Chemical synthesis and functional characterization of a new class of ceramide analogues as anti-cancer agents

Article abstract of DOI:10.1016/j.bmc.2019.02.030

Deregulation of ceramide metabolism is a hallmark of human cancer. Ceramide analogues thereby represent a new class of anti-cancer agents. We aimed at developing effective and low toxic ceramide analogues and synthesized a new class of ceramide analogues starting from L-threonine. Several analogues exhibit potent cytotoxicity against human cancer cells in vitro with IC₅₀ as low as 4.8 μM. These ceramide analogues decreased xIAP and Bcl-xL level and exhibited significant sensitization activity to overcome human cancer cell resistance to TRAIL, a cancer-selective agent that are being tested in human clinical trials. Furthermore, we determined that these ceramide analogues effectively suppress human cancer xenograft growth in vivo with no significant toxicity at the efficacious dose. Therefore, we have developed a simple and effective method to synthesize functional ceramide analogues using L-threonine as starting material and these analogues have the great potential to be further developed as anti-cancer agents in human cancer therapy.

Full text of DOI:10.1016/j.bmc.2019.02.030

Accepted Manuscript
Chemical Synthesis and Functional Characterization of A New Class of Ce-
ramide Analogues as Anti-Cancer Agents
Qianqian Liu, Xia Li, Yong-Sheng Bao, Jingxin Lu, Hua Li, Zhizhen Huang,
Feiyan Liu
PII:
S0968-0896(18)32085-6
https://doi.org/10.1016/j.bmc.2019.02.030
BMC 14766
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 December 2018
5 February 2019
15 February 2019
Please cite this article as: Liu, Q., Li, X., Bao, Y-S., Lu, J., Li, H., Huang, Z., Liu, F., Chemical Synthesis and
Functional Characterization of A New Class of Ceramide Analogues as Anti-Cancer Agents, Bioorganic &
Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.bmc.2019.02.030
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Chemical Synthesis and Functional Characterization of A New Class of
Ceramide Analogues as Anti-Cancer Agents
Qianqian Liu1#, Xia Li2#, Yong-Sheng Bao3#, Jingxin Lu2, Hua Li2, Zhizhen Huang3,4,
and Feiyan Liu2*
1Zhejiang University Hospital, Zhejiang University, Hangzhou 310058, P. R. China.
2Research Centre of Siyuan Natural Pharmacy and Biotoxicology, College of Life
3
Sciences, Zhejiang University, Hangzhou 310058, P. R. China. School of Chemistry
and Chemical Engineering, Nanjing University, Nanjing 210093, P. R. China.
4Department of Chemistry, Zhejiang University, Hangzhou 310028, P. R. China.
# equal contributions
Running title: Synthesis and function of anti-cancer ceramide analogueues
*Correspondence: Feiyan Liu, College of Life Sciences, Zhejiang University,
Hangzhou 310058, P. R. China, e-mail: liuf64@zju.edu.cn;

Abstract
Deregulation of ceramide metabolism is a hallmark of human cancer. Ceramide
analogues thereby represent a new class of anti-cancer agents. We aimed at
developing effective and low toxic ceramide analogues and synthesized a new class of
ceramide analogues starting from L-threonine. Several analogues exhibit potent
cytotoxicity against human cancer cells in vitro with IC50 as low as 4.8 M. These
ceramide analogues decreased xIAP and Bcl-xL level and exhibited significant
sensitization activity to overcome human cancer cell resistance to TRAIL, a cancer-
selective agent that are being tested in human clinical trials. Furthermore, we
determined that these ceramide analogues effectively suppress human cancer
xenograft growth in vivo with no significant toxicity at the efficacious dose.
Therefore, we have developed a simple and effective method to synthesize functional
ceramide analogues using L-threonine as starting material and these analogues have
the great potential to be further developed as anti-cancer agents in human cancer
therapy.
Key words: Anti-tumor agents; Ceramide Analogues; xIAP; L-threonine; TRAIL.

1. Introduction
Sphingolipids are a family of ubiquitous membrane components that are the
most structurally diverse and complex lipids owing to their numerous variations in the
sphingosine bases, N-acyl linked fatty acids and head groups. Sphingolipids were
originally thought to be solely the cellular energy source and membrane structure
molecules. The revelation that sphingolipids are key bioeffectors that regulate a broad
range of cellular processes, including cell proliferation, apoptosis, motility,
differentiation, stress responses, protein synthesis, carbohydrate metabolism,
immunity, and angiogenesis , is widely regarded as one of the major advances in
modern biology in the past two decades [1, 2]. The two major sphingolipid metabolites,
ceramide and sphingosine-1-phosphate (S1P), are the central mediators of the
sphingolipid metabolism pathways. Interestingly, ceramide and S1P exhibit opposing
bioactivities with ceramide being pro-apoptotic and S1P generally promoting cell
survival [3,4]
.
Compelling experimental data from mouse models and human patients have
shown that sphingolipid deregulation, namely the unbalance between ceramide and
S1P, is a key factor in tumor pathogenesis, progression and cancer cell resistance to
chemotherapeutic agents and radiation [5, 6]. The crucial roles of ceramide and S1P in
tumor development and cancer cell responses to chemotherapy and radiation have led
to extensive efforts to target the ceramide-S1P metabolism signaling pathways for
anticancer therapy. For the last two decades, extensive efforts have be devoted to

