Cellular Stress and p53-Associated Apoptosis by Juniperus communis L. Berry Extract Treatment in the Human SH-SY5Y Neuroblastoma Cells

Plant phenolics have shown to activate apoptotic cell death in different tumourigenic cell lines. In this study, we evaluated the effects of juniper berry extract (Juniperus communis L.) on p53 protein, gene expression and DNA fragmentation in human neuroblastoma SH-SY5Y cells. In addition, we analyzed the phenolic composition of the extract. We found that juniper berry extract activated cellular relocalization of p53 and DNA fragmentation-dependent cell death. Differentially expressed genes between treated and non-treated cells were evaluated with the cDNA-RDA (representational difference analysis) method at the early time point of apoptotic process when p53 started to be activated and no caspase activity was detected. Twenty one overexpressed genes related to cellular stress, protein synthesis, cell survival and death were detected. Interestingly, they included endoplasmic reticulum (ER) stress inducer and sensor HSPA5 and other ER stress-related genes CALM2 and YKT6 indicating that ER stress response was involved in juniper berry extract mediated cell death. In composition analysis, we identified and quantified low concentrations of fifteen phenolic compounds. The main groups of them were flavones, flavonols, phenolic acids, flavanol and biflavonoid including glycosides of quercetin, apigenin, isoscutellarein and hypolaetin. It is suggested that juniper berry extract induced the p53-associated apoptosis through the potentiation and synergism by several phenolic compounds.


Introduction
Cell signaling pathways related to apoptosis and cell cycle have an essential role in development and progression of complex diseases such as cancer. One of the key orchestrators of those cellular functions is the tumour protein p53 [1]. The fact that p53 dysfunctions in most cancers indicates its essential role in tumour suppression: p53 has been found to be mutated in half of the cases while other cases often possess dysregulation of its upstream signaling pathways [2]. Cellular stress, including DNA damage, ribosomal and endoplasmic reticulum stress activates p53 whereas its levels are strictly maintained low under normal conditions. Activated p53 translocates into the nucleus where it modulates the expression of over hundred genes [3]. Furthermore, p53 acts in cytoplasm where it regulates the mitochondrial membrane permeabilization and directly interacts with other proteins [4,5]. In addition to these well-known tumour suppressing activities of p53, recent studies have revealed that it has a central role in tumour-related metabolism, cell-cell communication and metastasis as well [1].
Juniper (Juniperus communis L., Cupressaceae) is an evergreen coniferous shrub or small tree growing on the temperate regions of the northern hemisphere. Mature female cones of juniper are generally called berries for their "berry-like" appearance, and they are used to flavour game meat and alcoholic beverages, e.g., gin and beer. In traditional herbal medicine juniper has been used for many purposes as e.g., treating wounds, pain, fevers, rheumatism, snakebites, swellings, gastrointestinal infections, bronchitis and cancers [6,7] and it has been claimed to possess also diuretic, antiseptic, carminative, stomachic and antirheumatic properties [8].
Although juniper berries are mainly used for their aromatic properties, they also contain bioactive plant phenolics, e.g., quercetin glycosides [9], which might explain at least some of the claimed health-promoting effects of juniper. It is known that naturally derived phenolic compounds can affect different cell signalling pathways inducing both cell cycle progression and apoptosis [10]. The mechanisms of single compounds have been studied more closely but the interest on the mixtures of compounds or plant extracts has been raised over the last years [11]. Suggested benefits of using combinations of different therapeutic agents include reduced toxicity based on the lower-dose usage of drugs and decreased development of drug resistance [12].
We have shown earlier that the juniper berry extract can induce a p53-dependent cell death in human SH-SY5Y neuroblastoma cells [13]. In addition, the anti-and pro-oxidant capacities of the extract have been analysed in biochemical test models [14]. In these studies, the juniper berry extract was prepared using a hydrodistillation process to remove volatile compounds [15], and their absence was verified with chromatographic analysis. Therefore, the observed anti-and/or pro-oxidant and cell death-inducing effects of the extract did not result from toxicity of volatile compounds but rather by specific non-volatile compounds mediated cellular mechanisms [13,14]. Studies on the bioactivity of juniper berry extracts without volatile components are rare in literature.
In the present work, we have studied the phenolic composition and biological effects of aqueous juniper berry extract in more detail. Therefore, we have examined the mechanisms of juniper extract induced apoptosis by analyzing the p53 translocation, and gene expression and DNA fragmentation. The identification and quantification of phenolic compounds was performed using chromatographic methods. The results showed that the juniper extract contained several compounds with cell cycle and apoptosis regulating effects. The extract treatment of human SH-SY5Y neuroblastoma cells resulted in the increased DNA damage, p53 translocation from the cytoplasmic to nuclear compartment and overexpression of several genes in parallel with the reactivation of impaired apoptosis.

