TNF-α is a 17 kDa protein consisting of 156 amino acids which acts as a cellular mediator provoking inflammation, proliferation, tumorigenesis and even apoptosis(Reference Balkwill1). This cytokine plays a key role in the pathogenesis of diseases such as cancer, rheumatoid arthritis, Crohn's disease, diabetes and psoriasis(Reference Locksley, Killeen and Lenardo2). TNF-α binds with TNF receptor I (p55 receptor) or II (p75 receptor) and associates with adaptor proteins to trigger downstream kinases such as phosphoinositide 3-kinase (PI3K)–protein kinase B (Akt) and mitogen-activated protein kinases. Many inflammatory proteins are expressed through activation of transcription factors such as activator protein-1 (AP-1) and NF-κB by upstream signalling. Therefore, inhibition of the TNF-α signalling pathway may help prevent these diseases(Reference MacEwan3).
Vascular endothelial growth factor (VEGF) is a multifunctional cytokine that is a key regulator of angiogenesis(Reference Ferrara, Gerber and LeCouter4). VEGF expression is associated with the pathology of tumorigenesis, CVD, rheumatoid arthritis and psoriasis. Tumour cells require blood vessels because of nutrient and oxygen deficiency, and VEGF promotes tumour cell survival via angiogenesis. VEGF is involved in tumour progression by controlling angiogenesis as well as tumour promotion. The Ha-ras gene mutation in a chemical-induced two-stage carcinogenesis model has been shown to enhance VEGF expression and benign papilloma development(Reference Larcher, Murillas and Bolontrade5). Therefore, inhibition of VEGF expression might be a promising approach to protect against carcinogenesis.
Cocoa contains high amounts of polyphenols(Reference Borchers, Keen and Hannum6) and contributes to a healthy life through strong antioxidant and anti-inflammatory activities and a increase of NO bioavailability(Reference Corti, Flammer and Hollenberg7, Reference Serafini, Bugianesi and Maiani8). Recently, the health benefits of dark chocolate with reduced energy and an increased proportion of cocoa were revealed(Reference McShea, Ramiro-Puig and Munro9). Our previous studies show that cocoa contains more polyphenols and a higher antioxidant capacity than tea and red wine(Reference Lee, Kim and Lee10). It exerted protective effects against H2O2-induced inhibition of gap junction intercellular communication in rat epithelial cells(Reference Lee, Hwang and Joo11) and cocoa polyphenol extract (CPE) inhibited phorbol ester-induced inflammation through suppression of cyclo-oxygenase-2 expression in mouse skin(Reference Lee, Kundu and Sue12). Recently, we reported that CPE inhibits thrombin-induced expression and activation of matrix metalloproteinase-2 (MMP-2) in vascular smooth muscle cells and phorbol ester-induced neoplastic transformation of promotion-sensitive (P+) JB6 mouse epidermal cells through direct inhibition of mitogen-activated protein kinase kinase-1 (MEK1)(Reference Lee, Kang and Oak13, Reference Kang, Lee and Lee14). However, the chemopreventive effects of cocoa involve mechanisms besides the inhibition of MEK1. Here, we tested whether cocoa polyphenols inhibit TNF-α-induced VEGF expression and we suggested a novel molecular target of CPE, PI3K.
Materials and methods
Eagle's minimum essential medium (MEM), gentamicin, fetal bovine serum (FBS) and l-glutamine were purchased from GIBCO BRL (Carlsbad, CA, USA). TNF-α was purchased from ProSpec-Tany TechnoGene (Rehovot, Israel). The ELISA kit for VEGF was purchased from R&D Systems (Minneapolis, MN, USA). The antibodies against phosphorylated (p-) MEK (Ser217/221), total extracellular signal-regulated kinase (ERK), p-p90 kDa ribosomal S6 kinases (p90RSK; Thr359/Ser363), total p90RSK, p-p70 kDa ribosomal protein S6 kinase (p70S6K; Thr389) and total p70S6K were obtained from Cell Signaling Biotechnology (Beverly, MA, USA). p-Akt (Ser 473), p-ERK (Thr202/Tyr204), total ERK and total MEK1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The MEK1 activity assay kit and PI3K active protein were purchased from Upstate Biotechnology (Lake Placid, NY, USA). CNBr-Sepharose 4B, glutathione-Sepharose 4B and [γ-32P]ATP were from GE Healthcare (Piscataway, NJ, USA), and the protein assay kit was purchased from Bio-Rad Laboratories (Hercules, CA, USA). G418 and the luciferase assay substrate were from Promega (Madison, WI, USA).
