Hydrogel-mediated delivery of celastrol and doxorubicin induces a synergistic effect on tumor regression via upregulation of ceramides

Nihal Medatwal ab, Mohammad Nafees Ansari a, Sandeep Kumar ab, Sanjay Pal ac, Somesh Kumar Jha a, Priyanka Verma a, Kajal Rana a, Ujjaini Dasgupta *d and Avinash Bajaj *a
aLaboratory of Nanotechnology and Chemical Biology, Regional Centre for Biotechnology, 3rd Milestone, Faridabad-Gurgaon Expressway, NCR Biotech Science Cluster, Faridabad-121001, Haryana, India. E-mail: bajaj@rcb.res.in
bManipal Academy of Higher Education, Manipal-576104, Karnataka, India
cKalinga Institute of Industrial Technology, Bhubaneswar-751024, Odisha, India
dAmity Institute of Integrative Sciences and Health, Amity University, Panchgaon, Manesar, Gurgaon-122413, Haryana, India. E-mail: udasgupta@ggn.amity.edu

Received 7th February 2020 , Accepted 3rd August 2020

First published on 5th August 2020


Abstract

The release of anticancer drugs in systemic circulation and their associated toxicity are responsible for the poor efficacy of chemotherapy. Therefore, the identification of new chemotherapeutic combinations designed to be released near the tumor site in a sustained manner has the potential to enhance the efficacy and reduce the toxicity associated with chemotherapy. Here, we present the identification of a combination of doxorubicin, a DNA-binding topoisomerase inhibitor, with a naturally occurring triterpenoid, celastrol, that induces a synergistic effect on the apoptosis of colon cancer cells. Hydrogel-mediated sustained release of a combination of doxorubicin and celastrol in a murine tumor model abrogates tumor proliferation, and increases the median survival with enhanced apoptosis and concurrent reduction in proliferation. Sphingolipid profiling (LC-MS/MS) of treated tumors showed that the combination of celastrol and doxorubicin induces global changes in the expression of sphingolipids with an increase in levels of ceramides. We further demonstrate that this dual drug combination induces a significant increase in the expression of ceramide synthase 1, 4, and 6, thereby increasing the level of ceramides that contribute to the synergistic apoptotic effect. Therefore, hydrogel-mediated localized delivery of a combination of celastrol and doxorubicin provides a new therapeutic combination that induces a sphingolipid-mediated synergistic effect against colon cancer.


Introduction

Cancer chemotherapy is often challenged with low efficacy and high toxicity of anticancer drugs due to their lack of specificity for cancer cells leading to poor patient survival.1 Therefore, combination therapy is usually preferred in clinical settings as it helps in reducing the toxicity owing to the lower dosage of chemotherapeutics.2 Combination therapies also provide advantages of high efficacy, enhanced patient survival, and ability to combat drug resistance as a combination of drugs can modulate multiple signalling pathways in cancer cells.3 However, numerous challenges are associated with the use of combination therapy like poor knowledge of an appropriate combination of drugs and their dosage, bioavailability of drugs in the desired ratio at the tumor site, their non-specific targeting, varying pharmacokinetics among different drugs, and lack of understanding of their mode of action.4

Many natural products like taxanes have been clinically approved for cancer treatment in combination with other therapeutic regimens.5 Celastrol (CEL), a quinine methide triterpenoid extracted from Tripterygium wilfordii Hook. f., is a traditional Chinese medicine.6,7 Recent studies have shown that CEL can enhance leptin sensitivity,8 and can also activate the heat shock factor 1 (HSF1) that enhances the energy expenditure and mitochondrial functions for the treatment of obesity.9 CEL is known to inhibit cancer cell proliferation by inducing the expression of pro-apoptotic proteins like Bax and cytochrome c, and by enhancing the Bax/Bcl-2 ratio.10,11 CEL reduces the colon tumor progression where CEL-mediated activation of LKB1 activates AMPKα and phosphorylates YAP, leading to the degradation of β-catenin.12 Administration of CEL also inhibits ulcerative colitis-induced colorectal cancer by preventing the upregulation of β-catenin, and downregulating the expression of inflammatory cytokines.13 Therefore, a combination of CEL with other clinically approved chemotherapeutics can be explored as new strategies for cancer treatment.14

The use of nanoparticle-mediated delivery of chemotherapeutics has achieved massive success in preclinical studies, but it could not completely eradicate the toxicity associated with the burst release of drugs in systemic circulation.15 Recent studies have shown the use of localized and injectable low molecular weight hydrogels (LMWHs) for maintaining the sustained delivery of chemotherapeutic drugs at the targeted site in the case of arthritis,16 organ transplantation,17 cancer,18 and inflammatory bowel disease.19 We have recently shown that hydrogel-mediated sustained release of a combination of doxorubicin, combretastatin A4, and dexamethasone targeting proliferation, inflammation, and angiogenesis can induce significant tumor regression along with an increase in survival in a murine tumor model.20 This unique combination of drugs altered the sphingolipid metabolism via post-transcriptional regulation, thereby altering the levels of glucosylceramides that assist in combating drug resistance.21 Therefore, identification of a suitable combination of drugs that can be delivered in the desired ratio at the tumor site, and understanding their mode of action are necessary for successful chemotherapy.

Cancer chemotherapeutics are known to induce apoptosis through alteration of the sphingolipid metabolism as upregulation of ceramides can activate serine/threonine protein phosphatases and proline-directed protein kinase C to induce apoptosis.22 Ceramides can also induce apoptosis by directly inhibiting the mitochondrial respiratory chain complex III, thereby causing the release of cytochrome c.23 Besides, ceramides elicit their apoptotic effect by activating thioredoxin interacting protein (Txnip)24 and transcription factor 4 (ATF-4), C/EBP homologous protein,24 or by regulating the alternative splicing of caspase 9 and Bcl-x.25–30 As doxorubicin (DOX), a topoisomerase inhibitor, is one of the first-line chemotherapeutic drugs, we hypothesize that hydrogel-mediated delivery of a combination of CEL and DOX may induce a synergistic effect on tumor regression in a murine colon cancer model through changes in sphingolipid metabolism (Fig. 1A). Therefore, we screened different combinations of CEL and DOX in colon cancer cells to identify an appropriate combination ratio that induces a synergistic apoptotic effect. This synergistic drug combination of CEL and DOX was then entrapped in a LMWH, and was explored for its ability to combat tumor regression in the murine model followed by in-depth mechanistic studies.


image file: d0nr01066a-f1.tif
Fig. 1 (A) Study plan where the hydrogel-mediated delivery of a combination of doxorubicin (DOX) and celastrol (CEL) is explored for combating tumor progression. (B) Experimental design used for investigating the synergistic effect of a combination of CEL and DOX. (C and D) Heat map showing the fraction of cells affected (C) and combination index (D) using MTT assay in response to a combination of CEL and DOX at different concentrations against murine (CT26) and human (HCT-8, DLD-1 and HCT-116) colon cancer cells.

