C6 Ceramide (d18:1/6:0) as a Novel Treatment of Cutaneous T Cell Lymphoma

Simple Summary There is no curative treatment for mycosis fungoides and Sézary syndrome, which are the most frequent forms of cutaneous T cell lymphoma (CTCL). Short-chain ceramides like C6 Ceramide are known to induce cell death by both apoptosis and necrosis. Here, we demonstrate that C6 Ceramide strongly reduced cell viability and induced cell death in CTCL cell lines but not in HaCaT keratinocytes and primary human keratinocytes. C6 Ceramide was rapidly metabolized by both keratinocyte cell types but not by CTCL cells. These results provide the basis for further clinical trials with topical applicated C6 Ceramide against mycosis fungoides and Sézary syndrome. Abstract Cutaneous T cell lymphomas (CTCLs) represent a heterogeneous group of T cell lymphomas that primarily affect the skin. The most frequent forms of CTCL are mycosis fungoides and Sézary syndrome. Both are characterized by frequent recurrence, developing chronic conditions and high mortality with a lack of a curative treatment. In this study, we evaluated the effect of short-chain, cell-permeable C6 Ceramide (C6Cer) on CTCL cell lines and keratinocytes. C6Cer significantly reduced cell viability of CTCL cell lines and induced cell death via apoptosis and necrosis. In contrast, primary human keratinocytes and HaCaT keratinocytes were less affected by C6Cer. Both keratinocyte cell lines showed higher expressions of ceramide catabolizing enzymes and HaCaT keratinocytes were able to metabolize C6Cer faster and more efficiently than CTCL cell lines, which might explain the observed protective effects. Along with other existing skin-directed therapies, C6Cer could be a novel well-tolerated drug for the topical treatment of CTCL.


Introduction
Cutaneous T cell lymphomas (CTCLs) are a group of extranodal non-Hodgkin T cell lymphomas that especially affect the skin and show an increasing standardized incidence of 1-9 per million people with regional variations [1,2]. The most frequent clinicopathological short-chain ceramides were used in one phase 2 study to treat cutaneous breast cancer by topical application [31].
Mycosis fungoides is histopathologically characterized by epidermotropism of malignant T cells and is therefore predestined for topical treatments [6]. Since short-chain ceramides are highly skin permeable, they are suitable for topical applications and therefore promising to treat these forms of CTCLs [32]. Hence, we wanted to analyze and compare the effects of C6Cer on keratinocytes and two CTCL cell lines, MyLa (as a model for mycosis fungoides) and HuT78 (as a model for Sézary syndrome), to provide the basis for the employment of C6Cer as a treatment for cutaneous cancer and further in vivo studies.

Effect of C6 Ceramide on Cell Viability of Keratinocytes and Cutaneous T Cell Lymphoma Cell Lines
We investigated the effect of C6Cer on the cell viability of primary human keratinocytes ( Figure 1A 46.7%, and 63.9%, respectively, and, after the treatment with 100 µM for 6 h, 16 h, and 24 h by 52.4%, 77.1%, and 79.8% respectively (mean reduction of cell viability after 6 h, 16 h, and 24 h (%)).