develop ceramide analogues to mimic natural ceramide; small molecular inhibitors for
enzymes that catalize ceramide catabolism or its conversion; and S1P receptor
[7,
antagonists
8]. Among these various approaches, development of ceramide
analogues by far is the most commonly one. Numerous ceramide analogues with
different chemical and biological properties have been developed [7, 9-20], and various
ceramide analogues have been tested in human clinical trials against cancer. However,
the facts that ceramide is one of the most structurally diverse and complex lipids and
that ceramide analogues can potentially interact with numerous cellular targets in the
sphingolipid signaling pathways make development of effective and yet low toxic
ceramide analogues a challenging one and greatly limit its clinical use in human
cancer therapy. Therefore, development of new ceramide analogues with effective
anticancer activity and yet low toxicity are still in urgent need.
TNF-related apoptosis-inducing ligand (TRAIL) is a member of the TNF
superfamily. Ever since its discovery in 1995, TRAIL has been under intense study
since it preferentially induces apoptosis in tumor cells but not in normal cells.
However, the success of TRAIL-based cancer therapy so far is limited because cancer
cells, especially metastatic cancer cells, often exhibit a TRAIL-resistance phenotype.
Therefore, identifying novel TRAIL sensitizers with low toxicity and high
sensitization activity is also in urgent need.
For the synthesis of ceramide analogues, L-serine has been widely used as a
starting material [10,11]. L-threonine, one of essential amino acids, has one more methyl
and one more chiral carbon as compared to L-serine. Based on the medicinal

chemistry phenomenon of "magic methyl" effect that links the dramatic change in
affinity to the addition of a single methyl group in just the right place, we
hypothesized that L-threonine is a better substrate for synthesis of ceramide analogues
with higher solubility and greater bioactivity. To test this hypothesis, we made use of
L-threonine as a starting material for the synthesis of novel ceramide analogues.
Functional assays determined that these ceramide analogues exhibit potent
cytotoxicity against multiple types of human cancer cells with IC50 as low as 4.8 M
in vitro. Furthermore, sublethal doses of these ceramide analogues effective overcame
human cancer cell chemoresistance. Biochemical analysis validated that these new
ceramide analogues effectively decrease anti-apoptosis IAP and Bcl-xL proteins in
human cancer cells. Most importantly, several of these ceramide analogues exhibited
effective tumor suppressive activity at doses that cause no significant toxicity in vivo.
Therefore, these ceramide analogues represent new candidates that have the potential
to be developed as effective and yet low toxic anticancer agents.
2. Materials and Methods
2.1 Chemistry: All chemicals were purchased from Aladdin Chemical Co. All
reactions were carried out under anhydrous conditions. Chromatography was
performed on HPTLC silica gel 60 F254. HNMR spectra were recorded on a Bruker
AVANCE 300MHz spectrometer, and CNMR spectra were recorded on Bruker
AVANCE 400MHz spectrometer. Mass spectral data were recorded on LCMS-2020
(SHIMADZU) mass spectrometer.

2.1.1 Synthetic procedure for 4a-d: To a solution of N-Boc-L-threonine 2 (2.19g,
10mmol) in anhydrous THF (60mL) 1-hydroxybenzotriazole hydrate (HOBt) (1.35g,
10mmol) was added at 0 ºC. The pH value of the solution was adjusted to 8-9 with 4-
methylmorpholine. After the reaction mixture was stirred for 5 min, 1-ethyl-3-(3-
dimethyllaminopropyl) carbodiimide hydrochloride (EDC·HCl) (2.20g, 11mmol) and
a solution of alkyl amine in anhydrous THF (5mL) was added. The reaction mixture
was stirred at 0 ºC for 2 h and at 20ºC overnight. After evaporation of the mixture in
vacuo, the residue was dissolved in ethyl acetate (60mL), and the solution was
washed successively with saturated NaHCO3, 5% KHSO4, and saturated NaCl. The
organic phase was separated and dried over anhydrous MgSO4. After filtration and
evaporation in vacuo, desired (2S,3R)-tert-butyl-3-hydroxy-1-oxo-1-(alkyl)butan-2-
ylcarbamate.3a-d was obtained. To a solution of (2S,3R)-tert-butyl-3-hydroxy-1-
oxo-1-(alkyl)butan-2-ylcarbamate 3 in CH2Cl2 (70mL) trifluoroacetic acid (10.0mL)
was added at 0 ºC. The reaction mixture was stirred at 20ºC for 4 h. After evaporation
in vacuo, the residue was dripped into the 1N NaOH (100mL), and generate
precipitation. Filtration and dried in vacuo led to desired L-threoninamide 4a-d.
2.1.2 Synthetic procedure for 5xa-xj: To a solution of carboxylic acid (1.0mmol)
in anhydrous THF (20mL) were added (2S,3R)-2-amino-3-hydroxy-N-
alkylbutanamide 4a-d (1.0mmol) and HOBt (1.0mmol, 0.135g) at 0ºC. After the
reaction mixture was stirred for 5 min, EDC·HCl (0.220g, 1.1mmol) was added. The

