Novel peptide dermaseptin‐PS1 exhibits anticancer activity via induction of intrinsic apoptosis signalling

Abstract Antimicrobial peptides (AMP) secreted by the granular glands of frog skin have been widely reported to exhibit strong bacteriostatic and bactericidal activities. Many of them have been documented with potent antiproliferative effects on multiple cancer cells, many studies also suggested that AMPs exert their functions via disrupting cell membranes. However, whether and how other cell death induction mechanism is involved in mammalian cancer cells has rarely been investigated. In this study, a novel AMP named Dermaseptin‐PS1 was isolated and identified from Phyllomedusa sauvagei, it showed strong antimicrobial activities against three types of microorganisms. In vitro antiproliferative studies on human glioblastoma U‐251 MG cells indicated that Dermaseptin‐PS1 disrupted cell membranes at the concentrations of 10−5 M and above, while the cell membrane integrity was not affected when concentrations were decreased to 10−6 M or lower. Further examinations revealed that, at the relatively low concentration (10−6 M), Dermaseptin‐PS1 induced apoptosis through mitochondrial‐related signal pathway in U‐251 MG cells. Thus, for the first time, we report a novel frog skin derived AMP with anticancer property by distinct mechanisms, which largely depends on its concentration. Together, our study provides new insights into the mechanism‐illustrated drug design and the optimisation of dose control for cancer treatment in clinic.

cell membrane and eukaryotic mitochondrial membrane possess high transmembrane potentials. By contrast, the zwitterionic phospholipid, which is the main composition of eukaryotic membrane, possesses low transmembrane potential. [4][5][6][7] As a consequence, the majority of AMPs prefer to corrupt the prokaryotic membrane and eukaryotic mitochondrial membrane rather than the eukaryotic membrane.
Thus, those AMPs with distinct membrane-penetration mechanism on tumour cells that allows the receptor-regulated internalisation, followed by induction of programmed cell death, could serve as unique and supplementary options for cancer therapy.
Apoptosis is a programmed process that plays a pivotal role in modulating physiological and pathological cell death. 8 Extensive scrutiny has been focused on the investigation of apoptosis-related cancer research, in particular the disorder of cancer cells to undergo apoptosis could lead to severe malignant progression and chemotherapy resistance. 9 The whole process of apoptosis composed of either of the two distinguished signalling transduction pathways and a number of proteins that located in multiple organelles. 10 Compared with death receptor (extrinsic) pathway, the mitochondrial (intrinsic) pathway of apoptosis is most commonly deregulated in cancer cells, and most stimuli induce the initiation of apoptosis through mitochondrial signalling that governs the release of cytochrome c to the cytosol, which in response regulates the activity of caspase protease. 11,12 Upon activation, caspases cleave a number of proteins, resulting in cell death, and the alterations in biochemistry and morphology occur. 13 In this study, using human glioblastoma U-251 MG cells, we investigated the tumour suppression effect of the novel isolated AMP Dermaseptin-PS1 from P. sauvagei. We provide the evidence that the concentration is critical for anticancer agents to exert biological functions through differentiated mechanisms, which support the further drug design and development for targeting the intervention of mitochondrial apoptosis in cancer therapy.

| Specimen biodata and skin secretion harvesting
Specimens of P. sauvagei (n = 3, two males 5 cm snout-to-vent length, one female 7 cm snout-to-vent length), originally from South America, were obtained from a commercial source in the United States. The skin secretion produced by the holocrine glands of frogs was obtained by mild transdermal electrical stimulation (5 V, 50 Hz, 4 ms plus width) as previously described. 14

| "Shotgun" cloning of novel Dermaseptin-like peptide encoding cDNAs from lyophilized skin secretion
Five milligrams of lyophilised P. sauvagei secretion powder were dissolved in 1 mL cell Lysis/Binding buffer to isolate polyadenylated mRNA by using magnetic oligo-dT beads in Dynabeads ® mRNA DIRECT ™ Kit (Dynal Biotech, Liverpool, UK). Then the reverse-transcribed cDNA library was subjected to 3′-RACE PCR procedures to acquire the full length of preproprotein nucleic acid sequences using a SMART-RACE kit (Clontech, Palo Alto, CA, USA) essentially as described by the manufacturer. Briefly, the 3′-RACE reactions employed a UPM primer (supplied with the kit) and degenerate sense primers (S1; 5′-ACTTTCYGAWTTRYAAGMCCAAABATG-3′, was designed to a segment of the 5′-untranslated region of phylloxin cDNA from Phyllomedusa bicolor (EMBL accession no. AJ251876) and the opioid peptide cDNA from Pachymedusa dacnicolor (EMBL accession no. AJ005443). 15 PCR cycling procedures were carried out as follows: initial denaturation step: 90 seconds at 94°C; 35 cycles: denaturation 30 seconds at 94°C, primer annealing for 30 seconds at 58°C; extension for 180 seconds at 72°C. PCR products were gelpurified and cloned using a pGEM-T vector system (Promega Corporation, Southampton, UK), and selected samples were sequenced by an ABI 3730 automated sequencer.

