Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

Contents lists available at ScienceDirect

Journal of Photochemistry and Photobiology B: Biology journal homepage: www.elsevier.com/locate/jphotobiol

Differential susceptibility of primary cultured human skin cells to hypericin PDT in an in vitro model A. Popovic, T. Wiggins, L.M. Davids ⇑ Redox Laboratory, Dept Human Biology, Rm 6.02.2, Level 6, Anatomy Bldg, University of Cape Town Medical School, Anzio Rd, Observatory 7925, Cape Town, South Africa

a r t i c l e

i n f o

Article history: Received 2 March 2015 Received in revised form 2 June 2015 Accepted 13 June 2015 Available online 14 June 2015 Keywords: Human skin cells Hypericin Photodynamic therapy ROS Apoptosis

a b s t r a c t Skin cancer is the most common cancer worldwide, and its incidence rate in South Africa is increasing. Photodynamic therapy (PDT) has been shown to be an effective treatment modality, through topical administration, for treatment of non-melanoma skin cancers. Our group investigates hypericin-induced PDT (HYP-PDT) for the treatment of both non-melanoma and melanoma skin cancers. However, a prerequisite for effective cancer treatments is efficient and selective targeting of the tumoral cells with minimal collateral damage to the surrounding normal cells, as it is well established that cancer therapies have bystander effects on normal cells in the body, often causing undesirable side effects. The aim of this study was to investigate the cellular and molecular effects of HYP-PDT on normal primary human keratinocytes (Kc), melanocytes (Mc) and fibroblasts (Fb) in an in vitro tissue culture model which represented both the epidermal and dermal cellular compartments of human skin. Cell viability analysis revealed a differential cytotoxic response to a range of HYP-PDT doses in all the human skin cell types, showing that Fb (LD50 = 1.75 lM) were the most susceptible to HYP-PDT, followed by Mc (LD50 = 3.5 lM) and Kc (LD50 > 4 lM HYP-PDT) These results correlated with the morphological analysis which displayed distinct morphological changes in Fb and Mc, 24 h post treatment with non-lethal (1 lM) and lethal (3 lM) doses of HYP-PDT, but the highest HYP-PDT doses had no effect on Kc morphology. Fluorescent microscopy displayed cytoplasmic localization of HYP in all the 3 skin cell types and additionally, HYP was excluded from the nuclei in all the cell types. Intracellular ROS levels measured in Fb at 3 lM HYP-PDT, displayed a significant 3.8 fold (p < 0.05) increase in ROS, but no significant difference in ROS levels occurred in Mc or Kc. Furthermore, 64% (p < 0.005) early apoptotic Fb and 20% (p < 0.05) early apoptotic Mc were evident; using fluorescence activated cell sorting (FACS), 24 h post 3 lM HYP-PDT. These results depict a differential response to HYP-PDT by different human skin cells thus highlighting the efficacy and indeed, the potential bystander effect of if administered in vivo. This study contributes toward our knowledge of the cellular response of the epidermis to photodynamic therapies and will possibly enhance the efficacy of future photobiological treatments. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Photodynamic therapy (PDT) has been touted as a novel, effective alternative anti-cancer therapy [1]. Defined as a therapy that uses a photosensitizing compound, activated by light of a specific wavelength in the presence of molecular oxygen [2,3] it has been shown to be effective in reducing tumorigenicity in a number of cancers both in vivo and in vitro [4]. Recently, its application to superficial basal cell carcinoma (sBCC), Bowen’s disease and actinic keratosis skin cancers demonstrated an impressive 84% clearance rate at 1 year [5]. With increasing prevalence of skin cancers

⇑ Corresponding author. http://dx.doi.org/10.1016/j.jphotobiol.2015.06.009 1011-1344/Ó 2015 Elsevier B.V. All rights reserved.

worldwide and the fact that South Africa ranks second behind Australia in the world incidence of skin cancers [6]. PDT is an attractive minimally invasive treatment option. Furthermore, the recent advent of daylight-mediated PDT for topical PDT, reduces the medical costs, making it an even more attractive treatment option [7–9]. Currently, porfimer sodium (PhotofrinÒ), 5-aminolevulinic acid (5-ALA) (LevulanÒ) and its methyl ester (MAL) (MetvixÒ), VisudyneÒ and FoscanÒ are the most commonly used FDA-approved photosensitizers (PS) however, other promising types of PS include chlorins, phthalocyanines and naphthodianthrones [4,10–12]. We have recently reported on the increased efficacy of the napthodianthrone compound, hypericin, in killing both non-melanoma and melanoma cells in vitro [13–18]. Due to its hydrophobic nature, hypericin passively enters all cells via their

250

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

cell membranes and therefore, in addition to it being a potentially effective photosensitizer of tumor tissue, peri-tumoral cells are also prone to photodamage [19,20]. Moreover, it has been shown to preferentially accumulate in tumors, suggested to be due to increased lipoprotein receptors and increased tumorigenic metabolism of the tumor cells [21–23]. Although hypericin-based PDT (HYP-PDT) has been shown to be effective in a number of cancers a dearth of information is available on its effect on the surrounding cells in the 3-dimensional space [24–30]. Hadjur et al. were the first to highlight the effect of HYP-PDT on human cells and although their focus was on the efficacy if PDT against melanoma, they also included a fibroblast cell line, MRC5 in their study. This cell line was however an immortalised line [31]. This initial study was followed by Bernd et al. who reported on the cytotoxic effect of a hypericum extract on cultures of human keratinocytes [32]. Similarly, this group used an immortal keratinocyte cell line (HaCaT). More recently, Kashef et al. [51] in showing the microbial efficacy of HYP-PDT, used primary human fibroblasts to highlight that these skin cells can be preserved by keeping the HYP concentration below 0.6 lg/ml and the light dose below 48 J/cm2. In none of these studies was the focus on the HYP-PDT effect on the surrounding cellular milieu i.e. the melanocytes, keratinocytes and fibroblasts existing in the layers of the skin. That the cellular and stromal components surrounding the malignant lesion is inescapable from the photodynamic treatment, is a given. This effect has been loosely defined as the ‘‘bystander’’ effect where indirect damage is induced into adjacent/ peri-lesional cells either via intercellular gap junctions or via diffusible ROS released into the microenvironment [33,34]. To this end, the current study set out to investigate the HYP-PDT effect on human skin cells normally found in the close vicinity of a skin cancer. Moreover, these cells were obtained from fresh human skin samples and represent primary cultures of the epidermis (melanocytes and keratinocytes) and dermis (fibroblasts). We found a distinct differential response to HYP-PDT between the different skin cell types correlating to their levels of cytotoxicity and ROS. Morphologically, the results corroborated both the cytotoxicity and ROS measurements. Very little evidence of apoptosis was evident in the cells representing the upper epidermal layers (melanocytes and keratinocytes) contrasting to the massive destruction of the cells representing the deeper dermal layer (fibroblasts). This study represents a novel approach to identifying the effect of PDT on the tumor microenvironment and could represent a significant step toward improving phototherapies. 2. Material and methods 2.1. Materials Hypericin (HYP) and all other reagents were purchased from Sigma–Aldrich, unless otherwise stated. Tissue culture dishes were obtained from Greiner Bio-One and media was purchased from Highveld Biological (PTY) LTD, unless otherwise stated. As HYP is sensitive to auto-oxidation, experiments were conducted under subdued light conditions. Unactivated (unirradiated) and untreated controls were plated in a separate unirradiated plate or dish. All plates/dishes, tubes and microcentrifuge tubes containing HYP were protected from light at all times. 2.2. Primary skin cell culture Primary human skin cells were isolated from neonatal or adult foreskins and human tissue obtained from plastic and reconstructive surgeries. Primary human melanocytes (Mc) keratinocytes (Kc) were isolated from the epidermal skin layer by cutting the skin tissue into

