after glucose starvation to assess basal glycolytic and

respiratory activity. This was followed by sequential injection of D-glucose (Glc) and oligomycin (oligom.) to determine glycolysis and respiratory activity after

injection of saturating concentrations of glucose and glycolytic reserve upon inhibition of oxidative phosphorylation, respectively. The last injection of the

glycolysis inhibitor 2-deoxy-D-glucose (2-DG) served as control. The OCR (oxygen consumption rate)/ECAR (extracellular acidiﬁcation rate) ratio is displayed

over time as mean SEM, performed in triplicate (unpaired t test with Welch’s correction). An increase of the OCR/ECAR ratio, indicating a decreased reliance on

aerobic glycolysis of TPC2 KO cells, is visible.

¨

(B) A scheme that illustrates the relation between ATP levels and FRET (Forster resonance energy transfer). ATP is sandwiched between the FRET pair and

corresponds to higher FRET efﬁciency (E).

(C and D) ATP levels of (C) RIL175 WT and (D) TPC2 KO cells were monitored before and after injection of 2-DG, suggesting a higher dependence on aerobic

glycolysis for energy production of WT cells. Bar graphs, line charts, and images from a representative experiment performed in triplicate are shown (unpaired t

test with Welch’s correction).

into a phenyl group (SG-122 (15) versus SG-157 (14)), whereas

Thus, the two simplest BTHIQs, SG-005 (2) and SG-094 (3),

shifting a benzyloxy residue from position C-7 to C-6 had no in- either bearing two unsubstituted benzyl (SG-005 (2)) or two

ﬂuence (Z3 (7) versus Z5 (16)). Furthermore, the impact of mod- phenyl (SG-094 (3)) ethers, were subjected to further investiga-

iﬁcations of the 1-benzyl residue was investigated. Variation of tion. SG-005 (2) (IC₅₀2.4 mM) and SG-094 (3) (IC₅₀3.7 mM)

substitution patterns, mostly differing in meta or para position, both displayed markedly enhanced antiproliferative effects

did not markedly affect antiproliferative activity (compounds 2, compared with tetrandrine (1) (Figures 4A and 4B). Furthermore,

8, 17–22, IC₅₀1.2–4.8 mM). Miscellaneous diaryl ethers at the both compounds exerted similar or increased antiproliferative

1-benzyl residue were synthesized bearing both electron- potencies against various other cancer cell lines (Figure S3A

donating and -releasing substituents, mainly in meta or para po- and Table S3) and moderately induced apoptosis of RIL175 cells

sition (compounds 23–29). In most cases, no changes in antipro- (Figure S3B). By applying the endolysosomal patch-clamp tech-

liferative potency were observed (IC₅₀3.9–5.6 mM). However, nique (Chen et al., 2017), we analyzed the potential of com-

cyano and chlorine substituents slightly reduced it (SG-164 pounds 2 and 3 to block PI(3,5)P₂-elicited TPC2 currents.

(28), SG-165 (29), IC₅₀8.5–9.1 mM) and the pyridine moiety Notably, the bis-phenyl ether SG-094 (3) signiﬁcantly increased

strongly reduced it (SG-163 (30), IC₅₀22.0 mM). Deletion of the percentage of channel inhibition (75%) compared with tet-

benzyl or diaryl ether moieties resulted in signiﬁcant reduction randrine (1) (54%), whereas SG-005 (2) was virtually equipotent

(SG-132 (31), IC₅₀11.0 mM) or complete loss of activity (SG- (44%) with tetrandrine (1), the related alkaloid cepharanthine

089 (10), SG-121 (32), SG-121-NP (33), IC₅₀>50 mM). In sum, (34), and the natural seco-analog dauricine (4) (Figure 4C). The

all BTHIQs carrying two aryl or benzyl ether groups and a basic inhibitory effect of tetrandrine (1), SG-005 (2), and SG-094 (3)

amine inhibited proliferation of RIL175 cells to a similar extent on TPC2 was further conﬁrmed by single cell Ca²⁺imaging using

as or more efﬁciently than tetrandrine (1).

a plasma membrane version of TPC2 and the small-molecule

4

Cell Chemical Biology 28, 1–13, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 3. Structures of the (bis)benzyltetra-

hydroisoquinoline library

(A) Bisbenzyltetrahydroisoquinoline alkaloids and

derivatives, which were applied in enantiopure form.

(B) Structures of monomeric 1-benzyltetrahy-

droisoquinolines, consisting of synthesized tetran-

drine congeners (SG compounds) and substances

from other sources (Z compounds), which were

used in racemic form.

and 5D) and TPC2-A1-N (Gerndt et al.,

2020) (Figures 5B and 5E). Consequently,

the racemates were used for all subse-

quent experiments. Due to the strong

dependence of channel activation on the

transfection level, single cell Ca²⁺imaging

rather represents a qualitative than a quan-

titative approach. Obviously, the presence

of a simpliﬁed, monomeric BTHIQ structure

bearing two aryl ethers (the additional

phenyl rings most likely mimicking parts

of the second benzylisoquinoline unit of

tetrandrine (1)) appears to be sufﬁcient to

induce inhibitory effects on TPC2 and cell

proliferation, while both effects are at least

as strong as those of the more complex

lead structure tetrandrine (1).

Biological characterization of SG-

005 (2) and SG-094 (3)

Next, we endeavored to characterize the

advanced candidates SG-005 (2) and SG-

094 (3) in more detail in a cellular setting

by validating known biological effects of

TPC2 inhibition by determining their

cellular and lysosomal uptake and their ef-

fects on glucose metabolism, all in direct

comparison with tetrandrine (1). First, we

wanted to conﬁrm their on-target activities

activator TPC2-A1-N (Gerndt et al., 2020) (Figure 4D). In by investigating their inﬂuence on the VEGF receptor 2 (VEGFR2)

contrast, SG-132 (31) and SG-145 (11), lacking either both signaling pathway. As described by others, impaired TPC2 func-

essential aromatic substituents or a basic nitrogen atom, tion reduces VEGF-evoked activation of eNOS, JNK, MAPK, and

respectively, were not able to inhibit TPC2 activation (Figure 4E). AKT (Favia et al., 2014) in human umbilical vein endothelial cells

SG-005 (2) additionally blocked the activation of other lysosomal (HUVECs). We used SG-132 (31) and SG-145 (11), which do not

cation channels (TRPML1 and TPC1), whereas SG-094 had no block TPC2, as negative controls. Immunoblotting revealed that

considerable inhibitory effect on TRPML1, only on TPC1 only SG-005 (2) and SG-094 (3) were capable of signiﬁcantly

(Figure S4).

reducing VEGF-induced phosphorylation of several downstream

To identify potential eutomers, we separated the racemates targets (Figures 6A and 6B) without decreasing their total protein

SG-005 (2) and SG-094 (3) by semi-preparative chiral high-per- levels (Figures S5A and S5B), indicating that both molecules also

formance liquid chromatography (HPLC). The absolute conﬁgu- act as TPC2 inhibitors in a cellular setting. As expected, SG-132

ration of the enantiomers was determined using electronic circu- (31) or SG-145 (11) did not result in reduced phosphorylation of

lar dichroism spectra and conﬁrmed by computational the investigated VEGFR2 downstream targets. Surprisingly, tet-

calculations (Data S1). To account for the large variety of ro- randrine (1) also had no effect under the chosen treatment con-

tamers, we used the "Torsional Sampling MCMM" method of ditions. As tetrandrine (1) and SG-005 (2) displayed a similar

MacroModel followed by optimization and TD calculations on inhibitory effect on TPC2 in patch-clamp analysis (Figure 4C),

the B3LYP/6-331G(d,p) level of theory. Notably, Ca²⁺imaging the question arose as to whether they differ in their pharmacoki-

revealed that both enantiomers of SG-005 (2) and SG-094 (3) netic properties. Therefore, we investigated cellular uptake of

inhibit TPC2 upon activation with both TPC2-A1-P (Figures 5A tetrandrine (1), SG-005 (2), and SG-094 (3) by analytical HPLC.

Cell Chemical Biology 28, 1–13, June 17, 2021

5

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 4. TPC2 inhibitor screening

(A) Antiproliferative effects of tetrandrine (1) and its more potent analogs SG-005 (2) and SG-094 (3) on RIL175 cells were determined by CellTiter-Blue assay (72 h,

normalized to DMSO control, displayed as mean SEM of three independent experiments).

(B) Structures of tetrandrine (1), SG-005 (2), and SG-094 (3). Dispensable substructures for biological activity of tetrandrine (1) are shaded. Benzyl and phenyl

moieties of SG-005 (2) and SG-094 (3) are highlighted in purple.

(C) Inhibition percentage of TPC2 blockers (10 mM) upon activation with PI(3,5)P₂on isolated endolysosomes from HEK293 cells expressing TPC2-EGFP. The bar

graph indicates mean SEM of at least two independent experiments (unpaired t test). A representative current density-voltage relation is shown for a recording

of SG-094 (3).

(D) Ca²⁺imaging results of tetrandrine (1), SG-005 (2), SG-094 (3), and a DMSO control in HEK293 cells expressing TPC2^L11A/L12A-RFP. After applying the inhibitor

(10 mM), cells were stimulated with the activator TPC2-A1-N (10 mM). Representative Ca²⁺signals comparing SG-094 (3) and DMSO (means highlighted, single

cell traces shaded) are shown.

(E) SG-132 (31) and SG-145 (11) were investigated as in (D).

In (D) and (E), experiments were performed at least in triplicate, and statistical analyses of the maximal change in Fura-2 ratio (mean SEM, unpaired t test)

are shown.

For the conditions used to investigate the VEGF-dependent reduction of LysoTracker Red ﬂuorescence by SG-005 (2) and

signaling pathway (10 mM, 75 min), no signiﬁcant differences SG-094 (3) was larger than that caused by tetrandrine (1) in

were found (Figure S5C). These data indicate that the driver for HUVECs and RIL175 cells, indicating stronger accumulation of

improved efﬁcacy of SG-005 (2) or SG-094 (3) to hinder VEGF- the synthetic compounds in acidic vesicles (Figures 6D

induced phosphorylation of eNOS, JNK, and AKT, as compared and 6E). This decrease was similarly observed in RIL175 TPC2

with tetrandrine (1), is not an altered uptake into the cells. Next, KO cells (Figure S5E). Taken together, analysis of VEGFR2

we determined the uptake of the TPC2 blockers into RIL175 downstream targets suggested improved on-target activity

cells. Neither SG-005 (2) nor SG-094 (3) was internalized to a on TPC2 for the tetrandrine congeners SG-005 (2) and SG-094

greater extent than tetrandrine (1) (Figure S5D). Of note, when (3). While initial cellular uptake does not markedly differ, accumu-

comparing their uptake into primary endothelial and cancerous lation of SG-005 (2) and SG-094 (3) in their intended cellular

liver cells, all three molecules showed a preference for RIL175 target organelles is enhanced compared with tetrandrine (1).

cells versus HUVECs (Figure 6C). Thereafter, their uptake into ly- Additionally, known TPC2 inhibitors were investigated. Naringe-

sosomes was examined by their propensity to diminish ﬂuores- nin and Ned-19 required high-micromolar concentrations to

cence of LysoTracker Red, as shown by others for basic com- achieve an antiproliferative effect (R75 mM) (Figures S6A and

pounds (Kazmi et al., 2013; Petersen et al., 2013). Interestingly, S6B), whereas ﬂuphenazine and raloxifene (IC₅₀8.6 mM and

6

Cell Chemical Biology 28, 1–13, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 5. Enantiomerically pure (S)-SG-005 ((S)-2), (R)-SG-005 ((R)-2), (S)-SG-094 ((S)-3), and (R)-SG-094 ((R)-3) have inhibitory effects on

TPC2, as determined by Fluo-4-based Ca²⁺imaging

After separation of the racemic mixtures SG-005 (2) and SG-094 (3) (C and F) by semi-preparative chiral HPLC, inhibitory effects of the enantiomers and ra-

cemates were probed by Fluo-4-based Ca²⁺imaging. Cells were sequentially treated with the TPC2 inhibitors and the TPC2 activators (A and D) TPC2-A1-P, or (B

and E) TPC2-A1-N. Line graphs show mean SEM of at least three independent experiments.

15.0 mM, respectively) were more potent (Figure S6C). Further- lation of mitochondrial mass was detected for longer incubation

more, ﬂuphenazine and raloxifene, but not Ned-19, reduced times and high-micromolar doses of SG-005 (2) and SG-094 (3)

LysoTracker Red ﬂuorescence of RIL175 cells and HUVECs (Figure S7D), a phenomenon described in the literature as a

(Figures S6D and S6E), indicating lysosomal uptake. How- response to mitochondrial dysfunction (Kogot-Levin et al.,

ever, only raloxifene signiﬁcantly impaired VEGF-evoked activa- 2016). To further decipher the effects of pharmacological

tion of JNK but not of other downstream targets (Figures TPC2 inhibition on cellular energy metabolism, we visualized

S6F–S6I).

ATP levels by FRET-based live cell monitoring. As observed

for genetic TPC2 ablation, SG-005-treated (2 mM, 1 h) RIL175

cells were less susceptible to inhibition of the glycolytic

pathway (Figure 6H). In contrast, ATP levels in untreated WT

cells were signiﬁcantly decreased after addition of 2-DG (Fig-

ure 6G), suggesting a stronger dependence on glycolysis for

energy production. Consequently, pharmacological inhibition

of TPC2 can lead to a metabolic shift toward a less glycolytic

phenotype.

SG-005 (2) and SG-094 (3) inﬂuence cellular energy

metabolism

Next, we aimed to test the inhibitors for their effect on glucose

metabolism, which we identiﬁed as one pathway that is altered

as a result of TPC2 deﬁciency. Metabolic proﬁling using glycol-

ysis stress tests revealed that, upon short-term pretreatment (1

h) with low-micromolar doses of tetrandrine (1), SG-005 (2), and

SG-094 (3), a shift toward higher oxygen consumption after

glucose starvation occurred (Figure 6F). Applying higher con- SG-005 (2) and SG-094 (3) are less toxic than tetrandrine

centrations (10–50 mM, 1 h) and extending treatment duration (1) to non-cancerous cells

(to 24 h) also caused a metabolic shift toward lower glycolysis As outlined previously, tetrandrine (1) treatment can be harmful

for tetrandrine (1), whereas the opposite effect was observed to lungs (Jin et al., 2011; Tian et al., 2016) and liver (Li et al.,

for SG-005 (2) and SG-094 (3) in both cases (Figures S7A 2019a; Qi et al., 2013; Tainlin et al., 1982; Yan et al., 2006) in vivo,

and S7B). We partly linked this to a non-TPC2-related impair- which is related to metabolic activation by cytochrome P450

ment of mitochondrial function, indicated by mitochondrial (CYP) enzymes (Jin et al., 2011; Qi et al., 2013; Tian et al.,

membrane depolarization, which was most evident after SG- 2016) or mitochondrial pathways (Yan et al., 2006). As we sought

005 (2) treatment (Figure S7C). Loss of mitochondrial mem- to cope with this drawback by simpliﬁcation to the relevant phar-

brane potential results in a decoupling of the respiratory chain macophores, we probed tetrandrine (1) and its congeners SG-

(Zorova et al., 2018), whereas cytosolic glycolysis is likely to 005 (2) and SG-094 (3) for their toxic effect on the non-cancerous

remain unaffected. Additionally, strong compensatory upregu- hepatic stem cell line HepaRG^TM. First, HepaRG^TMprogenitor

Cell Chemical Biology 28, 1–13, June 17, 2021

7

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 6. Biological investigation of SG-005 (2) and SG-094 (3)

(A and B) Only SG-005 (2) and SG-094 (3) impair phosphorylation of VEGFR2 downstream targets. (A) After serum starvation, HUVECs were treated with 10 mM

tetrandrine (1), SG-005 (2), SG-094 (3), SG-132 (31), or SG-145 (11) for 1 h before stimulation with VEGF-A₁₆₅for 15 min. Phosphorylated proteins were analyzed

by western blotting. The experiment was performed in triplicate, and a representative experiment is shown. tet, tetrandrine. (B) Quantitative evaluation of protein

levels from (A). Bands were normalized to whole lane protein and are presented as percentage relative to DMSO control. The bar graphs indicate mean SEM of

three independent experiments (one-way ANOVA, Dunnett’s test).

(C) Tetrandrine (1), SG-005 (2), and SG-094 (3) are preferably taken up by RIL175 cells (determined by analytical HPLC and normalized to cell number). Bar graphs

indicate mean SEM from at least three independent experiments (unpaired t test with Welch’s correction).

(D and E) Reduction of LysoTracker Red ﬂuorescence indicates lysosomal uptake of TPC2 inhibitors into (D) RIL175 WT cells and (E) HUVECs. Geometric means

of ﬂuorescence intensity (gMFI) were determined by ﬂow cytometry and normalized to DMSO control. Concanamycin A (ConcA) served as control. Bar graphs

indicate mean SEM of three independent experiments (one-way ANOVA, Tukey’s test).

(legend continued on next page)

8

Cell Chemical Biology 28, 1–13, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

cells, which possess various characteristics of primary human DISCUSSION

hepatocytes (Vlach et al., 2019), were exposed to 24-h treatment

to assess acute toxicity. Tetrandrine (1) treatment caused a sig- The identiﬁcation of druggable targets represents a key step in

niﬁcant reduction of cell viability at 10 and 20 mM, unlike SG-005 the improvement of tumor therapy. Analysis of differentially ex-

(2) and SG-094 (3) (Figure 7A). Second, differentiated HepaRG pressed gene signatures revealed that high gene expression of

cells, characterized by peak levels of CYP activities (Pernelle TPCN2 acts as negative predictive value for postoperative

et al., 2011), were used. Treatment with 10 mM tetrandrine (1) biochemical recurrence in prostate cancer patients (Li et al.,

induced a stronger effect on cell viability compared with SG- 2019b). Moreover, TPC2 was shown to be involved in neoangio-

005 (2) and SG-094 (3), whereas 20 mM caused similar cytotox- genesis (Favia et al., 2014) as well as migration and metastasis

icities (Figure 7B). Toxicity was also probed against peripheral formation (Nguyen et al., 2017), providing incentive for a further

blood mononuclear cells (PBMCs). At 10 mM, the strongest investigation of its role in cancer. Our ﬁndings clearly illustrate

impact on cell viability was observed for tetrandrine (1) in com- that TPC2 is a cancer driver, as genetic ablation of the channel

parison with the congeners (Figure 7C). Following treatment reduces proliferation of liver cancer cells in vitro and effectively

with 20 mM, SG-094 (3) had the least inﬂuence of the three tested prevents tumor growth in vivo. Apart from induction of blood

compounds on cell viability. Together, removal of substructures vessel formation and activation of invasion and metastasis,

from tetrandrine (1) slightly (SG-005 (2)) or substantially (SG-094 chronic proliferation doubtlessly represents a key hallmark capa-

(3)) improved the toxicity proﬁle for non-cancerous liver and bility that enables tumor growth (Hanahan and Weinberg, 2011).

blood cells. Additionally, toxicity proﬁles of tetrandrine (1),

In addition to the frequent dysregulation of cellular prolifera-

SG-005 (2), and SG-094 (3) on non-cancerous cells were related tion, cancer cells often reprogram their energy metabolism in or-

to their IC₅₀values in cancer cells, and an overview is given in der to fuel extensive cell division (Hanahan and Weinberg, 2011).