Results and discussion

Study design

It is always desirable to find a combination of drugs that can be used at lower concentrations to get a similar or even better therapeutic effect than a single drug at a higher concentration. A combination of two drugs at lower concentrations will also be advantageous as both drugs will act on different signalling pathways, and there will be less chance of emergence of drug resistance. Therefore, we first performed MTT assay to determine the IC50 (minimum inhibitory concentrations at which 50% cell death is achieved) of CEL and DOX against murine (CT26) and different human (HCT-8, HCT-116, and DLD-1) colon cancer cell lines. Cytotoxicity assay at constant and non-constant ratios of CEL and DOX was then used to determine the effect of their combination on cell survival (Fig. 1B).31 Cells were treated with CEL alone (at 0, 0.125, 0.25, 0.5, 1.0, and 2.0 μM), DOX alone (at 0, 0.125, 0.25, 0.5, 1.0, and 2.0 μM), and a combination of CEL and DOX at different concentrations. Cell viability in terms of fraction affected (fraction of cell death) was assessed using MTT assay after 48 h of treatment. We analyzed the combination index (CI) at different combination ratios using Calcu-Syn® software to determine the additive, synergistic, or antagonistic effect of the combination of CEL and DOX (Fig. 1B).32 Drug combinations at which CI is less than 0.8 were considered to impart a synergetic effect, whereas drug combinations with CI greater than 1 were considered as antagonistic, and those between 0.8–1.0 were considered to have an additive effect.33

Combination of CEL and DOX induces a synergistic effect on colon cancer cell death

MTT assay revealed that the IC50 of CEL is 0.46 μM for CT-26, 0.71 μM for HCT-8, 0.45 μM for DLD-1, and 0.74 μM for HCT-116 cells, and that of DOX is 0.35 μM for CT-26, 0.23 μM for HCT-8, 0.49 μM for DLD-1, and 0.92 μM for HCT-116 cells (Fig. S1, Table S1). Heat maps of fraction affected (fraction of cell death) and combination index (CI) at different combination ratios of CEL and DOX showed a dose-dependent increase in fraction affected for all the cell lines where CT26 and HCT-8 were more sensitive to the combination of CEL and DOX as compared to other cell types (Fig. 1C). The combination of CEL and DOX is less toxic to HCT-116 cells as compared to CT-26, DLD-1, and HCT-8 cells (Fig. 1C). We observed ∼76, 60, and ∼75% cell death on using a combination of 0.5[thin space (1/6-em)]:[thin space (1/6-em)]0.5 (μM) of CEL and DOX in CT26, DLD1, and HCT-8 cells whereas HCT-116 cells showed only ∼40% cell death. Similarly, there was ∼88, ∼80, and ∼85% cell death on using a combination of 1[thin space (1/6-em)]:[thin space (1/6-em)]1 (μM) of CEL and DOX in CT26, DLD1, and HCT-8 cells, whereas HCT-116 cells showed only ∼74% cell death (Fig. 1C). A comparison of CI at different combination ratios for different cell types present few interesting features. Murine colon cancer cells (CT26) show a synergistic effect at maximum combination ratios, whereas a synergistic effect was observed only at selective combination ratios for HCT-116 cells. HCT-8, DLD-1, and CT26 cells showed a synergistic effect at some common combination treatment regimens with >80% cytotoxicity (Fig. 1C and D).

As CT26 cells can induce syngeneic murine tumor in BALB/c mice, we used CT26 cells for all further studies. The fraction affected (fraction of cell death) for CT26 cells after treatment with a combination of CEL (0.5 μM) and DOX (0.5 μM), or of CEL (1.0 μM) and DOX (1.0 μM) showed significantly more cell death as compared to individual treatments (Fig. 2A). CI at different CEL[thin space (1/6-em)]:[thin space (1/6-em)]DOX ratios range from 0.4–0.8 confirming their synergistic effect (Fig. 2B).


image file: d0nr01066a-f2.tif
Fig. 2 (A and B) Fraction affected (fraction cell death) (A) and combination index (CI) (B) on using a combination of CEL and DOX at constant ratios against murine colon cancer cells (CT26). (C) Percentage of apoptotic cells after 48 h treatment of CT26 cells with CEL, DOX, and a combination of CEL and DOX at different concentrations. (D) Percentage of DiOC6(3)-negative cells after treatment with CEL, DOX, and a combination of CEL and DOX after 24 h. (E) Representative micrographs of DiOC6(3)-stained CT26 cells after treatment with CEL (0.5 μM), DOX (0.5 μM) and a combination of CEL (0.5 μM) and DOX (0.5 μM) after 24 h. Data are presented as mean ± SD of three independent experiments. Data in A, C, and D were analyzed using unpaired two-tailed Student's t-test.

To further confirm the synergy between CEL and DOX, we performed Annexin-V and propidium iodide (PI) based apoptotic assay. CT26 cells were treated with CEL (0.5 μM, 1.0 μM), DOX (0.5 μM, 1.0 μM), or with a combination of CEL and DOX (0.5 μM or 1.0 μM each) for 48 h, and were analyzed by flow cytometry after staining with Annexin-FITC and PI. CEL induces ∼30% apoptosis at 0.5 μM, whereas DOX causes 40% apoptosis at 0.5 μM in CT26 cells after 48 h (Fig. 2C and Fig. S2A). In contrast, we observed >70% apoptosis on using a combination of CEL (0.5 μM) and DOX (0.5 μM) (Fig. 2C). In contrast, the combination of CEL (1.0 μM) and DOX (1.0 μM) did not induce a synergistic effect as only DOX (1.0 μM) was sufficient to induce significant apoptosis which was not further enhanced on combination treatment (Fig. 2C). Disruption of the mitochondrial membrane potential is one of the key targets of apoptotic pathways. We, therefore, investigated the effect of the combination of CEL and DOX on mitochondrial membrane potential. CT26 cells were then treated with CEL (0.5 μM or 1.0 μM), DOX (0.5 μM or 1.0 μM) or a combination of CEL and DOX (0.5 μM or 1.0 μM each) for 24 h. CT26 cells were then incubated with DiOC6(3) (3,3′-dihexyloxacarbocyanine iodide) that readily gets accumulated in active mitochondria due to its negative charge and shows enhanced green fluorescence.34 Flow cytometry data confirmed a significant increase in the number of DiOC6(3)-negative cells (>60%) on treatment with the combination of CEL and DOX (0.5 μM each) as compared to only CEL- (∼30%) and DOX-treated cells (∼40%) at 0.5 μM (Fig. 2D and Fig. S2B), thereby confirming the synergistic effect. In contrast, the combination of CEL (1.0 μM) and DOX (1.0 μM) did not induce a synergistic effect as only DOX (1.0 μM) was sufficient to induce a significant change in the mitochondrial membrane potential (Fig. 2D). Fluorescence micrographs of DiOC6(3)-stained CT-26 cells on different treatments confirm that a combination of CEL and DOX (0.5 μM each) induces disruption of the mitochondrial membrane potential, thereby causing loss of DiOC6(3)-staining due to increase in cell apoptosis (Fig. 2E).