| Identification and structural characterisation of the novel Dermaseptin-like peptide
An aliquot sample of the lyophilised P. sauvagei skin secretion was dissolved in 1 mL of trifluoroacetic acid (TFA)/water (0.05:99.95, v/v) and clarified by centrifugation. One millilitre of clear supernatant was carefully decanted and pumped directly into a reverse-phase HPLC column (C-18, 300 Å, 5 μm, 4.6 mm × 250 mm; Phenomenex, Cheshire, UK). The elution gradient formed from 0.05/99.5 (v/v) TFA/ water to 0.05/19.95/80.0 (v/v/v) TFA/water/acetonitrile in 240 minutes at a flow rate of 1 mL/min and the effluent was detected by UV absorbance at 214 nm and 280 nm. An automatic fraction collector (GE Healthcare, Little Chalfont, UK) was used to collect the fractions at 1-minute interval. All fractions were interrogated by matrixassisted laser desorption ionisation time-of-flight (MALDI-TOF) mass spectrometry in positive detection mode using alpha-cyano-4-hydroxycinnamic acid (CHCA) as matrix. The fractions with masses coincident with the putative peptide from molecular cloning were subjected to Liquid Chromatography Quadruple (LCQ)-Fleet electrospray ion-trap mass spectrometer (Thermo Fisher Scientific, San Francisco, CA, USA) for primary structural analysis.

| Solid-phase peptide synthesis
Following unambiguous confirmation of the primary structure through both molecular cloning strategy and LCQ-Fleet mass spectrometry, the peptide was chemically synthesised by Tribute ™ LONG ET AL. | 1301 automated solid phase peptide synthesizer 4 (Protein Technologies, Tucson, AZ, USA). The synthesised peptide replicates were then purified by reverse-phase HPLC and confirmed by MALDI-TOF mass spectrometry prior to use. where c is the molar concentration of the sample (mol/L) and l is the cuvette path length in centimetre (cm). DICHROWEB webserver (http://dichroweb.cryst.bbk.ac.uk/html/home.shtml) was used to estimate the contents of different secondary structures. [16][17][18]

| Antimicrobial and haemolytic assays
The antimicrobial assays for evaluating the minimum inhibitory con-

| MTT cell proliferation and LDH cell membrane integrity evaluations
Five thousand cells/well in 100 μL full growth DMEM medium were planted into 96-well plates and allowed to attach for 24 hours. Cells were starved by serum-free medium for 6 hours before treated with peptide concentration gradient (10 −4 to 10 −9 M, n = 5) for 24 hours.
The cell membrane integrity was measured by using lactate dehydrogenase (LDH) assay with Pierce LDH Cytotoxicity Assay Kit (Thermo Scientific, Loughborough, UK) according to the manufacturer's instruction with the same treatment conditions. Hill, MA, USA) was used to fix cells and one drop of anti-fade mounting medium (Thermo Fisher, UK) was added to each well before coverslip seal using nail polish. The slides were immediately analysed by Leica DMi8 fluorescent microscope imaging system.

PCR analysis
Cells were planted into 6-well plates at a density of 2 × 10 5 cells/2 mL/ well and allowed to attach for 24 hours. After 6 hours starvation, cells were treated by peptide concentration gradient (10 −4 to 10 −7 M) for 16 hours. Then, 500 μL of TRIzol ® Reagent (Thermo Fisher, Loughborough, UK) was added into each well and RNA was extracted by 100 μL chloroform and purified by 200 μL isopropanol. The solutions were centrifuged at 12 000 × g for 10 minutes and the supernatant was removed before the pellets were washed with 75% ethanol. After which, the RNA pellets were resuspended in 20 μL diethyl pyrocarbonate (DEPC) water and quantified using Nanodrop One c system (Thermofisher, UK). One microgram RNA from each sample was reverse