pH value of the solution was adjusted to 8-9 with 4-methylmorpholine. The reaction
mixture was stirred at 0 ºC for 2 h and overnight at room temperature. After
evaporation of the mixture in vacuo, the residue was dissolved in ethyl acetate
(50mL). The solution was washed successively with saturated NaHCO3, 5% KHSO4,
and saturated NaCl. The organic phase was separated and dried over anhydrous
MgSO4. After filtration and evaporation in vacuo, residue was purified by
recrystallization in petroleum ester/ethyl acetate to give the desired ceramide
analogues 5xa-xj.
2.1.3 Synthetic procedure for 5xk-xm: At 0ºC, to a solution of 1.0mmol of (2S,3R)-
2-amino-3-hydroxy-N-alkylbutanamide (4a-4d) and 2.0mmol of pyridine in
anhydrous CH2Cl2 (20mL), 10mL CH2Cl2 solution of heterocyclic sulfonyl chloride or
aryl sulfonyl chloride was dripped added. The mixture was stirred at 0 ºC for 2 h and
overnight at room temperature. After evaporation in vacuo,the residue was dissolved
in ethyl acetate(50ml). The solution was washed successively with saturated NaHCO3,
5% KHSO4, and saturated NaCl，and the organic phase was separated and dried over
anhydrous MgSO4 for 2 h. After filtration and evaporation in vacuo, residue was
purified by recrystallization in petroleum ester/ethyl acetate to give the desired
ceramide analogues 5xk-xm .
2.1.4
Synthetic procedure for control compound 6： For comparing carbon
skeleton of L-threoninamide with L-serinamide, control compound 6 was prepared by

the same protocol with ceramide analogues 5.
2.2 Cell viability assay: The effect of ceramide analogues on cell viability was
measured by MTT assay according to the manufacturer’s instructions (ATCC,
Manassas VA). Briefly, six human cancer cells (Liver Cancer HepG2， Squamous
Carcinoma A431，Pancreatic Cancer SA，Breast Cancer MCF-7，Ovarian Cancer
SiHa and Colon Cancer SW620) were seeded in 96-well plates (7x103 cells/well) for
24 h. 47 Ceramide analogues were then added to final concentrations of 0 to 200
μM, and the cells were cultured for another 48h, followed by MTT assay.
2.3 Apoptosis assay: Human Colon Cancer cells SW620 were treated with the
indicated ceramide analogues at doses of their IC50 for 8 h or 24h. Cells were then
fixed in ice-cold 70% ethanol for 15 min, washed with PBS and stained with Hoechst
33342 (Sigma) for 20 min in the dark. Morphologic changes were analyzed under a
fluorescence microscope. DMSO used as vehicle control.
2.4 TRAIL-mediated apoptosis: Five human cancer cells(Liver Cancer HepG2，
Colon Cancer SW620、 LS411N， Breast Cancer MCF-7， Cervical Cancer Hela)
were seeded in 96-well plates and cultured for 24 h. Ceramide analogues, TRAIL
(100 ng/ml) or both ceramide analogue and TRAIL were added to the culture for 48 h.
Cell death was measured by MTT assay.

2.5 Western blotting analysis: Western blotting analysis was performed essentially
as previously described [32]. Anti-xIAP antibody was obtained from Cell signaling,
anti-Bcl-xL was obtained from BD Biosciences, and anti-β-actin was from Sigma.
2.6 Human cancer xenograft growth inhibition: The metastatic human colon
carcinoma SW620 cells were injected to athymic mice subcultaneously. Tumor-
bearing mice were randomly grouped into 7 groups with 6 mice in each group. The
control group was injected with 200 L 1% polyoxythylenated castor oil, and the rest
four groups of mice were treated with ceramide analogues 4c, 5ch, 5ck, and 5cm,
respectively, at a dose of 100 mg/kg body weight on alternate days for 15. Tumor
sizes were measured with a micrometer caliper. Serum samples were collected at days
15 and analyzed for creatine kinase level using the creatine kinase ELISA kit
(Shangshai Biological Technology Co. LTD).
Statistical analysis. Where indicated, data were represented as the means ± SD.
Statistical analysis was performed using two-sided t test, with p-values<0.05
considered statistically significant.
3. Results
3.1 Chemical Synthesis of a group of new ceramide analogues
The general synthetic procedure is outlined in Figure 1. Initially, L-
threonine 1 was reacted with (Boc)2O in the mixed solvent of H2O-dioxane, affording

N-Boc protected threonine 2 in 94% yield. Because long-chain ceramide analogues
often exhibit greater antitumor activity due to optimum lipid solubility [10, 21], we
employed long-chain amines to react with Boc-L-threonine 2 in the presence of
[22]
EDCl/HOBt, giving corresponding Boc-L-threoninamide 3 in 82-88% yields
.
[23]
After the deprotection of Boc group with trifluoroacetic acid in dichloromethane
,
four types of L-threoninamides 4a-d (Fig. 1B) were obtained in 75-85% yields. Under
the catalysis of EDCl/HOBt, L-threoninamides 4a-d was carboxylated in THF to give
desired ceramide analogues 5xa-xj in 64-85% yields. We synthesized 37 ceramide
analogues 5xa-xj with each bearing two amide groups (Fig. 1B). X representatives 4
types of long-chains containing 8, 12, 14, or 18 carbons in its nitrogen. L-
Threoninamides 4a-d, on the other hand, reacted with aryl or heteroaryl sulfonyl
chloride in the presence of pyridine to give ceramide analogues 5xk-xm with each
bearing one amide group and one sufonamide group. (Fig.1B). Therefore, a total of 45
ceramide analogues 5 were synthesized (Table S1).

Figure1. Ceramide analogue synthesis outline. A. Natural C16-ceramide (a), ceramide analogue
(b) and ceramide analogues synthesized in this study (c). B. Summary of ceramide synthesis. C.
The structure of control compound 6. The different functional groups are indicated in the
bottom.