Lithocholic acid-derived hydrogel can entrap the combination of CEL and DOX

In our recent study, we have shown that the conjugation of a dipeptide (glycine–glycine) to lithocholic acid helps the lipidated-dipeptide (LCA-GG, A13) to self-assemble into a supramolecular injectable hydrogel (Fig. 3A).20 A13 gel is biodegradable as the esterase-sensitive ester linkage is used between lithocholic acid and benzyl/peptide groups. As esterases are over expressed in tumor tissues, hydrogels get degraded with time by cleavage of the ester bond.20 We further showed that this biocompatible and biodegradable hydrogel could maintain a sustained release of a combination of chemotherapeutic drugs at the tumor site.20 Therefore, we explored the ability of LCA-GG (A13) gelator to entrap a combination of CEL and DOX, and found that A13 gel can easily entrap ∼30 mg of DOX or ∼10 mg of CEL in 70 mg of gelator in 1 mL of water. We then selected an optimised dose of 5 mg kg−1 of DOX as per our previous studies.20 As a combination of DOX and CEL in a 1[thin space (1/6-em)]:[thin space (1/6-em)]1 ratio showed a synergistic effect on cell apoptosis, we also selected 5 mg kg−1 of CEL for our animal studies. Drug entrapment studies found that A13 gel (200 μL used for one mice) can entrap CEL (100 μg) (called C-Gel), DOX (100 μg) (called D-Gel), and a combination of CEL (100 μg) and DOX (100 μg) (called CD-Gel) effectively (Fig. 3B). Rheological characterization of CD-Gel loaded with a combination of hydrophilic (DOX) and hydrophobic (CEL) drugs suggested that the gel can maintain the elasticity on drug entrapment (Fig. 3C–F and Fig. S3). Analysis of rheological parameters such as crossover of G′ and G′′ over applied strain supported the fact that CEL being a rigid scaffold enhanced the elasticity of the A13 hydrogel due to hydrophobic interactions (Fig. 3D and Fig. S3A). C-Gel showed a crossover point at 64.7% as compared to the A13 hydrogel (5.7%) (Fig. 3D), and D-Gel showed a crossover point at 10.5% of strain applied (Fig. 3E and Fig. S3). Entrapment of a combination of CEL and DOX (CD-Gel) maintained the viscoelastic nature of the A13 hydrogel (Fig. 3F). Biomaterials with appropriate stiffness having G′ > G′′ offer the possibility of in vivo delivery of the desired payload by syringe injection. We observed no crossover of G and G′′ at higher strain suggesting the ability of the gel to maintain injectable properties (Fig. 3F and Fig. S3C). The injectable nature of CD-Gel is further supported by the comparable elastic modulus and gel disintegration properties at higher strain (51%) applied during amplitude and frequency sweep measurements. Scanning electron micrographs showed a porous filamentous mesh architecture of C-Gel, D-Gel, and CD-Gel (Fig. 3G–I).
image file: d0nr01066a-f3.tif
Fig. 3 (A) Molecular structure of the glycine–glycine conjugated lithocholic acid (LCA-GG, A13) amphiphile that can form the hydrogel. (B) Gelation properties of the LCA-GG (A13) amphiphile and its ability to entrap CEL (C-Gel), DOX (D-Gel), and a combination of CEL and DOX (CD-Gel). (C) Rheological characterization of A13-Gel, C-Gel, D-Gel, and CD-Gel. (D–F) Amplitude sweep profiles of C-Gel (D), D-Gel (E) and CD-Gel (F) on the entrapment of CEL, DOX, or a combination of CEL and DOX. (G–I) Scanning electron micrographs of C-Gel (G), D-Gel (H), and CD-Gel (I). (J and K) Hemolytic activities of C-Gel, D-Gel, and CD-Gel after the incubation of the respective gels with red blood cells for 2 h (J) and 6 h (K). (L) Drug release kinetics showing the % cumulative release of DOX and CEL from CD-Gel.

Next, we tested the biocompatibility of the C-Gel, D-Gel, and CD-Gel by quantification of the haemolysis after incubation with red blood cells for 2 and 6 h (Fig. 3J and K). We observed no significant haemolysis on incubation of RBCs with any of the gels suggesting the biocompatible nature of the gel (Fig. 3J and K). To decipher the in vitro drug release profile, we incubated the CD Gel with 1× PBS at 37 °C. We estimated the concentration of drugs in release media by LC-MS/MS in MRM mode. Being hydrophobic, CEL showed minimal release until 2 weeks, thereby suggesting the gel degradation-mediated release of CEL (Fig. 3L). However, DOX showed burst release within 24 h by diffusion from the gel matrix due to its high aqueous solubility, followed by the slow release (Fig. 3L). Therefore, these results confirm that CD-Gel can maintain the sequential and sustained release of DOX and CEL.

CD-Gel induces a synergistic effect in a murine tumor model

To test the biocompatibility of the CD-Gel on implantation in mice, we quantified the infiltration of leukocytes at the gel site after 7 and 14 days of implantation of CD-Gel by flow cytometry. A single cell suspension of the excised gel-tissue interface was stained with CD45, followed by flow cytometry analysis. We did not see any significant increase in the number of infiltrated leukocytes (Fig. 4A and Fig. S4A), thereby confirming the inability of the CD-Gel to induce any innate immune response.
image file: d0nr01066a-f4.tif
Fig. 4 (A) Percentage of infiltrated CD45+ cells at the gel injected site. (B) Schema showing the plan of anticancer studies in mice. (C) Tumor growth kinetics (mean ± SEM, n = 6 per group) of CT26 tumor-bearing mice, (D) final tumor volume (mean ± SEM, n = 6 per group) on day 20, (E) Kaplan–Meier curve showing the percentage survival of mice, and (F) change in body weight of mice on different treatments. (G–I) Flow cytometric analysis for % apoptotic cells (G), proliferating cells (H), and leukocytes (I) on different treatments (mean ± SD, n = 4 per group). Data were analyzed using two-way ANOVA (C and F), unpaired two-tailed Student's t-test (D and G–I) and log-rank Mantel-Cox test (E). UT means untreated.