Table S3.

We unraveled that TPC2 inﬂuences glucose utilization of liver

cancer cells for energy production, expanding the number of

cancer hallmarks affected by TPC2. Healthy cells convert

SG-094 (3) suppresses tumor growth in vivo

Apart from the outlined toxicity issues, in vivo use of tetrandrine glucose to pyruvate in the cytosol by glycolysis and then further

(1) is limited due to its poor solubility, and some authors oxidize it to carbon dioxide in mitochondria. Many cancer cells,

describe the need for dissolving it in acidiﬁed solutions before however, preferentially use glycolysis for ATP production,

dosage (Huang et al., 1997; Jin et al., 2011; Lin et al., 2009; despite the presence of oxygen, a phenomenon that is termed

Sakurai et al., 2015). Hence, the tetrandrine congeners SG- aerobic glycolysis or the Warburg effect (Gatenby and Gillies,

005 (2) and SG-094 (3) were evaluated for their therapeutic po- 2004). Upon TPC2 KO or pharmacological inhibition, a switch to-

tential in a syngeneic ectopic mouse model. Dosing was ward lower glycolysis occurred and, hence, metabolic reprog-

adapted from a published dose of 100 mg/kg body weight ramming of liver cancer cells was partly reversed. Although

(161 nmol/kg) for tetrandrine (1) (Jin et al., 2011) and adjusted glycolysis produces substantially less ATP per mol of glucose

by their reduced IC₅₀values (Figure 4A), plus a second regimen than mitochondrial OXPHOS (Hanahan and Weinberg, 2011), it

with a 50% higher dose (SG-005 (2): 38 nmol/kg and 57 nmol/ is hypothesized that it represents a distinct growth advantage

kg, SG-094 (3): 59 nmol/kg and 90 nmol/kg). As mouse weight and seems to be linked to tumor aggressiveness. Constantly up-

remained constant among all groups over the initial dose- regulated glycolysis provides faster ATP production and meta-

ﬁnding period and no solubility issues without acidiﬁcation of bolic intermediates for biosynthesis of macromolecules required

the injection solutions occurred, SG-005 (2) was ﬁnally admin- for rapid proliferation as well as enabling the adaptation to hyp-

istered intraperitoneally at 57 nmol/kg and SG-094 (3) at oxic conditions that frequently occur in tumor regions (Gatenby

90 nmol/kg every 2–3 days over a 10-day timescale (Figures and Gillies, 2004). Several preclinical studies showed the efﬁ-

7D–7G). Due to the known toxicity problems, tetrandrine (1) cacy of disrupting glycolytic energy metabolism of cancer cells,

was not used in vivo. To exclude effects of SG-005 (2) or SG- but direct inhibition of the glycolytic pathway can cause severe

094 (3) treatment on the accumulation of D-luciferin in the en- toxicity (Abdel-Wahab et al., 2019). By targeting TPC2, we pre-

grafted tumors related to altered blood supply, we measured sent an alternative approach to indirectly interfere with aerobic

tumor volume simultaneously using a caliper (Figure 7E) in glycolysis.

addition to bioluminescence imaging (Figures 7D and 7G).

Importantly, we have underlined that TPC2 is a druggable

Notably, 10 days post engraftment of RIL175 cells, average target, as small-molecule blockers, SG-005 (2) and SG-094 (3),

bioluminescence signals in the SG-094 (3) arm were decreased with antiproliferative activity have been developed in the course

by 42% (Figures 7D and 7G) and average tumor volume in the of this investigation. Reduced proliferation upon TPC2 silencing

SG-094 (3) arm was decreased by 33% (Figure 7E). Concur- is not limited to cancer cells, as non-cancerous hepatocytes

rently, no weight loss (Figure 7F) or changes in health or were also affected in this study. However, taking into consider-

behavior of the mice were observed. In contrast, the adminis- ation that TPC2 KO mice are viable (Favia et al., 2014; Grimm

tered dose of SG-005 (2) was not sufﬁcient to markedly sup- et al., 2014), our study suggests that pharmacological interfer-

press tumor growth (Figures 7D and 7E).

ence with TPC2 function is sufﬁcient to induce antitumor effects

(F) TPC2 inhibition increases OCR/ECAR ratios of RIL175 WT cells after glucose starvation. The experiment was performed as described for Figure 2A. OCR/

ECAR ratios are displayed over time as mean SEM of three independent experiments, and statistical analyses are shown for 2 mM (repeated-measures ANOVA,

Dunnett’s test).

(G and H) Real-time ATP monitoring in RIL175 WT cells after treatment with (G) DMSO or (H) SG-005 (2) (2 mM) as described for Figures 2C and 2D.

Cell Chemical Biology 28, 1–13, June 17, 2021

9

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Figure 7. SG-094 (3) is less toxic to non-cancerous cells in vitro and reduces tumor growth in vivo

(A–C) Toxicity of tetrandrine (1), SG-005 (2), and SG-094 (3) was probed by treating (A) HepaRG progenitor, (B) differentiated HepaRG, and (C) peripheral blood

mononuclear cells (PBMCs). Cell viability was assessed by (A and B) CellTiter-Blue assay and normalized to DMSO control or by (C) propidium iodide staining. Bar

graphs indicate mean SEM of three independent experiments (one-way ANOVA, Tukey’s test).

(D–G) RIL175 WT cells were implanted into the ﬂanks of C57Bl/6-Tyr mice. SG-005 (2) and SG-094 (3) were administered intraperitoneally at 57 and 90 nmol/kg,

respectively, every 2–3 days. Line graphs indicate (D) bioluminescence signal intensities and (E) tumor volumes over time. In (D) and (E), engrafted tumors derived

from TPC2 KO cells are displayed as dashed line for comparison. In (D), data are presented as mean SEM (n = 9 for WT PBS/DMSO, n = 10 for other treatment

arms, Grubbs’ outlier test [a = 0.05], one-way ANOVA, Dunnett’s test). In (E), data are presented as mean SEM (n = 9 for SG-094 (3), n = 10 for other treatment

arms, Grubbs’ outlier test [a = 0.05], one-way ANOVA, Dunnett’s test). (F) Mouse weight from (D) and (E) over time as mean SEM (n = 10). (G) Bioluminescence

images from day 10. The outlier is marked with an x. One excised tumor is shown for each group.

while it can generally be tolerated by mammalian organisms. Be- (100 mM) were required (Favia et al., 2014). Similarly, 500 mM

sides, preferential uptake of the blockers into cancer cells could naringenin was applied to prevent VEGF-induced vessel forma-

contribute to the observed tumor selectivity.

tion (Pafumi et al., 2017). Despite their improved antiprolifera-

Literature-known TPC2 inhibitors naringenin, Ned-19, tive properties in comparison with naringenin and Ned-19 and

ﬂuphenazine, and raloxifene were mechanistically studied by their rapid uptake into lysosomes, only raloxifene, not ﬂuphen-

us and others with respect to the involvement of TPC2 in azine, was able to partly reduce phosphorylation of SAPK/JNK.

VEGF-induced angiogenesis, which is necessary for vasculari- Of note, we were able to show that a low concentration of SG-

zation of solid tumors (Favia et al., 2014; Pafumi et al., 2017). 005 (2) and SG-094 (3) (10 mM) was sufﬁcient to impair phos-

In detail, TPC2 knockdown abolished tube formation of phorylation of VEGFR2 downstream targets, whereas tetran-

HUVECs in vitro and vessel formation was abrogated in TPC2 drine (1) did not show an effect. Thus, an improved on-target

KO, but not in TPC1 KO mice, suggesting an isoform-speciﬁc effect of SG-005 (2) and SG-094 (3) was found in intact whole

role for TPC2. This effect was mimicked with Ned-19 and it cells, which we partly linked to enforced accumulation in

impaired proangiogenic signaling, albeit high-micromolar doses lysosomes.

10 Cell Chemical Biology 28, 1–13, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

In line with reported low potency to impair vessel formation, 094 (3) has antitumor efﬁcacy in vivo, positioning it as a

Ned-19 and naringenin displayed weak antiproliferative effects promising preclinical candidate.

on RIL175 liver cancer cells, starting at high-micromolar concen-

trations (R75 mM). Instead, we chose tetrandrine (1) as lead

molecule because it could successfully reduce Ebola virus infec-

STAR+METHODS

tion of HeLa cells (%2 mM) (Sakurai et al., 2015) via TPC inhibi-

Detailed methods are provided in the online version of this paper

and include the following:

tion. Although tetrandrine (1) also considerably inhibited prolifer-

ation of RIL175 cells (IC₅₀9.1 mM), antiproliferative effects were

markedly enhanced for SG-005 (2) and SG-094 (3) (IC₅₀2.4 and

3.7 mM, respectively).

d KEY RESOURCES TABLE

d RESOURCE AVAILABILITY

B Lead contact

Tetrandrine (1) has serious deﬁciencies, which include its

structural complexity and its toxicity (Jin et al., 2011; Li et al.,

2019a; Qi et al., 2013; Tainlin et al., 1982; Yan et al., 2006). To

cope with these limitations, we synthesized and characterized

various analogs of the lead molecule. Apart from enabling chem-

ical accessibility, structural simpliﬁcation was proved to be an

effective strategy to overcome drawbacks of natural compounds

with respect to drug-likeness and potency (Wang et al., 2019).

Here, we demonstrate that the rigid, macrocyclic BBTHIQ struc-

ture of tetrandrine (1) can be reduced to the less complex BTHIQ

scaffold, in case additional aromatic residues are introduced. In

comparison with the very complex synthetic routes to tetran-

B Materials availability

B Data and code availability

d EXPERIMENTAL MODEL AND SUBJECT DETAILS

B Mouse experiments

B Cell lines and culture

d METHOD DETAILS

B Endolysosomal patch-clamp method

B Fura-2 Ca²⁺imaging

B Fluo-4 Ca²⁺imaging

B Generation of a TPC2 knockout cell line

B Quantitative real-time PCR

B Wound healing assay

€

drine (1) (Inubushi et al., 1968; Schutz et al., 2020), SG-005 (2)

and SG-094 (3) are accessible via a short, inexpensive synthesis.

Moreover, simpliﬁcation resulted in similar (SG-005 (2)) or even

enhanced inhibition (SG-094 (3)) of TPC2 currents. Concurrently,

toxicity to non-cancerous cells was reduced. Since SG-005 (2)

was more toxic to PBMCs than SG-094 (3), a lower dose was

applied in a mouse model, but this dose was not sufﬁcient to

translate the antiproliferative in vitro potency into in vivo efﬁcacy.

Nevertheless, due to the enhanced on-target effect on TPC2

versus tetrandrine (1), SG-005 (2) might have the potential to

treat other conditions that are connected to TPC2, such as car-

diac dysfunction and viral infections (Ou et al., 2020; Patel and

Kilpatrick, 2018). The other optimized tetrandrine congener,

SG-094 (3), successfully managed to reduce tumor growth in vivo

at tolerated doses and thereby represents a promising drug

candidate for cancer therapy.

B Proliferation assays

B Colony formation

B In vivo experiments

B Seahorse extracellular ﬂux analysis

B Cellular uptake assay

B Lysosomal uptake assay

B FLIM-FRET imaging

B Toxicity assays

B Western blot analysis and antibodies

B Apoptosis assay

B Mitochondrial mass

B Mitochondrial membrane potential

B Chemistry

B Computational methods

d QUANTIFICATION AND STATISTICAL ANALYSIS

SUPPLEMENTAL INFORMATION

SIGNIFICANCE

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

chembiol.2021.01.023.

The identiﬁcation of druggable targets presents a key strat-

egy to improve cancer therapy. This study unveils that ge-

netic knockout of the endolysosomal cation channel TPC2

reduces cancer cell proliferation and abrogates tumor

growth, thus making the ion channel a valid target. The num-

ber of currently available TPC2 blockers is scarce and their

therapeutic relevance is either limited by low potency or

toxicity. By combining chemical synthesis and state-of-

the-art screening methods, two easily accessible, potent

TPC2 blockers (SG-005 (2) and SG-094 (3)) were developed.

Both molecules are simpliﬁed analogs of the natural alkaloid

tetrandrine. Chemical simpliﬁcation to the relevant pharma-

cophores enhanced their lysosomal uptake, increased their

antiproliferative potencies against tumor cells, and

improved their on-target effects on TPC2 in a cell-based

setting. Concurrently, toxicity to non-malignant cells could

be reduced, allowing in vivo administration. Of note, SG-

ACKNOWLEDGMENTS

This work was funded by the German Research Foundation (DFG VO 376/19-

1, BR 1034/7-1, GR 4315/2-1, BI 484/6-1). We thank Henry Dube for providing

the HPLC, CD-spectrometer, and calculation resources. We thank Kerstin

Loske for her help with the animal experiments and Anna Niedrig, Bernadette

€

Grohs, Julian Fradrich, Jana Peliskova, Silvia Schnegg, Moritz Schroll, and

Benedikt Spitzelberger for technical assistance.

AUTHOR CONTRIBUTIONS

C.G., F.B., A.M.V., M.M., S.G., Y.-K.C., K.B., M.B., M.S., and O.N.P.N.

conceived the project and designed experiments. F.B. and S.G. designed

and synthesized compounds. S.G., Y.-K.C., N.U., and C.-C.C. performed

Ca²⁺imaging, patch-clamp, and data analysis. M.M., T.Z., O.N.P.N., C.M.,

F.A.G., and F.G. carried out cell-based experiments and data analysis. A.G.

performed computational calculations. C.O. carried out animal experiments.

Cell Chemical Biology 28, 1–13, June 17, 2021 11

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

M.M. and S.G. wrote the manuscript. C.M., C.G., F.B., A.M.V., S.Z., and K.B.

critically revised the manuscript.

Gawley, R.E., and Smith, G.A. (2011). Formal synthesis of the bisbenzylisoqui-

noline alkaloid berbamunine by asymmetric substitution of chiral organolithium

compounds. ARKIVOC 2011 , 167–179.

Gerndt, S., Chen, C.C., Chao, Y.K., Yuan, Y., Burgstaller, S., Scotto Rosato, A.,

Krogsaeter, E., Urban, N., Jacob, K., Nguyen, O.N.P., et al. (2020). Agonist-

mediated switching of ion selectivity in TPC2 differentially promotes lysosomal

function. eLife 9, https://doi.org/10.7554/eLife.54712.

DECLARATION OF INTERESTS

The authors declare no competing interests.

Received: July 15, 2020

Grimm, C., Jors, S., Saldanha, S.A., Obukhov, A.G., Pan, B., Oshima, K.,

Cuajungco, M.P., Chase, P., Hodder, P., and Heller, S. (2010). Small molecule

activators of TRPML3. Chem. Biol. 17 , 135–148.

Revised: January 8, 2021

Accepted: January 28, 2021

Published: February 23, 2021

€

Grimm, C., Holdt, L.M., Chen, C.C., Hassan, S., Muller, C., Jors, S., Cuny, H.,

Kissing, S., Schroder, B., Butz, E., et al. (2014). High susceptibility to fatty liver

REFERENCES

disease in two-pore channel 2-deﬁcient mice. Nat. Commun. 5 , 4699.

Gunaratne, G.S., Yang, Y., Li, F., Walseth, T.F., and Marchant, J.S. (2018).

NAADP-dependent Ca(2+) signaling regulates Middle East respiratory syn-

drome-coronavirus pseudovirus translocation through the endolysosomal

system. Cell Calcium 75 , 30–41.

Abdel-Wahab, A.F., Mahmoud, W., and Al-Harizy, R.M. (2019). Targeting

glucose metabolism to suppress cancer progression: prospective of anti-

glycolytic cancer therapy. Pharmacol. Res. 150 , 104511.

€

¨

Ardelt, M.A., Frohlich, T., Martini, E., Muller, M., Kanitz, V., Atzberger, C.,

Cantonati, P., Messner, M., Posselt, L., Lehr, T., et al. (2018). Inhibition of cy-

clin-dependent kinase 5 - a novel strategy to improve sorafenib response in

HCC therapy. Hepatology 69 , 376–393.

Haber, M., Norris, M.D., Kavallaris, M., Bell, D.R., Davey, R.A., White, L., and

Stewart, B.W. (1989). Atypical multidrug resistance in a therapy-induced drug-

resistant human leukemia cell line (LALW-2): resistance to Vinca alkaloids in-

dependent of P-glycoprotein. Cancer Res. 49 , 5281–5287.

Awuah, E., and Capretta, A. (2010). Strategies and synthetic methods directed

toward the preparation of libraries of substituted isoquinolines. J. Org. Chem.

75 , 5627–5634.

Haeussler, M., Schonig, K., Eckert, H., Eschstruth, A., Mianne, J., Renaud,

J.B., Schneider-Maunoury, S., Shkumatava, A., Teboul, L., Kent, J., et al.

(2016). Evaluation of off-target and on-target scoring algorithms and integra-

tion into the guide RNA selection tool CRISPOR. Genome Biol. 17 , 148.

Bauer, D.E., Canver, M.C., and Orkin, S.H. (2015). Generation of genomic de-

letions in mammalian cell lines via CRISPR/Cas9. J. Vis. Exp. e52118.

Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next gener-

Bergers, G., and Benjamin, L.E. (2003). Tumorigenesis and the angiogenic

ation. Cell 144 , 646–674.

switch. Nat. Rev. Cancer 3 , 401–410.

Huang, Y.T., Liu, T.B., Lin, H.C., Yang, M.C., and Hong, C.Y. (1997).