3.2 Ceramide analogues exhibits potent cytotoxicity against human cancer cells
in vitro
To functional determine the bioactivities of these new ceramide analogues in
suppression of human cancer cell growth, we performed cancer cell growth inhibition
assays using six types of human cancer cells. It is clear that most of the newly
synthesized analogues exhibit potent anti-cancer activity against all six types of
human cancer cells. Interestingly, it seems that these analogues are relatively more
cytotoxic to breast cancer cell MCF-7 cells (Table 1). A structure-activity-relation
was observed between length of the alkyl chain and cytotoxicity of these analogues.
Because of various kinds of acyl groups coupled with four different length carbon
chain cores, these analogues didn’t exhibit alkyl chain length dependent on
cytotoxicity. For example, 5bd has most cytotoxicity in benzofuran-2-carboxamides,
but 5dd, 5dc, 5dh are less cytotoxic than analogues with shorter alkyl chains (Table
1). However in general, ‘b’ serial (N-dodecyl) and ‘c’ serial (N-tetradecyl) have
higher cytotoxicity against all tested cancer cell lines than ‘a’ serial (N-octyl) and ‘d’
serial (N-octadecyl) . For instance, 5bd, 5cd have higher cytotoxicity than 5ad, 5dd.
These associations thus indicate that only L-threoninamide coupled with middle long
carbon chain in the amide group rather than original amino group have greater anti-
cancer cytotoxicity (Table 1, Table S1).
A structure-activity-relation was also found that heteroaryl carbonyl groups
substituted L-threoninamides were more potent than aromatic carbonyl groups or

aliphatic carbonyl groups substituted L-threoninamides. For example, 5cd, 5ce, 5cf,
5cg, 5ch are superior to 5ca, 5ci, 5cj in all six types of (HepG2, A431, MES-SA, Siha,
MCF-7, SW620) human cancer cell lines. Above all, 5ch containing indole ring
emerged as one of the most potent analogue tested in this study with lowest IC50 of
6.81µM in MCF-7 and 4.80µM in SiHa, respectively (Table1). These results indicate
that heteroaryl carbonyl group could significantly increase anti-cancer cytotoxicity of
ceramide analogues. In addition, 5cg has one more methyl than L-serinamide
derivative 6, and its cytotoxicity is higher than L-serinamide derivative 6 for six types
of human cancer cells, suggesting that carbon skeleton of L-threoninamide is superior
to L-serinamide after coupling with the same heteroaryl carbonyl group (Table1).

Table 1．Cytotoxicity of 47 novel ceramide analogues against six human cancer cells in
vitro .Cells (1x104 ml-1) were incubated with various concentrations of 47 novel ceramide
analogues for 48h. Cell viability was measured by MTT assay. IC50 represents concentration
causing a 50% growth inhibition. Data are mean ± SD, N = 4

To determine whether the cytotoxicity observed above is due to apoptosis
induction, we next selected four ceramide analogues which have higher cytotoxicity
and tested their apoptosis-inducing activity in SW620 cells. Morphologic changes of
apoptotic cells were analyzed by staining cells with Hoechst 33342 and observed
under a fluorescence microscope. Shining cells were considered as apoptotic cells. It
is clear that 5ch，5ck, 5cm, 5cg especially 5ch show greater pro-apoptotic activity
with more shining cells as compared to control. Moreover, 5cg which has one more
methyl than L-serinamide derivative 6 shows greater pro-apoptotic activity with more
shining cells as compared to control compound 6 (Fig.2). Therefore, these
observations validate the above cytotoxicity assays that the pro-apoptotic action of L-
threoninamide is greater than L-serinamide after coupling with same heteroaryl
carbonyl group.

Figure 2. Ceramide analogues induce human cancer cell apoptosis. SW620 cells were treated
with the indicated ceramide analogues at doses of their IC50. Cells analyzed for apoptosis at 8 and
24 h. Cells were fixed and stained with Hoechst 33342. Shown are representative images of one of
three experiments.
3.3 Ceramide analogues decrease Bcl-xL and xIAP levels in human cancer cells
Ceramide regulates cellular proliferation and apoptosis through various
signaling pathways. However, compelling experimental data indicate that ceramide
promotes apoptosis through modulating key apoptosis regulators in both the extrinsic
and intrinsic apoptosis signaling pathways [24-29]. Specifically, ceramide targets the
Bcl-xL and xIAP to directly induce apoptosis or to increase tumor cell sensitivity to
apoptosis induction [24, 30, 31]. Therefore, we examined the effects of these new
ceramide analogues on Bcl-xL and xIAP as biochemical markers of these ceramide
analogues’ bioactivity. Indeed, all four ceramide analogues tested which has lower
IC50 decreased Bcl-xL and/or xIAP protein level in the metastatic human colon
carcinoma cells (Fig. 3). However, although all four analogues decreased Bcl-xL
protein level, only two of the four analogues (4c and 5ch) decreased xIAP protein
level (Fig. 3).