We tested the effect of the combination of CEL and DOX on tumor regression in a syngeneic colon tumor model in BALB/c mice using CT26 cells (Fig. 4B). Tumor-bearing mice were randomized into four different groups where group 1 mice were untreated. Group 2 mice were treated with a single dose of C-Gel (200 μL) near the tumor site carrying 5 mg kg−1 of CEL (100 μg). A single dose of D-Gel (200 μL) carrying 5 mg kg−1 of DOX (100 μg) was administered near the tumor site to group 3 mice. Mice in group 4 were treated with CD-Gel (200 μL) having a combination of CEL (100 μg) and DOX (100 μg) (Fig. 4B). Tumor growth kinetic curves showed a significant reduction of tumor growth on CD-Gel treatment as compared to untreated, C-Gel, and D-Gel treatments (Fig. 4C). CD-Gel treatment showed a ∼10-fold decrease in tumor volume as compared to untreated mice after 20 days (Fig. 4D). We observed a 4-fold reduction in tumor volume on CD-Gel treatment as compared to C-Gel and a ∼2-fold reduction in tumor volume as compared to D-Gel treated mice after 20 days (Fig. 4D). Survival studies showed a 5-day increase in median survival of mice on CD-Gel treatment as compared to untreated, C-Gel and D-Gel treatments (Fig. 4E). Change in body weight on different treatment showed only ∼5% decrease of body weight on CD-Gel treatment as we used sub-toxic dose of the drugs (Fig. 4F). Therefore, these results confirm that the combination of CEL and DOX is highly effective in tumor regression with an increase in survivability in a murine colon tumor model.

Next, we quantified the effect of C-Gel, D-Gel, and CD-Gel on proliferation, apoptosis and inflammation at the tumor site by flow cytometry, where single-cell suspension of tumor tissues from different treatment groups were stained with specific antibodies and analysed (Fig. S4B). CD-Gel induced a significant increase in the number of apoptotic cells (Fig. 4G) along with a concomitant decrease in the number of proliferating cells (Fig. 4H) as compared to other treatments. Staining of single cells with CD45 specific antibody showed a significant reduction in % leukocytes in the tumor microenvironment on CD-Gel treatment (Fig. 4I).

To elucidate the effect of sequential and sustained release of DOX and CEL near the tumor site, we compared the effect of CD-Gel with the localized (tumor site) delivery of the combination of DOX and CEL (CD-TS) without using the hydrogel as it would not be able to maintain sequential and sustained release of drugs. Tumor growth kinetic showed significant reduction in growth kinetics on CD-Gel treatment as compared to CD-TS treated mice (Fig. S5A and B). We observed ∼4.8-fold and ∼2-fold decrease in CD-Gel treated mice as compared to untreated and CD-TS treated mice respectively (Fig. S5C) without a major change in the body weight of mice (Fig. S5D).

CD-Gel induces the upregulation of ceramides and downregulation of glucosylceramides

Ceramides, glucosylceramides, and sphingomyelins are critical lipids in the sphingolipid metabolic pathway where increased ceramide levels, generated by the de novo pathway or by the salvage pathway, are known to enhance apoptosis.35 In contrast, increased glucosylceramide levels mediate enhanced expression of multidrug resistance genes.36 The de novo pathway involves the addition of different fatty acid chains to the sphinganine backbone mediated by ceramide synthases (CerS1CerS6). The salvage pathway allows for the hydrolytic degradation of sphingomyelins and glucosylceramides to ceramides.27 DOX is known to induce the activation of ceramide synthases leading to the enhanced expression of ceramides.28,29 Sub-toxic levels of DOX can cause activation of sphingomyelinase that results in the degradation of sphingomyelins to ceramides followed by ceramide-enriched membrane domains facilitating the DR5 clustering and apoptosis.30 Therefore, to decipher the effect of CD-Gel on the sphingolipid metabolism, we quantified the changes in different sphingolipid species in response to C-Gel, D-Gel and CD-Gel treatments using LC-MS/MS (Tables S2 and S3). Heat map showing the normalized levels of ceramides, glucosylceramides and sphingomyelins showed a clear increase in the levels of ceramides with a concomitant decrease in glucosylceramides, whereas sphingomyelins did not show any significant alterations (Fig. 5A, Table S3). There was a 1.5–2-fold increase in the level of C14 (p = 0.086), C18 (p < 0.05) and C20 (p < 0.05) ceramide species on CD-Gel treatment as compared to untreated tumor tissues (Fig. 5B–D). In contrast, C-Gel and D-Gel did not show any significant change in ceramides (Fig. 5B–D). We observed a 2–3-fold decrease in the expression of C22, C24 and C24:1 glucosylceramide species on CD-Gel treatment as compared to untreated tumors (Fig. 5E–G). Therefore, these results suggest that the combination of CEL and DOX in CD-Gel can induce an increase in ceramides that may be responsible for their synergistic effect on tumor regression.
image file: d0nr01066a-f5.tif
Fig. 5 (A) Heat map representing the altered profile of fatty acyl chain-specific ceramides, glucosylceramides, sphingomyelins and lactosylceramides on C-Gel, D-Gel, and CD-Gel treatments as compared to untreated tumors (UT). 1–4 represent four different mice in a group. (B–D) Quantification of chain specific ceramides (pmol mg−1 protein) (mean ± SD, n = 4) in UT, C-Gel, D-Gel, and CD-Gel treated tumors. (E–G) Quantification of chain specific glucosylceramides (pmol mg−1 protein) (mean ± SD, n = 4) in UT, C-Gel, D-Gel and CD-Gel treated tumors. Data were analyzed by unpaired two-tailed Student's t-test (B–G).