The JB6 P+ mouse epidermal cell lines were cultured in monolayers in 5 % FBS–MEM, 1000 units of penicillin and 1 mg/ml of streptomycin at 37°C in a 5 % CO2 incubator. The JB6 mouse epidermal cell line stably transfected with dominant negative p85 (DN-p85), NF-κB, and AP-1 luciferase reporter plasmid was a gift from Dr Zigang Dong(Reference Kwon, Lee and Kim15) and maintained in 5 % FBS–MEM and 200 μg/ml G418.
Preparation of cocoa polyphenol extract
Cocoa polyphenols were extracted from commercial cocoa powder (50 g) produced by LOTTE Confectionery Co. Ltd (Seoul, Korea) with 500 ml of 50 % (v/v) aqueous ethanol under reflux for 6 h. After the extraction, the solution was filtered twice to collect the extract. The collected cocoa extract was loaded onto a styrene-based adsorption resin column (60 × 450 mm, HP-20; Mitsubishi Chemical Co., Tokyo, Japan), washed with 20 % (v/v) aqueous ethanol, and then eluted with 60 % (v/v) aqueous ethanol. The eluted CPE was concentrated at 50°C under reduced pressure, frozen, and dried.
Determination of vascular endothelial growth factor production
JB6 P+ cells (5 × 103) were cultured in ninety-six wells and incubated for 48 h. After incubation, the cells were pretreated with CPE at the indicated concentrations for 1 h before incubation with TNF-α (4 ng/ml) for 18 h, and then the culture medium was harvested. VEGF was quantified according to the manufacturer's instructions.
Measurement of cell viability
Cell viability was measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, which is based on the ability of live cells to convert tetrazolium salt into purple formazan. JB6 P+ cells (5 × 103) were cultured in ninety-six wells and incubated for 48 h. After incubation, the cells were pretreated with CPE at the indicated concentrations for 1 h before incubation with TNF-α (4 ng/ml) for 18 h, followed by the addition of 20 μl MTT stock solution (5 mg/ml; Sigma, St Louis, MO, USA) to each well and the plates were further incubated for 4 h at 37°C. The supernatant fraction was removed and 200 μl dimethyl sulfoxide was added to each well to solubilise the water-insoluble purple formazan crystals. The absorbency at the wavelength of 570 nm was measured with a microplate reader (Molecular Devices, Sunnyvale, CA, USA). All the measurements were performed in triplicate. Results are expressed as the percentage proliferation with respect to untreated cells.
Luciferase assay for NF-κB and activator protein-1 transcription activity
Confluent monolayers of JB6 P+ cells stably transfected with an AP-1 or NF-κB luciferase reporter plasmid were trypsinised, and 5 × 103 viable cells suspended in 100 μl 5 % FBS–MEM was added to each well of a ninety-six-well plate. Plates were incubated at 37°C in a humidified atmosphere of 5 % CO2. When cells reached 80–90 % confluence, they were starved by culturing them in 0·1 % FBS–MEM for another 24 h. The cells were then treated for 15 min with CPE and exposed to TNF-α (4 ng/ml) for an additional 4 h. After treatment, cells were disrupted with 100 μl lysis buffer (0·1 m-potassium phosphate buffer (pH 7·8); 1 % Triton X-100; 1 mm-dithiothreitol (DTT); 2 mm-EDTA), and the luciferase activity was measured using a luminometer (Microlumat Plus LB 96V; Berthold Technologies, Bad Wildbach, Germany).