Influence of C6 Ceramide on Necrosis, Apoptosis, and Autophagy in MyLa and HuT78 Cells
Next, we investigated the details of the C6Cer triggered cytotoxic effect. Necrosis is a lytic cell death modality, accompanied by osmotic imbalances and early membrane ruptures. As an indicator for necrosis, we measured lactate dehydrogenase (LDH) release in the cell culture supernatant after treatment with 25, 50, and 100 µM C6Cer for 24 h (Figure 2A). In line with the results obtained by MTS assay, C6Cer treatment was cytotoxic for CTCL cells and HaCaT keratinocytes in a dose-dependent manner. After treatment with 25 µM C6Cer for 24 h, LDH assay results showed cytotoxicity of 30.9 ± 5.41% (MyLa), 48.8 ± 2.65% (HuT78), with 100 µM C6Cer for 24 h 56.5 ± 5.94% (MyLa) and 60.8 ± 4.35% (HuT78). In contrast, C6Cer treatment of HaCaT cells for 24 h led to cytotoxicity of only 9.16 ± 6.94% with 25 µM C6Cer and 28.9 ± 4.76% with 100 µM C6Cer treatment (mean ± SEM). As C6Cer led to low viability and high cytotoxicity in T cells but not in keratinocytes, we further investigated the dose-dependent effect of C6Cer on caspase-dependent poly (ADP-ribose) polymerase-1 (PARP1) cleavage as an indicator for apoptosis [33] in HuT78 ( Figure 2B) and MyLa ( Figure 2C) cells. Treatment with 25, 50, and 100 µM C6Cer for 24 h led to a significant increase of cleaved PARP1 (cPARP1) compared to the vehicle-treated control group. In contrast to the CTCL cells, HaCaT keratinocytes showed no significant increase in PARP1 cleavage ( Figure 2D). 1 µM staurosporine (STS), an extensively used model compound for induction of apoptosis, served as a positive control for caspase-3 induced apoptosis. The whole blots showing all the bands and molecular weight markers are shown in Figure S1. To distinguish between apoptosis and necrosis in CTCL cells, we analyzed Annexin V (ANXA5) staining and propidium iodide (PI) incorporation in MyLa and HuT78 cells by flow cytometry after treatment with 100 µM C6Cer for 24 h. A significant decrease of the ANXA-, PI-cell population (healthy), and a significant increase of the ANXA5+ and PI+ (late apoptotic/necrotic) cell populations were measured. This reflects a reduction of healthy cells and an increase of necrotic or late apoptotic cells in both cell lines after the treatment with 100 µM C6Cer for 24 h ( Figure 2E). There were no significant alterations in the early apoptotic cell population (ANXA+, PI-) after treatment with 100 µM C6Cer for 24 h. Furthermore, we looked for microtubule-associated proteins 1A/1B light chain 3B (LC3B) in T cells and HaCaT keratinocytes as a marker for autophagy flux and autophagosome biogenesis. T cells were treated with 25 µM C6Cer for 24 h and LC3B mRNA expression was measured by TaqMan ® . A significant increase of LC3B mRNA expression was detected after treatment with 25 µM C6Cer for 24 h in HuT78 cells but not in MyLa cells and HaCaT keratinocytes ( Figure 2F). Western Blot analysis of HuT78 cells treated with 25, 50, and 100 µM C6Cer for 24 h confirmed an increase of LC3B lipidation ( Figure 2G). The whole blots are shown in Figure S2. Taken together, C6Cer treatment led to necrosis and apoptosis in MyLa and HuT78 cells and autophagy in HuT78 cells but not in HaCaT keratinocytes.

Metabolism of C6 Ceramide in HaCaT Keratinocytes and Cutaneous T Cell Lymphoma Cell Lines
Since HaCaT keratinocytes proved to be more resistant against C6Cer compared to the CTCL cell line, we investigated the lipid metabolism and looked for differences in the efficacy of C6Cer degradation, which could protect against cell death. Therefore, Ha-CaT cells were incubated with 25 µM C6Cer for 0.5 h or 24 h and CTCL cell lines were treated with 12.5 µM C6Cer for the same time. C6Cer concentration was measured in the supernatant ( Figure 3A) and the cell pellet ( Figure 3B) by LC-MS/MS. There was a reduction of the C6Cer level in the supernatant after 24 h compared to 0.5 h of 51.9% (3113 ± 40.5 to 1495 ± 44.2 ng/mL) in HuT78 cells and 47.9% (3055 ± 13.3 to 1591 ± 54.0 ng/mL) in MyLa cells. In HaCaT cells, LC/MS-MS results showed a reduction of at least 98.3% (357 ± 37.3 to less than the lowest limit of quantification (LLOQ) of 6 ng/mL ± 0). The C6Cer level in the cell pellet of both CTCL cell lines after 0.5 h and 24 h were above the upper limit of quantification of 180 ng/sample. In HaCaT cells, LC/MS-MS results showed a reduction of 94.0% (7.5 ± 0.95 to 0.44 ± 0.05 ng/5 × 10 5 cells) (mean 0.5 h ± SEM to mean 24 h ± SEM). To elucidate if enzymatic equipment is the reason for the differences in ceramide catabolism between keratinocytes and CTCL cell lines, we measured basal mRNA expression of involved enzymes ( Figure 3C). Interestingly, mRNA expression of acid ceramidase (ASAH) 1 in primary human keratinocytes and HaCaT cells was threefold higher compared to T cells. Sphingosine kinase 1 (SphK1) mRNA was eightfold higher expressed in both keratinocyte cell lines compared to the CTCL cell lines. Sphingomyelin synthase (SGMS) 2 was barely expressed in T cells, whereas it was well expressed in both keratinocyte cell lines. However, absolute mRNA expressions of SPHK1 and SGMS2 were lower compared to ASAH1 expression. Ceramide glucosyltransferase (UGCG) was highly expressed in all cell lines except for MyLa cells. The mRNA expression of ASAH2, SPHK2, SGMS1, ceramide kinase (CERK), and alkaline ceramidase (ACER) 1-3 showed low expressions in all cell lines or no considerable distinctions between keratinocytes and CTCL cell lines ( Figure 3D).