CD-Gel induces the overexpression of ceramide synthases

Next, we investigated the effect of C-Gel, D-Gel and CD-Gel on the expression of genes in the sphingolipid biosynthetic pathway.37 Tumor-bearing mice were subjected to different treatments (untreated, C-Gel, D-Gel and CD-Gel) and changes in the expression of sphingolipid-metabolizing genes were quantified from RNA isolated from tumor tissues after 20 days (Fig. 6A, Table S4). Ceramide synthases (Cers1Cers6) help in tethering of specific alkyl chains on the sphinganine backbone to generate ceramides (Fig. 6B).38 Among ceramide synthase genes, we observed a ∼5-fold increase in the expression of Cers1, >2.5-fold increase in the expression of Cers4, and >2-fold increase in the expression of CerS6 in response to CD-Gel treatment as compared to untreated tumors (Fig. 6A, C–E, Table S4). In contrast, C-Gel and D-Gel did not induce any significant change in the expression of ceramide synthases. These results suggest that combination therapy upregulates the de novo synthesis of ceramides via the enhanced expression of ceramide synthases 1, 4, and 6 as we observed a concomitant increase in C14:0 (short-chain), C18:0 and C20:0 (long-chain) ceramides. This increase in ceramides may be responsible for inducing apoptosis and inhibiting cell proliferation of tumor cells.39–41
image file: d0nr01066a-f6.tif
Fig. 6 (A) qRT-PCR based quantification showing fold change in the expression of different sphingolipid genes on C-Gel, D-Gel and CD-Gel treatments. (B) Schematic presentation showing the sphingolipid biosynthetic pathway. (C–H) qRT-PCR based changes in the expression of ceramide synthase 1 (Cers1), 4 (Cers4), 6 (Cer6), and glucosylceramide synthase (Ugcg), beta-1,4-galactosyltransferase 6 (B4galnt6) (G) and galactosidase beta 1 (Glb1) (H) on C-Gel, D-Gel, and CD-Gel treatments.

Ceramidases are key enzymes of the sphingolipid salvage pathway where acid ceramidase 1 (Asah1), neutral ceramidase (Asah2), and alkaline ceramidase 1 (Acer1/Asah3), 2 (Acer2/Asa3L) and 3 (Acer3/Phca) help in breaking ceramides to sphingosine (Fig. 6B).42 We observed that CD-Gel induces a ∼2.5-fold increase in the expression of Acer1 that catalyses the breakdown of only very long-chain ceramides (>C24:1) to sphingosine (Fig. 6A, Table S4).42 There was no significant change in any other ceramidases, thereby reducing the chances of ceramides getting hydrolysed. Therefore, the ceramide levels are maintained at high concentrations at the tumor site by treatment with a combination of CEL and DOX. Different acid and neutral sphingomyelinases (Smpds) hydrolyse the sphingomyelins to ceramides, thereby elevating the pool of ceramides.43 We observed that CD-Gel did not induce any significant change in the expression of any of the Smpd genes, whereas D-Gel decreased the expression of Smpd2 and Smpd3, and C-Gel enhanced the expression of Smpd3 and Smpd5 (Fig. 6A, Table S4). No significant upregulation of any of the Smpd genes on CD-Gel treatment suggest that the sphingomyelin hydrolysis pathway is not involved for enhanced ceramide levels.44

Sphingosine kinase 1 and 2 (Sphk1 and Sphk2) cause the phosphorylation of sphingosine to sphingosine-1-phosphate which plays a crucial role in proliferation, inflammation, and migration.45 We observed that CD-Gel did not induce any significant change in the expression of Sphk1 and Sphk2 (Fig. 6A, Table S4). Glucosylceramides are synthesized from ceramides by glucosylceramide synthase (Gcs) and can be hydrolysed back to ceramides by glucosidases (Gba1) (Fig. 6B).46 Glucosylceramides are key sphingolipids that generate lactosylceramides by B4galt6-mediated galactose transfer followed by synthesis of higher gangliosides. Upregulation of glucosylceramides can potentially promote the synthesis of higher gangliosides that are known to play a role in acquired drug resistance through activation of multi-drug resistance genes.47 We observed that D-Gel, C-Gel, and CD-Gel induced a ∼2-fold decrease in the expression of Ugcg accountable for the synthesis of glucosylceramides (Fig. 6F). A significant decrease in the level of glucosylceramides on CD-Gel treatment validates the gene expression studies (Fig. 5E–G). As DOX is known to increase the expression of Ugcg and induce drug resistance, slow release of DOX and CEL probably helps in lowering the expression of Ugcg, thereby not allowing the tumor to develop drug resistance (Fig. 6F). We observed a 2-fold decrease in the expression of B4galt6 (Fig. 6G), and a ∼2-fold increase in the expression of Glb1 (Fig. 6H), thereby suggesting no significant change in levels of lactosylceramides which is also confirmed by lipid quantitation (Fig. 5A). Therefore, gene expression studies validated by lipid profiling confirm that CD-Gel activates the de novo synthesis of ceramides probably responsible for apoptosis at the tumor site by upregulating the gene expression of ceramide synthases.

Conclusions

In summary, we screened the effect of the combination of DOX and CEL at different drug concentrations against colon cancer cell lines. We observed that specific combinations of CEL and DOX induce a synergistic apoptotic effect on colon cancer cells. We showed that a lithocholic acid-dipeptide conjugate based hydrogel could entrap the desired concentration of a combination of CEL and DOX. We further demonstrated that hydrogel-mediated delivery of a combination of CEL and DOX near the tumor site induces a synergistic effect on tumor regression, and causes an increase in the median survival of mice. Lipidomics studies validated by gene expression analysis confirmed that the combination of CEL and DOX induces an increase in the level of ceramides through activation of ceramide synthases that contribute to an increase in apoptosis. In conclusion, we showed a hydrogel-mediated synergistic therapeutic effect of the combination of CEL and DOX against colon cancer through an increase in ceramide levels that opens up prospects for the use of a cocktail of CEL with other chemotherapeutics. Our data also suggest that the modulation of the sphingolipid pathway is one of the effective strategies that combination therapies employ to target tumor regression, and ceramide synthases have the potential to be used as new targets for cancer therapy.

Experimental section

Materials

Doxorubicin, DiOC(6)3, celastrol, propidium iodide, DMEM media, antibiotics, and Annexin-FITC apoptosis kit were purchased from Sigma. FBS was purchased from HyClone. MS Grade methanol, acetonitrile, water and chloroform were purchased from Honeywell. Formic acid and ammonium formate were purchased from FLUKA. Glacial acetic acid was from Merck. Kinetex® UPLC C8, 2.1 × 50 mm column was purchased from Phenomenex®. Ceramide/Sphingoid Internal Standard Mixture II ((Cat #LM6005) was obtained from Avanti Polar Lipids, USA. iScript cDNA synthesis kit, iTaq Universal SYBR Green Supermix, Hard-Shell 96 well thin wall PCR plate, and Microseal® ‘B’ seal were purchased from Bio-Rad.

Cell culture

CT26, HCT-8 and DLD-1 cells were cultured in RPMI-1640 media, and HCT-116 cells were cultured in McCoy's medium where media were supplemented with 10% FBS, penicillin (100 U mL−1) and streptomycin (100 μg mL−1). Cells were maintained in a humidified environment at 37 °C under 5% CO2.