| Statistical analysis
All results are presented as mean ± SEM determined by two-tailed Student's t tests or one-way ANOVA. Pair comparisons of the means were made, and P < 0.05 was taken as a significant difference. The Bonferroni method was used to adjust the observed significance levels for the fact the multiple contrasts were being tested. showed a mixed conformation of random coil (48%) and β-sheet (48%) with a negative band presented at 198 nm. By contrast, in 50% TFE-10 mmol/L NH 4 Ac solution, the α-helical structure was increased to 25%, and random coil increased to 60%, while β-sheet decreased to 15% ( Table 2). The results suggested that the Dermaseptin-PS1 was partially structured to a membrane-penetrated β-sheet structure in membranes mimic surroundings, but largely remained unstructured in aqueous solution, indicating that

| Dermaseptin-PS1 induced U-251 MG cell death at 10 −6 M via induction of apoptosis
Apoptosis and autophagy are two major programmed cell-destructive processes, both of them could be elicited by exogenous stress or ligand. 26 We firstly used Western blotting to examine whether Dermaseptin-PS1 caused cell viability decrease was through the activation of either or both of apoptosis and autophagy. As the dose assessment shown in Figure  Annexin V nor PI signal increase ( Figure 4D).
Next, we examined the related marker genes for both intrinsic and extrinsic cascades of apoptosis at transcriptional level. As shown in Figure 4E, the pro-apoptotic genes were increased significantly, while the anti-apoptotic Bcl-2 gene remained unaltered, which lead to the elevation of Bax/Bcl-2 ratio. The up-regulation of Bax/Bcl-2 ratio contributes to the activation of apoptotic cascade. 28 Together, we confirmed that Dermaseptin-PS1 disrupted the U-251 MG cell membranes at 10 −5 M and higher concentrations, and induced apoptosis at 10 −6 M to exert the antiproliferative effects.

| Dermaseptin-PS1 induced U-251 MG cell apoptosis through intrinsic cascade
To investigate the specific intracellular mechanism by which Dermaseptin-PS1 exerts the apoptosis induction effect, we examined the marker genes that related to the intrinsic or extrinsic apoptotic cascades. As shown in Figure

| DISCUSSION
Although, in the past decades, significant progress has been made for the treatment of cancers via induction of apoptosis, many challenges still remain, which are not only in the field of expanding the application of targeting apoptosis in multiple cancer cells, but also preventing cancer cells resistant to therapies. 12 A novel type of drugs, named "BH-3 mimetics," has been developed and applied clinically, they directly inhibit Bcl-2 family members for apoptosis activation, with the aim of overcoming the upstream proteins initiated therapeutic resistance. 29 However, these intrinsic apoptotic pathway targeted drugs also face the problems that toxicities need to be improved in the meantime of keeping or enhancing the potency on killing cancer cells. [30][31][32] In this study, we focused on investigating the potential anticancer mechanism of the novel char-  Figure 3B) and LDH ( Figure 3C) analysis, we presumed that these effects were resulted from necrosis caused by Dermaseptin-PS1, which is a non-specific cell membrane disruption mechanism, and this could be further unambiguously determined by the most reliable transmission electron microscopic detection. 33 While the U-251 MG cell viability decrease after being exposed to 10 −6 M peptide attracted our attention, as all the results implied that no mechanic cell membrane damage was likely to be involved. As a consequence, we examined the apoptosis activation. One of the interesting issues In addition, the marker proteins from the intrinsic apoptotic cascade ( Figure 6A) indicated that the intrinsic apoptotic cascade was responsible for low concentration of Dermaseptin-PS1 induced cell death. Although the cleaved caspase 9 levels were increased after exposure to Dermaseptin-PS1, the caspase 9 level was also increased compared with the positive control etoposide treatment. We suspect that, apart from activation of caspase 9, Dermaseptin-PS1 could also increase the expression of this gene, which culminates in elevation of both caspase 9 and its cleaved form.
Overall, our study not only provides new candidate for the treatment of malignant glioblastoma, but also illuminates  Table 1. The mRNA expression of 18S was used as a calibration standard. Data were analysed with one-way ANOVA using GraphPad Prism 5 software. Values are the mean ± SEM for three independent experiments. ***P < 0.001 vs negative control LONG ET AL. The detection of GAPDH protein was used as an internal control. The signal intensity was quantified by Image Lab software and GraphPad Prism 5 software was used for statistical comparison. NC, negative control. *or # P < 0.01, ** or ## P < 0.05 and *** or ### P < 0.001 vs NC. B, Cytochrome c release from the mitochondria into the cytosol was measured via Western blot. COX IV was used as a loading control for mitochondrial fractions. The signal intensity was quantified by Image Lab software and GraphPad Prism 5 software was used for statistical comparison. NC, negative control; PS1, Dermaseptin-PS1; Eto, etoposide. *P < 0.05 vs protein expression in cytosol after etoposide treatment; ### P < 0.001 vs protein expression in cytosol after Dermaseptin-PS1 treatment