Localization of Protein p53 in Cytoplasm and Nucleus
The accumulation of p53 after a 12 h-treatment with the juniper berry extract in human neuroblastoma SH-SY5Y cells was about two fold, i.e., in the same level than in our previous study [13] with statistically significant increases. Here we show that the accumulated p53 starts to translocate into the nucleus 12 h after treatment, and the highest nuclear p53 concentrations were detected after 24, 36 and 48 h ( Figure 2). Translocation of p53 declines after 72 h of treatment. Observations were statistically significant after 36 and 48 h of treatment. The amount of cytoplasmic p53 is stable within the treatments when compared to the untreated control cells although the variation of nuclear amounts was detected between treatments. These observations suggest the transcriptional activation of p53 after treatment with juniper berry extract.

Localization of Protein p53 in Cytoplasm and Nucleus
The accumulation of p53 after a 12 h-treatment with the juniper berry extract in human neuroblastoma SH-SY5Y cells was about two fold, i.e., in the same level than in our previous study [13] with statistically significant increases. Here we show that the accumulated p53 starts to translocate into the nucleus 12 h after treatment, and the highest nuclear p53 concentrations were detected after 24, 36 and 48 h ( Figure 2). Translocation of p53 declines after 72 h of treatment. Observations were statistically significant after 36 and 48 h of treatment. The amount of cytoplasmic p53 is stable within the treatments when compared to the untreated control cells although the variation of nuclear amounts was detected between treatments. These observations suggest the transcriptional activation of p53 after treatment with juniper berry extract.

DNA Fragmentation and Morphology
Fragmented DNA was detected first after 24 h-treatment but the most abundant laddering was detected after a 72 h-treatment ( Figure 3A). The DNA of non-treated control cell was also analyzed (results not shown) and some minor breakage of DNA was detected later after 48 and 72 h but not 24 h after treatment. In addition, the visible DNA ladders were remarkably weaker compared to ladders detected from treated cells. Morphological changes in the treated and non-treated cells were observed by a microscopic examination. After the 12 h-treatment before the cell death occurred, the amount of 56% (SD 2.5, median 56) of treated cells were slightly shrunk and membranes slightly disrupted compared to the non-treated control cells but the amount of floating dead cells was comparable to the non-treated cells ( Figure 2B,C). The detachment of cells increased gradually 24, 36, 48 and 72 h after the treatment. The breakage of DNA is one of the final steps of apoptotic cell death [26], and it is described as a hallmark of apoptosis [27]. These results suggest that the juniper berry extract induced the apoptotic cell death after 24 h of treatment.

DNA Fragmentation and Morphology
Fragmented DNA was detected first after 24 h-treatment but the most abundant laddering was detected after a 72 h-treatment ( Figure 3A). The DNA of non-treated control cell was also analyzed (results not shown) and some minor breakage of DNA was detected later after 48 and 72 h but not 24 h after treatment. In addition, the visible DNA ladders were remarkably weaker compared to ladders detected from treated cells. Morphological changes in the treated and non-treated cells were observed by a microscopic examination. After the 12 h-treatment before the cell death occurred, the amount of 56% (SD 2.5, median 56) of treated cells were slightly shrunk and membranes slightly disrupted compared to the non-treated control cells but the amount of floating dead cells was comparable to the non-treated cells ( Figure 2B,C). The detachment of cells increased gradually 24, 36, 48 and 72 h after the treatment. The breakage of DNA is one of the final steps of apoptotic cell death [26], and it is described as a hallmark of apoptosis [27]. These results suggest that the juniper berry extract induced the apoptotic cell death after 24 h of treatment.