Western blotting analysis
After JB6 P+ cells (1·0 × 106) were cultured in a 10 cm diameter dish for 48 h, they were starved in serum-free medium for another 24 h to eliminate the influence of FBS on the activation of kinases. The cells were then treated with CPE (0–20 μg/ml) for 1 h before being exposed to TNF-α (4 ng/ml) for 15 min. After centrifugation, cell lysis was performed at 4°C for 30 min in cell lysis buffer (20 mm-2-amino-2-hydroxymethyl-propane-1,3-diol (Tris)-HCl (pH 7·5), 150 mm-NaCl, 1 mm-Na2EDTA, 1 mm-ethylene-glycol-bis(a-aminoethyl)-N,N,N′,N′-tetra-acetic acid (EGTA), 1 % Triton, 2·5 mm-sodium pyrophosphate, 1 mm-β-glycerophosphate, 1 mm-Na3VO4, leupeptin (1 μg/ml) and 1 mm-phenylmethanesulfonyl fluoride (PMSF)). The cell lysates were centrifuged at 23 000 g for 15 min, and the resulting supernatant fraction was stored at − 70°C before Western blot analysis. The protein concentration in each sample was measured by subjecting lysate protein (30 μg) to 10 % SDS-PAGE, with the protein electrophoretically transferred to a nitrocellulose membrane (Whatman, Clifton, NJ, USA). Protein bands were visualised by a chemiluminescence detection kit (Amersham, Piscataway, NJ, USA) after hybridisation with the horseradish peroxidase-conjugated secondary antibody. Quantification of band intensity was measured using Image J software (National Institutes of Health, Bethesda, MD, USA).
Mitogen-activated protein kinase kinase-1 immunoprecipitation and activity assay
JB6 P+ cells were cultured to 80 % confluence and then serum-starved in 0·1 % FBS–MEM for 24 h at 37°C. Cells were either treated or not treated with CPE (0–20 μg/ml) for 1 h before being exposed to TNF-α (4 ng/ml) for 15 min, disrupted with lysis buffer (20 mm-Tris–HCl (pH 7·4), 1 mm-EDTA, 150 mm-NaCl, 1 mm-EGTA, 1 % Triton X-100, 1 mm-β-glycerophosphate, leupeptin (1 mg/ml), 1 mm-Na3VO4 and 1 mm-phenylmethylsulfonyl fluoride), and finally centrifuged at 16 000 g for 10 min in a microcentrifuge. The lysates containing 500 μg protein were used for immunoprecipitation with an antibody against MEK1 and incubated at 4°C overnight. Protein A/G Plus agarose beads were then added and the mixture was continuously rotated for another 3 h at 4°C. The beads were washed three times with kinase buffer (20 mm-3-(N-morpholino)propanesulfonic acid (MOPS; pH 7·2), 25 mm-β-glycerol phosphate, 5 mm-EGTA, 1 mm-sodium orthovanadate and 1 mm-DTT), and then re-suspended in 20 μl of 1 × kinase buffer supplemented with 1 μg of inactive ERK2 and incubated for an additional 30 min at 30°C. Then myelin basic protein (20 μg) and 10 μl of diluted [γ-32P]ATP solution were added and the mixture was incubated for 10 min at 30°C. A 20 μl sample was transferred onto p81 paper and washed three times with 0·75 % phosphoric acid for 5 min per wash and once with acetone for 2 min. The radioactive incorporation was determined using a scintillation counter.
Phosphoinositide 3-kinase activity assay
An active PI3K protein (100 ng) was incubated with CPE for 10 min at 30°C. Then, the mixtures were incubated with 20 μl phosphatidylinositol (0·5 mg/ml) (Avanti Polar Lipids, Alabaster, AL, USA) for 5 min at room temperature, followed by incubation with reaction buffer (100 mm-HEPES (pH 7·6), 50 mm-MgCl2, 250 μm-ATP containing 0·37 MBq (10 μCi) of [γ-32P]ATP) for an additional 10 min at 30°C. The reaction was stopped by adding 15 μl of 4 m-HCl and 130 μl chloroform–methanol (1:1, v/v). After vortexing, 30 μl of the lower chloroform phase was spotted onto a 1 % potassium oxalate-coated silica gel plate that was previously activated for 1 h at 110°C. The resulting 32P-labelled phosphatidylinositol-3-phosphate was separated by TLC and radiolabelled spots were visualised by autoradiography.