Influence of C6 Ceramide on Necrosis, Apoptosis, and Autophagy in MyLa and HuT78 Cells
Next, we investigated the details of the C6Cer triggered cytotoxic effect. Necrosis is a lytic cell death modality, accompanied by osmotic imbalances and early membrane ruptures.

The Role of Acid Ceramidase 1 in C6 Ceramide Metabolism
As mass spectrometry results showed that the synthetical C6Cer was metabolized by the cells, we wanted to know whether the reason for the resistance of keratinocytes to C6Cer-induced cell death is due to the ceramide degrading enzyme ASAH1. HaCaT cells were treated with C6Cer for 24 h and ASAH 1 protein expression was measured. Interestingly, the protein expression of ASAH1 was significantly upregulated compared to the vehicle group ( Figure 4A). The whole blots are shown in Figure S3. Because ASAH1 could play a critical role in protecting keratinocytes from C6Cer-induced cell death, we knocked down the enzyme by siRNA. Cells were incubated with ASAH1 siRNA and mock siRNA for 24 h before they were stimulated with C6Cer for 24 h. Knockdown was validated by Western Blot analysis. The whole blots are shown in Figure S4. There were no significant differences in cell viability between mock and ASAH1 siRNA treated groups ( Figure 4B). To verify the data by a different approach, HaCaT cells were stimulated with 25 µM C6Cer, 15 µM of the highly potent ASAH1 inhibitor HCFU (1-hexylcarbamoyl-5-fluorouracil or Carmofur) or both. After an incubation time of 24 h, cell viability was determined by MTS assay. There was, again, no significant decrease in cell viability observed when treated with the ASAH1 inhibitor plus C6Cer (co: 100 ± 0; 25 µM C6Cer = 91.3 ±

The Role of Acid Ceramidase 1 in C6 Ceramide Metabolism
As mass spectrometry results showed that the synthetical C6Cer was metabolized by the cells, we wanted to know whether the reason for the resistance of keratinocytes to C6Cer-induced cell death is due to the ceramide degrading enzyme ASAH1. HaCaT cells were treated with C6Cer for 24 h and ASAH 1 protein expression was measured. Interestingly, the protein expression of ASAH1 was significantly upregulated compared to the vehicle group ( Figure 4A). The whole blots are shown in Figure S3. Because ASAH1 could play a critical role in protecting keratinocytes from C6Cer-induced cell death, we knocked down the enzyme by siRNA. Cells were incubated with ASAH1 siRNA and mock siRNA for 24 h before they were stimulated with C6Cer for 24 h. Knockdown was validated by Western Blot analysis. The whole blots are shown in Figure S4. There were no significant differences in cell viability between mock and ASAH1 siRNA treated groups ( Figure 4B). To verify the data by a different approach, HaCaT cells were stimulated with 25 µM C6Cer, 15 µM of the highly potent ASAH1 inhibitor HCFU (1-hexylcarbamoyl-5-fluorouracil or Carmofur) or both. After an incubation time of 24 h, cell viability was determined by MTS assay. There was, again, no significant decrease in cell viability observed when treated with the ASAH1 inhibitor plus C6Cer (co: 100 ± 0; 25 µM C6Cer = 91.3 ± 6.65%; 15 µM HCFU = 114 ± 6.72%; 25 µM C6Cer and 15 µM HCFU = 101 ± 5.79% (mean ± SEM)).