Hemodynamic effects of chronic tetrandrine and propranolol administration

on portal hypertensive rats. Pharmacology 54 , 225–231.

Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R.L., Torre, L.A., and Jemal, A.

(2018). Global cancer statistics 2018: GLOBOCAN estimates of incidence

and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin.

68 , 394–424.

Imamura, H., Nhat, K.P., Togawa, H., Saito, K., Iino, R., Kato-Yamada, Y.,

Nagai, T., and Noji, H. (2009). Visualization of ATP levels inside single living

cells with ﬂuorescence resonance energy transfer-based genetically encoded

indicators. Proc. Natl. Acad. Sci. U S A 106 , 15651–15656.

Cava, M.P., and Buck, K.T. (1969). Some synthetic studies in the isoquinoline

series. Tetrahedron 25 , 2795–2805.

Chao, Y.K., Schludi, V., Chen, C.C., Butz, E., Nguyen, O.N.P., Muller, M.,

Kruger, J., Kammerbauer, C., Ben-Johny, M., Vollmar, A.M., et al. (2017).

TPC2 polymorphisms associated with a hair pigmentation phenotype in hu-

mans result in gain of channel function by independent mechanisms. Proc.

Natl. Acad. Sci. U S A 114 , E8595–E8602.

Inubushi, Y., Masaki, Y., Matsumoto, S., and Takami, F. (1968). Total synthe-

ses of optically active natural isotetrandrine, phaeanthine and tetrandrine.

Tetrahedron Lett. 3399–3402.

´

Iturriaga-Vasquez, P., Miquel, R., Ivorra, M.D., D’Ocon, M.P., and Cassels,

B.K. (2003). Simpliﬁed tetrandrine congeners as possible antihypertensive

Chen, C.C., Cang, C., Fenske, S., Butz, E., Chao, Y.K., Biel, M., Ren, D., Wahl-

Schott, C., and Grimm, C. (2017). Patch-clamp technique to characterize ion

channels in enlarged individual endolysosomes. Nat. Protoc. 12 , 1639–1658.

agents with a dual mechanism of action. J. Nat. Prod. 66 , 954–957.

Jin, H., Li, L., Zhong, D., Liu, J., Chen, X., and Zheng, J. (2011). Pulmonary

toxicity and metabolic activation of tetrandrine in CD-1 mice. Chem. Res.

Toxicol. 24 , 2142–2152.

Colella, A.D., Chegenii, N., Tea, M.N., Gibbins, I.L., Williams, K.A., and

Chataway, T.K. (2012). Comparison of stain-free gels with traditional immuno-

blot loading control methodology. Anal. Biochem. 430 , 108–110.

Kawabata, Y., Naito, Y., Saitoh, T., Kawa, K., Fuchigami, T., and Nishiyama, S.

(2014). Synthesis of (+)-O-methylthalibrine by employing a stereocontrolled

Bischler-Napieralski reaction and an electrochemically generated diaryl ether.

Eur. J. Org. Chem. 2014 , 99–104.

Denko, N.C. (2008). Hypoxia, HIF1 and glucose metabolism in the solid

tumour. Nat. Rev. Cancer 8 , 705–713.

Dehairs, J., Talebi, A., Cheriﬁ, Y., and Swinnen, J.V. (2016). CRISP-ID: decod-

Kazmi, F., Hensley, T., Pope, C., Funk, R.S., Loewen, G.J., Buckley, D.B., and

Parkinson, A. (2013). Lysosomal sequestration (trapping) of lipophilic amine

(cationic amphiphilic) drugs in immortalized human hepatocytes (Fa2N-4

cells). Drug Metab. Dispos. 41 , 897–905.

ing CRISPR mediated indels by Sanger sequencing. Sci. Rep. 6 , 28973.

Divac, N., Prostran, M., Jakovcevski, I., and Cerovac, N. (2014). Second-gen-

eration antipsychotics and extrapyramidal adverse effects. Biomed. Res. Int.

2014 , 656370.

Kogot-Levin, A., Saada, A., Leibowitz, G., Soiferman, D., Douiev, L., Raz, I.,

and Weksler-Zangen, S. (2016). Upregulation of mitochondrial content in cyto-

chrome c oxidase deﬁcient ﬁbroblasts. PLoS One 11 , e0165417.

Evans, D.A., Katz, J.L., and West, T.R. (1998). Synthesis of diaryl ethers

through the copper-promoted arylation of phenols with arylboronic acids. An

expedient synthesis of thyroxine. Tetrahedron Lett. 39 , 2937–2940.

Li, F., Dong, Y.Z., Zhang, D., Zhang, X.M., Lin, Z.J., and Zhang, B. (2019a).

Molecular mechanisms involved in drug-induced liver injury caused by ur-

ate-lowering Chinese herbs: a network pharmacology study and biology ex-

periments. PLoS One 14 , e0216948.

Favia, A., Desideri, M., Gambara, G., D’Alessio, A., Ruas, M., Esposito, B., Del

Bufalo, D., Parrington, J., Ziparo, E., Palombi, F., et al. (2014). VEGF-induced

neoangiogenesis is mediated by NAADP and two-pore channel-2-dependent

Ca²⁺signaling. Proc. Natl. Acad. Sci. U S A 111 , E4706–E4715.

Li, F., Ji, J.P., Xu, Y., and Liu, R.L. (2019b). Identiﬁcation a novel set of 6 differ-

ential expressed genes in prostate cancer that can potentially predict

biochemical recurrence after curative surgery. Clin. Transl Oncol. 21 ,

1067–1075.

Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A.,

Cheeseman, J.R., Scalmani, G., Barone, V., Petersson, G.A., Nakatsuji, H.,

et al. (2016). Gaussian 16 Revision C.01 (Gaussin Inc).

Gatenby, R.A., and Gillies, R.J. (2004). Why do cancers have high aerobic

Li, W., Qiu, Y., Hao, J., Zhao, C., Deng, X., and Shu, G. (2018). Dauricine upre-

gulates the chemosensitivity of hepatocellular carcinoma cells: role of

glycolysis? Nat. Rev. Cancer 4 , 891–899.

12 Cell Chemical Biology 28, 1–13, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

repressing glycolysis via miR-199a:HK2/PKM2 modulation. Food Chem.

Riccardi, C., and Nicoletti, I. (2006). Analysis of apoptosis by propidium iodide

Toxicol. 121 , 156–165.

staining and ﬂow cytometry. Nat. Protoc. 1 , 1458–1461.

Lin, T.Y., Tseng, S.H., Li, S.J., Chen, J.C., Shieh, J.S., and Chen, Y. (2009).

Tetrandrine increased the survival rate of mice with lipopolysaccharide-

induced endotoxemia. J. Trauma 66 , 411–417.

Ruas, M., Rietdorf, K., Arredouani, A., Davis, L.C., Lloyd-Evans, E., Koegel, H.,

Funnell, T.M., Morgan, A.J., Ward, J.A., Watanabe, K., et al. (2010). Puriﬁed

TPC isoforms form NAADP receptors with distinct roles for Ca(2+) signaling

and endolysosomal trafﬁcking. Curr. Biol. 20 , 703–709.

Lu, S.-T., Wu, Y.-C., and Leou, S.-P. (1987). The oxidation of isoquinoline alka-

Sakurai, Y., Kolokoltsov, A.A., Chen, C.C., Tidwell, M.W., Bauta, W.E.,

Klugbauer, N., Grimm, C., Wahl-Schott, C., Biel, M., and Davey, R.A. (2015).

Ebola virus. Two-pore channels control Ebola virus host cell entry and are

drug targets for disease treatment. Science 347 , 995–998.

loids withm-chloroperbenzoic acid. J. Chin. Chem. Soc. 34 , 33–42.

Nakano, M., Imamura, H., Nagai, T., and Noji, H. (2011). Ca(2)(+) regulation of

mitochondrial ATP synthesis visualized at the single cell level. ACS Chem. Biol.

6 , 709–715.

Santos, R., Ursu, O., Gaulton, A., Bento, A.P., Donadi, R.S., Bologa, C.G.,

Karlsson, A., Al-Lazikani, B., Hersey, A., Oprea, T.I., et al. (2017). A compre-

hensive map of molecular drug targets. Nat. Rev. Drug Discov. 16 , 19–34.

Nguyen, O.N., Grimm, C., Schneider, L.S., Chao, Y.K., Atzberger, C., Bartel,

K., Watermann, A., Ulrich, M., Mayr, D., Wahl-Schott, C., et al. (2017). Two-

pore channel function is crucial for the migration of invasive cancer cells.

Cancer Res. 77 , 1427–1438.

Schrittwieser, J.H., Resch, V., Wallner, S., Lienhart, W.D., Sattler, J.H., Resch,

J., Macheroux, P., and Kroutil, W. (2011). Biocatalytic organic synthesis of

optically pure (S)-scoulerine and berbine and benzylisoquinoline alkaloids.

J. Org. Chem. 76 , 6703–6714.

Orci, L.A., Lacotte, S., Delaune, V., Slits, F., Oldani, G., Lazarevic, V., Rossetti,

C., Rubbia-Brandt, L., Morel, P., and Toso, C. (2018). Effects of the gut-liver

axis on ischaemia-mediated hepatocellular carcinoma recurrence in the

mouse liver. J. Hepatol. 68 , 978–985.

¨

Schrodinger (2019). S. Release 2019-2: MacroModel (Schrodinger, LLC).

¨

€

Schutz, R., Meixner, M., Antes, I., and Bracher, F. (2020). A modular approach

Ou, X., Liu, Y., Lei, X., Li, P., Mi, D., Ren, L., Guo, L., Guo, R., Chen, T., Hu, J.,

et al. (2020). Characterization of spike glycoprotein of SARS-CoV-2 on virus

entry and its immune cross-reactivity with SARS-CoV. Nat. Commun.

11 , 1620.

to the bisbenzylisoquinoline alkaloids tetrandrine and isotetrandrine. Org.

Biomol. Chem. 18 , 3047–3068.

Tainlin, L., Tingyi, H., Changqi, Z., Peipei, Y., and Qiong, Z. (1982). Studies of

the chronic toxicity of tetrandrine in dogs: an inhibitor of silicosis. Ecotoxicol

Environ. Saf. 6 , 528–534.

Pafumi, I., Festa, M., Papacci, F., Lagostena, L., Giunta, C., Gutla, V., Cornara,

L., Favia, A., Palombi, F., Gambale, F., et al. (2017). Naringenin impairs two-

pore channel 2 activity and inhibits VEGF-induced angiogenesis. Sci. Rep.

7 , 5121.

Tian, Y., Shen, S., Jiang, Y., Shen, Q., Zeng, S., and Zheng, J. (2016). CYP3A5

mediates bioactivation and cytotoxicity of tetrandrine. Arch. Toxicol. 90 ,

1737–1748.

Patel, S., and Kilpatrick, B.S. (2018). Two-pore channels and disease.

Biochim. Biophys. Acta Mol. Cell Res. 1865 , 1678–1686.

Vlach, M., Quesnot, N., Dubois-Pot-Schneider, H., Ribault, C., Verres, Y.,

Petitjean, K., Rauch, C., Morel, F., Robin, M.A., Corlu, A., et al. (2019).

Cytochrome P450 1A1/2, 2B6 and 3A4 HepaRG cell-based biosensors to

monitor hepatocyte differentiation, drug metabolism and toxicity. Sensors

(Basel) 19, https://doi.org/10.3390/s19102245.

Penny, C.J., Vassileva, K., Jha, A., Yuan, Y., Chee, X., Yates, E., Mazzon, M.,

Kilpatrick, B.S., Muallem, S., Marsh, M., et al. (2019). Mining of Ebola virus en-

try inhibitors identiﬁes approved drugs as two-pore channel pore blockers.

Biochim. Biophys. Acta Mol. Cell Res. 1866 , 1151–1161.

Wang, S., Dong, G., and Sheng, C. (2019). Structural simpliﬁcation of natural

Pernelle, K., Le Guevel, R., Glaise, D., Stasio, C.G., Le Charpentier, T., Bouaita,

B., Corlu, A., and Guguen-Guillouzo, C. (2011). Automated detection of hepa-

totoxic compounds in human hepatocytes using HepaRG cells and image-

based analysis of mitochondrial dysfunction with JC-1 dye. Toxicol. Appl.

Pharmacol. 254 , 256–266.

products. Chem. Rev. 119 , 4180–4220.

Workman, P., and Al-Lazikani, B. (2013). Drugging cancer genomes. Nat. Rev.

Drug Discov. 12 , 889–890.

Wu, H., Ding, Z., Hu, D., Sun, F., Dai, C., Xie, J., and Hu, X. (2012). Central role

of lactic acidosis in cancer cell resistance to glucose deprivation-induced cell

death. J. Pathol. 227 , 189–199.

Petersen, N.H., Olsen, O.D., Groth-Pedersen, L., Ellegaard, A.M., Bilgin, M.,

Redmer, S., Ostenfeld, M.S., Ulanet, D., Dovmark, T.H., Lonborg, A., et al.

(2013). Transformation-associated changes in sphingolipid metabolism sensi-

tize cells to lysosomal cell death induced by inhibitors of acid sphingomyeli-

nase. Cancer Cell 24 , 379–393.

Yamaguchi, S., Jha, A., Li, Q., Soyombo, A.A., Dickinson, G.D., Churamani, D.,

Brailoiu, E., Patel, S., and Muallem, S. (2011). Transient receptor potential mu-

colipin 1 (TRPML1) and two-pore channels are functionally independent or-

ganellar ion channels. J. Biol. Chem. 286 , 22934–22942.

Pouysegu, L., Avellan, A.V., and Quideau, S. (2002). Iodine(III)-mediated gen-

eration of nitrogen-tethered orthoquinol acetates for the construction of

oxygenated indole, quinoline, and phenanthridine alkaloid motifs. J. Org.

Chem. 67 , 3425–3436.

Yan, C., Xin-Ming, Q., Li-Kun, G., Lin-Lin, L., Fang-Ping, C., Ying, X., Xiong-Fei,

W., Xiang-Hong, L., and Jin, R. (2006). Tetrandrine-induced apoptosis in rat

primary hepatocytes is initiated from mitochondria: caspases and endonu-

clease G (Endo G) pathway. Toxicology 218 , 1–12.

Qi, X.M., Miao, L.L., Cai, Y., Gong, L.K., and Ren, J. (2013). ROS generated by

CYP450, especially CYP2E1, mediate mitochondrial dysfunction induced by

tetrandrine in rat hepatocytes. Acta Pharmacol. Sin. 34 , 1229–1236.

Zhang, J., Nuebel, E., Wisidagama, D.R., Setoguchi, K., Hong, J.S., Van Horn,

C.M., Imam, S.S., Vergnes, L., Malone, C.S., Koehler, C.M., et al. (2012).

Measuring energy metabolism in cultured cells, including human pluripotent

stem cells and differentiated cells. Nat. Protoc. 7 , 1068–1085.

Rossi, L.M.M., Copes, R.M., Dal Osto, L.C., Flores, C., Comim, F.V., and

Premaor, M.O. (2018). Factors related with osteoporosis treatment in post-

menopausal women. Medicine (Baltimore) 97 , e11524.

Zorova, L.D., Popkov, V.A., Plotnikov, E.Y., Silachev, D.N., Pevzner, I.B.,

Jankauskas, S.S., Babenko, V.A., Zorov, S.D., Balakireva, A.V., Juhaszova,

M., et al. (2018). Mitochondrial membrane potential. Anal. Biochem.

552 , 50–59.

Ran, F.A., Hsu, P.D., Wright, J., Agarwala, V., Scott, D.A., and Zhang, F. (2013).

Genome engineering using the CRISPR-Cas9 system. Nat. Protoc. 8 ,

2281–2308.

Cell Chemical Biology 28, 1–13, June 17, 2021 13

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE

Antibodies

SOURCE

IDENTIFIER

Akt antibody

Cell Signaling Technologies

Santa Cruz Biotechnology

Cell Signaling Technologies

Cat# 9272; RRID: AB_329827

Cat# sc-654; RRID: AB_631423

Cat# 9102: RRID: AB_330744

Cat# 9271; RRID: AB_329825

Cat# 9571: RRID: AB_329837

Cat# 9106; RRID: AB_331768

NOS3 antibody

p44/p42 MAPK antibody

phospho-Akt (Ser473) antibody

phospho-eNOS (Ser1177) antibody

phospho-p44/p42 MAPK (Thr202/Tyr204)

antibody

phospho-p54/p46 SAPK/JNK (Thr183/

Tyr185) antibody

Cell Signaling Technologies

Cat# 9251; RRID: AB_331659

Cat# 9252; RRID: AB_2250373

SAPK/JNK antibody

Biological Samples

Primary Peripheral Blood Mononuclear

Cells (PBMC), Normal, Human

ATCC

Cat# PCS-800-011ꢀ

Chemicals, Peptides, and Recombinant Proteins

( )-naringenin

Sigma Aldrich

Fisher Scientiﬁc

PAN Biotech

Cat# N5893

2-deoxy-D-glucose

Cat# D8375

Acetonitrile (HPLC grade)

Amphotericin B

Cat# A998-1

Cat# P06-01250P

Cat# 552843

Anti-Mouse Ig, k/Negative Control (FBS)

BD Biosciences

Compensation Particles Set

ATP

Sigma Aldrich

Cat# A9187

Cat# 557721

BD Alexa Fluorꢁ 488 Mouse IgG1 k Isotype

Control

BD Biosciences

BD PE Mouse IgG1, k/Isotype Control

Berbamine dihydrochloride

Cepharanthine

BD Biosciences

Sigma Aldrich

Cat# 555749

Cat# 547190

Cat# FC66045

Cat# 04574834001

Cat# sc-202111

Cat# FD66056

N/A

Carbosynth

cOmpleteꢁ EDTA-free

Concanamycin A

Roche Diagnostics

Santa Cruz Biotechnology

Carbosynth

Dauricine

Deionized water

self-prepared with an in-house ion

exchanger

DMEM

PAN Biotech

Sigma Aldrich

Cat# P04-03550

Cat# D5030

DMEM 5030 powder without glucose,

L-glutamine, phenol red, sodium pyruvate

and sodium bicarbonate, powder, suitable

for cell culture

EGCM Kit enhanced

Endothelial cell growth medium

Fetal calf serum

PeloBiotech

Cat# PB-SH-100-2199

Cat# PB-BH-100-9806

Cat# A15-751

Cat# 14202

PeloBiotech

PAA Laboratories

Life Technologies

Sigma Aldrich

Thermo Fisher

Biotium

Fluo-4 AM

Fluphenazine dihydrochloride

Fura-2 AM

Cat# F4765

Cat# F1221

GelRedꢁ Nucleic Acid Gel Stain

GeneRuler 1 kb Plus DNA Ladder

GlutaMAX

Cat# 41003

Thermo Fisher

Life Technologies

Cat# 10787018

Cat# 35050061

(Continued on next page)

e1 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Continued

REAGENT or RESOURCE

SOURCE

IDENTIFIER

HepaRG Maintenance/Metabolism Medium

Supplement

Life Technologies

Cat# HPRG620

HepaRG Thaw, Plate & General Purpose

Medium Supplement

Life Technologies

Cat# HPRG670

Cat# 10600023

Hybond PVDF (polyvinylidene diﬂuoride)

membrane

Amersham Bioscience

ibidi 8 well m-slides with glass bottoms

Ibidi

Cat# 80827

Cat# T3168

JC-1 Dye (Mitochondrial Membrane

Potential Probe)