Figure 3. Ceramide analogues decrease Bcl-xL and xIAP in human colon carcinoma cells.
The metastatic human colon carcinoma SW620 cells were treated with the indicated ceramide
analogues at the indicated concentrations for 48 h, and analyzed for Bcl-xL and xIAP protein
levels by Western blotting analysis.
3.4 Ceramide analogues effectively overcome human cancer cell resistance to
TRAIL
TRAIL is a cancer-selective agent that has been extensively tested in clinical
trials as an anti-cancer agent. However, the clinical outcome so far is poor, which is
diminishing the hope to develop TRAIL or TRAIL receptor agonists as anti-cancer
agents. This outcome, however, is not totally surprising since it is known that human
cancer cells are often resistant to TRAIL-induced apoptosis [32, 33]. Because TRAIL-
induced apoptosis depends on the intrinsic apoptosis pathway [34], and ceramide
analogues decreases Bcl-xL and xIAP [24, 30, 31], we reasoned that these ceramide
analogues should be effective in sensitization of human cancer cells to TRAIL-
induced apoptosis, and screened some of these new ceramide analogues with IC50

greater than 50µM. This is because we classify these new ceramide analogues into
two groups: 1) highly cytotoxic compounds (HCC) and low cytotoxic compounds
(LCC). For HCC the objective is to determine their mechanism of action as potential
monotherapeutic agent for cancer therapy. For LCC, the objective is to develop their
agents as adjuvant agents to enhance the efficacy of approved anti-cancer agents.
Human cancer LS411N, SW620, MCF-7, HeLa and HepG2 cells were cultured in the
presence of TRAIL, ceramide analogue at a dose of their IC50, or both TRAIL and
ceramide analogue. Cell growth inhibition was then measured. As expected, human
cancer cells are resistant to TRAIL (Fig. 4). Interestingly, seven ceramide analogues
tested significantly increased the sensitivity of these 5 types of human cancer cells to
TRAIL-mediated growth inhibition (Fig. 4).

Figure 4. Ceramide analogues overcame human cancer cell resistance to TRAIL. The
indicated human cancer cells were seeded in 96-well plate for 24 h. Cells were then treated with
TRAIL (100 ng/ml), ceramide analogue at a dose of its IC50, or both TRAIL and ceramide
analogue for another 48 h. Cell viability was measured by MTT assay. Growth inhibition was

calculated by the formula: OD570 of untreated cells – OD570 of treated cells. Column: mean,
bar:SD.
3.5 Inhibitory activity of ceramide analogues against human cancer xenograft in
vivo
To determine whether the above observed in vitro anti-tumor cell cytotoxicity
can be translated into tumor growth inhibition in vivo, we examined the efficacy of
these newly synthesized ceramide analogues in suppression of human tumor growth
in vivo. The metastatic human colon carcinoma SW620 cells were injected sc to
athymic mice. Tumor-bearing mice were then treated with four of the ceramide
analogues, respectively. Tumor growth was monitored over time. All 4 ceramide
analogues exhibited tumor growth suppression activity. Among them, three analogues
(4c, 5ch and 5cm) exhibited potent suppressive efficacy against the established tumor
xenograft, while the other one (5ck) exhibited mild tumor growth suppression activity
(Fig. 5).

Figure 5. Ceramide analogues suppress human cancer growth in vivo. The metastatic human
colon carcinoma SW620 cells (3x106 cells/mouse) were injected subcutaneously to athymic mice.
Mice were randomized into experimental groups (n=6 for each group) when the tumors 60-80
mm3. The tumor-bearing mice were treated with saline or the indicated ceramide analogues (100
mg/kg body weight) on alternate days for a total of 15 days. Tumor size was measured in three
diameter with micrometer caliper at the indicate times to permit calculation of tumor volume.
3.6 Ceramide analogues exhibit no significant toxicity in vivo
We next sought to determine the toxicity of these ceramide analogues in mice.
Three of the ceramide analogues that were tested for in vivo with potent anti-cancer
activity were administered to mice, and serum level of creatine kinase, a commonly
used biochemical marker of myocardial infarction, rhabdomyolysis, muscular
dystrophy, autoimmune myositides and and acute renal failure, was measured as in
vivo toxicity indicator. Serum was collected from tumor-bearing mice after the last
treatment as in Figure 5, and measured for creatine kinase level. None of the ceramide
analogues caused significant increase in serum creatine kinase level (Table 2).
Table 2. Evaluation of the toxicity of four novel ceramide analogues in vivo. Serum samples
of tumor-bearing nude mice were collected at days 15 after injection, and analyzed for
creatine kinase level using the creatine kinase ELISA kit. Data are mean ± SD, N = 6

4. Discussion
For the synthesis of ceramide analogues, L-serine has been widely used as a
starting material [10, 11]. A series of novel ceramide analogues have been synthesized
by changing sphingosine backbone to a long-chain L-serinamide and introducing
amide or imine substituents on the original nitrogen [14]. These analogues exhibit high
[15,
efficacy as anti-proliferative agents against human cancer cell lines
16]. L-
threonine, one of essential amino acids, has one more methyl and one more chiral
carbon as compared to L-serine. Based on the medicinal chemistry phenomenon of
"magic methyl" effect that links the dramatic change in affinity to the addition of a
single methyl group in just the right place, we hypothesized that L-threonine is a
better substrate for synthesis of ceramide analogues with higher solubility and greater
bioactivity and made use of L-threonine as a starting material for the synthesis of
novel ceramide analogues. We successfully synthesized 47 novel ceramide analogues
with great solubility and high yield. Therefore, we have established a simple and yet
efficient synthesis procedure to chemically synthesize new ceramide analogues
starting from L-threonine.
In vitro cytotoxicity analysis revealed that the majority of these 47 new ceramide
analogues are bioactive as they can effectively suppress human cancer cell growth.
Several analogues (i.e. 5ch and 5cm) exhibited super ant-proliferative activity with
IC50 of 4.8-11 M. Initial testing of four of these new ceramide analogues indicate
that they are also effective in suppression of established human cancer xenograft
growth in vivo. However, it is interesting to notice that the analogue with the most