Discussion
To date, there is no curative treatment of mycosis fungoides or Sézary syndrome [9]. As CTCLs are chronic diseases with frequent recurrence, patients must undergo treatment for their entire remaining life. Therefore, it is important to find new efficient treatments.
In our study, we demonstrated the potential of short-chain C6Cer to treat mycosis fungoides and Sézary syndrome. C6Cer treatment led to apoptosis and necrosis in MyLa and HuT78 cell lines. HuT78 cells showed a dose-dependent increase of autophagic flux in presence of C6Cer, which indicates vigorous cell stress and damage induced by C6Cer [34,35]. The results obtained from LDH assay and flow cytometry analysis indicate that the major cause for cell death is due to necrosis (Figure 2). In our experiments a significant increase in apoptotic PARP cleavage was observed, and, therefore, it cannot be excluded that the detected necrosis represents a late phase of apoptotic cell death. The predominance of necrosis was already shown in C6Cer treated Jurkat cells, a T cell leukemia cell line [26].
Further, we showed that HaCaT cells and primary human keratinocytes were less sensitive to C6Cer compared to CTCL cells. No significant increase in PARP cleavage or autophagic flux was observed in HaCaT keratinocytes. In this context, we were able to find differences in the efficiency of catabolizing C6Cer between HaCaT keratinocytes and CTCL cell lines. For lipid analysis, we treated HaCaT keratinocytes with 25 µM C6Cer and CTCL cell lines only with 12.5 µM to save most cells from dying. Already after 30 min, HaCaT cells catabolized most of C6Cer and after 24 h the concentrations in supernatant and pellet were dramatically decreased. Although CTCL cell lines were stimulated with less C6Cer, the lipid concentrations measured in samples from CTCL cell lines were always higher compared to the values measured in samples from HaCaT cells. Hence, Ha-CaT keratinocytes might be less sensitive to C6Cer-induced cell death due to their ability to catabolize it more efficiently than CTCL cell lines. Although ASAH1 mRNA expression