Thermo Fisher

LipofectamineTM 3000

Invitrogen

Cat# L3000008

Cat# 13778030

LipofectamineTM RNAiMAX Transfection

Reagent

Thermo Fisher

LysoTracker-RedTM DND-99

MEM Eagle

Thermo Fisher

PAN Biotech

Cat# L7528

Cat# P04-09500

Cat# A452-1

Methanol (HPLC grade)

Fisher Scientiﬁc

Merck Millipore

Millipore Expressꢁ PLUS membrane

ﬁlter (0.22 mM)

Cat# GPWP09050

MitoTrackerꢀ Green FM

ML-SA1

Thermo Fisher

Merck

Cat# M7514

Cat# 648493

Non-targeting siRNA (Control siRNA-A)

Oxyacanthine sulfate

Santa Cruz Biotechnology

abcr

Cat# sc-37007

Cat# AB167724

Cat# P06-08025P

Penicillin-Streptomycin, 10,000 U/ml

Penicillin, 10 mg/ml Streptomycin

PAN Biotech

peqGOLD agarose, universal

PI(3,5)P₂

VWR International

Echelon Biosciences

Thermo Fisher

Sigma Aldrich

Roth

Cat# 732-2789DE

Cat# P-3508

Cat# 24040032

Cat# P6407

Cat# P8920

Cat# CN74.1

Cat# M0491L

Cat #S1227

Cat# 100-20

Cat# P04-16500

N/A

Pluronic acid

Poly-D-lysine hydrobromide

poly-L-lysine

Propidium Iodide

Q5ꢁ High-Fidelity DNA Polymerase

Raloxifene

New England Biolabs

Selleckchem

Recombinat Human VEGF165

RPMI 1640

Peprotech

PAN Biotech

Small molecules labeled as SG#

This paper; Bracher lab

Small molecules labeled as Z# (1-benzyl-

1,2,3,4-tetrahydro-isoquinolines)

molecules were a gift from Prof. Meinhart

Zenk (Prof. Meinhart Zenk passed away in

July 2011)

N/A

Sodium hydroxide solution 1 M

reagent grade

Sigma Aldrich

Cat# 79724

T4 DNA ligase

Tetrandrine

Thermo Fisher

Cat# 15224090

N/A

Prof. Peter Pachaly kindly provided

tetrandrine from a natural source (Prof.

Peter Pachaly passed away in

January 2019)

Thermo Scientiﬁc Phusion Green Hot Start

II High-Fidelity Polymerase

Thermo Fisher

Cat# F530L

TPC2 siRNA (human)

Santa Cruz Biotechnology

Cat# sc-96434

TPC2-A1-N (TPC2 activator)

Gerndt et al., 2020

Bracher Lab

PMID: 32167471

TPC2-A1-P (TPC2 activator)

Gerndt et al., 2020

Bracher Lab

PMID: 32167471

Cat# 3954

trans-Ned19

Tocris Bioscience

(Continued on next page)

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e2

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Continued

REAGENT or RESOURCE

TurboFect Transfection Reagent

Williams’ medium E

SOURCE

IDENTIFIER

Thermo Fisher

Life Technologies

Cat# R0531

Cat# A1217601

Critical Commercial Assays

CellTiter-Blueꢁ Assay

Promega

Cat# G8080

Cat# C34570

Cat# 4368813

CellTrace^TMCFSE Cell Proliferation Kit

Invitrogen

High Capacity cDNA Reverse

Transcription Kit

Applied Biosystems

QIAGEN Plasmid Maxiprep Kit

QIAGEN RNeasy Mini Kit

QIAGEN

Cat# 12165

QIAGEN

Cat# 74126

QIAquick Gel Extraction Kit

Seahorse XFe96 sensor cartridge

Seahorse XFGlycolysis Stress Test Kit

T7 DNA Endonuclease I assay

XFe96 microplate

QIAGEN

Cat# 28706

Agilent Technologies

New England Biolabs

Agilent Technologies

Cat# 102416-100

Cat# 103020-100

Cat# M0302L

Cat# 101085-004

Deposited Data

DNA sequences from CRISPR-Cas9 off-

target screening of RIL175 TPC2 KO

clone 1D1

This paper;

N/A

Supplemental Information, Table S1

DNA sequences of RIL175 TPC2 KO clones

This paper;

N/A

Supplemental Information, Figure S1

ECD spectra and related data generated

form computational methods

This paper;

Supplemental Information (Computational

methods)

Spectroscopic data of synthesized small

molecules

This paper;

N/A

STAR Methods (synthetic procedures; 1H

and 13C NMR, IR, HRMS data) and

Supplemental Information (NMR spectra)

Experimental Models: Cell Lines

HCT-15

ATCC

Cat# CCL-225

Cat# ACC 305

Cat# HB-8064

Cat# HPRGC10

Cat# ACC 180

Cat# C-12203

HEK293

DSMZ

Hep3B

HepaRG^TM, cryopreserved

ATCC

Life Technologies

DSMZ

HepG2

Human Umbilical Vein Endothelial Cells

(HUVEC c-pooled)

Promocell

RIL175

gift from Prof. Simon Rothenfußer (Center

for Integrated Protein Science Munich

(CIPS-M) and Division of Clinical

PMID: 29331341

Pharmacology, Department of Internal

€

Medicine IV, Klinikum der Universitat

€

Munchen, Munich, Germany); described by

Orci et al. (Orci et al., 2018)

Vincristine-resistant (VCR-R) CEM

gift from Prof. Maria Kavallaris (University of

New South Wales, Kensington, Australia);

described by Haber et al. (Haber

et al., 1989)

PMID: 2569932

Experimental Models: Organisms/Strains

C57Bl/6-Tyr mice

Envigo

N/A

Cat# 103

N/A

Oligonucleotides

as more than 10 oligonucleotides were

used, they are provided in Table S4

(Continued on next page)

e3 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Continued

REAGENT or RESOURCE

Recombinant DNA

SOURCE

IDENTIFIER

N/A

eSpCas9_2A_Blasti plasmid

plasmid was a gift from Dr. Lars Ko¨ nig

€

(Klinikum der Universitat Munchen, Munich,

€

Germany)

GO-ATeam (FRET Biosensor)

plasmid was a gift from Dr. Hiromi Imamura

(Nakano et al., 2011)

PMID: 21488691

hsTPC1-RFP

hsTPC2^L11A/L12A–RFP

Sakurai et al., 2015

Gerndt et al., 2020

Chao et al., 2017

PMID: 25722412

PMID: 32167471

PMID: 28923947

PMID: 21540176

PMID: 20189104

hsTPC2-YFP

hsTRPML1^LLAA-GFP

hsTRPML1-YFP

Yamaguchi et al., 2011

Grimm et al., 2010

pSpCas9(BB)-2A-Puro plasmid

plasmid was a gift from Feng Zhang (Ran

et al., 2013) and purchased from addgene

Cat# 62988 ; http://n2t.net/addgene:62988;

RRID:Addgene_62988

Software and Algorithms

Agilent ChemStationꢁ software

Rev. B04.02

Agilent Technologies

https://www.agilent.com/cs/library/

usermanuals/Public/G2170-

90234_upgradetoB.04.02-SP1.pdf

CRISP-ID

Dehairs et al., 2016

Haeussler et al., 2016

BD Biosciences

http://crispid.gbiomed.kuleuven.be/

http://crispor.tefor.net/

CRISPOR TEFOR

FlowJo 7.6

https://www.ﬂowjo.com/

Gaussian16 Revision A.03 program

package

Frisch et al., 2016

https://gaussian.com/gaussian16/

Image Lab 6.0.1

Bio-Rad

https://www.bio-rad.com/de-de/product/

image-lab-software?ID=KRE6P5E8Z

ImageJ

NIH

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

Living Image 4.4

Perkin Elmer

https://www.perkinelmer.com/de/lab-

products-and-services/resources/in-vivo-

imaging-software-downloads.html

MacroModel Software suite

Schro¨ dinger, LLC

Mestrelab Research S.L.

PicoQuant

www.schrodinger.com/macromodel

MestReNova, version 12.0.1-20560

Picoquant SymPhoTime 64 (V2.1)

https://mestrelab.com/download/mnova/

https://www.picoquant.com/products/

category/software/symphotime-64-

ﬂuorescence-lifetime-imaging-and-

correlation-software

Prism 8

GraphPad

https://www.graphpad.com/scientiﬁc-

software/prism/

Wave 2.6.0 software

Agilent Technologies

https://www.agilent.com/en/products/cell-

analysis/software-download-for-wave-

desktop

xCELLigence RTCA software

ACEA Biosciences

N/A

RESOURCE AVAILABILITY

Lead contact

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

Bartel (karin.bartel@cup.uni-muenchen.de).

Materials availability

Compounds and cell lines generated in this study will be made available on request, but we may require a payment and/or a

completed Materials Transfer Agreement, if there is potential for commercial application.

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e4

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Data and code availability

The published article includes all analyzed sequences of RIL175 TPC2 KO clones that were generated, it lists all the sequences from

the off-target screening of the RIL175 TPC2 KO clone 1D1 and it contains all structures of small molecules that were synthesized and

data that were generated from Computational methods during this study.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Mouse experiments

All animal experiments were approved by the District Government of Upper Bavaria, in accordance with institutional guidelines and

the German Animal Welfare. In accordance with the German law (x7a Abs. 2 Nr TierSchG together with x31 TierSchVersV), animal

experiments need to be essential, without any alternative. As stated and approved in the governmental permission by the Bavarian

government (ROZ 55.2-1-545-2532-107-13), there is still an unmet need to identify novel therapeutic strategies for tumor therapy and

to overcome therapeutic drawbacks and relapse. In that regard, basic research is mandatory and animal experiments are imperative

for the development of novel therapies (x22 AMG). Nevertheless, the suffering and number of animals used should be kept as limited

as possible. All animal experiments were performed in accordance with the 3 R principle (replace, reduce, reﬁne) in the institute’s

animal facility (Department Pharmazie, Haus G, quarantine room, Butenandtstr. 5-13, 81377 Munich). C57Bl/6-Tyr mice could be

used as RIL175 cells originate from them. To enable in vivo bioluminescence imaging of RIL175 cells that contain a stable luciferase

knock-in, the albino phenotype was necessary. Female C57Bl/6-Tyr mice (5-6 weeks old with a weight of around 20 g) were pur-

chased from Envigo. Mice were kept in a temperature-controlled facility with a usual 12 h/12 h light and dark schedule. In the facility,

humidity was kept at 60% and the room temperature at 22^ꢀC, both of which were continuously monitored by operating technology.

Light intensity in the facility was kept at around 120 lux, in the racks at around 40-60 lux. Mice were maintained in cages in groups

consisting of up to 5 animals in cages that can be ventilated individually and which had an area of 700 cm²(IVC, Type 2 long, System

Techniplast). Softwood ﬁbers (Ssniff, Soest) were used as embedding material. Additionally, recycled and autoclaved paper tissues

with a size of 24 cm x 23 cm (Papernet, 404882) were added to the cages, which are frequently used the mice for playing, nest-build-

ing and for hiding. A special diet for rats/mice (R/M-H, 10 mm; Ssniff, Soest) served as animal feed. Drinking water was acidiﬁed with

phosphoric acid and autoclaved prior to application. Animal feed and drinking water were applied ad libitum. To protect mice from

noise, volume did not exceed 60 dB. The animal facility’s rooms are dedicated facilities for animal maintenance and access is limited

to fulﬁl the appropriate hygiene standards. Hygiene monitoring was performed by an external laboratory on a quarterly basis in accor-

dance with the FELASA guidelines. Prior to the experiments, mice were randomized to experimental groups. For the dissemination

experiment, a total number of 20 mice was used, for the dose ﬁnding test, a total number of 12 mice was used and for the ectopic

tumor growth experiment, a total number of 40 mice was used.

Cell lines and culture

RIL175 cells (female) were kindly provided by Prof. Simon Rothenfußer (Center of Integrated Protein Science Munich (CIPS-M) and

€

Division of Clinical Pharmacology, Department of Internal Medicine IV, Klinikum der Universitat Munchen, Munich, Germany) and

they were described by Orci et al. (Orci et al., 2018). Vincristine-resistant (VCR-R) CEM cells (female) were a kind gift from Prof. Maria

Kavallaris (University of New South Wales, Australia). VCR-R CEM cells that are mainly resistant to vincristine, but also cross-resistant

to a variety of other antitumor agents, were generated and described by Haber et al. (Haber et al., 1989). HepG2 cells (male) were

obtained from German Research Centre of Biological Material (DSMZ). HCT-15 cells (male) and Hep3B cells (male) were purchased

from ATCC. HEK293 cells (female) were purchased from DSMZ (ACC 305). RIL175 and HepG2 cells were cultivated with DMEM (PAN

Biotech), HCT-15 and VCR-R CEM cells were cultivated with RPMI 1640 (PAN Biotech), supplemented with 10% fetal calf serum

(FCS) (PAA Laboratories). Hep3B cells were cultured in MEM Eagle (PAN Biotech), supplemented with 10% FCS (PAA Laboratories).

Human umbilical vein endothelial cells (HUVECs, female) were purchased from Promocell and cultivated with ECGM Kit enhanced

(PELO Biotech) supplemented with 10% (FCS) (PAA Laboratories) and 1% penicillin/streptomycin/amphotericin B (all purchased

from PAN Biotech). HUVECs were cultured for a maximum of six passages. HEK293 cells were cultured in DMEM (Gibco), supple-

mented with 10% FCS, 100 U/mL penicillin and 0.1 mg/mL streptomycin. HepaRG^TMcells were purchased from Life Technologies

and they were plated and maintained in Williams’ medium E supplemented with GlutaMAX and HepaRG Thaw, Plate & General Pur-

pose Medium Supplement (Thaw, Plate, & General Purpose Working Medium) (all purchased from Life Technologies). All cells were

cultured at 37 ^ꢀC with 5% CO₂with constant humidity. None of the cell lines used are listed in the database of commonly misidentiﬁed

cell lines maintained by ICLAC. All cells are proven to be mycoplasma-free on a quarterly basis.

METHOD DETAILS

Endolysosomal patch-clamp method

For conﬁrmation of loss of channel function in RIL175 cells, whole endolysosomal patch-clamp recordings were conducted using

modiﬁed conventional patch clamp. RIL175 cells were plated onto collagen G (0.001% in PBS, Biochrom)-coated glass coverslips,

grown overnight and subsequently treated with YM201636 (800 nM) for at least 2 h to enlarge endolysosomes. Intact endolysosomes

were isolated and current recordings and data analysis were performed as described before (Chen et al., 2017). In detail, currents

were recorded with an EPC-10 patch-clamp ampliﬁer (HEKA, Lambrecht, Germany) and PatchMaster acquisition software

e5 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

(HEKA). Data were digitalized at 40 kHz and ﬁltered at 2.8 kHz. Recording pipettes had a resistance of 8-10 MU and liquid junction

potential was corrected. Pipette solution (corresponding to luminal endolysosomal solution) contained 140 mmol/L Na-MSA, 5 mmol/

L potassium K-MSA, 2 mmol/L Ca-MSA, 1 mmol/L CaCl₂, 10 mmol/L HEPES and 10 mmol/L MES, pH 4.6. Bath solution (correspond-

ing to cytosolic solution) contained 140 mmol/L K-MSA, 5 mmol/L KOH, 4 mmol/L NaCl, 0.39 mmol/L CaCl₂, 1 mmol/L EGTA, and

10 mmol/L HEPES, pH7.2. PI(3,5)P₂was used in a water-soluble diC8 form (A.G. Scientiﬁc). pH values of bath solution and

pipette solution were adjusted with KOH and MSA, respectively. 500-ms voltage ramps from -100 to +100 mV were applied every

5 s, holding potential at 0 mV. For determination of the inhibitory effect of benzylisoquinolines on TPC2, TPC1 and TRPML1,

TPC2-YFP (Chao et al., 2017), TPC1-RFP (Sakurai et al., 2015) or TRPML1-YFP (Grimm et al., 2010) were transiently transfected

into HEK293 cells that were previously plated onto poly-L-lysine (Sigma Aldrich)-coated glass cover slips (overnight) using TurboFect

Transfection Reagent (Thermo Fisher) according to the manufacturer’s instructions. Isolation of intact endolysosomes from HEK293

cells after vacuolin treatment for at least 2 h, current recordings and data analysis were performed as described for the conﬁrmation

of loss of function in RIL175 cells with the exception of TPC1 measurements. For TPC1 measurements, a one-step protocol was

applied (+140 mV over 2 s, holding potential of -70 mV) and the cytoplasmic solution contained 140 mM Na-gluconate, 5 mM

NaOH, 4 mM KCl, 2 mM MgCl₂, 0.39 mM CaCl₂, 1 mM EGTA and 10 mM HEPES (pH 7.2). The luminal solution contained

140 mM Na-MSA, 5 mM K-MSA, 2 mM Ca-MSA, 1 mM CaCl₂, 10 mM HEPES and 10 mM MES (pH was adjusted to 4.6 with

MSA) . 10 mM of each inhibitor and 1 mM of PI(3,5)P₂or ML-SA1 were applied. Inhibition percentage was calculated as follows

(Equation 1):

current ðPIð3; 5ÞP2 + inhibitor

inhibition percentage ð%Þ = 100% ꢁ

3 100%:

(Equation 1)

current ðPIð3; 5ÞP2Þ

Fura-2 Ca²⁺imaging

Single cell Ca²⁺imaging experiments were performed using Fura-2. HEK293 stably expressing TPC2^L11A/L12A–RFP (targeted to the

plasma membrane by mutation/deletion of the endo-lysosomal targeting motifs) (Gerndt et al., 2020) or TRPML1^LLAA-GFP (Yamagu-

chi et al., 2011) were cultured at 37^ꢀC with 5% of CO₂in Dulbecco’s modiﬁed Eagle medium (Thermo Fisher), supplemented with

10% fetal bovine serum, 100 U/mL penicillin, and 0.1 mg/mL streptomycin. Cells were plated onto poly-L-lysine (Sigma Aldrich)-

coated glass coverslips and grown for 2-3 days. For Ca²⁺imaging experiments, cells were loaded for 1 h at room temperature

with Fura-2 AM (4.0 mM) and 0.005% (v/v) pluronic acid (both from Thermo Fisher) in HEPES-buffered solution 1 (HBS1) comprising

138 mM NaCl, 6 mM KCl, 2 mM MgCl₂, 2 mM CaCl₂, 10 mM HEPES, and 5.5 mM D-glucose (adjusted to pH 7.4 with NaOH). After

loading, cells were washed in HBS, and mounted in an imaging chamber. All recordings were performed in HBS1. Calcium imaging

was performed using a Leica DMi8 live cell microscope. Fura-2 was excited at 340 nm/380 nm. Emitted ﬂuorescence was captured

using 515 nm long-pass ﬁlter. Compounds were pre diluted in DMSO and stored as 10 mM stock solutions not exceeding three

months. Working solutions were prepared directly before using by dilution with HBS.