fragments (5 mm  2.5 mm) and sub-merging the pieces in 5 mg/ml of dispase solution overnight, at 4 °C. Following the separation of epidermis from the dermis, the epidermis was submerged in TEG (0.25% trypsin, 0.05% EDTA, 0.1% glucose) and cut into smaller pieces. After a 15 min incubation in a 37 °C water bath, a cell suspension was visible and fetal calf serum (FCS) was used to inactivate trypsinisation. Cells were then centrifuged at 5250 g for 5 min. Thereafter, the cell pellet were triturated and added to cell specific medium to yield either Mc or Kc. Mc were cultured in Hams F10 medium supplemented with: 2% (v/v) FCS, 0.02% basic fibroblast growth factor, 0.02% endothelin, 0.04% TPA, 1% penicillin/1% streptomycin, 1% ULTROSER G and 1% IBMX. Kc were cultured in Keratinocyte specific medium (KSFM) (GibcoÒ, Life technologies) supplemented with bovine pituitary extract and epidermal growth factor according to manufacturer’s instructions (GibcoÒ, Life Technologies). Primary human fibroblasts (Fb) were isolated from the dermal skin layer by placing the dermal tissue directly underneath a glass coverslip immersed in Dulbecco’s modified medium supplemented with 10% FCS and 1% penicillin/1% streptomycin. Cells were maintained in their respective specific media in a tissue culture incubator (MCO-175M, Sanyo, United Scientific) at 37 °C in a humidified atmosphere with 5% CO2. Cells used in all experiments did not exceed passage 12 and were seeded and maintained during experiments in their respective media. Routine weekly mycoplasma tests were employed with Hoechst nuclear dye to maintain sterility in all cultures. 2.3. HYP-PDT treatment HYP extract from the species Hypericum perforatum L. was prepared in 1 ml of 100% dimethyl sulfoxide (Merck) to obtain a 2 mM stock solution. HYP was activated with a diode pumped solid state, continuous wave, tunable laser that emits 561 nm yellow/green light. The fluence and power were kept constant at 5 J/cm2 and 20 mW, respectively. Primary human skin cells were seeded at 80% confluency (cell numbers of Fb, Mc and Kc varied due to different seeding efficiencies and growth rates) and allowed to adhere for 24 h. Thereafter, cells were incubated for 4 h with various doses of HYP (0.25–4 lM), prior to light activation. Following the incubation period, cells were rinsed twice with PBS and irradiated in PBS. All experiments that involved HYP-PDT treatment were protected from light at all times. Post-irradiation, the cells were replenished with fresh medium. The following controls were included when treating cells with HYP-PDT: C = untreated control (cells treated with neither HYP nor light); VC = vehicle control (cells treated with 0.15% DMSO and no light); H = HYP only control (cells treated with unactivated HYP and no light); L = light control (cells treated with light only); VL = vehicle light control (cells treated with 0.15% DMSO and light). At least 3 biological repeats (n = 3) were carried out using each cell type for each experiment that involved HYP-PDT treatment. 2.4. Cell viability assay Cells (2  104 Fb, 7  104 Mc and 7  104 Kc) were seeded in triplicate on 2 separate 96 well plates (TRPÒ, Switzerland: 92096). Following 24 h, cells were treated with a range of HYP-PDT doses (0.25 lM; 0.5 lM, 0.75 lM, 1 lM; 2 lM; 3 lM; 4 lM). Cell viability was assessed 24 h post-HYP-PDT treatment; using the Cell Proliferation Kit II (XTT) (Roche). Raw data for each cell type was normalized to its own vehicle control prior to statistical analysis. 2.5. Reactive oxygen species (ROS) assay Cells (2  104 Fb, 7  104 Mc and 7  104 Kc) were seeded per well, in triplicate, on 2 separate 96 well white plates. Following