Differentially Expressed Genes
Changes in gene expression caused by the 12 h-treatment with juniper berry extract were examined by the cDNA representational difference analysis (RDA) to detect overexpressed genes in treated cells in comparison to non-treated control cells. This polymerase chain reaction-based, qualitative technique is able to detect very small mRNA-level differences in gene expression. The gene expression was analyzed at the time point when the amount of p53 starts to accumulate but there are no detectable apoptotic features of cell death.
Twenty one overexpressed protein-coding genes were identified by BLAST search (Table 2). To clarify the varied nomenclature of genes and proteins, their preferred names and synonyms are listed in Table 3. To analyze the functions of proteins encoded by differentially expressed genes, they were assessed using literature search and UniProt KB and SwissProt databases (Table 4). Proteins were subdivided to seven different functional groups: cell death and cell survival proteins, cell cycle proteins, cellular stress proteins, cell shape, motility and polarity proteins, protein synthesis proteins,

Differentially Expressed Genes
Changes in gene expression caused by the 12 h-treatment with juniper berry extract were examined by the cDNA representational difference analysis (RDA) to detect overexpressed genes in treated cells in comparison to non-treated control cells. This polymerase chain reaction-based, qualitative technique is able to detect very small mRNA-level differences in gene expression. The gene expression was analyzed at the time point when the amount of p53 starts to accumulate but there are no detectable apoptotic features of cell death.
Twenty one overexpressed protein-coding genes were identified by BLAST search (Table 2). To clarify the varied nomenclature of genes and proteins, their preferred names and synonyms are listed in Table 3. To analyze the functions of proteins encoded by differentially expressed genes, they were assessed using literature search and UniProt KB and SwissProt databases (Table 4). Proteins were subdivided to seven different functional groups: cell death and cell survival proteins, cell cycle proteins, cellular stress proteins, cell shape, motility and polarity proteins, protein synthesis proteins, Ca 2+ -signaling proteins and proteins with enzymatic and protein-protein interactions. Further, the interactions of proteins with each other, p53 and cell death or survival were refined by STRING analysis accompanied with literature search (Figures 4 and 5).
binds with several proteins including Mink1 [45] and calmodulin [4] TTC37 Subunit of a SKI complex which assists exosome in a mRNA degradation UP and associate with PAF in transcription [57]
At the time when these genes were detected, the cells contained both pro-apoptotic and pro-survival signals and several proteins operating bifunctionally in cell survival and death either activating or inhibiting, i.e., CALM2 (calmodulin) and RAC1 (Rac1/TC25). Interestingly, genes related to ER stress (CALM2, HSPA5, YKT6), ribosomal stress (RPL6, RPLP0) and genomic stress (MORF4L1) were over-expressed. Direct interactions with p53 were found for ER stress sensor HSPA5 (BiP, GRP78), which is suggested to inactivate p53 via binding [23] and for ribosomal protein RPL6, which activates 53 via inhibiting Mdm2 [43]. STRADA and RAC1 modulate activations of p53 via LKB1. STRADA mediates G1 cell cycle arrest [44] whereas RAC1 promotes cell survival via PAKs [54]. The major regulator of calcium-mediated signaling CALM2 (calmodulin) modulates cell survival and  Table 4. Bold lines represent plasma and nuclear membranes.