Cytotoxicity assay

Cells were seeded in 96 well plates for 24 h at a density of 5000 cells per well and treated with doxorubicin and celastrol at different concentrations. After 48 h of incubation, cell viability was measured using MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide) assay.48 Single drug and combination drug treatments were performed, and CI values were calculated with the help of CalcuSyn software (Biosoft).32 Combination index (CI) was calculated using the equation:
image file: d0nr01066a-t1.tif

In this equation, D1 and D2 are the concentrations of drug 1 (doxorubicin) and drug 2 (celastrol) alone at which drugs produce 50% effect on cancer cells whereas d1 and d2 are the doses of drugs 1 (doxorubicin) and 2 (celastrol) used in combination producing the same result.

Apoptosis assay

Cells (2.5 × 105 cells per well) were seeded in a 6 well plate for 24 h and treated with celastrol (0.5 or 1.0 μM), doxorubicin (0.5 or 1.0 μM), and combination of celastrol and doxorubicin (0.5 or 1.0 μM each) for 48 h. Cells were then stained with the Annexin V FITC/propidium iodide (PI) kit and analysed by flow cytometry. Data were collected and analyzed with BD FACSuite software.

Mitochondrial membrane potential assay

CT26 cells (0.5 × 106 per well) were seeded in a 12 well plate for 24 h, and treated with celastrol (0.5 or 1.0 μM), doxorubicin (0.5 or 1.0 μM) and a combination of celastrol and doxorubicin (0.5 or 1.0 μM each). After 24 h of treatment, cells were harvested, washed, and incubated in 1 mL of 0.1 μM DiOC(6)3 for 30 min at 37 °C in the dark. Cells were washed twice with DPBS and data were collected by BD FACS verse and analyzed with FlowJo_V10.7.

For microscopy, cells were washed twice with DPBS and counterstained with 0.2 mL of Hoechst 33258 (1[thin space (1/6-em)]:[thin space (1/6-em)]5000 dilution from 5 mg mL−1 stock) for 1 min, and washed twice with DPBS. Coverslips were taken out of wells and mounted on slides with Prolong Gold, and kept in the dark overnight to dry. Images were taken on Leica TCS SP5 at 40× oil and processed using LAS AF software.

Characterization of drug-loaded hydrogels

For drug encapsulation studies, the A13 hydrogelator (7% w/v) was mixed with either a single or a combination of celastrol and doxorubicin hydrochloride in water (1 mL). The resultant solution was heated until a clear solution was obtained using a hot air gun. Vials were allowed to cool at room temperature to form respective C-Gel, D-Gel, and CD-Gel. Rheological characterization of all hydrogels and scanning electron microscopy studies were performed as per the published protocol.20

Drug release studies

The concentrations of doxorubicin and celastrol released from the hydrogel were determined by LC-MS-MS. Briefly, CD-Gel (1 mL) was incubated with 2 mL of PBS (pH 7.4 with 0.05% Tween 80) at 37 °C. Release media were taken and replaced by an equivalent volume (1 mL) at predetermined time points and dried using a high vacuum speed rotator. Samples were resuspended in a mixture of acetonitrile and water (2[thin space (1/6-em)]:[thin space (1/6-em)]1) and sonicated for 10 minutes. Each sample was diluted 2000 times in methanol. Each independent biological replicate with three technical replicates of each sample was run with two blank runs between the samples. Standard curves were generated using different concentration ranges (0.078125 ng ml−1 to 200 ng ml−1) of each drug. Quantitation of doxorubicin and celastrol in each sample was performed using a Linear Ion Trap Quadrupole (QTRAP 4500, SCIEX, USA) LCMS/MS system coupled with a Turbo V™ source and electrospray ionization (ESI) probe in positive ion mode. Multiple reaction monitoring (MRM) with enhanced product ion (EPI) mode was used for selective quantitation, where Q1/Q3 transition 544.2/396.9 for doxorubicin and 451.3/201.0 for celastrol were optimized. Liquid chromatographic separation was performed on high pressure UHPLC (ExionLC™ AC, SCIEX, USA) using a Phenomenex® C8 column, 2.1 × 50 mm, 1.7 μm particle size. Solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in methanol) were used as the mobile phase for liquid chromatographic separation with the column temperature fixed at 60 °C. A gradient flow of solvent A and B (80[thin space (1/6-em)]:[thin space (1/6-em)]20) for 0.0 to 1.5 min, solvent A[thin space (1/6-em)]:[thin space (1/6-em)]B (20[thin space (1/6-em)]:[thin space (1/6-em)]80) from 1.50 to 4.1 min, solvent A[thin space (1/6-em)]:[thin space (1/6-em)]B (10[thin space (1/6-em)]:[thin space (1/6-em)]90) from 4.1 to 8.1 min, and solvent A[thin space (1/6-em)]:[thin space (1/6-em)]B (80[thin space (1/6-em)]:[thin space (1/6-em)]20) from 8.1 to 10 min was used. The flow rate was 0.3 mL min−1 for a total run time of 10 minutes.

Ethics statement

We used BALB/c or C57/BL6 mice (6–8 weeks old, weighing 18–20 g). All the animal experiments were carried out at the small animal facility of Regional Centre for Biotechnology, Faridabad as per the guidelines of Department of Biotechnology and Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA) Govt. of India, and after due approval of the protocols (RCB/IAEC/2016/002, RCB/IAEC/2016/011) by Institutional Animal Ethical Committee of Regional Centre for Biotechnology. Impact of the CD-Gel on the infiltration of leukocytes in mice was studied as per a previous protocol.20

Anticancer activities

All anticancer studies were performed using the syngeneic colon carcinoma (CT26) model in BALB/c mice. We shaved the flank region of mice to remove hair and injected CT26 cells (1.5 × 106) suspended in FBS[thin space (1/6-em)]:[thin space (1/6-em)]Matrigel (1[thin space (1/6-em)]:[thin space (1/6-em)]1, 200 μL per mouse) subcutaneously in the flank region of mice. We randomized the mice after 3 days when tumors reached ∼30–40 mm3 into four different groups (n = 7 per group), and subjected them to different treatments. There was no treatment given to mice in group 1. We gave a single subcutaneous injection of celastrol (5 mg kg−1) entrapped in A13 gel and C-Gel (200 μL of 70 mg mL−1) near the tumor site of mice in group 2, and of doxorubicin (5 mg kg−1), D-Gel (200 μL of 70 mg mL−1) to mice in group 3. We injected a single subcutaneous dose of the combination of celastrol (5 mg kg−1) and doxorubicin (5 mg kg−1) entrapped in A13 gel (called CD-Gel, 200 μL) near the tumor site of group 4 mice. Tumor volume and body weight of mice were measured on alternate days. Tumor volume was calculated using the formula V = W2 × L/2, where W and L refer to the shortest and the longest diameters. Mice were sacrificed on day 21, and tumor tissues were excised for lipidomics and gene expression studies. Another set of mice were observed for survival analysis.