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

24 h, Fb and Mc were treated with 1 lM and 3 lM HYP-PDT and Kc were treated with 3 lM and 4 lM HYP-PDT. 20 ,70 -Dichlorofluores cin diacetate (5 lM), prepared in medium, was added to each well immediately after treatment. Following a 20 min incubation, cells were washed twice with PBS and fresh PBS was added to each well prior to fluorescence detection; using the Cary Eclipse e104043731 fluorimeter (excitation wavelength = 488 nm; emission wavelength = 535 nm). Raw data for each cell type was normalized to its own vehicle control prior to statistical analysis. 2.6. Cell morphology analysis Cells (1  105 Fb, 1  105 Mc and 1.5  105 Kc) were cultured from 2 to 7 days on glass coverslips (Marienfeld) in 35 mm tissue culture dishes. Once Fb and Mc were treated with1 lM and 3 lM HYP-PDT and Kc were treated with 3 lM and 4 lM HYP-PDT, 1 lg/ml of Hoechst Live 33342 (Invitrogen, Molecular Probes) nuclear dye was added directly to the recovery medium for 20 min, at 24 h post-treatment. Following 2 washes with PBS, cells were fixed with 4% paraformaldehyde for 20 min and mounted on glass slides (Marienfeld) with 50 ll of mowiol containing an anti-fade compound (N-propyl gallate). Phase contrast images and their complementary multi-acquisition fluorescent images were acquired using the Axiovert 200M inverted fluorescent microscope (Zeiss). 2.7. Apoptosis analysis using fluorescent activated cell sorting (FACS) Cells (1  105 Fb, 1  105 Mc and 1.5  105 Kc) were seeded in duplicate on 2 separate 24 well plates., and treated with HYP-PDT lethal and non-lethal doses as stipulated in 2.6. Twenty four hours post treatment, cells were harvested from both the recovery medium and the plate and centrifuged at 5250 g for 10 min to obtain cell pellets. The cell pellets were re-suspended in 1 ml of ice cold PBS and further centrifuged at 5250 for 5 min. The pellets were then re-suspended in 300 ll of Annexin V-FITC binding buffer, prior to adding 4 ll Annexin V-FITC working solution (BD Biosciences) and 10 ll propidium iodide (PI) (50 lg/ml) (Roche). The samples were allowed to stand at room temperature for 15 min before FACS was conducted using the FACSAria 1 cell sorting machine which counted 10 000 events (BD Biosciences, USA). FACS data was analyzed using FlowJo Software Version 10.0.7.0. Raw data for each cell type was normalized to its own vehicle control prior to statistical analysis. 2.8. Data analysis Graphpad Prism (Version 5, Graphpad Software Inc.) was used to analyze raw data. Raw data was normalized taking into account the mean and standard error of the mean (mean ± SEM). Statistical differences were elicited using a one way ANOVA and Dunnet multiple comparison post-test. A One way analysis of variance (ANOVA) with Bonferroni post-tests was used to compare groups of data sets when comparing different cell types to each other. P-values less than: 0.05; 0.01 and 0.001 indicated significantly different values.

251

on any of the skin cells (data not shown but available on request). However, hypericin-PDT (HYP-PDT) induced a dose-dependent response in fibroblasts (Fb, Fig. 1), and a significant difference (p < 0.001) in cell viability was observed compared to the vehicle control, following different doses of HYP-PDT: 1 lM (56% viable cells), 2 lM (48%), 3 lM (33%) and 4 lM (34%). On the other hand, in melanocytes (Mc) HYP-PDT induced a significant difference in cell viability at doses of 2 lM (60%) (p < 0.05), 3 lM (61%) (p < 0.05) and 4 lM (38%) (p < 0.001) HYP-PDT (Fig. 1). However, in keratinocytes (Kc), HYP-PDT induced an initial significant difference (p < 0.05) in cell viability (79%) at a dose of 4 lM HYP-PDT (Fig. 1). The LD50 for HYP-PDT treated Fb and Mc occurred at 1.75 lM and 3.5 lM HYP-PDT respectively, but the LD50 for Kc occurred at a greater dose than 4 lM HYP-PDT (Table 1). Furthermore, these different cytotoxic profiles were confirmed statistically as the cell viability of Fb and Kc was significantly different at concentration greater than 1 lM HYP-PDT. Additionally, a significant difference between Mc and Kc occurred at a dose of 4 lM (p < 0.05) HYP-PDT. 3.2. Cell morphology 24 h post HYP-PDT Fb and Mc were treated with a non-lethal (1 lM) and a lethal dose (3 lM) of HYP-PDT, but Kc were treated with the highest HYP-PDT doses (3 lM and 4 lM), 24 h prior to staining with Hoechst nuclear dye and fixing the cells in 4% PFA. Phase contrast images of Fb (Fig. 2) revealed, a change in morphology in treated Fb (1 lM and 3 lM HYP-PDT) compared to the untreated control. Morphological alterations included cell shrinkage and pronounced vacuolation in the cytoplasm (Fig. 2, see white arrows), but no evidence of apoptotic nuclei was observed. Phase contrast images of Mc treated with 1 lM HYP-PDT (Fig. 2) displayed Mc with retracted dendrites that underwent shrinkage compared to the elegant Mc with long slender dendrites observed in untreated Mc. Furthermore, there was evidence of cytoplasmic vacuoles (Fig. 2, see white arrows). Interestingly, Mc which were treated with 3 lM HYP-PDT, exhibited slight dendritic shrinkage and cytoplasmic vacuolation but maintained cell integrity (Fig. 2, see white arrows). Despite being subjected to the highest HYP-PDT doses (3 lM and 4 lM), Kc looked unaffected and remained 80–90% confluent, maintaining cell shape and integrity, thereby looking very similar to the untreated control (Fig. 2). The corresponding fluorescent images of all the investigated cell types showed cytoplasmic localization of HYP being displayed as bright red punctae (Fig. 2, see white arrows) in Fb and Mc, whereas HYP seemed more evenly distributed in the cytoplasm and displayed less fluorescence in Kc. In addition, Hoechst-stained nuclei (Fig. 2) appeared shrunken 24 h post HYP-PDT in both the Fb and Mc. Interestingly, at the 24 h timepoint, HYP was excluded from all nuclei. 3.3. Intracellular ROS levels 30 min post HYP-PDT

3. Results

The measurement of intracellular ROS is reflective of intracellular hydrogen peroxide, which was measured in human skin cells, using the H2DCF-DA fluorescence based ROS assay, 20 min post HYP-PDT treatment. A significant 3.8 fold increase (p < 0.05) in ROS levels occurred in Fb at a dose of 3lM HYP-PDT compared to the vehicle control (Fig. 3). No significant differences in intracellular ROS levels were observed in both Mc and Kc (Fig. 3).