Fluo-4 Ca²⁺imaging

Fluo-4 based calcium imaging experiments were performed using a custom-made ﬂuorescence imaging plate reader built into a ro-

L11A/L12A

€

botic liquid handling station (Freedom Evo 150, Tecan, Mannedorf, Switzerland). HEK293 stably expressing TPC2

–RFP

(Gerndt et al., 2020) were cultured at 37^ꢀC with 5% of CO₂in Dulbecco’s modiﬁed Eagle medium (Thermo Fisher), supplemented

with 10% fetal calf serum (Biochrom, Berlin, Germany), 2 mM L-glutamine, 100 U/mL penicillin, 0.1 mg/mL streptomycin, and

400-800 mg/mL G418. Cells were seeded on black-walled, clear-bottom 384-well plates (Greiner, Germany) and incubated with

Fluo-4 AM (4 mM; Life Technologies, Eugene, OR) for 30 minutes at 37^ꢀC, washed and resuspended in HEPES-buffered solution 2

(HBS2) comprising 132 mM NaCl, 6 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂, 10 mM HEPES, and 5.5 mM D-glucose (adjusted to pH

7.4 with NaOH). For concentration-effect experiments cells were sequentially stimulated with different concentrations of inhibitor

(0.1 to 100 mM) and the TPC2 activators TPC2-A1-N and TPC2-A1-P (Gerndt et al., 2020) (10 mM each). Concentration-effect relation-

ships were plotted using GraphPad Prism 5 and ﬁtted to the Hill equation.

Generation of a TPC2 knockout cell line

The knockout of TPCN2 in murine RIL175 cells with the CRISPR-Cas9 system was conducted as described earlier (Bauer et al., 2015;

Ran et al., 2013). As a strategy, we deleted exon 2 of TPCN2 after having sequenced the region of interest (TPC2 sequencing primer:

5’-CCACATTTACCTCATGACATCC-3’). Then, single guide RNAs (sgRNAs) were designed as described previously (Haeussler et al.,

2016). Three sgRNAs upstream (5’) of exon 2 each were used for further experiments and cloned into the pSpCas9(BB)-2A-Puro

plasmid (pSpCas9(BB)-2A-Puro (PX459) V2.0 was a gift from Feng Zhang (Addgene plasmid # 62988; http://n2t.net/addgene:62988;

RRID:Addgene_62988, Ran et al., 2013) using the T4 DNA ligase protocol as indicated by the manufacturer (Thermo Fisher). Similarly,

¨

two sgRNAs downstream (3’) of exon 2 were cloned into the eSpCas9_2A_Blasti plasmid (provided by Lars Konig, Klinikum der Uni-

€

versitat Munchen, Munich, Germany). All constructs were transformed into competent DH5a-E.coli, then plasmids were prepared

using the QIAGEN Plasmid Maxiprep Kit according to the manufacturer’s instructions. Correct insertions were conﬁrmed by Sanger

sequencing starting from the U6 promotor. Subsequently, RIL175 WT cells were transfected with three pairs of sgRNAs using Lip-

ofectamine^TM3000 (Invitrogen) transfection reagent according to the manufacturer’s instructions. Selection was performed with pu-

romycin (2 mg/mL) and blasticidin S (6 mg/mL). By applying the T7 DNA Endonuclease I assay (New England Biolabs), the sgRNA pair

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e6

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

with the highest editing efﬁciency was determined (sgRNA pair 1, see Figure S1A, sgRNA sequences are listed in Table S4) and used

for subsequent single cell dilution and clonal expansion.

Successful exon 2 deletion was conﬁrmed by standard PCR (Thermo Scientiﬁc Phusion Green Hot Start II High-Fidelity Polymer-

ase, Thermo Fisher), agarose gel analysis (peqGOLD agarose, universal, VWR International) and Sanger sequencing (sequences of

sequencing primers are listed in Table S4). To determine sequences of individual alleles of several clones, amplicons were PCR-

ampliﬁed (Thermo Scientiﬁc Phusion Green Hot Start II High-Fidelity Polymerase, Thermo Fisher; primer pair: TPC2 spanning ex2)

and puriﬁed with the QIAquick Gel Extraction Kit (QIAGEN) without previous separation of PCR products on an agarose gel. Subse-

quently, puriﬁed PCR products were analyzed by Sanger sequencing (sequences of sequencing primers are listed in Table S4). To

deconvolute overlapping spectra from different alleles, CRISPR-ID (http://crispid.gbiomed.kuleuven.be/) was used as indicated by

the Dehairs and coworkers (Dehairs et al., 2016). For the clone 1D1, further referred to as RIL175 TPC2 KO, the ten predicted off-

target sequences present in exons or introns with the highest cutting frequency determination (CFD) scores according to the

CRISPOR TEFOR in silico tool (Haeussler et al., 2016) for each sgRNA were ampliﬁed by PCR (Thermo Scientiﬁc Phusion Green

Hot Start II High-Fidelity Polymerase, Thermo Fisher or Q5ꢁ High-Fidelity DNA Polymerase, New England Biolabs), analyzed by

Sanger sequencing and compared with the WT control. The investigated off-targets, sequences and the used primer pairs and poly-

merases in Table S1. Cloning oligos and sequencing services were provided by Euroﬁns Genomics. Primers were purchased from

Metabion.

Quantitative real-time PCR

Isolation of mRNA (QIAGEN RNeasy Mini Kit), synthesis of cDNA templates (High Capacity cDNA Reverse Transcription Kit, Applied

Biosystems) and quantitative real-time PCR (ABI 7300 Real-Time PCR System, Applied Biosystems) with actin serving as house-

keeping gene were performed as described previously (Ardelt et al., 2018). Primers were purchased from Metabion. Sequences

of primers used for qPCR analysis are listed in Table S4).

After quantitative real-time PCR, some PCR products were analyzed on an agarose gel (peqGOLD agarose, universal, VWR Inter-

national) as indicated. GeneRuler 1 kb Plus DNA Ladder (Thermo Fisher) was used as molecular size marker. GelRed (Biotium) was

used to detected DNA.

Wound healing assay

2.5 3 10⁴RIL175 WT and TPC2 KO cells were seeded into the wells of a 96-well plate. A wound was scratched into the cell monolayer

with a 96-well scratching device when cells reached conﬂuency. PBS (containing Mg⁺and Ca⁺) was used to remove detached cells

and each scratch was examined under a light microscope. After incubation at 37^ꢀC for 18 h, cells were stained with crystal violet,

which was performed by aspiration of medium, washing with PBS, ﬁxation with methanol for 10 min and addition of crystal violet

solution for min. Crystal violet solution was carefully removed and washed with water. The plate was dried overnight before images

were taken with a LeicaDmil microscope with 4x magniﬁcation. Cells that were incubated with medium without FCS served as nega-

tive control. Wound closure was quantiﬁed with ImageJ software (NIH) and it was calculated with Equation 2:

wound area ð18 hÞ

Wound closure ð%Þ = 100% ꢁ

3 100%:

(Equation 2)

wound area ð0 hÞ

Proliferation assays

General

Cellular proliferation of RIL175 WT and TPC2 KO cells was assessed by impedance measurement and by CellTrace^TMCFSE staining.

Cellular proliferation of HepaRG^TMand Hep3B cells was assessed by impedance measurement. Antiproliferative effects of the com-

pounds from the library were determined by CellTiter-Blueꢁ assay.

Impedance measurement

With the xCELLigence RTCA device (ACEA Biosciences), the cell index, a dimensionless parameter that is proportional to the cell

number, can be determined through impedance measurement. RIL175 WT and TPC2 KO cells were seeded at a density of 2 3

10³cells per well into an equilibrated 16-well E-plate. Hep3B and HepaRG^TMcells were seeded at a density of 1 3 10⁴cells per

well into an equilibrated 16-well E-plate. 24 h after seeding, Hep3B and HepaRG^TMcells were transfected with non-targeting or

TPC2 siRNA (Santa Cruz Biotechnology, sc-96434) using Lipofectamine^TMRNAiMAX Transfection Reagent (Thermo Fisher) as

described by the manufacturer. Slopes were calculated using the xCELLigence RTCA software (ACEA Biosciences) for each cell

line until reaching the plateau phase using Equation 3:

ꢀ ꢁ

1

Cell Index ꢁ intercept

time ðhÞ

Slope

=

:

(Equation 3)

h

CellTrace^TMCFSE staining

Additionally, the CellTrace^TMCFSE Cell Proliferation Kit (Invitrogen) was used to assess proliferation visually. It relies on a ﬂuorescent

CellTrace^TMreagent that diffuses into cells and binds covalently to intracellular amines. Cell proliferation can be followed by ﬂow cy-

tometry for several generations. Labelling of the cells was conducted as indicated by the manufacturer using 10 mM CFSE. CFSE

ﬂuorescence was analyzed directly after labelling and after a 72 h incubation time by ﬂow cytometry on a BD FACS Canto II (BD

e7 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Biosciences). Forward versus side scatter (FSC vs SSC) gating was used to exclude cell debris using FlowJo 7.6 (BD Biosciences).

CFSE ﬂuorescence was analyzed visually in the FITC-A channel as described by the manufacturer.

Compound screening by CellTiter-Blueꢁ assay

Antiproliferative effects of compounds from the benzylisoquinoline library were determined by CellTiter-Blueꢁ cell viability assay

(Promega). Adherent cells were seeded at the indicated cell densities (RIL175: 2 3 10³cells per well, HepG2: 5 3 10³cells per

well, HCT-15: 3 3 10³cells per well) into wells of a 96-well plate and allowed to adhere overnight. 4 h after seeding, initial metabolic

activity was determined and used as zero value. Treatment was performed for 72 h at the indicated concentrations. VCR-R CEM cells

were seeded at a density of 20 3 10³cells per well of a 96-well plate, incubated for 4 h before stimulation with the indicated concen-

trations of compounds for 48 h. 2 h before the end of stimulation time, CellTiter-Blueꢁ reagent was added and ﬂuorescence at

590 nm was detected with a Sunrise ELISA reader (Tecan). Half-maximal inhibitory concentrations (IC₅₀) values were calculated

by nonlinear regression using GraphPad Prism 8.4.0 software.

Colony formation

Hep3B cells were seeded at a density of 0.1 3 10⁶cells per well of a 6-well plate and incubated for 24 h prior to transfection with non-

targeting (Santa Cruz Biotechnology, sc-37007) or TPC2 siRNA (Santa Cruz Biotechnology, sc-96434) using Lipofectamine^TM

RNAiMAX Transfection Reagent (Thermo Fisher) as described by the manufacturer. Following another incubation step for 24 h, cells

were seeded at a density of 5 3 10³cells per well of a 6-well plate and incubated for 7 d. Cells were stained with crystal violet solution

for 10 min. Excess crystal violet solution was removed with water and the plate was dried overnight. Subsequently, crystal violet was

extracted with trisodium citrate buffer (50 mM in 50/50 (V/V) ethanol/water) and absorption was detected with a Sunrise ELISA

reader (Tecan).

In vivo experiments

Dissemination assay

20 C57Bl/6-Tyr mice (Envigo), female, 6 weeks old, were used for intravenous injection of either 2 3 10⁵RIL175 WT or TPC2 KO cells

into the tail vein. Bioluminescence imaging using the IVIS Lumina system (PerkinElmer) was conducted on day 3 after intraperitoneal

injection of 6 mg/mL luciferin per mouse. Previously, mice were put under anesthesia with 3% isoﬂurane in oxygen. Imaging of mice

was performed in ventrodorsal position and mice were kept under narcosis with 2% isoﬂurane in oxygen. Hypothermia was pre-

vented by a heating plate (37^ꢀC). The tumor signal per deﬁned region of interest was calculated as photons/second/cm (total ﬂux/

area) using the Living Image 4.4 software (Perkin Elmer)

Dose-ﬁnding test

Initially, for determination of the doses of SG-005 and SG-094, 12 female C57Bl/6-Tyr mice (Envigo), 6 weeks old, were used. SG-005

was applied at 38 nmol/kg and 57 nmol/kg and SG-094 at 59 nmol/kg and 90 nmol/kg daily on three consecutive days. Compounds

were dissolved in a solvent consisting of 5% DMSO, 10% solutol and 85% PBS and they were injected intraperitoneally (100 ml in-

jection volume). Due to the known toxicity problems of tetrandrine in the literature, the compound was not used in vivo

Ectoptic tumor model

40 C57Bl/6-Tyr mice (Envigo), female, 5 weeks old, were injected with either 2 3 10⁶RIL175 WT or TPC2 KO cells into the ﬂank. Treat-

ment of mice with vehicle control, SG-005 (57 nmol/kg) or SG-094 (90 nmol/kg) was performed on day 0, 2, 4, 7 and 9 after implan-

tation of cells. Compounds were dissolved as described for the dose-ﬁnding test. Tumor volumes were measured by caliper on day 4,

7, 9 and 10 and calculated with the formula p/6 3 L 3 W 3 H. Bioluminescence imaging was performed on day 2, 4, 7 and 10 after

implantation of cells following intraperitoneal injection of 6 mg/mL luciferin per mouse. Anesthesia, narcosis and calculation of tumor

signal per deﬁned region were performed as described for the dissemination assay.

Seahorse extracellular ﬂux analysis

Preparation of reagents

Assay medium was prepared by supplementing DMEM 5030 powder (Sigma Aldrich) with L-Glutamine (4 mM) and NaCl (143 mM),

followed by adjustment of pH to 7.35 0.05 and sterile-ﬁltration through a Millipore Expressꢁ PLUS membrane ﬁlter (0.22 mM, Merck

Millipore). Poly-D-lysine coating solution was prepared by diluting 5 mg poly-D-lysine hydrobromide (Sigma Aldrich) in 50 mL ster-

ile water

Metabolic proﬁling

Before cell seeding, the XF^e96 microplate (Agilent Technologies) was coated by adding 15 ml poly-D-lysine coating solution per well,

followed by 5 min incubation at room temperature. Subsequently, poly-D-lysine coating solution was removed and wells were rinsed

twice with sterile water. The XF^e96 microplate was dried for 2 h prior to adding collagen G coating solution (0.001% in PBS, Bio-

ꢀ

chrom). The plate was incubated with collagen G coating solution for 30 min at 37 C prior to seeding of cells. For comparison of

RIL175 WT and TPC2 KO cells, cells were seeded at a density of 1 3 10⁴cells per well of the XF^e96 microplate plate and allowed

to adhere overnight. For determining the inﬂuence of TPC2 inhibitors, RIL175 WT cells were seeded at a density of 2 3 10⁴cells

per well and allowed to adhere for at least 6 h. Cells were either pre-treated for 24 h in culture medium or for 1 h in assay medium

without glucose as indicated prior to the assay. The Seahorse XF^e96 sensor cartridge was hydrated according to the manufacturer’s

instructions. Extracellular ﬂux analysis was performed as indicated by the manufacturer on a Seahorse the XF^e96 Analyzer (Agilent

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e8

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Technologies). Data were analyzed with Wave 2.6.0 software (Agilent Technologies). For data evaluation, OCR/ECAR (oxygen con-

sumption rate/extracellular acidiﬁcation rate) ratios were calculated as described before (Zhang et al., 2012).

Cellular uptake assay

Pharmakokinetics of tetrandrine, SG-005 and SG-094 were determined by HPLC-DAD. RIL175 WT and HUVECs were incubated with

these compounds at 2 and 10 mM. RIL175 WT and HUVEC cells were seeded at a density of 0.3 3 10⁶cells per well of a 6-well plate

and allowed to adhere overnight. Cells were treated with 2 and 10 mM of the respective compounds for 75 min. Before trypsinization,

cells were washed twice with PBS (2 mL). For normalization, cells were counted using a ViCell XR cell counter (Beckman). After har-

vesting, cell pellets were centrifuged (Hettich universal 320, 13,000 rpm, 5 min, 4^ꢀC), washed with ice-cold PBS (1 mL), centrifuged

(Hettich universal 320, 13,000 rpm, 5 min, 4^ꢀC) and resuspended in 50 ml acetonitrile. For cell lysis by bead beating, glass beads (acid-

washed, Sigma Aldrich) were added and cell lysis was conducted by shaking on an orbital shaker at 4 ^ꢀC for at least 30 min. Reaction

mixtures were centrifuged again (Hettich universal 320, 13,000 rpm, 5 min, 4 ^ꢀC) and the supernatants (40 ml) were subjected to

analytical HPLC analysis. The supernatant was analyzed by HPLC-DAD using an Agilent Series 1100 HPLC system (Agilent Technol-

ogies) consisting of a quaternary pump system (G1311 A QuatPump), an autosampler (G1329 A ALS), a column oven (G1316 A Col-

Comp) and a UV-DAD detector (G1315 A DAD). Chromatographic separation was carried out with an Agilent Zorbax Eclipse Plus C18

(150 3 4.6 mm, i.d. 5.0 mm) column (Agilent Technologies) and a mobile phase of methanol and water (adjusted with 1 M sodium

hydroxide solution to pH 9) 80:20 (v/v). The total run time was 18 min with an isocratic ﬂow rate at 1.2 mL/min, and an injection volume

of 10 ml. The column oven was set at 30 ^ꢀC. The UV detection wavelength was set at 210 nm. Data analysis and instrument control was

carried out with Agilent ChemStationꢁ software Rev. B04.02 (Agilent Technologies). The retention time of tetrandrine was 4.5 min, of

SG-005 was 10.4 min, and of SG-094 10.0 min. The concentration of each compound was determined according to an external stan-

dard calibration.