Functions Genes Specific Functions for Proteins Encoded by Differentially Expressed Genes
Cell death and cell survival

ARHGAP35
Positive regulator of cell cycle in lung carcinoma cells [39] CALM2 Controls cell cycle progression [28] CSDE1 Activates IRES-mediated translation of cell cycle regulator PITSLRE during mitosis [40] MAT2A Inhibits the growth of liver cancer cells via SAMe [41] MORF4L1 Activator of tumour suppressor-mediated cell cycle arrest UP accompanied with DNA repair [42] RPL6 Inhibition of cell cycle via stabilization of p53 by suppressing MDM2 activity [43] STRADA Mediates G1 cell cycle arrest via activating tumor suppressor LKB1 [44] STRN4

Required for the completion of cytokinesis (abscission) while interacting with Mink1 [45]
YKT6 Over-expression alters cell cycle by increasing mitotic index, DNA synthesis and decreasing cell size [46] Cellular stress

SF3B2
Splices mRNA UP

TTC37
As a subunit of SKI complex, mediates RNA surveillance with PAF complex [57] and assists exosome in mRNA degradation UP Ca 2+ -signaling

Functions Genes Specific Functions for Proteins Encoded by Differentially Expressed Genes
Enzymatic and protein-protein interactions

Discussion
Numerous plant phenolics have been discovered to cause p53 accumulation in cells and this effect has been detected also in studies with flavonoid-rich plant extracts. Plant extracts are complicated mixtures of numerous bioactive compounds, which makes their testing challenging. In the present study, we have shown that the juniper extract can induce the p53 translocation from the cytosol into the nucleus and this occurred before the DNA fragmentation and cell death in the human neuroblastoma SH-SY5Y cells. The cleavage of DNA molecule into smaller fragments was detected 24, 48 and 72 h after the treatment while the most abundant accumulation and nuclear translocation of p53 occurred already before the DNA damage after 24-48 h. The breakage of DNA is one of the final steps of apoptotic cell death [26], and it is described as a hallmark of apoptosis [27]. Therefore, the p53-mediated DNA fragmentation suggests that the juniper berry extract activated cell death was mediated by active apoptotic mechanisms. Here, these changes were seen already at the early time point before a fully activated cell death when no caspase activity was detected but p53 started to be activated.
The human neuroblastoma cells are shown to possess sensitivity on toxicity treatments [25] and therefore chosen for testing of plant extracts in our studies [13]. The data concerning the role of p53 in neuroblastoma is controversial but the SH-SY5Y cell line contains wild-type p53 capable to transcription activity [58][59][60]. Mdm2 is the cytosolic main inactivator of p53 and the translocation of p53 in nucleus requires a breakage of a Mdm2-p53 complex accompanied with acetylation and phosphorylation [61]. Pro-apoptotic target genes of p53 include Pum, Noxa, Bax and Bid [62] but the major transcription target is p21 which contribute cell cycle arrest in G1 and G2 phases [63]. Transcription-independent actions of p53 include the interactions of p53 with different proteins in cytosol. For example p53 interacts with anti-apoptotic proteins Bcl-2 and Bcl-xL to enable pro-apoptotic proteins Bax and Bid to translocate on mitochondrial outer membrane [64].
To understand the molecular mechanisms related to the observed cell death, differentially expressed genes were identified and their functions analyzed. One aim of differential expression analysis was to discover the specific genes involved in the cellular response for the juniper extract. After treatment of neuroblastoma cells with 10 µg/mL of juniper berry extract, differentially expressed genes were analyzed by cDNA RDA method. Gene expression was evaluated 12 h after treatment with juniper berry extract, which reveals the upregulated genes before p53-mediated and DNA fragmentation-related cell death occurred. We detected twenty-one genes upregulated in response to the juniper extract ( Table 2). The genes were classified to different functional groups (Table 4) to alleviate their analysis. The functions of detected genes were involved in cell death and survival (e.g., apoptosis and autophagy), cellular stress, cell cycle regulation, Ca 2+ -signaling, protein synthesis and protein-protein interactions.
Endoplasmic reticulum (ER) stress is the cellular response to the toxic or harmful stimulus. With this response the cells try to survive dangerous period and after removing the toxic stimulus, the cells restore normal activity. However, when the stimulus is longer, the cells die via apoptosis. Cells exploit ER stress caused by i.e., oxidative stress [65] to reduce misfolded and aggregated proteins, which is reported to trigger cyclosporine A-induced autophagy and apoptosis in malignant glioma cells with the inhibition mTOR/p70S6K1 pathway [66]. ER stress-induced glucose responsive protein BiP/GRP78 (HSPA5) is a heat shock protein, which belongs to chaperones; proteins involved in folding of macromolecules and expressed by ER stress [30]. BiP inhibits unfolded protein response (UPR) pathway and mediates both prosurvival and proapoptotic effects, respectively. It binds overexpressed wild-type p53 in nasopharyngeal carcinoma cells [23] and inhibits apoptosis via PERK/eIF2/NF-κB pathway [32]. In addition, it activates caspase 12 [67] and PERK/eIF2/ATF4/CHOP-mediated apoptosis [32].