3.1. Cell viability 24 h post HYP-PDT

3.4. Apoptosis analysis 24 h post HYP-PDT

Skin cells were treated with a range of hypericin (HYP) concentrations (0.25 lM; 0.5 lM 1 lM; 2 lM; 3 lM; 4 lM) and cell viability was evaluated 24 h post treatment, using the XTT cell viability assay. Unactivated HYP did not yield a significant effect

Fluorescent activated cell sorting (FACS) analysis was conducted 24 h post HYP-PDT using Annexin V-FITC, an early apoptosis marker, and propidium iodide (PI) which is indicative of cell membrane integrity. In the Fb (Fig. 4), a significant (p < 0.001) early

252

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

Fig. 1. Cell viability 24 h post HYP-PDT assessed using the XTT cell viability assay in Fb, Mc and Kc. VC = vehicle control. LC = light control. Significant difference = ⁄=p < 0.05; ⁄⁄ = p < 0.01; ⁄⁄⁄ = p < 0.001. (Mean; SEM) (n = 3).

Table 1 Summary of cellular responses to HYP-PDT in human skin cells. Cell

Fb Mc Kc

Response to HYP-PDT Cell viability

Morphology

Intracellular ROS

Apoptosis

LD50 = 1.75 lM HYP-PDT LD50 = 3.5 lM HYP-PDT LD50 > 4 lM HYP-PDT

Distinctive changes Distinctive changes Non-distinctive changes

Increase

60%

No difference No difference

20% 14% post unactivated 3 lM HYP

apoptotic population (64%) was evident after treatment with 3 lM HYP-PDT (Fig. 4). Furthermore, a smaller, significant (p < 0.05) early apoptotic Mc population (20%) was prominent as a result of 3 lM HYP-PDT treatment (Fig. 4). Interestingly, 3 lM unactivated HYP resulted in a significant (p < 0.05) early apoptotic (14%) Kc population (Fig. 4) but HYP-PDT treatment did not result in a significant apoptotic Kc population. Although late apoptotic and necrotic populations were evident, none exhibited significant differences when compared to their vehicle control. 4. Discussion With the fact that the surrounding peri-lesional normal tissue would be inescapable to photosensitizer (PS) exposure and irradiation during PDT [35–38], this in vitro study investigated the effects of specifically hypericin-PDT (HYP-PDT) on representative cell types, representing the peri-lesional cells. The premise was that as these peri-lesional cells are in the same intimate environment; their resultant response to treatment could have bearing on the success or efficacy of the outcome – a concept known as the ‘‘bystander effect’’. This study therefore focused on the in vitro post-treatment effects of HYP-PDT on primary human melanocytes (Mc), keratinocytes (Kc) and fibroblasts (Fb) with respect to cell viability and morphology, intracellular ROS production and the induction of apoptosis. To date we and others, using the plant-based photosensitiser hypericin (HYP), have highlighted its powerful cytotoxic effects in several cancer cells including melanoma and non-melanoma skin cancer, cervical cancer cells; human myeloid leukemia cells; hepatocellular liver carcinoma cells and breast cancer cells [30,39–42]. Previously, we have shown that 3 lM activated HYP used during HYP-PDT induces cytotoxic effects in human melanoma cancer cells [17,18]. Therefore, a 4 h incubation time for HYP implemented in this study, was adjusted to correlate with optimal accumulation times of HYP in melanoma cells. Herewith, our results showed that the cytotoxic profile for human skin cells treated with unactivated HYP did not result in

any deleterious effects on any of the human skin cells after 24 h. This result was not surprising as it had been previously shown that HYP on its own has a cytostatic effect on cells but not a cytotoxic effect in the dark [20,43]. However, activated HYP-PDT resulted in varying LD50 values for treated Fb (1.75 lM), Mc (3.5 lM) and Kc (>4 lM) (Table 1). As a lower LD50 is indicative of a higher susceptibility to a drug or treatment, these different cell types clearly displayed a differential susceptibility to HYP-PDT – a very important point to consider when attempting to optimize and indeed increase the efficaciousness of anti-cancer treatment. The susceptibility profile of the three cell types (Table 1) also follows a logical ‘‘skin protection’’ layering from the upper epidermal area (Kc) through the basal layers (Mc) to the dermis (Fb). Moreover, reports suggest increased endogenous antioxidant levels or sophisticated antioxidant responses in the Kc and Mc, residing in the epidermis, suggesting that protection against ROS-based HYP-PDT induced cytotoxicity is greater in the epidermis than the dermis [44,45]. While not recorded for the purposes of this study in addition to the fact that a number of cultures were pooled to increase the amount of primary cells, the vast majority of cells were obtained from foreskins of Fitzpatrick skin types II–III. Interestingly, although the pigmentation levels in Mc were not quantified in this study, presence of melanin pigment could be a major contributor to the protection of Mc from PDT-induced ROS through the inherent antioxidant nature of this pigment [46–48]. This suggestion is further supported by our ROS assay data which showed a significant 3.8-fold increase in intracellular ROS in the Fb (Table 1) and no significant intracellular ROS production in Mc and Kc at the 3 lM HYP-PDT dose, despite 3 lM HYP-PDT having a significant cytotoxic effect in Mc. It is known that melanin has a dual role with regards to anti- and pro-oxidative functions [49]. Therefore, in the case of Mc, we postulate that both functions may be at play. Initially the melanin could function as an antioxidant and reduce the ROS by directing/sinking the HYP in melanosomes, as we have previously shown that HYP localizes to melanosomal membranes [17]. However, with increased oxidative stress an overwhelming oxidative burden may affect the integrity of the melanosomes and hence causes leakage of the toxic melanin intermediates, thereby leading to cell death. Our group is currently investigating the anti-and pro-oxidative functions of melanin in human skin cells, in response to HYP-PDT. In addition, this ties in with earlier work from our group that showed a decrease of melanin in melanoma cells leads to increasing susceptibility to cell death after HYP-PDT [14,16]. The potential correlation between the levels of melanin and susceptibility to cell death in peri-lesional melanocytes is certainly an avenue needing further exploration. Moreover, the low levels of intracellular ROS observed in Kc post treatment, correlated with their resistance to 3 lM HYP-PDT which correlated with their cell viability and cell

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

Control

1µM HYP-PDT

253

3µM HYP-PDT

Fb

Mc

Kc

Fig. 2. Human skin cells were stained with Hoechst nuclear dye (blue) 24 h post HYP-PDT. HYP has an endogenous red fluorescence. Phase contrast images are above the corresponding fluorescent image for each cell type. White arrows point to vacuoles and HYP punctae in phase and fluorescent images, respectively. Magnification = 200 for Fb and Mc and magnification = 100 for Kc. (n = 3).