Lysosomal uptake assay

Lysosomal uptake of TPC2 inhibitors and control compounds was determined by LysoTracker-Red^TMDND-99 (Thermo Fisher) stain-

ing and ﬂow cytometry as described previously (Petersen et al., 2013). In detail, RIL175 WT, RIL175 TPC2 KO and HUVEC cells were

seeded at a density of 0.06 3 10⁶cells per well of a 24-well plate and allowed to adhere overnight. On the subsequent day, cells were

treated with 2 and 10 mM of the respective compounds (or with 10 nM of the V-ATPase inhibitor concanamycin, Santa Cruz Biotech-

nology) for 60 min. LysoTracker-Red^TMDND-99 (25 nM) was present during the last 15 min. Fluorescence was determined on a BD

FACS Canto II (BD Biosciences) and ﬂuorescence intensities were normalized to the DMSO control using FlowJo 7.6 (BD

Biosciences).

FLIM-FRET imaging

¨

A Forster resonance energy transfer (FRET)-based ﬂuorescent ATP probe, named GO-ATeam, was recently developed by Nakano et

and coworkers (Nakano et al., 2011) for ATP imaging in single living cells, in which the ε subunit of the bacterial F₀F₁-ATP synthase is

sandwiched between a green ﬂuorescent protein (GFP, cp173-mEGFP) and an orange ﬂuorescent protein (OFP, mKOk) that are used

as a FRET pair (Nakano et al., 2011). Speciﬁc binding of ATP to the ε subunit 13 in GO-ATeam induces the conformational change of

the ε subunit from extended to retracted form, and thereby FRET efﬁciency in GO-ATeam increases, enabling the monitoring of ATP

concentrations in living cells at a subcellular spatial resolution (Nakano et al., 2011). The use of the current GO-ATeam probe (Nakano

et al., 2011) offers several advantages over the originally developed ATeams (Imamura et al., 2009). Speciﬁcally, the signal is more

stable against acidiﬁcation, which allows ATP imaging inside acidic intracellular compartments (Nakano et al., 2011). Also, the GO-

ATeam longer excitation wavelength is much less phototoxic to cells, making the probe suitable for longer times of observation (Na-

kano et al., 2011). The dynamics of intracellular ATP levels in RIL175 hepatocellular carcinoma cells transfected with the GO-ATeam

probe were successfully visualized. In these experiments, higher ATP levels correspond to higher FRET efﬁciency as deﬁned by the

differences in the lifetime of the GO-ATeam probe. Furthermore, sequential false colour Heatmap images for each time point showing

the distribution of ATP, via FRET efﬁciency, inside the cells were generated. Speciﬁcally, two separate data sets comparing the dif-

ferences in ATP levels were collected, one between RIL175 WT and RIL175 TPC2 KO cells and the other between SG-005 pre-treated

or non-treated RIL175 WT cells. For that purpose, RIL175 WT and TPC2 KO cells were seeded at different densities ranging from 0.5

to 2 3 10⁴cells per well of ibidi 8 well m-slides with glass bottoms (ibidi) and incubated overnight. Before imaging, cells were sup-

plemented with new DMEM (PAN Biotech), supplemented with 10% fetal calf serum (FCS) (PAA Laboratories). To determine the in-

ﬂuence of SG-005 on ATP levels, RIL175 WT cells were treated with SG-005 (2 mM) or vehicle for 1 h prior to imaging. For the com-

parison of RIL175 WT and TPC2 KO cells, cells were left untreated. FLIM (ﬂuorescence lifetime imaging)-FRET measurements were

performed on a Leica TCS SP8 SMD microscope combined with a Picoquant LSM Upgrade Kit using a 470 pulse laser at 40 MHz

frequency and a time-correlated single-photon counting (TCSPC) approach (SI). For all measurements, a 633 Zeiss water immersion

lens was used. A complete FLIM image of single cells was recorded until an average of 1000 photon counts per pixel was collected.

The ﬂuorescence decay curve was calculated from the recorded FLIM image that was then analyzed by deconvolution ﬁt using the

instrument response function (IRF). This process was repeated for twelve different timepoints (0, 10, 20, 30, 35, 40, 45, 60, 75, 90, 105

and 120 min) with 5 mM of 2-deoxy-D-glucose (2-DG) (Sigma Aldrich) being added after the 30 min timepoint measurement. Data

acquisition and analysis was performed using the Picoquant SymPhoTime 64 (V2.1). In control experiments, the ﬂuorescence decay

time of the donor in absence of the acceptor (t_D) was determined by applying a single-exponential decay function. For the FRET

e9 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

samples, a bi-exponential ﬁt function was assumed. Here, the ﬁrst time component was ﬁxed according to t_Dpreviously obtained

from the control experiments and reﬂecting the non-interacting donor fraction and the second time component t_DAwas freely ﬁtted

reﬂecting the donor fraction interacting with the acceptor. The quality of the ﬁt was assessed by randomly distributed residuals and

by low c_Rvalues. From the interacting fraction, the FRET efﬁciency (E) was calculated according to Equation 4:

tDA

E = 1 ꢁ

(Equation 4)

tD

Toxicity assays

General

For assessing the toxicity of tetrandrine, SG-005 and SG-094, several cellular models and approaches were used: HepaRG^TMpro-

genitor cells, differentiated HepaRG^TMcells and peripheral blood mononuclear cells (PBMCs).

HepaRG^TMprogenitor cells

HepaRG^TMcells, cultivated in Thaw, Plate, & General Purpose Working Medium, were seeded at a density of 10 3 10³cells per well of

a 96-well plate and allowed to adhere overnight. After 24 h treatment with the respective compound concentrations, cell viability was

determined by CellTiter-Blueꢁ cell viability assay.

Differentiated HepaRG^TMcells

HepaRG^TMcells were differentiated in Williams’ medium E supplemented with GlutaMAX and HepaRG Maintenance/Metabolism

Medium Supplement (Maintenance/Metabolism Working Medium) (all purchased from Life Technologies) as indicated by the manu-

facturer. Cell viability was determined as described in the previous section

PBMCs

PBMCs were purchased from ATCC and cultivated in RPMI 1640 supplemented with 20% FCS and 1% penicillin/streptomycin (all

purchased from PAN Biotech). 4 hours after seeding, cells were treated as indicated for 48 h. Cell death was analyzed by propidium

iodide (5 mg/mL in PBS; Carl Roth) staining and ﬂow cytometry using a BD FACS Canto II (BD Biosciences). For evaluation, cell debris

was excluded and propidium iodide positive cells were determined using FlowJo 7.6 (BD Biosciences) as shown in Figure S7E.

Western blot analysis and antibodies

Human umbilical vein endothelial cells (HUVECs) were seeded at a density of 0.5 3 10⁶cells per well of a 6-well plate and allowed to

adhere overnight. After serum starvation for 4 h using ECGM medium (PELO Biotech) without supplements, cells were treated with

vehicle, tetrandrine, SG005 or SG-094 (10 mM) for 1 h. Subsequently, cells were either stimulated with VEGF-A₁₆₅(0.1 mg/mL) (Pe-

protech) or left untreated for 15 min. Cell lysis was performed with RIPA (radioimmunoprecipiation) buffer (50 mM Tris/HCl, 150 mM

NaCl, 1 % Nonidet NP-40, 0.25% Sodium deoxycholate, 0.10% SDS, 4 mM cOmpleteꢁ EDTA-free from Roche Diagnostics, 0.5 mM

PMSF, 2 mM Na₃VO₄). Proteins were separated by SDS-PAGE, transferred to Hybond PVDF (polyvinylidene diﬂuoride) membranes

(Amersham Bioscience) and incubated with primary antibodies. Proteins were visualized using horseradish peroxidase coupled sec-

ondary antibodies and ECL solution containing 2.5 mM luminol. Chemiluminescence was detected on a Chemidoc^TMTouch Imaging

System (Bio-Rad). The following antibodies were used: phospho-eNOS (Ser1177) (1:1000, Cell Signaling Technologies, #9571),

phospho-p54/p46 SAPK/JNK (Thr183/Tyr185) (1:1000, Cell Signaling Technologies, #9251), phospho-p44/p42 MAPK (Thr202/

Tyr204) (1:1000, Cell Signaling Technologies, #9106), phospho-Akt (Ser473) (1:1000, Cell Signaling Technologies, #9271), p44/

p42 MAPK (1:1000, Cell Signaling Technologies, #9102), SAPK/JNK (1:1000, Cell Signaling Technologies, #9252), NOS3 (1:1000,

Santa Cruz Biotechnology, sc-654), Akt (1:1000, Cell Signaling Technologies, #9272). The stain free technology (Bio-Rad) served

as loading control and enabled normalization of protein bands to whole lane protein (Colella et al., 2012) using Image Lab 6.0.1

(Bio-Rad).

Apoptosis assay

To determine apoptosis, RIL175 WT cells were seeded at a density of 0.05 3 10⁶cells per well of a 24-well plate and allowed to

adhere overnight. Cells were with the respective compounds for 48 h at the indicated concentrations. For staining with propidium

iodide, cells were detached and stained with hypotonic ﬂuorochrome solution (propidium iodide, 50 mg/mL in 0.1% sodium citrate

(w/v), 0.1% Triton X-100 (V/V) in deionized water) as described previously (Riccardi and Nicoletti, 2006) for at least 1 h in the dark.

Forward versus side scatter (FSC vs SSC) gating was used to exclude cell debris as shown below for determination of mitochondrial

mass using FlowJo 7.6 (BD Biosciences). Propidium iodide ﬂuorescence was analyzed on a BD FACS Canto II (BD Biosciences).

Mitochondrial mass

To determine mitochondrial mass, RIL175 WT cells were seeded at a density of 0.1 3 10⁶cells per well of a 24-well plate and allowed

to adhere overnight. Cells were either with the respective compounds for 1 h or for 24 h at the indicated concentrations. After detach-

ment, cells were incubated with 250 ml MitoTrackerꢀ Green FM (Thermo Fisher) in DMEM (PAN Biotech) without supplements for

3 min at 37 ^ꢀC as indicated by the manufacturer. Subsequently, medium was discarded and cells were resuspended in 250 ml

PBS and analyzed by ﬂow cytometry. Forward versus side scatter (FSC vs SSC) gating was used to exclude cell debris as shown

below using FlowJo 7.6 (BD Biosciences). MitoTrackerꢀ Green ﬂuorescence was analyzed in the Alexa Fluor 488-A channel as

shown in in Figure S7F on a BD FACS Canto II (BD Biosciences).

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e10

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Mitochondrial membrane potential

The cationic dye JC-1 (Thermo Fisher) was used to monitor mitochondrial health and, more speciﬁcally, to measure mitochondrial

membrane potential (DJm). When DJm is intact, JC-1 accumulates in mitochondria, yielding a red ﬂuorescence. In contrast, mito-

chondrial depolarization is indicated by a decrease in the red/green ﬂuorescence intensity. RIL175 WT cells were seeded at a density

of 0.1 3 10⁶cells per well of a 24-well plate and allowed to adhere for at least 6 h. Subsequently, cells were treated for 1 h or 24 h with

the indicated concentrations of the respective compounds. Then, JC-1 (1 mg/mL) was added and incubated for 1 h. For 1 h treatment

with compounds, staining with JC-1 was performed simultaneously. Cells were harvested, washed with PBS and resuspended in

PBS for ﬂow cytometric analysis on a BD FACS Canto II (BD Biosciences). In parallel, compensation samples were prepared using

Anti-Mouse Ig, k/Negative Control (FBS) Compensation Particles Set (BD Biosciences) and a BD PE Mouse IgG1, k/ Isotype Control

(BD Biosciences #555749) and a BD Alexa Fluorꢁ 488 Mouse IgG1 k Isotype Control (BD Biosciences #557721) as described by the

manufacturer. Prior to analysis of cellular samples, compensation of spectral overlap was performed on a BD FACS Canto II (BD Bio-

sciences) as indicated below. The percentage of PE-A positive populations (orange) and of Alexa-Fluor-488-A positive populations

was determined using FlowJo 7.6 (BD Biosciences) as shown in Figure S7G.

Chemistry

Materials and instruments

All chemicals used were of analytical grade and were obtained from abcr (Karlsruhe, Germany), Fischer Scientiﬁc (Schwerte, Ger-

many), Sigma-Aldrich (now Merck, Darmstadt, Germany), TCI (Eschborn, Germany) or Th. Geyer (Renningen, Germany). HPLC grade

and dry solvents were purchased from VWR (Darmstadt, Germany) or Sigma-Aldrich, all other solvents were puriﬁed by distillation. All

reactions were monitored by thin-layer chromatography (TLC) using pre-coated plastic sheets POLYGRAMꢁ SIL G/UV254 from Ma-

€

cherey-Nagel (Duren, Germany) and detected by irradiation with UV light (254 nm or 366 nm). Furthermore, reactions were monitored

by atmospheric pressure solids analysis probe (ASAP) via atmospheric-pressure chemical ionization (APCI) on an expression^LCMS

device (Advion, Ithaca, USA). Flash column chromatography (FCC) was performed on Merck silica gel Si 60 (0.015 – 0.040 mm). Pre-

parative and semipreparative (chiral) HPLC was performed on a Shimadzu HPLC system consisting of a LC-20AP solvent delivery

module, a CTO-20A column oven, a SPD-M20A photodiode array UV/vis detector and a CBM-20A system controller using a semi-

preparative CHIRALPAKꢁ IC column (particle size 5 mm, Diacel) or a preparative NUCLEODURꢁ 100-5 column (particle size 5 mm,

Macherey-Nagel). NMR spectra (¹H, ¹³C, DEPT, COSY, HMQC, HMBC) were recorded at 23 ^ꢀC on an Avance III 400 MHz Bruker

BioSpin or Avance III 500 MHz Bruker BioSpin instrument, unless otherwise speciﬁed. Chemical shifts d are stated in parts per million

(ppm) and are calibrated using residual protic solvents as an internal reference for proton (CDCl₃: d = 7.26 ppm, (CD₃)₂SO: d = 2.50

ppm, C₂D₂Cl₄: d = 5.91 ppm, CD₃OD: d = 3.31 ppm) and for carbon the central carbon resonance of the solvent (CDCl₃: d = 77.16

ppm, (CD₃)₂SO: d = 39.52 ppm, C₂D₂Cl₄: d = 74.20 ppm, CD₃OD: d = 49.00 ppm). Multiplicity is deﬁned as s = singlet, d = doublet,

t = triplet, q = quartet, m = multiplet. NMR spectra were analyzed with NMR software MestReNova, version 12.0.1-20560 (Mestrelab

Research S.L.). High resolution mass spectra were performed by the LMU Mass Spectrometry Service applying a Thermo Finnigan

MAT 95 or Joel MStation Sektorfeld instrument at a core temperature of 250 ^ꢀC and 70 eV for EI or a Thermo Finnigan LTQ FT Ultra

Fourier Transform Ion Cyclotron Resonance device at 250 ^ꢀC for ESI. IR spectra were recorded on a Perkin Elmer FT-IR Paragon 1000

instrument as neat materials. Absorption bands were reported in wave number (cm^-1) with ATR PRO450-S. Melting points were

€

determined by the open tube capillary method on a Buchi melting point B-540 apparatus and are uncorrected. HPLC purities

were determined using an HP Agilent 1100 HPLC with a diode array detector and an Agilent Poroshell column (120 EC-C18;

3.0 3 100 mm; 2.7 micron) with acetonitrile/water as eluent. Electronic circular dichroism (ECD) spectra were measured on a Jasco

J-810 Spectropolarimeter. Optical rotations were further measured at the given temperature (T in [^ꢀC]) on a Perkin Elmer 241 Polar-

imeter instrument using a sodium lamp (Na ^D-line, 589 nm). Measurements were carried out in a cell with path lengths (l) of 1.0 dm.

Concentrations are given in g/100 mL

General procedures for compound preparation

General Procedure A – Protection. Following a general procedure published by Schrittwieser et al. (Schrittwieser et al., 2011), the

ꢀ

appropriate amines or phenols (1.0 eq.) were dispersed in dry CH₂Cl₂and cooled to 0 C, followed by dropwise addition of NEt₃

(1.5-3.5 eq.) and ethyl chloroformate (1.3-2.5 eq.). The reaction mixture was stirred at rt for 4-72 h, before water was added. The

mixture was extracted thrice with CH₂Cl₂, the combined organic layers were dried over Na₂SO₄and concentrated in vacuo. Puriﬁ-

cation was accomplished by ﬂash column chromatography (FCC) to yield carbamates and carbonates.

General Procedure B – Wittig reaction. (Methoxymethyl)-triphenylphosphonium chloride (1.2 eq.) was suspended in dry THF and

cooled to -4 ^ꢀ_ꢀC, before LDA (20% in THF/ethylbenzene/heptane, 1.0-1.2 eq.) was added dropwise. The mixture was stirred for

20 min. at -4 C, followed by slow, dropwise addition of a solution of the appropriate aldehyde (1.0 eq.) in THF. The solution was

stirred for 16-50 h, then poured on water and the mixture extracted thrice with CH₂Cl₂. The combined organic layers were dried

over Na₂SO₄, ﬁltered and concentrated in vacuo. The crude product was puriﬁed by FCC to give desired cis/trans-enol ethers.

The products were directly used for the next step without further analysis.

General Procedure C – N-acyl Pictet-Spengler reaction. Carbamate (1.0 eq.) and enol ether (1.0-1.2 eq.) were dissolved in dry

CH₂Cl₂, 4 A molecular sieves (1.00 g per 20 mL solvent) was added and the mixture was cooled to 0 ^ꢀC, before TFA (10-11 eq.)

˚

was added dropwise. The mixture was allowed to warm to rt and stirred for 18-90 h, before the molecular sieves was removed by

e11 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

ﬁltration and sat. aq. NaHCO₃solution was added. The mixture was extracted thrice with CH₂Cl₂, the combined organic layers were

washed with 2M aq. HCl solution and brine, dried over Na₂SO₄, ﬁltered and concentrated in vacuo. Puriﬁcation was accomplished by

FCC to yield desired racemic tetrahydroisoquinolines.