Discussion
To date, there is no curative treatment of mycosis fungoides or Sézary syndrome [9]. As CTCLs are chronic diseases with frequent recurrence, patients must undergo treatment for their entire remaining life. Therefore, it is important to find new efficient treatments.
In our study, we demonstrated the potential of short-chain C6Cer to treat mycosis fungoides and Sézary syndrome. C6Cer treatment led to apoptosis and necrosis in MyLa and HuT78 cell lines. HuT78 cells showed a dose-dependent increase of autophagic flux in presence of C6Cer, which indicates vigorous cell stress and damage induced by C6Cer [34,35]. The results obtained from LDH assay and flow cytometry analysis indicate that the major cause for cell death is due to necrosis (Figure 2). In our experiments a significant increase in apoptotic PARP cleavage was observed, and, therefore, it cannot be excluded that the detected necrosis represents a late phase of apoptotic cell death. The predominance of necrosis was already shown in C6Cer treated Jurkat cells, a T cell leukemia cell line [26].
Further, we showed that HaCaT cells and primary human keratinocytes were less sensitive to C6Cer compared to CTCL cells. No significant increase in PARP cleavage or autophagic flux was observed in HaCaT keratinocytes. In this context, we were able to find differences in the efficiency of catabolizing C6Cer between HaCaT keratinocytes and CTCL cell lines. For lipid analysis, we treated HaCaT keratinocytes with 25 µM C6Cer and CTCL cell lines only with 12.5 µM to save most cells from dying. Already after 30 min, HaCaT cells catabolized most of C6Cer and after 24 h the concentrations in supernatant and pellet were dramatically decreased. Although CTCL cell lines were stimulated with less C6Cer, the lipid concentrations measured in samples from CTCL cell lines were always higher compared to the values measured in samples from HaCaT cells. Hence, HaCaT keratinocytes might be less sensitive to C6Cer-induced cell death due to their ability to catabolize it more efficiently than CTCL cell lines. Although ASAH1 mRNA expression was markedly upregulated in HaCaT keratinocytes after treatment with C6Cer, there was no impact on C6Cer-induced cell death by ASAH1 knockdown or pharmacological inhibition. However, other enzymes could compensate for the loss of ASAH1 and it is likely that several enzymes are involved in the metabolization of short-chain ceramides such as sphingomyelin synthases and ceramide glucosyltransferase [36]. Short-chain C6Cer is able to penetrate cells and skin and accumulates in cell membranes as well as in intracellular membranes [32,37]. Therefore, a topical C6Cer treatment of CTCL is technically possible. Topical application of short-chain ceramides as a treatment for cancer was already tested for C2Cer and C6Cer in phase 2 study of cutaneous breast cancer with twenty-five patients. Pruritis, rash, and dry skin were the most frequent side effects of short-chain ceramides, but no grade 3 or 4 toxicity was reported [31]. This is congruent with our in vitro data as HaCaT cells and primary human keratinocytes showed less cytotoxicity after C6Cer treatment compared to malignant T cells. Due to limited efficacy, the results of the cutaneous breast cancer trial were not promising enough to warrant further studies [31]. Because of high costs, they restricted their concentration to only a 1% mixture of C2Cer-and C6Cer [31] and transdermal drug application is limited by permeation of the stratum corneum, higher concentrations, and skin permeation enhancers, which induce a reversible decrease of the barrier resistance [38] and could be used to enhance the antitumoral effect of C6Cer in further in vivo trials.
Present topical therapy options consist above all of glucocorticoid and mechlorethamine [8,9]. Adverse effects of mechlorethamine include immediate hypersensitivity reactions (allergy), erythema, hyperpigmentation, and pruritis. Secondary malignancies like cutaneous melanoma, nonmelanoma skin cancer, and primary malignancies were also reported [39,40]. Glucocorticoids may trigger tachyphylaxis [41] and, as CTCL patients are treated for a long time, their therapeutic potential is therefore limited. Since glucocorticoids are an essential component in CTCL therapy [8], patients suffer from their local and systemic adverse effects. The most often adverse effects of topical steroids include cutaneous atrophy, local irritation, and skin depigmentation. Delayed wound healing and exacerbations of skin infections after treatment with glucocorticoids were also reported [42][43][44].
Therefore, it is necessary to find more efficient therapies with fewer adverse effects. From our data, we would suggest topical treatment with C6Cer as a potential topical treatment of mycosis fungoides and Sézary syndrome and as an alternative for patients with side effects or contraindications of other topical treatment options. Luckily adverse effects of C6Cer treatment were already investigated and seemed to be rather mild [31].
In conclusion, we could show that C6Cer significantly reduced cell viability and induced cell death in CTCL cell lines. Therefore, C6Cer offers a possible novel treatment of CTCL, which should be investigated in vivo. Furthermore, C6Cer treatment could be also studied for other skin associated diseases like dermal cancer and chronic inflammatory disorders with lymphocytic infiltrates.
As C6Cer treated cells were stimulated under aqueous cell culture conditions in this study, the results are limited to this in vitro model and the effect of topically applicated C6Cer on CTCLs has to be further investigated in vivo. In particular, it is questionable which concentrations can be applied for topical treatment of CTCLs. Furthermore, we focused on the comparison of CTCL cells with keratinocytes after the C6Cer treatment to elucidate the basis for clinical trials without affecting healthy skin. Therefore, this study is limited by the characterization of the detailed mechanism of C6Cer induced apoptosis in CTCL cell lines. We focused especially on the extrinsic pathway of apoptosis to compare the effect of C6Cer. Thus, we did not show the intrinsic pathway, alterations in mitochondria or an increase of reactive oxygen species, which has already been shown in previous studies [24][25][26]45].