Cell lysates of the TNF-α-treated cells used in the kinase assay (500 μg) or PI3K protein (2 μg) were incubated with CPE–Sepharose 4B (or Sepharose 4B alone as a control) beads (100 μl, 50 % slurry) in reaction buffer comprising 50 mm-Tris-HCl (pH 7·5), 5 mm-EDTA, 150 mm-NaCl, 1 mm-DTT, 0·01 % Nonidet P-40, 2 μg/ml bovine serum albumin, 0·02 mm-PMSF and 1 × protease inhibitor mixture. After incubation with gentle rocking overnight at 4°C, the beads were washed five times with buffer (50 mm-Tris-HCl (pH 7·5), 5 mm-EDTA, 150 mm-NaCl, 1 mm-DTT, 0·01 % Nonidet P-40 and 0·02 mm-PMSF), and proteins bound to the beads were analysed by immunoblotting.
Data are expressed as mean values and standard deviations. Student's t test was used for single statistical comparisons, with a probability value of P < 0·05 as the criterion for statistical significance.
Cocoa polyphenol extract inhibits TNF-α-induced vascular endothelial growth factor up-regulation in promotion-sensitive JB6 mouse epidermal cells
JB6 P+ cells were pretreated for 1 h with different concentrations of CPE and then stimulated with TNF-α (4 ng/ml) for 18 h. TNF-α-induced VEGF expression in JB6 P+ cells was determined by ELISA. CPE dose-dependently inhibited TNF-α-induced VEGF up-regulation (Fig. 1(a)). These effects were not due to the cytotoxicity of CPE, because CPE up to 20 μg/ml has no effect on cell viability (Fig. 1(b)).
Cocoa polyphenol extract suppresses TNF-α-induced transactivation of NF-κB and activator protein-1 in promotion-sensitive JB6 mouse epidermal cells
The eukaryotic transcription factors NF-κB and AP-1 are key regulators of VEGF expression(Reference Lee, Kang and Rogozin16). We next examined whether CPE attenuates TNF-α-induced transactivation of NF-κB and AP-1 using JB6 P+ cell lines stably transfected with an NF-κB and AP-1 luciferase reporter plasmid. Exposure to TNF-α (4 ng/ml) for 4 h increased NF-κB luciferase activity by about 2·5-fold compared with the untreated control, and CPE dose-dependently inhibited these effects (Fig. 2(a)). TNF-α also increased AP-1 luciferase activity by about 2-fold compared with the untreated control, which was attenuated by CPE treatment (Fig. 2(b)).
Effects of cocoa polyphenol extract on TNF-α-induced phosphorylation of protein kinase B and mitogen-activated protein kinase
We next examined the PI3K–Akt and mitogen-activated protein kinase pathways, which relate to TNF-α-induced NF-κB and AP-1 transactivation. We investigated the phosphorylation level of the kinase which is involved in these pathways by Western blot analysis. CPE suppressed TNF-α-induced phosphorylation of Akt, p70S6K, ERK, p90 kDa ribosomal S6 kinase (p90RSK), mitogen-activated protein kinase kinase 4 (MKK4) and c-Jun N-terminal kinase (JNK). Phosphorylation of MEK was not changed by CPE treatment and phosphorylation of p38 was increased slightly by CPE treatment (Fig. 3).
Cocoa polyphenol extract suppresses phosphoinositide 3-kinase activity and phosphoinositide 3-kinase is required for TNF-α-induced vascular endothelial growth factor expression
To elucidate whether CPE blocks the phosphorylation of Akt and its downstream kinase, p70S6K, to inhibit PI3K activity, a well-known upstream kinase of Akt, we further examined the effect of CPE on PI3K activity. CPE at 10 μg/ml (Fig. 4(a)) completely inhibited PI3K activity. To further confirm the mechanism of the inhibitory effect of CPE against PI3K activity, we investigated whether CPE directly binds with PI3K. PI3K was found in the CPE–Sepharose 4B beads (Fig. 4(b); lane 3), but not in the Sepharose 4B beads only (Fig. 4(b); lane 2).
Phosphoinositide 3-kinase is required for TNF-α-induced vascular endothelial growth factor expression
We next examined the involvement of the PI3K signalling pathway in the TNF-α-induced up-regulation of VEGF in JB6 P+ cells. LY294002, a PI3K inhibitor, effectively inhibited TNF-α-induced VEGF up-regulation in JB6 P+ cells (Fig. 5(a)). To confirm the role of PI3K in TNF-α-induced up-regulation of VEGF in JB6 P+ cells, we used dominant negative p85 (DN-p85) which is the regulatory domain of PI3K(Reference Nomura, Kaji and He17). Compared with mock vector transfected cells, TNF-α-induced up-regulation of VEGF markedly decreased in the DN-p85-expressing cells (Fig. 5(b)).