The percentage of apoptotic cells, proliferating cells, and leukocytes after different treatments from tumor tissues was quantified by flow cytometry as per a previous protocol.20

Lipidomics studies

Tumor tissues (∼50 mg) were homogenized by ceramic beads in 2 mL homogenizing tubes (Precellys homogenizer, Bertin Technologies, France) in PBS (200 μL) and further extraction was done as per a previous protocol.20 All parameters for LC-MS/MS analysis and quantification of lipids were performed as described previously with some changes as mentioned here.20 Sphingolipids were analysed by ultra-high-pressure liquid chromatography (ExionLC, SCIEX, USA) using a Kinetex® C8 (2.1 × 50 mm) column (Phenomenex®, USA), with a particle size of 1.7 μm and the oven temperature was maintained at 60 °C, coupled to a hybrid triple quadrupole/linear ion trap mass spectrometer (4500 Q TRAP, SCIEX, USA). All other parameters and data analysis methods were as described previously.20

Quantitative real-time PCR

For real-time PCR, cDNA was synthesized using 1 μg of RNA using the iScript cDNA synthesis kit (Bio-Rad). Gene expression analysis was done by quantitative real-time PCR (qRT-PCR) using iTaq Universal SYBR Green Supermix (Bio-Rad) on the AriaMax real-time PCR system (Agilent Technologies, USA) as described previously.20 Relative quantitation of gene expression was done using β-actin as the endogenous reference gene for normalization. All primers sequences used for qRT-PCR are listed in Table S5.

Author contributions

NM performed all in vitro studies and lipidomics studies. MNA performed all gene expression studies and analyzed the data. SK performed and analyzed the drug release, rheology and SEM experiments. SP and KR performed all animal experiments. SJ and PV performed biocompatibility studies. AB supervised animal studies. UD supervised lipidomics and gene expression studies. AB and UD wrote the manuscript, conceived the idea, and supervised the whole project.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

We thank Regional Centre for Biotechnology, and Amity University Haryana for intramural funding. AB thanks Department of Biotechnology, Department of Science and Technology, and Science and Engineering Research Board for financial support. UD thanks Department of Biotechnology (BT/PR19624/BIC/101/488/2016), and Department of Science and Technology, New Delhi (ECR/2016/001603) for financial support. We thank the small animal facility of Regional Centre for Biotechnology funded by DBT (BT/PR5480/INF/22/158/2012). We thank Amity Lipidomics Research Facility at Amity University Haryana funded by the DST-FIST grant (SR/FST/LSI-664/2016). We thank UGC (SP) and RCB (NM, SK) for research fellowships to students. We acknowledge the support of the DBT e-Library Consortium (DeLCON) for providing access to e-resources.