General Procedure D – Lithium alanate reductions of nitrooleﬁns, carbamates and carbonates. According to Cava et al. (Cava and

Buck, 1969), LiAlH₄(7-12 eq.) was suspended in dry THF and a solution of carbamate (or other starting material) (1.0 eq.) in dry THF

was added dropwise and the suspension was reﬂuxed for 3-20 h. The mixture was allowed to cool to room temperature and Na₂SO₄x

10 H₂O was added in small portions under vigorous stirring over 20 minutes. Water was added and the mixture was extracted thrice

with EtOAc. The combined organic layers were extracted with 2M aq. HCl solution, the acidic phase was neutralized with NaHCO₃

and extracted thrice with EtOAc again. This organic layer was dried over Na₂SO₄and concentrated in vacuo to give tertiary amines if

not otherwise described.

General Procedure E – Chan-Evans-Lam coupling. According to Evans et al. (Evans et al., 1998), phenol (1.0 eq), Cu(OAc)₂(1.1 eq.

˚

per phenolic group), boronic acid (3.0 eq. per phenolic group) and 4 A molecular sieves (1.00 g per 40 mL solvent) were suspended in

dry CH₂Cl₂and a mixture of NEt₃and pyridine (1:1, 2.5 eq. each) was added dropwise. The reaction mixture was stirred for 13-22 h at

rt under ambient atmosphere, followed by ﬁltration through a celite pad. The organic ﬁltrate was washed with sat. aq. NaHCO₃so-

lution, 10% aq. citric acid solution and brine, dried over Na₂SO₄, ﬁltered and concentrated in vacuo. The crude products were puriﬁed

by FCC or HPLC to obtain desired diaryl ethers.

General Procedure F – Formation of hydrochloride salts. To receive solid and stable compounds for biological experiments, the

tertiary amines were dissolved in 1.25 M methanolic HCl solution (3.0 mL per 100 mg amine) and concentrated in vacuo to yield

the hydrochloride salts as solids. Melting points were obtained from the hydrochloride salts, remaining analytical data from the

free amines, unless stated otherwise.

Synthetic procedures for compound preparation

Synthetic procedures are described in the following. NMR spectra are listed in Data S2.

(E)-1-(Benzyloxy)-2-methoxy-4-(2-nitrovinyl)benzene (37).

Following a procedure published by Pouysegu et al. (Pouysegu et al., 2002) 4-benzyloxy-3-methoxybenzaldehyde (10.0 g,

41.3 mmol, 1.0 eq.) and ammonium acetate (2.80 g, 36.4 mmol, 0.90 eq.) were dissolved glacial acetic acid (24 mL) and nitromethane

(16.0 mL, 299 mmol, 7.4 eq.) was added under nitrogen atmosphere. The mixture was heated to reﬂux for 3 h. Water (5.0 mL) was

added, the precipitate was collected by ﬁltration and was washed with cold MeOH. The crude product was recrystallized from MeOH

to yield nitrooleﬁn 37 (7.08 g, 24.8 mmol, 60%) as yellow crystals. Analytical data are in accordance with literature (Pouysegu et al.,

2002). R_f= 0.46 (4:1 hexanes/acetone). m.p.: 123 ^ꢀC. [lit. (Pouysegu et al., 2002): 124-125 ^ꢀC]. ¹H NMR (400 MHz, CDCl₃) d/ppm =

7.95 (d, J = 13.6 Hz, 1H, CHCHNO₂), 7.51 (d, J = 13.6 Hz, 1H, CHCHNO₂), 7.44 – 7.32 (m, 5H, PhCH₂), 7.10 (dd, J = 8.4, 2.0 Hz, 1H, 5-

H), 7.02 (d, J = 2.0 Hz, 1H, 3-H), 6.92 (d, J = 8.4 Hz, 1H, 6-H), 5.22 (s, 2H, PhCH₂), 3.93 (s, 3H, OCH₃). ¹³C NMR (101 MHz, CDCl₃) d/

ppm = 152.1 (C-1), 150.2 (C-2), 139.4 (CHCHNO₂), 136.2 (qPh), 135.4 (CHCHNO₂), 128.9 (Ph), 128.4 (Ph), 127.4 (Ph), 124.5 (C-5),

123.2 (C-4), 113.6 (C-3), 111.0 (C-6), 71.0 (PhCH₂), 56.3 (OCH₃). IR (ATR) ṽ_max/cm^-1= 3112, 3048, 2980, 2358, 1629, 1491, 1258,

1224, 1142, 976, 805, 746, 697. HRMS (EI): calcd. for C₁₆H₁₅NO₄(M)^,+285.0996; found 285.0994. Purity (HPLC): > 96 %

(l = 210 nm).

2-(4-(Benzyloxy)-3-methoxyphenyl)ethan-1-amine (38).

Similar to general procedure D, LiAlH₄(1.70 g, 44.8 mmol, 5.1 eq.) in dry THF (10 mL) and nitrovinyl compound 37 (2.53 g,

8.87 mmol, 1.0 eq.) in dry THF (20 mL) were used and the resulting mixture was stirred at 80 ^ꢀC for 18 h. Amine 38 (1.44 g,

5.60 mmol, 63%) was isolated as yellow oil. Analytical data are in accordance with literature (Pouysegu et al., 2002). R_f= 0.30

(9:1 CH₂Cl₂/MeOH). ¹H NMR (400 MHz, CDCl₃) d/ppm = 7.46 – 7.41 (m, 2H, 2’-H, 6’-H), 7.39 – 7.34 (m, 2H, 3’-H, 5’-H), 7.29

(ddd, J = 7.2, 4.8, 1.4 Hz, 1H, 4’-H), 6.82 (d, J = 8.1 Hz, 1H, 5-H), 6.75 (d, J = 2.0 Hz, 1H, 2-H), 6.67 (dd, J = 8.1, 2.0 Hz, 1H, 6-H),

5.13 (s, 2H, PhCH₂), 3.88 (s, 3H, OCH₃), 2.94 (t, J = 6.8 Hz, 2H, CH₂CH₂NH₂), 2.68 (t, J = 6.8 Hz, 2H, CH₂CH₂NH₂). ¹³C NMR (101

MHz, CDCl₃) d/ppm = 149.7 (C-3), 146.7 (C-4), 137.4 (C-1’), 133.0 (C-1), 128.5 (C-3’, C-5’), 127.8 (C-4’), 127.3 (C-2’, C-6’), 120.8

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e12

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

(C-6), 114.3 (C-5), 112.7 (C-2), 71.2 (PhCH₂), 56.0 (OCH₃), 43.6 (CH₂CH₂NH₂), 39.5 (CH₂CH₂NH₂). IR (ATR) ṽ_max/cm^-1= 3032, 2938,

2909, 2869, 2838, 1512, 1260, 1227, 1139, 1032, 738, 696. HRMS (ESI): calcd. for C₁₆H₂₀NO₂(M+H)⁺258.14886; found 258.14886.

Ethyl (4-(benzyloxy)-3-methoxyphenethyl)carbamate (39).

Following general procedure A, amine 38 (1.40 g, 5.44 mmol, 1.0 eq.) in CH₂Cl₂(20 mL), NEt₃(1.10 mL, 8.16 mmol, 1.5 eq.) and

ethyl chloroformate (680 mL, 7.07 mmol, 1.3 eq.) were used. The reaction was completed after 72 h and the product was puriﬁed by

FCC (4:1 hexanes/acetone) to yield carbamate 39 (861 mg, 2.61 mmol, 48%) as light yellow solid. Analytical data are in accordance

with literature (Schrittwieser et al., 2011). R_f= 0.28 (4:1 hexanes/acetone). m.p.: 76 ^ꢀC. [lit. (Schrittwieser et al., 2011): 80-81 ^ꢀC]. ¹H

NMR (500 MHz, CDCl₃) d/ppm = 7.43 (dt, J = 8.1, 1.9 Hz, 2H, 2’-H, 6’-H), 7.36 (ddt, J = 8.1, 6.4, 1.3 Hz, 2H, 3’-H, 5’-H), 7.32 – 7.28 (m,

1H, 4’-H), 6.82 (d, J = 8.1 Hz, 1H, 5-H), 6.76 – 6.70 (m, 1H, 2-H), 6.66 (dd, J = 8.1, 2.0 Hz, 1H, 6-H), 5.13 (s, 2H, PhCH₂), 4.65 (s, 1H, NH),

4.11 (q, J = 7.0 Hz, 2H, CH₂CH₃), 3.88 (s, 3H, OCH₃), 3.40 (q, J = 6.8 Hz, 2H, CH₂CH₂NH), 2.74 (t, J = 6.8 Hz, 2H, CH₂CH₂NH), 1.23 (t,

J = 7.0 Hz, 3H, CH₂CH₃). ¹³C NMR (126 MHz, CDCl₃) d/ppm = 156.7 (CO), 149.9 (C-3), 147.0 (C-4), 137.4 (C-1’), 132.1 (C-1), 128.7 (C-

3’, C-5’), 127.9 (C-4’), 127.4 (C-2’, C-6’), 120.8 (C-6), 114.4 (C-5), 112.6 (C-2), 71.3 (PhCH₂), 60.9 (CH₂CH₃), 56.1 (OCH₃), 42.3

(CH₂CH₂NH), 35.9 (CH₂CH₂NH), 14.8 (CH₂CH₃). IR (ATR) ṽ_max/cm^-1= 3333, 2980, 2936, 1682, 1515, 1252, 1228, 1138. HRMS

(ESI): calcd. for C₁₉H₂₄NO₄(M+H)⁺330.16998; found 330.17056. Purity (HPLC): > 96 % (l = 210 nm).

Ethyl (4-((ethoxycarbonyl)oxy)-3-methoxyphenethyl)carbamate (40).

Following general procedure A, 3-methoxytyramine hydrochloride (5.00 g, 24.5 mmol, 1.0 eq.) in CH₂Cl₂(25 mL), NEt₃(12.0 mL,

58.9 mmol, 3.5 eq.) and ethyl chloroformate (5.90 mL, 61.4 mmol, 2.5 eq.) were used. The reaction was completed after 16 h and the

product was puriﬁed by FCC (4:1 hexanes/acetone) to yield carbamate 40 (5.86 g, 18.8 mmol, 77%) as colorless solid. R_f= 0.10

(CH₂Cl₂). m.p.: 68 ^ꢀC. ¹H NMR (500 MHz, CDCl₃) d/ppm = 7.05 (d, J = 8.0 Hz, 1H, 5-H), 6.81 – 6.74 (m, 2H, 2-H, 6-H), 4.68 (s,

1H, NH), 4.31 (q, J = 7.0 Hz, 2H, OCOOCH₂CH₃), 4.11 (q, J = 7.0 Hz, 2H, NCOOCH₂CH₃), 3.84 (s, 3H, OCH₃), 3.43 (q, J = 6.5 Hz,

2H, CH₂CH₂NH), 2.80 (t, J = 6.5 Hz, 2H, CH₂CH₂NH), 1.38 (t, J = 7.0 Hz, 3H, OCOOCH₂CH₃), 1.23 (t, J = 7.0 Hz, 3H, NCOOCH₂CH₃).

¹³C NMR (126 MHz, CDCl₃) d/ppm = 156.7 (N-CO), 153.5 (O-CO), 151.3 (C-3), 138.9 (C-4), 138.1 (C-1), 122.5 (C-5), 120.9 (C-6), 113.2

(C-2), 65.1 (OCOOCH₂CH₃), 60.9 (NCOOCH₂CH₃), 56.1 (OCH₃), 42.1 (CH₂CH₂NH), 36.3 (CH₂CH₂NH), 14.8 (OCOOCH₂CH₃), 14.3

(NCOOCH₂CH₃). IR (ATR) ṽ_max/cm^-1= 3336, 2980, 2940, 2872, 1750, 1686, 1541, 1514, 1283, 1251, 1208, 1059, 1032. HRMS

(ESI): calcd. for C₁₅H₂₂NO₆(M+H)⁺312.14416; found 312.14425. Purity (HPLC): > 96 % (l = 210 nm), > 96 % (l = 254 nm).

Ethyl (4-hydroxy-3-methoxyphenethyl)carbamate (41).

Following general procedure A, 3-methoxytyramine hydrochloride (10.2 g, 50.0 mmol, 1.0 eq.) in CH₂Cl₂(50 mL), NEt₃(24.0 mL,

175 mmol, 3.5 eq.) and ethyl chloroformate (12.0 mL, 125 mmol, 2.5 eq.) were used. After 18 h, a 1 M ethanolic NaOH solution

(100 mL) was added and the mixture was stirred for 3.5 h. The reaction mixture was neutralized using 2 M aq. HCl solution, followed

by extraction with CH₂Cl₂(4 x 80 mL). The combined organic layers were dried over Na₂SO₄and concentrated in vacuo. Puriﬁcation

was accomplished by FCC (4:1 hexanes/acetone) to yield carbamate 41 (3.49 g, 14.6 mmol, 29%) as colorless solid. Analytical data

ꢀ

€

are in accordance with literature (Schutz et al., 2020). R_f= 0.18 (4:1 hexanes/acetone). m.p.: 98 C. [lit. (Schutz et al., 2020): 95.0 –

95.5 ^ꢀC]. ¹H NMR (400 MHz, CDCl₃) d/ppm = 6.85 (d, J = 8.4 Hz, 1H, 5-H), 6.74 – 6.63 (m, 2H, 2-H, 6-H), 5.50 (s, 1H, OH), 4.64 (s, 1H,

NH), 4.11 (q, J = 7.1 Hz, 2H, CH₂CH₃), 3.88 (s, 3H, OCH₃), 3.40 (q, J = 7.1 Hz, 2H, CH₂CH₂NH), 2.74 (t, J = 7.1 Hz, 2H, CH₂CH₂NH),

1.23 (t, J = 7.1 Hz, 3H, CH₂CH₃). ¹³C NMR (101 MHz, CDCl₃) d/ppm = 156.7 (CO), 146.7 (C-3), 144.4 (C-4), 130.8 (C-1), 121.5 (C-2 or

e13 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

6), 114.6 (C-5), 111.4 (C-2 or 6), 60.9 (CH₂CH₃), 56.0 (OCH₃), 42.4 (CH₂CH₂NH), 36.0 (CH₂CH₂NH), 14.8 (CH₂CH₃). IR (ATR) ṽ_max

/

cm^-1= 3276, 2979, 1688, 1516, 1237, 1033. HRMS (ESI): calcd. for C₁₂H₁₆NO₄(M-H)^-238.10848; found 238.10860. Purity

(HPLC): > 96 % (l = 210 nm).

Ethyl (3-methoxy-4-phenoxyphenethyl)carbamate (42).

Following general procedure E, carbamate 41 (2.00 g, 8.36 mmol, 1.0 eq.), phenylboronic acid (3.06 g, 25.1 mmol, 3.0 eq.), cop-

per(II) acetate (1.67 g, 9.20 mmol, 1.1 eq.), NEt₃(2.90 mL, 20.9 mmol, 2.5 eq.) and pyridine (1.70 mL, 20.9 mmol, 2.5 eq.) in CH₂Cl₂

(40 mL) were used. The reaction was completed after 22 h and the product was puriﬁed by FCC (4:1 hexanes/acetone) to yield diaryl

1

ꢀ

ether 42 (2.24 g, 8.36 mmol, 85%) as colorless solid. R_f= 0.35 (4:1 hexanes/acetone). m.p.: 40 C. H NMR (400 MHz, CDCl₃) d/

ppm = 7.32 – 7.26 (m, 2H, 3’-H, 5’-H), 7.07 – 7.01 (m, 1H, 4’-H), 6.96 – 6.92 (m, 2H, 2’-H, 6’-H), 6.90 (d, J = 8.1 Hz, 1H, 5-H), 6.83

(d, J = 1.8 Hz, 1H, 2-H), 6.74 (dd, J = 8.1, 1.8 Hz, 1H, 6-H), 4.69 (s, 1H, NH), 4.12 (q, J = 7.2 Hz, 2H, CH₂CH₃), 3.83 (s, 3H, OCH₃),

3.45 (q, J = 7.0 Hz, 2H, CH₂CH₂NH), 2.81 (t, J = 7.0 Hz, 2H, CH₂CH₂NH), 1.24 (t, J = 7.0 Hz, 3H, CH₂CH₃). ¹³C NMR (101 MHz,

CDCl₃) d/ppm = 158.1 (C-1’), 156.7 (CO), 151.6 (C-3), 143.8 (C-4), 135.7 (C-1), 129.6 (C-3’, C-5’), 122.6 (C-4’), 121.3 (C-5 or 6),

121.3 (C-5 or 6), 117.3 (C-2’, C-6’), 113.5 (C-2), 61.0 (CH₂CH₃), 56.2 (OCH₃), 42.3 (CH₂CH₂NH), 36.2 (CH₂CH₂NH), 14.8 (CH₂CH₃).

IR (ATR) ṽ_max/cm^-1= 3303, 2971, 1688, 1490, 1272, 1255, 1222, 1035, 754. HRMS (ESI): calcd. for C₁₈H₂₂NO₄(M+H)⁺

316.15433; found 316.15425. Purity (HPLC): > 96 % (l = 210 nm).

Ethyl (3,4-dimethoxyphenethyl)carbamate (43).

Following general procedure A, 3,4-dimethoxyphenylethylamine (5.00 mL, 29.6 mmol, 1.0 eq.) in CH₂Cl₂(30 mL), NEt₃(12.0 mL,

88.9 mmol, 3.0 eq.) and ethyl chloroformate (5.70 mL, 59.3 mmol, 2.0 eq.) were used. The mixture was heated to reﬂux for 4 h and

extraction yielded carbamate 43 (7.05 g, 27.8 mmol, 94%) as colorless solid. NMR data are in accordance with literature (Awuah and

Capretta, 2010). m.p.: 63 ^ꢀC. R_f= 0.25 (4:1 hexanes/acetone). ¹H NMR (400 MHz, CDCl₃) d/ppm = 6.81 (d, J = 8.0 Hz, 1H, 5-H), 6.78 –

6.65 (m, 2H, 2-H, 6-H), 4.65 (s, 1H, NH), 4.11 (q, J = 7.1 Hz, 2H, CH₂CH₃), 3.88 – 3.85 (m, 6H, 2x OCH₃), 3.41 (q, J = 6.8 Hz, 2H,

CH₂CH₂NH), 2.75 (t, J = 6.8 Hz, 2H, CH₂CH₂NH), 1.23 (t, J = 7.1 Hz, 3H, CH₂CH₃). ¹³C NMR (101 MHz, CDCl₃) d/ppm = 156.7

(CO), 149.2 (C-3), 147.8 (C-4), 131.5 (C-1), 120.8 (C-2), 112.1 (C-6), 111.5 (C-5), 60.9 (CH₂CH₃), 56.1 (OCH₃), 56.0 (OCH₃), 42.3

(CH₂CH₂NH), 35.9 (CH₂CH₂NH), 14.8 (CH₂CH₃). IR (ATR) ṽ_max/cm^-1= 3372, 2935, 1697, 1514, 1258, 1234, 1139, 1027. HRMS

(ESI): calcd. for C₁₃H₂₀NO₄(M+H)⁺254.13868; found 354.13885. Purity (HPLC): > 96 % (l = 210 nm).