254

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

Fig. 3. Intracellular ROS levels were measured 30 min post HYP-PDT in human skin cells using the H2DCF-DA fluorescence based ROS assay. VC = vehicle control; LC = light control; 3H = 3lM HYP; 3H + L = 3lM HYP-PDT. Significant difference = ⁄ = p < 0.05; p = ⁄⁄⁄ 3).

morphology. It is possible that the optimal uptake time of HYP in Kc may be different to the 4 h HYP uptake established in melanoma cells. Although we did not quantify HYP uptake in Kc in this study, HYP appeared less fluorescent in Kc compared to Mc and Fb, suggesting that Kc may have a different optimal uptake time. Therefore, the HYP taken up by Kc during the 4 h incubation period may have been insufficient to induce cytotoxicity or an increase in intracellular ROS levels. Overall, although we demonstrated in this study that 3 lM HYP-PDT is cytotoxic to Fb and Mc but not to Kc; it was pertinent to analyze the effect of this lethal dose (3 lM) compared to a non-lethal/control dose on cell morphology as even lower doses of HYP have been shown to cause a change in cellular morphology and thus reflect different cellular functions [27,40,50]. Non-lethal (1 lM) and lethal (3 lM) doses of HYP-PDT induced distinct morphological changes in both the Mc and the Fb (for a summary, see Table 1). Both these cell types exhibited evidence of cell shrinkage and vacuolation although the severity was different at 1 lM and 3 lM HYP-PDT. Interestingly, the Kc did not exhibit any aberrant morphological changes even at the highest HYP-PDT doses (3 lM and 4 lM, Table1). It was surprising that Fb morphology was also compromised at both lethal and non-lethal doses. However, we could not compare the morphology of Fb to other studies which investigated HYP-PDT effects, as they did not conduct morphological analysis [51,52]. In contrast, at the lethal dose, Mc looked healthier with few cytoplasmic vacuoles and longer dendrites compared to treatment with the non-lethal dose. One explanation for this, could be if some necrotic and late apoptotic cells were washed off the glass coverslips during the staining process. Furthermore, Mc may be reacting to the insult

of the lethal dose by upregulating their intracellular antioxidant system (catalase, superoxide dismutase and glutathione) and increasing production of melanin in response to the higher oxidative stress doses [48,49]. This is a well-known response in Mc and especially in their cancerous counterparts, melanoma [53,54]. Perhaps a tolerance to the HYP-PDT doses exists somewhere between 1 lM and 3 lM. It was not surprising that morphological changes were not evident in Kc as the highest HYP-PDT dose (4 lM) resulted in 79% viable cells. The enigma of why HYP does not enter the nucleus and effect damage is ongoing and it seems in agreement with published literature that HYP localization in the nucleus occurs at doses exceeding 20 lM HYP or after incubation time exceeding 8 h [19,20,55]. In our system (exposure to HYP for 4 h), fluorescent microscopy revealed cytoplasmic localization of HYP in distinct aggregates in all three cell types at a 24 h time point, Localization experiments at a much earlier time range using confocal microscopy may shed some light on the early nuclear tracking of hypericin – an important aspect to identify as it potentially contributes to DNA damage and cell death. As the consequence of high levels of intracellular ROS may lead to cell death induction, we investigated whether HYP-PDT-induced apoptosis in human skin cells [56–58]. Our FACS analysis confirmed significant early apoptotic Fb (64%) (p < 0.001) populations, 24 h post 3 lM HYP-PDT (Table 1). The early apoptotic Fb population correlated with these results in that the cells exhibited the highest susceptibility to treatment with a concomitant highest amount of intracellular ROS production. This result further correlated with literature stipulating that HYP-PDT causes accumulation of intracellular ROS which can lead to cell death via apoptosis and has been shown in several cancer cell types, including skin cancer cells [18–20,24,25,29,30,41,42,52,55,59]. A limitation of XTT cell viability assay, used to investigate cell viability post HYP-PDT, is that it measures metabolic activity as an indirect measure of cell viability. Therefore, some of the Fb (7%) and Mc (19%) may have not reached apoptosis and necrosis as indicated in Annexin V/PI analysis but may be undergoing another form of cell death such as autophagy induced cell death or necroptosis. Interestingly, the low levels of intracellular ROS measured in Mc and Kc were somewhat reflective in the low levels of early apoptotic population numbers in these cell types (20% and 14%, respectively). Low levels of apoptosis occur constitutively as a consequence of cellular turnover and these values (for the Mc and Kc) may simply be a reflection of these cellular processes [60–62]. In conclusion, this study highlighted the differential susceptibility of human skin cells to HYP-PDT. To the best of our knowledge, there have been no such studies investigating the effects of HYP-PDT in an in vitro system on all 3 human primary skin cell types. Our finding that 3 lM HYP-PDT is cytotoxic to Fb, to a lesser extent in Mc but completely refractory to Kc, further highlights the

Fig. 4. Apoptosis was analyzed in human skin cells using FACS, 24 h post HYP-PDT. Annexin V-FITC is a marker of early apoptosis and PI is a marker of membrane integrity. Gray lines = viable cells. Black = early apoptotic cells. White = late apoptotic/necrotic cells. VC = vehicle control; LC = light control; 3H = 3lM HYP; 3H + L = 3lM HYP-PDT. Significant difference = ⁄ = p < 0.05; ⁄⁄ = p < 0.01. (Mean; SEM) (n > 3).