potent in vitro cytotoxicity (i.e.5ch) does not exhibit the greatest suppressive efficacy
against human xenograft growth in vivo, suggesting that in addition to its cytotoxicity
potential, other factors, such as pharmacokinetics and pharmacodynamics of the
analogues may also be important determinants of in vivo anti-cancer activity. More
studies are apparently needed to expand the in vivo studies to more analogues to
identify the most effective and low toxic ceramide analogues among these 47
compounds for cancer therapy.
TRAIL is a cancer-selective agent that has been extensively tested in human
cancer patients for the last decade. However, so far, the test results in human cancer
patients are not encouraging and the hope for developing TRAIL or TRAIL receptor
agonist for human cancer therapy is fading. This outcome is not totally surprising
since it is known that most human cancer cells are resistant to TRAIL [33, 34]. We
screened some of these new ceramide analogues with IC50 greater than 50µM in five
types of human cancer cells, and observed that seven analogues are effective in
overcoming human cancer cell resistant to TRAIL. Our data thus indicate that these
new ceramide analogues have the potential to be developed as an adjunct agent for
TRAIL-based cancer therapy to improve the efficacy of TRAIL therapy.
Numerous studies have demonstrated that ceramide targets the Bcl-2 family
proteins and IAPs to induce cellular apoptosis [24-29]. We observed that indeed the new
ceramide analogues can significantly decrease the anti-apoptotic Bcl-xL and xIAP
protein level in human cancer cells [31]. However, the although degree of decrease in
xIAP and Bcl-xL protein level induced by a particular analogue is generally

associated with the in vitro cytotoxicity, the decrease of Bcl-xL and xIAP is only
partially associated with the in vivo tumor suppression efficacy of a particular
ceramide analogue. These observations suggest that these ceramide analogues
suppress tumor growth only partially through inhibiting these anti-apoptotic
mediators. Therefore, further studies are needed to further elucidate the molecular
mechanisms underlying ceramide-mediated apoptosis and tumor suppression.
In summary, we have developed a simple and efficient new procedure to
chemically synthesize ceramide analogues using L-threonine as substrate. A total of
47 novel ceramide analogues have been synthesized. Initial analysis indicates that
many of these ceramide analogues are bioactive as anti-cancer agents with IC50 as low
as 4.8 M. Furthermore, these ceramide analogues also exhibited potent sensitization
activity to overcome human cancer cell resistant to TRAIL. More significantly,
several of these new ceramide analogues suppressed established human cancer
xenograft growth in vivo without significant toxicity. These ceramide analogues thus
hold great potential to be developed as a class of new anti-cancer agents.
Acknowledgments
This study is supported by National Natural Science Foundation of China (No.
31570811) Natural Science Foundation of Zhejiang Province (LY15C020001) and Si-
Yuan Foundation.

References
[1] R.A. Wolff, R.T. Dobrowsky, A. Bielawska, L.M. Obeid, Y.A. Hannun, J Biol Chem, 1994, 269, 19605-
19609.
[2] B. Ogretmen, Y.A. Hannun, Nat Rev Cancer, 2004, 4, 604-616.
[3]Samy A. F. Morad and Myles C. Cabot, Nature Review/Cancer, 2013, 13, 51-65
[4]Sahar A. Saddoughi,, Besim Ogretmen, Advances in Cancer Research, 2013, 117, 37-58
[5] A. Apraiz, J.K. Idkowiak-Baldys, M.D. Boyano, G. Perez-Yarza, Y.A. Hannun, A. Asumendi, BMC
Cancer, 2011, 11, 477.
[6] J.C. Cheng, A. Bai, T.H. Beckham, S.T. Marrison, C.L. Yount, K. Young, P. Lu, A.M. Bartlett, B.X. Wu,
B.J. Keane, K.E. Armeson, D.T. Marshall, T.E. Keane, M.T. Smith, E.E. Jones, R.R. Drake, Jr., A.
Bielawska, J.S. Norris, X. Liu, Clin Invest, 2013, 123, 4344-4358.
[7] A. Bielawska, J. Bielawski, Z.M. Szulc, N. Mayroo, X. Liu, A. Bai, S. Elojeimy, B. Rembiesa, J. Pierce,
J.S. Norris, Y.A. Hannun, Bioorg Med Chem, 2008, 16, 1032-1049.
[8] B. Oskouian, J.D. Saba, Adv Exp Med Biol, 2010, 688, 185-205.
[9] A. Singh, H.J. Ha, J. Park, J.H. Kim, W.K. Lee, Bioorg Med Chem, 2011, 19, 6174-6181.
[10] H. Niiro, H. Azuma, S. Tanago, K. Matsumura, K. Shikata, T. Tachibana, K. Ogino, Bioorg Med
Chem, 2004, 12, 49-51.
[11] V. Gududuru, E. Hurh, G.G. Durgam, S.S. Hong, V.M. Sardar, H. Xu, J.T. Dalton, D.D. Miller, Bioorg
Med Chem Lett, 2004, 14, 4919-4923.
[12] E. Bieberich, T. Kawaguchi, R.K. Yu, J Biol Chem, 2000, 275, 177-181.
[13] K. Kim, J. Kang, S. Kim, S. Choi, S. Lim, C. Im, C. Yim, Arch Pharm Res, 2007, 30, 570-580.
[14] J.W. Antoon, J. Liu, M.M. Gestaut, M.E. Burow, B.S. Beckman, M. Foroozesh, Design, J Med Chem,
2009, 52, 5748-5752.
[15] J. Liu, J.W. Antoon, A. Ponnapakkam, B.S. Beckman, M. Foroozesh, Bioorg Med Chem, 2010, 18,
5316-5322.
[16] J.W. Antoon, B.S. Beckman, Bioorg Med Chem Lett, 2012, 22, 2624-2628.
[17] M. Selzner, A. Bielawska, M.A. Morse, H.A. Rudiger, D. Sindram, Y.A. Hannun, P.A. Clavien, Cancer
Res, 2001, 61, 1233-1240.
[18]K. Perez-Labrada, I. Brouard, I. Mendez and D. G. Rivera, J. Org. Chem. 2012, 77, 4660−4670.
[19]S. Barui, S. Saha, V. Yakati and A.Chaudhuri, Mol. Pharmaceutics 2016, 13, 404−419
[20]S. Vudhgiri,S. R. Routhu,C. G. Kumar,R. B. N. Prasad, R. C.R. Jala, Med Chem Res, 2017, DOI
10.1007/s00044-017-2049-9
[21] Y.T. Chang, J. Choi, S. Ding, E.E. Prieschl, T. Baumruker, J.M. Lee, S.K. Chung, P.G. Schultz, J Am
Chem Soc, 2002, 124, 1856-1857.
[22] R.S. Joshi, P.G. Mandhane, P.V. Badadhe, C.H. Gill, Ultrason Sonochem, 2011, 18, 735-738.
[23] Y. Kavanagh, C.M. Chaney, J. Muldoon, P. Evans, J Org Chem, 2008, 73, 8601-8604.
[24] C.E. Chalfant, K. Rathman, R.L. Pinkerman, R.E. Wood, L.M. Obeid, B. Ogretmen, Y.A. Hannun, J
Biol Chem, 2002, 277, 12587-12595.
[25] L.J. Siskind, T.D. Mullen, K. Romero Rosales, C.J. Clarke, M.J. Hernandez-Corbacho, A.L. Edinger,
L.M. Obeid, J Biol Chem, 2010, 285, 11818-11826.