Among other functions, ER is involved in Ca 2+ -homeostasis. Three overexpressed genes (CALM2, HSPA5, STRN4) involved in Ca 2+ -signaling and homeostasis advocate the role of calcium in stress response induced by the juniper berry extract. CALM2 (calmodulin) is a major regulator of calcium-mediated signaling, including the regulation of CaMKII-mediated apoptosis via JNK-pathway [28]. Interestingly, Rac1 has been found to promote JNK-mediated apoptosis as well [36] and SEMA3C (Sema E) is one of its activators [68]. Another overexpressed gene STRN4 (zinedin) regulates calmodulin-mediated signaling [57] and is claimed to act as a sensor for concentration changes of Ca 2+ [69]. Based on literature analysis and functional annotation of the list of genes, the juniper extract induces ER stress accompanied with accumulation of p53 in neuroblastoma cells. We also found that the apoptosis-related genes were activated by the juniper berry extract. Therefore, based on the gene expression profiling it is suggested that the treatment with 10 µg/mL of juniper berry extract induces ER stress in the neuroblastoma cells.
Recent studies have revealed that some of the over-expressed proteins comprise specific cellular stress, cell survival or death-mediating mechanisms beyond their well-known main functions (Table 4). Ribosomal proteins encoded by RPLP0 and RPL6 are involved in protein translation but they also possess extra-ribosomal functions. The upregulated RPL6 enhances cell growth and cell cycle progression [70] and RPLP0 has been found to by up-regulated after induction with doxorubicin and TRAIL [37]. The upregulated STRADA encodes STE20-related kinase adapter protein alpha (STRAD), which activates the tumour suppressor LKB1-A kinase involved in cell polarity, energy metabolism and cell growth [71]. LKB1 regulates cellular responses via different effectors and pathways including AMP-activated protein kinase (AMPK) activation after metabolic stress [72] leading to activation mTOR pathway and p53/p21-mediated cell cycle arrest and cell survival/apoptosis. Other tumour suppressor function of LKB1 involves the inhibition of oncogenic protein Yap [73].
In order to understand the cell death-inducing actions of aqueous juniper berry extract, we analyzed also its phenolic composition. To the best of our knowledge, this is the first report on the phenolic composition of an aqueous juniper berry extract. Six flavones, five flavonols, one flavanol, two phenolic acids and one biflavonoid were identified. We excluded the presence of volatile oil components by gas chromatography-mass spectrometry (GC-MS) method, because among phenolics volatile components of plants have shown to affect cellular mechanisms. Therefore, it is assumed that the aqueous nonvolatile oil containing juniper extract would be safer due to its lower toxicity in comparison with the volatile oil containing preparations. Otherwise, similar profiles of identified compounds have been detected by Innocenti et al. [9] with ethanol extract and Miceli et al. [18] with methanol extract. All of the identified compounds have been reported to possess different biological activities. Therefore, it is likely that their effects might explain also our observations partly.
Since the effects of juniper berry extract shown in this study might be explained by the activities of single phenolics, and the particular treatment concentrations were calculated for each phenolic compound found in the composition analysis (Table 1). Phenolic concentrations of our extract varied from 2.5 to 38.1 nM for single compounds. Literature search emphasized on neuroblastoma cells showed that cell death, cell cycle, DNA fragmentation and p53-related activities were described for identified compounds but usually at very high 100-1000-fold larger concentrations than in our study. The present results indicate that there might be synergistic or combinatorial effects involved explaining our results. e.g., 50 µM of apigenin decreased the SH-SY5Y cell viability after 24 h-treatment and increased the amount of subG1 apoptotic SK-N-DZ cells, caspase 3, caspase 8 and the proapoptotic ratio of Bax:Bcl-2 [74]. Apigenin concentration 1 µM decreased cell viability of SH-SY5Y cells 20% after 24 h-treatment but the increase of the ratio Bax:Bcl-2 and caspases 3 and 9 were detected with 50 µM-treatment [75]. In addition, quercetin has shown to activate wild-type p53 by breaking the Mdm2-p53 interaction [76].
There are only a few reports in literature to describe the biological effects of juniper berry extract. Naturally occurring plant phenolics and extracts are known to possess both antioxidant and pro-oxidant properties. The juniper berry extract used in this study has been evaluated for its antioxidant and pro-oxidant activity in our own laboratory [14,15]. It was shown that the extract possessed the pro-oxidant activity to stimulate protein degradation, and the antioxidant activity to chelate iron and to scavenge hydroxyl radicals. These properties might be involved in the cell death-inducing effects shown to plant phenolics. Tunón et al. [7] have tested traditional Swedish medicinal plants for their anti-inflammatory properties based on the ethnopharmacological use of plants. They showed that the extract from juniper berries possessed moderate inhibiting activity in both prostaglandin and platelet activating factor (PAF)-induced exocytosis tests in vitro.
The present results suggest that the juniper extract activates ER stress pathway in the human SH-SY5Y neuroblastoma cells accompanied with the increased amount and translocation of p53 from the cytosol to the nucleus, the expression of specific cell cycle and cell survival regulating genes, the DNA fragmentation and the apoptotic cell death. This activation was seen already before the caspase activity could be clearly detected. Therefore, it is concluded that the juniper berry extract is able to reactivate the impaired p53 function and apoptosis typically found in the human neuroblastoma cells. To the best of our knowledge, this is the first report of the phenolic composition of aqueous juniper berry extract. When the traditional use of plant extracts-including juniper berries-is often based on the aqueous extracts, these results improve the understanding of the possible mechanisms behind the traditional use of them.