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

specific, cellular functional differences between the cells of the skin. Fascinatingly, all these cells exist as a unit in the same environment. As a potential adjunctive therapy in both melanoma and non-melanoma therapies, it is imperative to know whether the dosages should be specific for the tumor environment or tailored for specific locations containing certain cell types. The skin exists as a milieu of different cell types and thus this study is one of the first to suggest that the skin cell types surrounding the tumor i.e. the bystander cells, have a their own inherent resistant mechanisms to therapy that are not generalized but very specific. Moreover, lineages of a number or all of these cells could contribute toward the stemness of the tumor which further translates into resistance mechanisms. These results thus broaden our knowledge with respect to the cellular response of the epidermis to photodynamic therapies and will possibly enhance the efficacy of future photobiological treatments. Ethics Isolation of skin cells from human skin samples and biopsies were obtained after full consent and this study was granted approval by our Human Research Ethics Institutional Board (REC REF 493/2009). Acknowledgement The Laser for the irradiation and activation of hypericin was obtained via a grant from the National Laser Centre at the South African Centre for Scientific and Industrial Research (CSIR, Grant number – NLC-LREHGOO-CON-001). References [1] P. Agostinis, K. Berg, K.A. Cengel, T.H. Foster, A.W. Girotti, S.O. Gollnick, et al., Photodynamic therapy of cancer: an update, CA. Cancer J. Clin. 61 (2011) 250– 281, http://dx.doi.org/10.3322/caac.20114. [2] C.A. Robertson, D.H. Evans, H. Abrahamse, Photodynamic therapy (PDT): a short review on cellular mechanisms and cancer research applications for PDT, J. Photochem. Photobiol. B 96 (2009) 1–8, http://dx.doi.org/10.1016/ j.jphotobiol.2009.04.001. [3] P.G. Calzavara- Pinton, P.G. Calzavara-Pinton, M. Venturini, R. Sala, Photodynamic therapy: update 2006. Part 1: photochemistry and photobiology, J. Eur. Acad. Dermatology Venereol. 21 (2007) 293–302, http:// dx.doi.org/10.1111/j.1468-3083.2006.01902.x. [4] P. Agostinis, K. Berg, K.A. Cengel, T.H. Foster, A.W. Girotti, S.O. Gollnick, et al., Photodynamic therapy of cancer: an update, CA. Cancer J. Clin. (2011) 250– 281, http://dx.doi.org/10.3322/caac.20114. [5] S.H. Ibbotson, J. Ferguson, Ambulatory photodynamic therapy using low irradiance inorganic light-emitting diodes for the treatment of non-melanoma skin cancer: an open study, Photodermatol. Photoimmunol. Photomed. 28 (2012) 235–239, http://dx.doi.org/10.1111/j.1600-0781.2012.00681.x. [6] S. Jessop, H. Stubbings, R. Sayed, J. Duncan-Smith, J.W. Schneider, H.F. Jordaan, Regional clinical registry data show increased incidence of cutaneous melanoma in Cape Town, South African Med. J. 98 (2008) 197–199. [7] L. Pérez-Pérez, J. García-Gavín, Y. Gilaberte, Daylight-mediated photodynamic therapy in Spain: advantages and disadvantages, Actas Dermosifiliogr. 105 (2014) 663–674, http://dx.doi.org/10.1016/j.ad.2013.10.021. [8] C.A. Morton, H.C. Wulf, R.M. Szeimies, Y. Gilaberte, N. Basset-Seguin, E. Sotiriou, et al., Practical approach to the use of daylight photodynamic therapy with topical methyl aminolevulinate for actinic keratosis: a European consensus, J. Eur. Acad. Dermatol. Venereol. (2015), http://dx.doi.org/ 10.1111/jdv.12974. [9] K.L. Lane, W. Hovenic, K. Ball, C.B. Zachary, Daylight photodynamic therapy: the Southern California experience, Lasers Surg. Med. (2015), http://dx.doi.org/ 10.1002/lsm.22323. [10] M.C.A. Issa, M. Manela-Azulay, Photodynamic therapy: a review of the literature and image documentation, An. Bras. Dermatol. 85 (2010) 501–511. [11] A. Castano, T. Demidova, M. Hamblin, Mechanisms in photodynamic therapy: part one—photosensitizers, photochemistry and cellular localization, Photodiagnosis Photodyn. Ther. 1 (2004) 279–293, http://dx.doi.org/10.1016/ S1572-1000(05)00007-4. [12] M.T. Wan, J.Y. Lin, Current evidence and applications of photodynamic therapy in dermatology, Clin. Cosmet. Investig. Dermatol. 7 (2014) 145–163, http:// dx.doi.org/10.2147/CCID.S35334.