Table 1.Cytotoxicity of 47 novel ceramide analogues against human cancer cells in vitro
Squamous
Cancer
Pancreatic
Cancer
Cancer cells
Live Cancer
Breast Cancer
Ovarian Cancer
Colon Cancer
4c
HepG2(μM)
N/A
A431(μM)
14.58±0.76
21.32±1.72
61.33±4.09
41.18±0.69
118.49±9.09
113.05±7.87
64.5±0.78
92.44±2.64
190.28±6.86
19.93±0.45
32.15±4.6
16.61±2.12
26.45±4.03
33.3±3.32
52.74±3.82
20.04±0.9
20.09±0.16
78.79±6.99
26.87±2.81
SA(μM)
MCF-7(μM)
10.69±0.98
27.42±1.48
111.47±4.66
36.57±8.86
24.77±6.75
31.75±6.55
91.84±4.21
67.49±4.88
104.63±5.03
21.38±2.4
SiHa(μM)
10.98±1.39
39.54±0.88
112.45±2.34
63.92±6.01
＞200
SW620(μM)
18.74±3.81
21.37±1.24
200
10.98±1.39
＞200
4d
44.52±3.49
136.48±5.92
＞400
5aa
5ba
5ca
5da
5ab
5bb
5cb
5db
5ac
5bc
5cc
5dc
5ad
5bd
5cd
5dd
5ae
34.62±4.17
75.45±6.24
56.04±7.43
＞400
200
54.21±3.95
＞400
200
38.22±2.52
124.01±8.78
64.32±1.68
127.1±7.41
13.35±1.05
91.85±0.59
23.85±3.65
63±0.25
200
192.16±11.09
146.86±22.28
＞400
64.65±0.5
93.49±3.13
148±11.63
52.55±6.02
60.93±4.29
23.28±3.33
21.22±3.78
53.49±5.32
40.78±3.63
10.49±1.31
9.61±0.56
161.96±4.55
22.91±2.59
200
87.5±7.58
200
42.29±3.73
149.26±8.7
30.1±3.84
88.4±15.53
91.96±7.76
81.26±9.13
22.85±1.18
23.38±1.19
＞200
＞200
72.46±3.37
30.43±2.8
136.32±42.57
29.19±0.57
78.44±11.128
200
36.07±0.27
25.73±0.47
47.22±7.3
116.6±3.9
66.48±2.5
12.85±2.07
14.23±1.84
＞200
200
20.89±0.54
23.38±1.19
126.36±7.19
63.21±4.73
21.32±1.28
26.7±2.48
＞200
88.19±1.92
65.99±5.48
67.75±0.099
[26] H. Lee, J.A. Rotolo, J. Mesicek, T. Penate-Medina, A. Rimner, W.C. Liao, X. Yin, G. Ragupathi, D.
Ehleiter, E. Gulbins, D. Zhai, J.C. Reed, A. Haimovitz-Friedman, Z. Fuks, R. Kolesnick, PLoS One, 2011, 6
e19783.
[27] J.E. Chipuk, G.P. McStay, A. Bharti, T. Kuwana, C.J. Clarke, L.J. Siskind, L.M. Obeid, D.R. Green,
Cell, 2012, 148, 988-1000.
[28] L.J. Beverly, L.A. Howell, M. Hernandez-Corbacho, L. Casson, J.E. Chipuk, L.J. Siskind, Biochem J,
2013, 492, 111-119.
[29] L. Casson, L. Howell, L.A. Mathews, M. Ferrer, N. Southall, R. Guha, J.M. Keller, C. Thomas, L.J.
Siskind, L.J. Beverly, PLoS One, 2013, 8, e54925.
[30] B.J. Kroesen, S. Jacobs, B.J. Pettus, H. Sietsma, J.W. Kok, Y.A. Hannun, L.F. de Leij, J Biol Chem,
2003, 278, 14723-14731.
[31] A.V. Paschall, M.A. Zimmerman, C.M. Torres, D. Yang, M.R. Chen, X. Li, E. Bieberich, A. Bai, J.
Bielawski, A. Bielawska, K. Liu, BMC Cancer, 2014, 14, 24.
[32] A. Ashkenazi, Nat Rev Drug Discov, 2008, 7, 1001-1012.
[33] C. Voelkel-Johnson, Y.A. Hannun, A. El-Zawahry, Mol Cancer Ther, 2005, 4, 1320-1327.
[34] A. Ashkenazi, V.M. Dixit, Science, 1998, 281, 1305-1308.