Cocoa polyphenol extract inhibits TNF-α-induced mitogen-activated protein kinase kinase-1 activity and directly binds mitogen-activated protein kinase kinase-1
We previously showed that CPE suppressed 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced cell transformation and MMP-2 expression though direct inhibition of MEK activity(Reference Lee, Kang and Oak18, Reference Kang, Lee and Lee19). Similarly here, CPE inhibited TNF-α-induced phosphorylation of ERK but not MEK. CPE dose-dependently suppressed TNF-α-induced MEK1 activity in JB6 P+ cells (Fig. 6(a)). To further confirm whether CPE directly interacts with MEK1, we used a CPE pull-down assay. MEK1 in the cell lysate of TNF-α-treated JB6 P+ cells bound to the CPE–Sepharose 4B beads (Fig. 6(b); lane 3), but not to the Sepharose 4B beads alone (Fig. 6(b); lane 2), indicating that CPE could directly bind to MEK1 and inhibit MEK1 activity.
TNF-α is a key regulator of inflammation and plays vital roles in inflammatory diseases such as cancer, rheumatoid arthritis, Crohn's disease, diabetes and psoriasis. TNF-α induces the anchorage-independent growth of mouse epidermal JB6 P+ cells(Reference Yan, Li and Ouyang20). Knockout mice (TNF-α − / − ) are more resistant to TPA-induced skin tumour promotion than wild-type mice(Reference Arnott, Scott and Moore21, Reference Moore, Owens and Stamp22). Neutralising TNF-α with an antibody also inhibits the development of skin tumours(Reference Scott, Moore and Arnott23). These results suggest that targeting TNF-α signalling can be an important strategy for delaying the promotion of skin tumours.
Cocoa contains large amount of flavonoids. It comprises mainly catechins, flavonol glycosides and procyanidins. Our previous study using the identical CPE shows that the CPE had 413 mg epicatechin equivalent flavonoids per g. This demonstrates that 40 % of CPE is comprised of flavonoids(Reference Lee, Kang and Oak13). We suggested that flavonoids can act as PI3K and MEK inhibitors(Reference Kwon, Lee and Kim15, Reference Kim, Kwon and Seo24, Reference Choi, Kim and Song25). Therefore, inhibitory effects of CPE on PI3K and MEK activity could be due to flavonoids contained in CPE.
When individuals consume dark chocolate, the substances epicatechin, catechin, procyanidin B2 or procyanidin B2 gallate are detected in their plasma(Reference Taubert, Roesen and Lehmann26). In a recent clinical study, 45 g dark chocolate containing up to 70 % cocoa was fed to human volunteers. Plasma epicatechin concentration of the volunteers was raised up to 155 ng/ml(Reference Spadafranca, Martinez Conesa and Sirini27). In this study only epicatechin concentration was measured. However, chocolate is comprised of many other flavonoids(Reference Schroeter, Heiss and Balzer28). The sum of flavonoid concentration would be much higher than that of epicatechin only. The CPE used in the present study was effective, starting from 5 μg/ml. Therefore, the concentration (5 to 20 μg/ml) of CPE which we used in the present study seems to be a reliable dose.
In the present study, we found that CPE (5 to 20 μg/ml) dose-dependently inhibited TNF-α-induced VEGF expression in JB6 P+ cells. Cocoa polyphenols exhibit free radical-scavenging activity, as determined by both 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical-scavenging assays and anti-inflammatory effects in mouse skin(Reference Lee, Kundu and Sue12). Cocoa polyphenols also suppressed the growth of Caco-2 human colon cancer cells through a blockade of the cell cycle at the G2/M phase(Reference Yamagishi, Natsume and Osakabe29). Further, cocoa polyphenols suppressed lung and pancreatic tumorigenesis in a male rat multi-organ carcinogenesis model(Reference Yamagishi, Natsume and Osakabe29, Reference Yamagishi, Natsume and Osakabe30), and TPA-induced tumour promotion in a mouse two-stage skin tumorigenesis model initiated with 7,12-dimethylbenz[a]anthrathene(Reference Osakabe, Yamagishi and Natsume31). Thus, the present results support the chemopreventive potential of cocoa polyphenols.