255

[13] L.M. Davids, B. Kleemann, S. Cooper, S.H. Kidson, Melanomas display increased cytoprotection to hypericin-mediated cytotoxicity through the induction of autophagy, Cell Biol. Int. 33 (2009) 1065–1072, http://dx.doi.org/10.1016/ j.cellbi.2009.06.026. [14] K.V. Sharma, N. Bowers, L.M. Davids, Photodynamic therapy-induced killing is enhanced in depigmented metastatic melanoma cells, Cell Biol. Int. 35 (2011) 939–944, http://dx.doi.org/10.1042/CBI20110103. [15] K.V. Sharma, L.M. Davids, Hypericin-PDT-induced rapid necrotic death in human squamous cell carcinoma cultures after multiple treatment, Cell Biol. Int. 36 (2012) 1261–1266, http://dx.doi.org/10.1042/CBI20120108. [16] K.V. Sharma, L.M. Davids, Depigmentation in melanomas increases the efficacy of hypericin-mediated photodynamic-induced cell death, Photodiagnosis Photodyn. Ther. 9 (2012) 156–163, http://dx.doi.org/10.1016/j.pdpdt.2011. 09.003. [17] B. Kleemann, B. Loos, T.J. Scriba, D. Lang, L.M. Davids, St John’s Wort (Hypericum perforatum L.) photomedicine: hypericin-photodynamic therapy induces metastatic melanoma cell death, PLoS ONE 9 (2014) e103762, http:// dx.doi.org/10.1371/journal.pone.0103762. [18] L.M. Davids, B. Kleemann, D. Kacerovská, K. Pizinger, S.H. Kidson, Hypericin phototoxicity induces different modes of cell death in melanoma and human skin cells, J. Photochem. Photobiol. B 91 (2008) 67–76, http://dx.doi.org/ 10.1016/j.jphotobiol.2008.01.011. [19] P. Agostinis, A. Vantieghem, W. Merlevede, P.A.M. de Witte, Hypericin in cancer treatment: more light on the way, Int. J. Biochem. Cell Biol. 34 (2002) 221–241, http://dx.doi.org/10.1016/S1357-2725(01)00126-1. [20] B. Krammer, T. Verwanger, Molecular response to hypericin-induced photodamage, Curr. Med. Chem. 19 (2012) 793–798. [21] G. Jori, E. Reddi, The role of lipoproteins in the delivery of tumour-targeting photosensitizers, Int. J. Biochem. 25 (1993) 1369–1375, http://dx.doi.org/ 10.1016/0020-711X(93)90684-7. [22] A. Karioti, A.R. Bilia, Hypericins as potential leads for new therapeutics, Int. J. Mol. Sci. 11 (2010) 562–594, http://dx.doi.org/10.3390/ijms11020562. [23] M. Van de Putte, T. Roskams, J.R. Vandenheede, P. Agostinis, P.A.M. de Witte, Elucidation of the tumoritropic principle of hypericin, Br. J. Cancer 92 (2005) 1406–1413, http://dx.doi.org/10.1038/sj.bjc.6602512. [24] T. Kiesslich, B. Krammer, K. Plaetzer, Cellular mechanisms and prospective applications of hypericin in photodynamic therapy, Curr. Med. Chem. 13 (2006) 2189–2204. [25] C.M. Schempp, B. Simon-haarhaus, C.C. Termeer, J.C. Simon, Hypericin photo-induced apoptosis involves the tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and activation of caspase-8, FEBS Lett. 1 (2001) 26–30, http://dx.doi.org/10.1016/S0014-5793(01)02268-2. [26] A.R. Kamuhabwa, P. Agostinis, M.A. D’Hallewin, A. Kasran, P.A. de Witte, Photodynamic activity of hypericin in human urinary bladder carcinoma cells, Anticancer Res. 20 (2000) 2579–2584. [27] J. Koval, J. Mikes, R. Jendzelovsky´, M. Kello, P. Solár, P. Fedorocko, Degradation of HER2 receptor through hypericin-mediated photodynamic therapy, Photochem. Photobiol. 86 (2010) 200–205, http://dx.doi.org/10.1111/j.17511097.2009.00639.x. [28] R. Sanovic, T. Verwanger, A. Hartl, B. Krammer, Low dose hypericin-PDT induces complete tumor regression in BALB/c mice bearing CT26 colon carcinoma, Photodiagnosis Photodyn. Ther. 8 (2011) 291–296, http:// dx.doi.org/10.1016/j.pdpdt.2011.04.003. [29] V. Huntosova, Z. Nadova, L. Dzurova, V. Jakusova, F. Sureau, P. Miskovsky, Cell death response of U87 glioma cells on hypericin photoactivation is mediated by dynamics of hypericin subcellular distribution and its aggregation in cellular organelles, Photochem. Photobiol. Sci. 11 (2012) 1428–1436, http:// dx.doi.org/10.1039/c2 pp05409d. [30] M. Barathan, V. Mariappan, E.M. Shankar, B.J.J. Abdullah, K.L. Goh, J. Vadivelu, Hypericin-photodynamic therapy leads to interleukin-6 secretion by HepG2 cells and their apoptosis via recruitment of BH3 interacting-domain death agonist and caspases, Cell Death Dis. 4 (2013) e697–710, http://dx.doi.org/ 10.1038/cddis.2013.219. [31] C. Hadjur, M.J. Richard, M.O. Parat, A. Favier, P. Jardon, Photodynamically induced cytotoxicity of hypericin dye on human fibroblast cell line MRC5, J. Photochem. Photobiol. B 27 (1995) 139–146. [32] A. Bernd, S. Simon, A. Ramirez Bosca, S. Kippenberger, J. Diaz Alperi, J. Miquel, et al., Phototoxic effects of hypericum extract in cultures of human keratinocytes compared with those of psoralen, Photochem. Photobiol. 69 (1999) 218–221. [33] J. Dahle, The bystander effect in photodynamic inactivation of cells, Biochim. Biophys. Acta – Gen. Subj. 1475 (2000) 273–280, http://dx.doi.org/10.1016/ S0304-4165(00)00077-5. [34] J. Dahle, O. Kaalhus, J. Moan, H.B. Steen, Cooperative effects of photodynamic treatment of cells in microcolonies, Proc. Natl. Acad. Sci. 94 (1997) 1773–1778, http://dx.doi.org/10.1073/pnas.94.5.1773. [35] L.R. Braathen, R.-M. Szeimies, N. Basset-Seguin, R. Bissonnette, P. Foley, D. Pariser, et al., Guidelines on the use of photodynamic therapy for nonmelanoma skin cancer: an international consensus, J. Am. Acad. Dermatol. 56 (2007) 125–143, http://dx.doi.org/10.1016/j.jaad.2006.06.006. International Society for Photodynamic Therapy in Dermatology, 2005. [36] K. Svanberg, T. Andersson, D. Killander, I. Wang, U. Stenram, S. AnderssonEngels, et al., Photodynamic therapy of non-melanoma malignant tumours of the skin using topical d-amino levulinic acid sensitization and laser irradiation, Br. J. Dermatol. 130 (1994) 743–751, http://dx.doi.org/10.1111/j.13652133.1994.tb03412.x.