Flow Cytometry after Annexin V and Propidium Iodide Staining
To measure the fraction of healthy, early apoptotic, and late apoptotic/necrotic cells, we used an apoptosis detection kit (ALX-850-253-KI02, Enzo Life Sciences, Inc., Farmingdale, NY, USA). 3 × 10 5 cells were seeded in 1 mL per well (6 well plate). After the respective treatment, 300 µL cell suspension was collected in a glass tube (Becton Dickinson, Franklin Lakes, NJ, USA).

Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS)
HaCaT cells were grown in 60 × 15 mm cell culture dishes. HuT78 and MyLa cells were grown in 6 well culture plates. After the respective treatment for the indicated time, supernatants of HaCaT cells were taken and cell pellets were washed once with cold Dulbecco's phosphate-buffered saline (DPBS) (Thermofisher, Waltham, MA, USA) after trypsinization and stored at −80 • C. T cells were centrifuged before supernatants were taken and cell pellets were washed once with DPBS and stored at −80 • C. Before storage, the amounts of healthy cells and total cells were determined with a TC20 TM automated cell counter (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Ceramide analyses were done using liquid chromatography-electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) as described elsewhere [47].
In brief, cell samples were resuspended in 200 µL extraction buffer (citric acid 30 mM, disodium hydrogen phosphate 40 mM) and mixed with 20 µL internal standard solution (2 µg/mL, Cer d18:1/8:0 and GlcCer d18:1/8:0 in chloroform:methanol (2:1, v/v)). The samples were then extracted once with 600 µL methanol:chloroform:HCl (15:83:2, v/v/v). The lower organic phase was transferred and evaporated at 45 • C under a gentle stream of nitrogen and reconstituted in 200 µL tetrahydrofuran:water (9:1, v/v) containing 0.2% formic acid and 10 mM ammonium formate. For the supernatant, a 50 µL sample was mixed with 150 µL extraction buffer and extracted as described above. For calibration standards and quality control samples preparation, 20 µL of the corresponding working solutions were processed as stated instead of sample. The quantification of all analytes was performed using a hybrid triple quadrupole-ion trap mass spectrometer QTRAP 5500 (Sciex, Darmstadt, Germany) equipped with a Turbo-V-source operating in positive ESI mode. Ceramides were separated using an Agilent 1290 HPLC system equipped with a Zorbax C18 Eclipse Plus UHPLC column (2.1 × 50 mm, 1.8 µm, Agilent technologies, Waldbronn, Germany). Quality control samples of three different concentration levels (low, middle, high) were run at the beginning and end of each run. Samples were processed using Analyst software 1.6 and the obtained concentrations were evaluated using MultiQuant Software 3.0 (both Sciex, Toronto, ON, Canada) using the internal standard method (isotope-dilution mass spectrometry).
The calibration curve was calculated by quadratic regression with 1/x 2 weighting. Variations in the accuracy of the calibration standards were lower than 15% over the range of calibration, except for the lower limit of quantification (LLOQ), where a limit of 20% was accepted. All measured data were collected in one LC-MS/MS measurement.

Knockdown Studies with Small Interfering RNA
Cells were transfected with small interfering RNA (siRNA) of ASAH1 and nontargeting siRNA as a negative control. (Hs.527412 (ASAH1) Assay ID 119213 Silencer ® and Silencer TM Negative Control No.1 siRNA, AM461 (all Thermo Fisher, Waltham, MA, USA)). The transfection tool was Lipofectamine TM 3000 Thermo Fisher, Waltham, MA, USA). Instructions of the manufacturer were followed. The success of the knockdown was verified by Western Blot analysis.

Statistics
Statistical significance was determined by an unpaired t-test. All statistics were evaluated with the program GraphPad ® Prism (Version 8.4.2 (464), GraphPad Software, San Diego, CA, USA).

Conclusions
In this study, we could show that C6Cer significantly reduced cell viability and induced cell death in CTCL cell lines but not in HaCaT keratinocytes and primary human keratinocytes. HaCaT keratinocytes seemed to be protected due to their ability to metabolize proapoptotic C6Cer more efficiently. Therefore, the application of C6Cer offers a possible novel topical treatment of CTCL.

Conflicts of Interest:
The authors declare no conflict of interest.