Plant Material, Solvents and Reference Substances
Ripe and dried berries of juniper (Juniperus communis L.) cultivated for spice were commercially obtained from Paulig Ltd., Helsinki, Finland [77]. Details of origin or cultivation information are not available for the test material. Acetonitrile and formic acid were of high performance liquid chromatography (HPLC) or LC-MS grade. Flavonoids and phenolic acids such as amentoflavone, apigenin-7-O-glucoside, hyperoside, kaempferol-3-O-glucoside, rutin, protocatechuic acid and rosmarinic acid were purchased from Extrasynthese (Genay Cedex, France).

Aqueous Extraction of Juniper Berries
Ground juniper berries were extracted as described previously [15]. In brief, the extraction was performed by boiling water during three-hour hydrodistillation, resulting in removal of volatile oil. Hydrodistillation was performed using a Clevenger-type apparatus reported in European Pharmacopoeia [78]. The aqueous extract was filtrated, freeze-dried and stored at 4˝C. The yield was 422 mg/g of ground berries [15].

Analysis of Volatile Oils by Gas Chromatography-Mass Spectrometry (GC-MS)
Volatile oils were removed from the extract during the extraction process by hydrodistillation. The absence of volatile oil compounds was verified from hexane and ethyl acetate extracts of aqueous juniper extract (50 mg/mL) by gas chromatography-mass spectrometry (GC-MS). Analyses were carried out with GC-MS on a semipolar NB-54 column (HNU-Nordion Ltd., Helsinki, Finland). Briefly, the column was combined to HP5890 GC coupled to an HP5970 quadrupole mass selective detector, and helium was used as a carrier gas. Column temperature shifted during analyses from 50 to 250˝C at 10˝C/min.