256

A. Popovic et al. / Journal of Photochemistry and Photobiology B: Biology 149 (2015) 249–256

[37] X.-L. Wang, H.-W. Wang, M.-X. Guo, S.-Z. Xu, Treatment of skin cancer and precancer using topical ALA-PDT–a single hospital experience, Photodiagnosis Photodyn. Ther. 5 (2008) 127–133, http://dx.doi.org/10.1016/j.pdpdt.2008.05. 003. [38] N. Zeitouni, A. Oseroff, S. Shieh, Photodynamic therapy for nonmelanoma skin cancers current review and update, Mol. Immunol. 39 (2003) 1133–1136, http://dx.doi.org/10.1016/S0161-5890(03)00083-X. [39] P. Ferenc, P. Solár, J. Kleban, J. Mikes, P. Fedorocko, Down-regulation of Bcl-2 and Akt induced by combination of photoactivated hypericin and genistein in human breast cancer cells, J. Photochem. Photobiol. B 98 (2010) 25–34, http:// dx.doi.org/10.1016/j.jphotobiol.2009.10.004. [40] V. Sacková, P. Fedorocko, B. Szilárdiová, J. Mikes, J. Kleban, Hypericin-induced photocytotoxicity is connected with G2/M arrest in HT-29 and S-phase arrest in U937 cells, Photochem. Photobiol. 82 (2006) 1285–1291, http://dx.doi.org/ 10.1562/2006-02-22-RA-806. [41] A. Vantieghem, Y. Xu, W. Declercq, P. Vandenabeele, G. Denecker, J.R. Vandenheede, et al., Different pathways mediate cytochrome c release after photodynamic therapy with hypericin, Photochem. Photobiol. 74 (2007) 133– 142, http://dx.doi.org/10.1562/0031-8655(2001)0740133DPMCCR2.0.CO2. [42] A. Vantieghem, Hypericin-induced photosensitization of HeLa cells leads to apoptosis or necrosis Involvement of cytochrome c and procaspase-3 activation in the mechanism of apoptosis, FEBS Lett. 440 (1998) 19–24, http://dx.doi.org/10.1016/S0014-5793(98)01416-1. [43] A. Kubin, F. Wierrani, U. Burner, G. Ualth, W. Grünberger, Hypericin-the facts about a controversial agent, Curr. Pharm. Des. 11 (2005) 233–253. [44] R. Kohen, Skin antioxidants: their role in aging and in oxidative stress–new approaches for their evaluation, Biomed. Pharmacother. 53 (1999) 181–192, http://dx.doi.org/10.1016/S0753-3322(99)80087-0. [45] Y. Shindo, E. Witt, D. Han, W. Epstein, L. Packer, Enzymic and non-enzymic antioxidants in epidermis and dermis of human skin, J. Invest. Dermatol. 102 (1994) 122–124, http://dx.doi.org/10.1111/1523-1747.ep12371744. [46] M.R. Chedekel, L. Zeise, Protection of hair and skin from environmental damage with melanin, SÖFW-J. 125 (1999) 14–18. [47] C.S. Sander, H. Chang, F. Hamm, P. Elsner, J.J. Thiele, Role of oxidative stress and the antioxidant network in cutaneous carcinogenesis, Int. J. Dermatol. 43 (2004) 326–335, http://dx.doi.org/10.1111/j.1365-4632.2004.02222.x. [48] H. Swalwell, J. Latimer, R.M. Haywood, M.A. Birch-Machin, Investigating the role of melanin in UVA/UVB- and hydrogen peroxide-induced cellular and mitochondrial ROS production and mitochondrial DNA damage in human melanoma cells, Free Radic. Biol. Med. 52 (2012) 626–634, http://dx.doi.org/ 10.1016/j.freeradbiomed.2011.11.019. [49] F.L. Meyskens Jr., P. Farmer, J.P. Fruehauf, Redox regulation in human melanocytes and melanoma, Chem. Res. Toxicol. 14 (2001) 148–154, http:// dx.doi.org/10.1034/j.1600-0749.2001.140303.x.

[50] A.L. Vandenbogaerde, J.F. Cuveele, P. Proot, B.E. Himpens, W.J. Merlevede, P.A. de Witte, Differential cytotoxic effects induced after photosensitization by hypericin, J. Photochem. Photobiol. B Biol. 38 (1997) 136–142, http:// dx.doi.org/10.1016/S1011-1344(96)07446-5. [51] N. Kashef, Y.S. Borghei, G.E. Djavid, Photodynamic effect of hypericin on the microorganisms and primary human fibroblasts, Photodiagnosis Photodyn. Ther. 10 (2013) 150–155, http://dx.doi.org/10.1016/j.pdpdt.2012.11.007. [52] A. Rezusta, P. López-Chicón, M.P. Paz-Cristobal, M. Alemany-Ribes, D. RoyoDíez, M. Agut, et al., In vitro fungicidal photodynamic effect of hypericin on Candida species, Photochem. Photobiol. 88 (2012) 613–619, http://dx.doi.org/ 10.1111/j.1751-1097.2011.01053.x. [53] C. Hadjur, M.J. Richard, M.O. Parat, P. Jardon, A. Favier, Photodynamic effects of hypericin on lipid peroxidation and antioxidant status in melanoma cells, Photochem. Photobiol. 64 (1996) 375–381. [54] M. Picardo, P. Grammatico, F. Roccella, M. Roccella, M. Grandinetti, G. del Porto, et al., Imbalance in the antioxidant pool in melanoma cells and normal melanocytes from patients with melanoma, J. Invest. Dermatol. 107 (1996) 322–326, http://dx.doi.org/10.1111/1523-1747.ep12363163. [55] T.A. Theodossiou, J.S. Hothersall, P.A. De Witte, A. Pantos, P. Agostinis, The multifaceted photocytotoxic profile of hypericin, Mol. Pharm. 6 (2009) 1775– 1789, http://dx.doi.org/10.1021/mp900166q. [56] K. Martin, J. Barrett, Reactive oxygen species as double-edged swords in cellular processes: low-dose cell signaling versus high-dose toxicity, Hum. Exp. Toxicol. 21 (2002) 71–75, http://dx.doi.org/10.1191/0960327102ht213oa. [57] H.-U. Simon, A. Haj-Yehia, F. Levi-Schaffer, Role of reactive oxygen species (ROS) in apoptosis induction, Apoptosis 5 (2000) 415–418, http://dx.doi.org/ 10.1023/A:1009616228304. [58] S. Tan, M. Wood, P. Maher, Oxidative stress induces a form of programmed cell death with characteristics of both apoptosis and necrosis in neuronal cells, J. Neurochem. 71 (2002) 95–105, http://dx.doi.org/10.1046/j.14714159.1998.71010095.x. [59] T. Verfaillie, A. Van Vliet, A.D. Garg, M. Dewaele, N. Rubio, S. Gupta, et al., Proapoptotic signaling induced by photo-oxidative ER stress is amplified by Noxa, not Bim, Biochem. Biophys. Res. Commun. 3 (2013) 500–556, http:// dx.doi.org/10.1016/j.bbrc.2013.07.107. [60] R.H. Engel, A.M. Evans, Oxidative stress and apoptosis: a new treatment paradigm in cancer, Front. Biosci. 11 (2006) 300, http://dx.doi.org/10.2741/ 1798. [61] J.M. Matés, J.A. Segura, F.J. Alonso, J. Márquez, Intracellular redox status and oxidative stress: implications for cell proliferation, apoptosis, and carcinogenesis, Arch. Toxicol. 82 (2008) 273–299, http://dx.doi.org/10.1007/ s00204-008-0304-z. [62] K.U. Schallreuter, Advances in melanocyte basic science research, Dermatol. Clin. 25 (2007) 283–291, http://dx.doi.org/10.1016/j.det.2007.04.010 (vii).