5be
5ce
5de
5bf
65.69±2.63
28.56±3.77
96.94±6.77
11.97±2.38
10.51±0.35
89.42±0.27
＞200
62.76±7.56
23.04±3.72
44.13±2.02
10.31±2.78
12.25±1.90
80.97±11.42
168.74±14.66
13.68±1.73
60.98±1.82
23.41±3.48
15.47±2.95
31.78±1.93
36.04±3.36
＞200
51.93±0.42
24.41±3.60
49.14±4.72
12.28±2.75
12.11±2.77
35.65±0.27
31.3±1.14
25.65±0.42
36.37±0.19
19.97±0.74
11.52±0.97
54.91±3.90
25.59±4.41
154.82±1.16
80.87±1.9
47.09±0.54
40.44±2.16
189.28±12.18
184.37±1.9
＞200
50.53±5.65
22.79±3.94
30.78±0.67
12.28±2.75
15.11±2.97
35.65±0.27
31.3±1.14
25.65±0.42
36.37±0.19
15.5±2.3
56.62±4.91
21.6±2.26
51.17±3.46
21.25±0.01
14.20±1.80
29.56±1.48
＞200
65.79±4.14
30.4±2.09
136.26±13.29
18.75±1.2
15.20±1.86
23.56±1.18
＞200
5cf
5df
5bg
5cg
5dg
5bh
5ch
5dh
5ai
34.45±2.34
＞200
26.52±2.33
112.98±12.02
13.6±1.02
4.8±0.29
21.52±3.33
123.98±13.82
22.02±2.86
21.965±0.71
103±7.48
61.35±0.27
200
22.61±2.89
21.11±1.92
57.57±2.69
55.55±0.64
＞200
6.81±1.73
37.33±1.82
35.06±0.2
＞200
65.85±6.45
41.41±5.96
＞200
5bi
5ci
150.18±4.37
127.61±9.04
101.68±4.07
＞400
64.86±4.87
177.08±7.17
58.24±0.11
171.45±12.08
＞200
64.57±4.54
39.22±2.21
148.86±3.24
37.1±4.11
27.76±0.14
79.63±2.34
31.12±1.99
12.89±1.5
60.12±3.2
19.1±1.65
18.66±1.2
＞200
35.82±0.68
35.94±2.64
123.22±2.5
＞200
200
5di
200
5aj
200
5bj
200
5cj
200
＞400
＞200
5dj
284.22±8.30
25.78±0.52
10.92±0.54
75.76±9.46
25.88±1.21
21.57±2.05
＞200
187.13±2.41
25.19±0.99
11.99±1.87
34.03±1.03
21.99±0.86
21.97±0.56
＞200
200
＞200
5bk
5ck
5dk
5bl
27.23±0.99
10.89±1.25
47.88±5.2
20.1±1.65
11.56±2.2
＞200
31.59±4.56
19.23±2.14
29.78±2.23
30.94±1.28
20.37±2.47
＞200
25.59±3.56
15.23±3.16
32.48±1.23
27.94±1.08
18.36±1.4
＞200
5cl
5dl
5cm
5dm
Compound
6
11.38±1.11
19.83±3.12
19.57±2.16
23.66±1.46
10.21±0.3
21.23±0.81
12.21±1.2
27.76±1.81
11.10±1.39
19.90±3.31
9.36±0.69
24.63±2.33
45.23±0.53
28.12±1.08
35.12±0.45
45.12±2.46
38.34±3.56
40.36±2.5
Table 2. Mouse serum toxicity profiles of ceramide analogues
Treatment
Serum Enzyme Level
(100mg/kg body weight)
CK(U/L)
Blood Collected Day
Control (n=6)
1.94±1.04
15
4c
(n=6)
2.98±0.79
2.44±0.61
2.32±0.34
15
15
15
5ch (n=6)
5ck (n=6)

5cm (n=6)
2.63±1.02
15
Measurements were performed 15 days after ceramide analogues injection.
CK, creatine kinase
Graphical Abstract
1. (Boc)2O / NaOH
O
O
O
2. RNH2
HO
HO
NHR
R'
NHR
HO
OH
NH2
R'COOH
EDC / BtOH
EDC / BtOH
3. CF3COOH
H3C
H3C
HN
H3C
NH2
O