References

  1. M. O. Palumbo, P. Kavan, W. H. Miller Jr., L. Panasci, S. Assouline, N. Johnson, V. Cohen, F. Patenaude, M. Pollak, R. T. Jagoe and G. B. Front, Front. Pharmacol., 2013, 4, 57 CrossRef .
  2. Q. Hu, W. Sun, C. Wang and Z. Gu, Adv. Drug Delivery Rev., 2016, 98, 19–34 CrossRef CAS .
  3. R. B. Mokhtari, T. S. Homayouni, N. Baluch, E. Morgatskaya, S. Kumar, B. Das and H. Yeger, Oncotarget, 2017, 8, 38022–38043 CrossRef .
  4. D. Miles, G. von Minckwitz and A. D. Seidman, Oncologist, 2002, 7, 13–19 CrossRef CAS .
  5. A. D. Kinghorn, Y. Chin and S. M. Swanson, Curr. Opin. Drug Discovery Dev., 2009, 12, 189–196 CAS .
  6. H.-Y. Li, J. Zhang, L. L. Sun, B.-H. Li, H.-L. Gao, T. Xie, N. Zhang and Z.-M. Ye, Cell Death Dis., 2015, 6, e1604 CrossRef CAS .
  7. S. Boridy, P. U. Le, K. Petrecca and D. Maysinger, Cell Death Dis., 2014, 5, e1216 CrossRef CAS .
  8. J. Liu, K. Lee, M. A. S. Hernandez, R. Mazitschek and U. Ozcan, Cell, 2015, 161, 999–1011 CrossRef CAS .
  9. X. Ma, L. Xu, A. T. Alberobello, O. Gavrilova, A. Bagattin, M. Skarulis, J. Liu, T. Finkel and E. Muller, Cell Metab., 2015, 22, 695–708 CrossRef CAS .
  10. B. Peng, L. Xu, F. Cao, T. Wei, C. Yang, G. Uzan and D. Zhang, Mol. Cancer, 2010, 9, 79 CrossRef .
  11. S. M. Raja, R. J. Clubb, C. Ortega-Cava, S. H. Williams, T. A. Bailey, L. Duan, X. Zhao, A. L. Reddi, A. M. Nyong, A. Natarajan, V. Band and H. Band, Cancer Biol. Ther., 2011, 11, 263–276 CrossRef CAS .
  12. S. Wang, K. Ma, C. Zhou, Y. Wang, G. Hu, L. Chen, Z. Li, C. Hu, Q. Xu, H. Zhu, M. Liu and N. Xu, Ther. Adv. Med. Oncol., 2019, 10, 1–19 Search PubMed .
  13. L. Lin, Y. Sun, D. Wang, S. Zheng, J. Zhang and C. Zheng, Front. Pharmacol., 2016, 6, 320 Search PubMed .
  14. Y. Xiao, J. Liu, M. Guo, H. Zhou, J. Jin, J. Liu, Y. Liu, Z. Zhang and C. Chen, Nanoscale, 2018, 10, 12639–12649 RSC .
  15. S. Sengupta and A. Kulkarni, ACS Nano, 2013, 7, 2878–2882 CrossRef CAS .
  16. N. Joshi, J. Yan, S. Levy, S. Bhagchandani, K. V. Slaughter, N. E. Sherman, J. Amirault, Y. Wang, L. Riegel, X. He, T. S. Rui, M. Valic, P. J. Vemula, O. R. Miranda, O. Levy, E. M. Gravallese, A. O. Aliprantis, J. Ermann and J. M. Karp, Nat. Commun., 2018, 9, 1275 CrossRef .
  17. T. Gajanayake, R. Olariu, F. M. Leclere, A. Dhayani, Z. Yang, A. K. Bongoni, Y. Banz, M. A. Constantinescu, J. M. Karp, P. K. Vemula, R. Rieben and E. Vogelin, Sci. Transl. Med., 2014, 6, 249ra110 CrossRef .
  18. M. Singh, S. Kundu, M. A. Reddy, V. Sreekanth, R. K. Motiani, S. Sengupta, A. Srivastava and A. Bajaj, Nanoscale, 2014, 6, 12849–12855 RSC .
  19. S. Zhang, J. Ermann, M. D. Succi, A. Zhou, M. J. Hamilton, B. Cao, J. R. Korzenik, J. N. Glickman, P. K. Vemula, L. H. Glimcher, G. Traverso, R. Langer and J. M. Karp, Sci. Transl. Med., 2015, 7, 300ra128 CrossRef .
  20. S. Pal, N. Medatwal, S. Kumar, A. Kar, V. Komalla, P. S. Yavvaari, D. Mishra, Z. A. Rizvi, S. Nandan, D. Malakar, M. Pillai, A. Awasthi, P. Das, R. D. Sharma, A. Srivastava, S. Sengupta, U. Dasgupta and A. Bajaj, ACS Cent. Sci., 2019, 5, 1648–1662 CrossRef CAS .
  21. Y. A. Hannun, Science, 1996, 274, 1855–1859 CrossRef CAS .
  22. T. I. Gudz, K. Y. Tserng and C. L. Hoppel, J. Biol. Chem., 1997, 272, 24154–24158 CrossRef CAS .
  23. C. L. Chen, C. F. Lin, W. T. Chang, W. C. Huang, C. F. Teng and Y. S. Lin, Blood, 2008, 111, 4365–4374 CrossRef CAS .
  24. A. Carracedo, M. Lorente, A. Egia, C. Blazquez, S. Garcia, V. Giroux, C. Malicet, R. Villuendas, M. Gironella, L. Gonzalez-Feria, M. A. Piris, J. L. Iovanna, M. Guzman and G. Velasco, Cancer Cell, 2006, 9, 301–312 CrossRef CAS .
  25. C. E. Chalfant, K. Rathman, R. L. Pinkerman, R. E. Wood, L. M. Obeid, B. Ogretmen and Y. A. Hannun, J. Biol. Chem., 2002, 277, 12587–12595 CrossRef CAS .
  26. A. Massiello and C. E. Chalfant, J. Lipid Res., 2006, 47, 892–897 CrossRef CAS .
  27. M. Stevens and S. Oltean, Front. Genet., 2019, 10, 804 CrossRef CAS .
  28. V. del Solar, D. Y. Lizardo, N. Li, J. J. Hurst, C. J. Brais and G. E. Atilla-Gokcumen, Chem. Biol., 2015, 22, 1662–1670 CrossRef CAS .
  29. G. A. Patwardhan and Y. Y. Liu, Prog. Lipid Res., 2011, 50, 104–114 CrossRef CAS .
  30. C. A. Dumitru, A. Carpinteiro, T. Trarbach, U. R. Hengge and E. Gulbins, Apoptosis, 2007, 12, 1533–1541 CrossRef CAS .
  31. J. Foucquier and M. Guedj, Pharmacol. Res. Perspect., 2015, 3, e00149 CrossRef .
  32. I. V. Bijnsdorp, E. Giovannetti and G. J. Peters, Methods Mol. Biol., 2011, 731, 421–434 CrossRef CAS .
  33. T. C. Chou, Cancer Res., 2010, 70, 440–446 CrossRef CAS .
  34. H. Chang, H. Huang, T. Huang, P. Yang, Y. Wang and H. Juan, Bio-Protoc., 2013, 3, 1 Search PubMed .
  35. B. Ogretmen and Y. A. Hannun, Nat. Rev. Cancer, 2004, 4, 604–616 CrossRef CAS .
  36. S. A. F. Morad and M. C. Cabot, Nat. Rev. Cancer, 2013, 13, 51–65 CrossRef CAS .
  37. C. R. Gault, L. M. Obeid and Y. A. Hannun, Adv. Exp. Med. Biol., 2010, 688, 1–23 CrossRef CAS .
  38. M. S. Wegner, S. Schiffman, M. J. Parnham, G. Geisslinger and S. Grösch, Prog. Lipid Res., 2016, 63, 93–119 CrossRef CAS .
  39. S. Spassieva, J. G. Seo, J. C. Jiang, J. Bielawski, F. Alvarez-Vasquez, S. M. Jazwinski, Y. A. Hannun and L. M. Obeid, J. Biol. Chem., 2006, 281, 33931–33938 CrossRef CAS .
  40. T. D. Mullen, S. Spassieva, R. W. Jenkins, K. Kitatani, J. Bielawski and Y. A. Hannun, J. Lipid Res., 2011, 52, 68–77 CrossRef CAS .
  41. D. Hartmann, J. Lucks, S. Fuchs, S. Schiffmann, Y. Schreiber, N. Ferreiros, J. Merkens, R. Marschalek, G. Geisslinger and S. Grosch, Int. J. Biochem. Cell Biol., 2012, 44, 620–628 CrossRef CAS .
  42. N. Coant, W. Sakamoto, C. Mao and Y. A. Hannun, Adv. Biol. Regul., 2017, 63, 122–131 CrossRef CAS .
  43. F. M. Goni and A. Alonso, FEBS Lett., 2002, 531, 38–46 CrossRef CAS .
  44. D. Hartmann, J. Lucks, S. Fuchs, S. Schiffmann, Y. Schreiber, N. Ferreiros, J. Merkens, R. Marschalek, G. Geisslinger and S. Grosch, Int. J. Biochem. Cell Biol., 2012, 44, 620–628 CrossRef CAS .
  45. L. A. Heffernan-Stroud and L. M. Obeid, Adv. Cancer Res., 2013, 117, 201–235 CrossRef CAS .
  46. A. Senchenkov, D. A. Litvak and M. C. Cabot, J. Natl. Cancer Inst., 2001, 93, 347–357 CrossRef CAS .
  47. Y. Y. Liu, V. Gupta, G. A. Patwardhan, K. Bhinge, Y. Zhao, J. Bao, H. Mehendale, M. C. Cabot, Y. T. Li and S. M. Jazwinski, Mol. Cancer, 2010, 9, 145 CrossRef .
  48. V. Sreekanth, N. Medatwal, S. Kumar, S. Pal, V. Malyla, A. Kar, P. Bhargava, A. Naaz, N. Kumar, S. Sengupta and A. Bajaj, Bioconjugate Chem., 2017, 28, 2942–2953 CrossRef CAS .

Footnotes

Electronic supplementary information (ESI) available: Supplementary figures and tables. See DOI: 10.1039/d0nr01066a
These authors contributed equally.

This journal is © The Royal Society of Chemistry 2020