Ethyl (4-formylphenyl) carbonate (44).

Following general procedure A, 4-hydroxybenzaldehyde (1.00 g, 8.19 mmol, 1.0 eq.) in CH₂Cl₂(10 mL), NEt₃(1.70 mL, 12.3 mmol,

1.5 eq.) and ethyl chloroformate (0.860 mL, 9.01 mmol, 1.1 eq.) were uses. The reaction was completed after 14 h and the product

was puriﬁed by FCC (CH₂Cl₂) to yield aldehyde 44 (1.53 g, 7.90 mmol, 97%) as colorless solid. R_f= 0.63 (CH₂Cl₂). m.p.: 121 ^ꢀC. ¹H

NMR (500 MHz, CDCl₃) d/ppm = 9.99 (s, 1H, CHO), 7.93 – 7.91 (m, 2H, 3-H, 5-H), 7.40 – 7.32 (m, 2H, 2-H, 6-H), 4.36 – 4.32 (m, 2H,

CH₂CH₃), 1.41 – 1.38 (m, 3H, CH₂CH₃). ¹³C NMR (126 MHz, CDCl₃) d/ppm = 190.9 (CHO), 155.7 (C-1), 152.9 (CO), 134.2 (C-4), 131.4

(C-3, C-5), 121.9 (C-2, C-6), 65.4 (CH₂), 14.3 (CH₃). IR (ATR) ṽ_max/cm^-1= 1757, 1668, 1602, 1254, 1210, 1159, 1055, 998, 840, 775.

HRMS (EI): calcd. for C₁₀H₁₀O₄(M) 194.0579; found 194.0579. Purity (HPLC): > 96 % (l = 210 nm), > 96 % (l = 254 nm).

Ethyl (4-(2-methoxyvinyl)phenyl) carbonate (45).

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e14

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

Following general procedure B, (methoxymethyl)-triphenylphosphonium chloride (9.53 g, 27.8 mmol, 1.2 eq.) and LDA (20% in

THF/ethylbenzene/heptane, 11.6 mL, 23.2 mmol, 1.0 eq.) in dry THF (50 mL) and aldehyde 44 (4.50 g, 23.2 mmol, 1.0 eq.) in dry

THF (50 mL) were used. The mixture was stirred for 16 h and the product was puriﬁed by FCC (9:1 hexanes/EtOAc) to yield desired

cis/trans-enol ether 45 (3.30 g, 14.8 mmol, 64%) as colorless oil. R_f= 0.28 (4:1 hexanes/EtOAc). ¹H NMR (cis/trans, 400 MHz, CDCl₃)

d/ppm = 7.59 – 7.55 (m, 2H, cis 3-H, 5-H), 7.24 – 7.20 (m, 2H, trans 3-H, 5-H), 7.10 – 7.06 (m, 4H, cis/trans 2-H, 6-H), 7.00 (d, J =

13.0 Hz, 1H, trans CHCH-O), 6.13 (d, J = 7.0 Hz, 1H, cis CHCH-O), 5.79 (d, J = 13.0 Hz, 1H, trans CHCH-O), 5.21 (d, J = 7.0 Hz,

1H, cis CHCH-O), 4.31 (qd, J = 7.1, 1.6 Hz, 4H, cis/trans CH₂CH₃), 3.78 (s, 3H, cis OCH₃), 3.68 (s, 3H, trans OCH₃), 1.38 (td, J =

7.1, 1.6 Hz, 6H, cis/trans CH₂CH₃). ¹³C NMR (cis/trans, 126 MHz, CDCl₃) d/ppm = 153.9 (trans CO), 153.8 (cis CO), 149.2 (trans

CHCH-O), 149.2 (trans C-1), 148.9 (cis C-1), 148.1 (cis CHCH-O), 134.5 (trans C-4), 134.0 (cis C-4), 129.3 (cis C-3, C-5), 126.1 (trans

C-3, C-5), 121.3 (trans C-2, C-6), 120.8 (cis C-2, C-6), 104.9 (cis CHCH-O), 104.3 (trans CHCH-O), 64.9 (trans CH₂CH₃), 60.8 (cis

CH₂CH₃), 56.7 (cis/trans CH₂CH₃), 14.4 (cis/trans CH₂CH₃). Cis/trans-ratio according to ¹H-NMR is 0.83. HRMS (EI): calcd. for

C₁₂H₁₄O₄(M)^,+222.0887; found 222.0879.

1-(2-Methoxyvinyl)-4-phenoxybenzene (46).

Following general procedure B, (methoxymethyl)triphenylphosphonium chloride (1.18 g, 3.43 mmol, 1.2 eq.) and LDA (20% in THF/

ethylbenzene/heptane, 1.90 mL, 2.86 mmol, 1.0 eq.) in dry THF (10 mL) and 4-phenoxybenzaldehyde (500 mL, 2.86 mmol, 1.0 eq.) in

dry THF (10 mL) were used. The mixture was stirred for 50 h and the product was puriﬁed by FCC (9:1 hexanes/EtOAc) to yield the

desired cis/trans-enol ether 46 (384 mg, 1.70 mmol, 59%) as colorless oil. R_f= 0.73 (4:1 hexanes/EtOAc). ¹H NMR (cis/trans, 400

MHz, CDCl₃) d/ppm = 7.60 – 7.53 (m, 2H, cis 3-H, 5-H), 7.36 – 7.28 (m, 4H, cis/trans Ph), 7.22 – 7.18 (m, 2H, trans 2-H, 6-H or 3-H, 5-

H), 7.12 – 7.05 (m, 2H, cis/trans Ph), 7.02 – 6.91 (m, 9H, 4x cis/trans Ph, trans CHCH-O, cis/trans 2-H, 6-H or/and 3-H, 5-H), 6.11 (d, J =

7.0 Hz, 1H, cis CHCH-O), 5.81 (d, J = 13.0 Hz, 1H, trans CHCH-O), 5.22 (d, J = 7.0 Hz, 1H, cis CHCH-O), 3.78 (s, 3H, cis OCH₃), 3.69 (s,

3H, trans OCH₃). ¹³C NMR (cis/trans, 101 MHz, CDCl₃) d/ppm = 157.8 (cis/trans qPh), 157.7 (cis/trans qPh), 155.2 (cis/trans C-1),

154.9 (cis/trans C-1), 148.5 (trans CHCH-O), 147.4 (cis CHCH-O), 131.8 (cis/trans C-4), 131.5 (cis/trans C-4), 129.9 (cis/trans Ph),

129.8 (cis/trans Ph), 129.7 (cis C-3, C-5), 126.5 (trans C-2, C-6 or C-3, C-5), 123.1 (cis/trans Ph), 123.0 (cis/trans Ph), 119.6 (cis/trans

Ph or C-2, C-6 or C-3, C-5), 119.1 (cis/trans Ph or C-2, C-6 or C-3, C-5), 118.7 (cis/trans Ph or C-2, C-6 or C-3, C-5), 118.6 (cis/trans Ph

or C-2, C-6 or C-3, C-5), 105.1 (cis CHCH-O), 104.5 (trans CHCH-O), 60.8 (cis OCH₃), 56.7 (trans OCH₃). Cis/trans-ratio according to

+

¹H-NMR is 0.75. IR (ATR) ṽ_max/cm^-1= 3036, 2935, 2831, 1488, 1234, 1150, 1092, 750, 691. HRMS (EI): calcd. for C₁₅H₁₄O₂(M)

226.0988; found 226.0987.

d

1-(Benzyloxy)-4-(2-methoxyvinyl)benzene (47).

Following general procedure B, (methoxymethyl)triphenylphosphonium chloride (4.11 g, 12.0 mmol, 1.2 eq.) and LDA (20% in THF/

ethylbenzene/heptane, 7.90 mL, 12.0 mmol, 1.2 eq.) in dry THF (50 mL) and 4-(benzoyloxy)benzaldehyde (2.12 g, 9.99 mmol, 1.0 eq.)

in dry THF (50 mL) were used. The mixture was stirred for 16 h and the product was puriﬁed by FCC (6:1 hexanes/EtOAc) to yield the

1

desired cis/trans-enol ether 47 (2.04 g, 8.49 mmol, 85%) as yellow solid. R_f= 0.68 (4:1 hexanes/EtOAc). H NMR (cis/trans, 400

MHz, CDCl₃) d/ppm = 7.53 – 7.49 (m, 2H, cis 3-H, 5-H), 7.45 – 7.29 (m, 10H, cis/trans PhCH₂), 7.18 – 7.13 (m, 2H, trans 3-H, 5-

H), 6.95 – 6.88 (m, 5H, cis/trans 2-H, 6-H, trans CHCH-O), 6.06 (d, J = 7.0 Hz, 1H, cis CHCH-O), 5.78 (d, J = 13.0 Hz, 1H, trans

CHCH-O), 5.18 (d, J = 7.0 Hz, 1H, cis CHCH-O), 5.09 – 5.03 (m, 4H, cis/trans PhCH₂), 3.76 (s, 3H, cis OCH₃), 3.66 (s, 3H, trans

OCH₃). ¹³C NMR (cis/trans, 101 MHz, CDCl₃) d/ppm = 157.2 (cis/trans C-1), 157.0 (cis/trans C-1), 147.8 (trans CHCH-O), 146.6

(cis CHCH-O), 137.3 (cis/trans qPh), 137.2 (cis/trans qPh), 129.5 (cis C-3, C-5), 129.3 (cis/trans C-4), 129.2 (cis/trans C-4), 128.8

(cis/trans Ph), 128.7 (cis/trans Ph), 128.1 (cis/trans Ph), 128.0 (cis/trans Ph), 127.6 (cis/trans Ph), 126.3 (trans C-3, C-5), 115.3 (trans

C-2, C-6), 114.8 (cis C-2, C-6), 105.4 (cis CHCH-O), 104.7 (trans CHCH-O), 70.2 (cis/trans PhCH₂), 70.1 (cis/trans PhCH₂), 60.6 (cis

OCH₃), 56.6 (trans OCH₃). Cis/trans-ratio according to ¹H-NMR is 1.1. HRMS (EI): calcd. for C₁₆H₁₆O₂(M) 240.1145; found

240.1146.

+

d

e15 Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021

€

Please cite this article in press as: Muller et al., Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and

tumor growth, Cell Chemical Biology (2021), https://doi.org/10.1016/j.chembiol.2021.01.023

ll

Article

( )-Ethyl 7-((ethoxycarbonyl)oxy)-1-(4-((ethoxycarbonyl)oxy)benzyl)-6-methoxy-3,4-dihy-droisoquinoline-2(1H)-carboxylate (48).

Following general procedure C, carbamate 40 (348 mg, 1.12 mmol, 1.0 eq.), enol ether 45 (248 mg, 1.12 mmol, 1.0 eq.) and TFA

(840 mL, 11.2 mmol, 10 eq.) were dissolved in CH₂Cl₂(15 mL). The reaction was completed after 18 h and the product was puriﬁed by

FCC (5:1 hexanes/acetone) to yield compound 48 (452 mg, 0.901 mmol, 81%) as colorless oil. R_f= 0.20 (4:1 hexanes/acetone). ¹H

NMR (400 MHz, C₂D₂Cl₄, 100 ^ꢀC) d/ppm = 7.03 (s, 4H, 2’-H, 3’-H, 5’-H, 6’-H), 6.69 – 6.63 (m, 2H, 5-H, 8-H), 5.19 (t, J = 6.6 Hz, 1H, 1-

H), 4.28 – 4.22 (m, 4H, 2 OCOOCH₂CH₃), 4.05 – 3.92 (m, 3H, NCOOCH₂CH₃, 3-H), 3.79 (s, 3H, OCH₃), 3.23 (ddd, J = 13.6, 9.7, 4.6 Hz,

1H, 3-H), 3.00 (h, J = 6.6 Hz, 2H, a-H), 2.81 (ddd, J = 15.8, 9.7, 5.8 Hz, 1H, 4-H), 2.61 – 2.54 (m, 1H, 4-H), 1.35 – 1.31 (m, 6H, 2

13

ꢀ

OCOOCH₂CH₃), 1.12 (t, J = 7.2 Hz, 3H, NCOOCH₂CH₃). C NMR (101 MHz, C₂D₂Cl₄, 100 C) d/ppm = 155.6 (N-CO), 153.6 (O-

CO), 153.3 (O-CO), 150.4 (C-4’), 150.3 (C-6), 139.1 (C-7), 136.0 (C-1’), 133.4 (C-4a), 130.7 (C-2’, C-6’or C-3’, C-5’), 129.3 (C-8a),

121.3 (C-5 or C-8), 120.9 (C-2’, C-6’ or C-3’, C-5’), 113.6 (C-5 or C-8), 65.1 (OCOOCH₂CH₃), 64.9 (OCOOCH₂CH₃), 61.5

(NCOOCH₂CH₃), 56.6 (OCH₃), 56.0 (C-1), 42.5 (C-a), 38.6 (C-3), 28.7 (C-4), 14.8 (NCOOCH₂CH₃), 14.4 (OCOOCH₂CH₃), 14.3

(OCOOCH₂CH₃). IR (ATR) ṽ_max/cm^-1= 2981, 2940, 1758, 1692, 1248, 1215, 1198. HRMS (ESI): calcd. for C₂₆H₃₅N₂O₉(M+NH₄)⁺

519.23371; found 519.23373. Purity (HPLC): > 96 % (l = 210 nm).

( )-Ethyl 1-(4-((ethoxycarbonyl)oxy)benzyl)-6,7-dimethoxy-3,4-dihydroisoquinoline-2(1H)-carboxylate – SG-089 (10).

Following general procedure C, carbamate 43 (5.60 g, 22.1 mmol, 1.0 eq.), enol ether 45 (4.92 g, 22.1 mmol, 1.0 eq.) and TFA

(18.0 mL, 243 mmol, 11 eq.) were dissolved in CH₂Cl₂(50 mL). The reaction was completed after 20 h and the product was puriﬁed

by FCC (5:1 hexanes/acetone) to give SG-089 (10, 7.50 g, 16.9 mmol, 77%) as colorless solid. R_f= 0.28 (4:1 hexanes/acetone). m.p.:

134 ^ꢀC. ¹H NMR (400 MHz, C₂D₂Cl₄, 100 ^ꢀC) d/ppm = 7.10 – 6.97 (m, 4H, 2’-H, 3’-H, 5’-H, 6’-H), 6.57 (s, 1H, 5-H), 6.23 (s, 1H, 8-H),

5.15 (t, J = 6.6 Hz, 1H, 1-H), 4.26 (q, J = 7.1 Hz, 2H, OCOOCH₂CH₃), 4.13 – 3.98 (m, 2H, NCOOCH₂CH₃), 3.98 – 3.82 (m, 1H, 3-H), 3.78

(s, 3H, 6-OCH₃), 3.61 (s, 3H, 7-OCH₃), 3.30 (ddd, J = 13.4, 9.4, 4.7 Hz, 1H, 3-H), 3.09 (dd, J = 13.4, 6.8 Hz, 1H, a-H), 2.94 (dd, J = 13.4,

6.8 Hz, 1H, a-H), 2.77 (ddd, J = 15.6, 9.4, 5.9 Hz, 1H, 4-H), 2.57 (dt, J = 15.6, 4.7 Hz, 1H, 4-H), 1.34 (t, J = 7.1 Hz, 3H, OCOOCH₂CH₃),

1.17 (t, J = 7.1 Hz, 3H, NCOOCH₂CH₃). ¹³C NMR (101 MHz, C₂D₂Cl₄, 100 ^ꢀC) d/ppm = 155.7 (N-CO), 153.6 (O-CO), 150.3 (C-4’),

148.8 (C-6), 148.1 (C-7), 136.4 (C-1’), 130.8 (C-2’, C-6’), 129.0 (C-8a), 127.0 (C-4a), 120.9 (C-3’, C-5’), 113.3 (C-5), 112.3 (C-8),

64.9 (OCOOCH₂CH₃), 61.5 (NCOOCH₂CH₃), 56.7 (OCH₃-6), 56.6 (OCH₃-7), 56.5 (C-1), 42.6 (C-a), 39.1 (C-3), 28.2 (C-4), 14.8

(OCOOCH₂CH₃), 14.4 (NCOOCH₂CH₃). IR (ATR) ṽ_max/cm^-1= 3006, 2918, 2849, 1754, 1673, 1281, 1260, 1245, 1206, 1096, 859.

HRMS (ESI): calcd. for C₂₄H₃₀NO₇(M+H)⁺444.20168; found 444.20143. Purity (HPLC): > 96 % (l = 210 nm), > 96 % (l = 254 nm).

( )-Ethyl 1-(4-((ethoxycarbonyl)oxy)benzyl)-6-methoxy-7-phenoxy-3,4-dihydroisoquino-line-2(1H)-carboxylate (49).

Following general procedure C, carbamate 42 (1.60 g, 5.06 mmol, 1.0 eq.), enol ether 45 (1.35 g, 6.07 mmol, 1.2 eq.) and TFA

(3.80 mL, 50.6 mmol, 10 eq.) were dissolved in CH₂Cl₂(50 mL). The reaction was completed after 90 h and the product was puriﬁed

Cell Chemical Biology 28, 1–13.e1–e27, June 17, 2021 e16

Article Doi

Gene editing and synthetically accessible inhibitors reveal role for TPC2 in HCC cell proliferation and tumor growth

DOI: 10.1016/j.chembiol.2021.01.023

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.chembiol.2021.01.023

Full text of DOI:10.1016/j.chembiol.2021.01.023

Products guided by the article

R&D Labs maybe for 1541762-89-3

Relevant to this article

Hot Product