NF-κB and AP-1 are important transcription factors in tumour promotion and are activated by many cytokines and chemicals, including TNF-α(Reference Szlosarek, Charles and Balkwill32), leading to VEGF transcription(Reference Jośko and Mazurek33). VEGF mRNA expression in MDA-MB-231 breast cancer cells is associated with NF-κB activation(Reference Shibata, Nagaya and Imai34). AP-1 regulated by ERK or Akt also mediates VEGF expression(Reference Lee, Chen and Tsai35, Reference Ouyang, Li and Shi36). In the present study the CPE suppressed TNF-α-induced NF-κB and AP-1 transactivation to block VEGF transcription.
The antioxidant effects of CPE may partially contribute to these effects, but it may also target other signalling molecules(Reference Sebolt-Leopold and English37). Some small molecules in food such as flavonoids also inhibit signalling molecules directly(Reference Lee, Kang and Rogozin16, Reference Walker, Pacold and Perisic38–Reference Kang, Lee and Rogozin40). Targeting PI3K or mitogen-activated protein kinase (ERK, JNK and p38) pathways is a good strategy for cancer prevention. PI3K phosphorylates the 3 position of the hydroxyl group of the inositol ring of phosphatidylinositol and controls cellular processes such as cell proliferation and apoptosis. It is activated in many tumours and small molecule inhibitors are being developed(Reference Raynaud, Eccles and Clarke41). In the present study the CPE strongly suppressed PI3K activity and bound PI3K directly. These results explain the inhibition of pAkt and p70S6K, which are downstream kinases of PI3K.
Our previous studies demonstrated that CPE inhibited TPA-induced neoplasmic transformation in mouse epidermal cells and suppressed thrombin-induced MMP-2 activation through direct inhibition of MEK1 activity(Reference Lee, Kang and Oak18, Reference Kang, Lee and Lee19). A pharmacological inhibitor of MEK suppressed TNF-α-induced VEGF expression in JB6 P+ cells(Reference Kim, Kwon and Lee42). CPE attenuated TNF-α-induced MEK1 activity, and subsequently phosphorylation of ERK, without affecting MEK phosphorylation. Therefore, inhibition of MEK1 contributes to inhibition of TNF-α-induced VEGF expression by CPE.
Our previous study showed that the MKK4–JNK pathway is required for TNF-α-induced VEGF expression using specific inhibitors(Reference Kim, Kwon and Lee42). In the present study the CPE inhibited TNF-α-induced phosphorylation of MKK4 and JNK. However, MEK1 and PI3K, which are revealed as targets of the CPE in the present study, cannot explain the phenomenon of inhibition on the MKK4–JNK pathway by the CPE. Increase of p38 phosphorylation is also hard to explain with MEK1 and PI3K inhibition. Further study must reveal new target proteins of CPE which control the TNF-α-induced MKK4–JNK and p38 pathway.
Multi-target kinase inhibitors have significant side effects. However, they are in the spotlight since the success of Gleevec targeting Bcr-Abl and KIT kinase due to its powerful effect and a variety of applications(Reference Vogt and Kang43). Food materials are usually regarded as safe because of their long-term use. Therefore, multi-targeting may be better than specific targeting for food materials. In summary, a CPE suppressed TNF-α-induced VEGF expression via targeting PI3K and MEK1 to inhibit phosphorylation of ERK and Akt in mouse epidermal cells. Taken together, these results indicated that CPE might have chemopreventive potential against pro-inflammatory cytokine-mediated skin cancer and inflammation.
The present study was supported by grants from the Basic Research Program (no. R01-2009-000-11957-0), Priority Research Centers Program (2009-0093824) and the WCU Program (R31-10056), through the National Research Foundation of Korea, funded by the Ministry of Education, Science, and Technology, Republic of Korea. K. W. L., C. Y. L. and H. J. L. designed the study and prepared the paper. J. E. K., J. E. S., S. K. J. and N. J. K. contributed to the successful execution of the experimental work. We declare that we have no conflict of interest.