Analysis of Phenolic Composition by High Performance Liquid Chromatography (HPLC)
The analyses were carried out using a Waters HPLC system (Waters Corp., Millford, MA, USA), consisting of a Millenium chromatography manager, 717 autosampler, 2996 photodiode array (PDA) detector, and 600 controller and a pump connected to an in-line degasser at room temperature. Compounds of juniper berry extract were separated in Phenomenex Synergi Fusion-RP column, 150 mmˆ4.6 mm, 4 µm particles (Phenomenex, Torrance, CA, USA), using 0.1% formic acid (A) and acetonitrile (B) as solvents. A gradient elution with a flow rate of 0.4 mL/min was 0-5 min, 5% B; 10 min, 90% A and 10% B; 15-18 min, 75% A and 25% B; 25-28 min, 65% A and 35% B; and 35-40 min, 100% B. The phenolic composition was determined by comparing the retention data and absorption maxima of the UV spectra (200-500 nm) of reference compounds with the values of sample components and earlier published data in literature.

LC-MS-UV Systems and Analyses
High performance liquid chromatography (HPLC) analyses were performed on Agilent 1100 HPLC system (Agilent Technologies, Santa Clara, CA, USA) fitted with the same column as in the phenolic composition analysis, and by using corresponding mobile phases, gradient elution and flow-rate, with an injection volume of 40 µL at room temperature. The UV detection was made at 280 and 350 nm. An Agilent 6410 triple quadrupole MS with electrospray ion (ESI) source was used in negative ion mode.
In the MS identification, full scan ESI-MS´spectra were acquired at a mass range of m/z 100-800 and at fragmentor voltage of 135 V. Juniper extract components were identified by comparing HPLC retention times, UV-vis spectra and MS fragmentation patterns with those of pure substances, and the data published in the literature.
In multiple reaction monitoring (MRM) technique, the fragmentation energy for parent Ñ product ion (m/z) transitions were used at a range between 120 and 210 V for pure phenolic acids and flavonoids. Quantitation of sample components was based on five-point calibration curves (r 2 = 0.991-0.999) prepared for reference substances. Other phenolics, not available as pure compounds, were quantified by using rutin as a standard.
Separation in ultra-performance liquid chromatography (UPLC) analyses was tested under different conditions regarding column, solvent A, gradient elution and flow rate. The analyses were performed on Waters Acquity UPLC™ combined with Waters QTOF Premier MS by using an Acquity™ BEH C18 column (100 mmˆ2.1 mm, 1.7 µm). Gradient solvent system consisted of water/acetonitrile/formic acid (95:5:0.1, solvent A) and acetonitrile (solvent B). The proportion of B increased from 15% to 50% in 17.5 min and to 100% in 2.5 min. The flow rate was 0.3 mL/min and injection volume 5 µL. MS data were collected in ESI´mode at a mass range of m/z 100-1000, and leucine enkephaline was used as the lock spray reagent.

Cell Culture and Treatments
Human SH-SY5Y neuroblastoma cells were cultured in Dulbeccos's modified eagle medium: nutrient mixture F-12 (1:1) containing 15 mM HEPES buffer and L-glutamine and supplemented with 15% heat-inactivated foetal bovine serum, antibiotic mixture of penicillin (170 U/mL) and streptomycin (170 µg/mL) and 1% non-essential amino acids. Continuous culturing of cells was performed at 37˝C in a humidified atmosphere containing 5% CO 2 in air. Cells were treated on 6 cm plates for DNA fragmentation assay and Western blot analysis, and on 10 cm plates for cDNA-RDA assay. For cell experiments, the juniper berry extract was dissolved in mQ-H 2 O at the concentration of 50 mg/mL.

Preparation of Protein Samples for Western Blot Analysis
The SH-SY5Y cells were collected using a cell scraper at different time points after treatment with 10 µg/mL of juniper berry extract, and spun down by centrifugation at 5000ˆg for 5 min. Cell pellets were stored at´80˝C until nuclear and cytoplasmic proteins were extracted by using a commercial kit (ProteoJET Cytoplasmic and Nuclear Protein Extraction Kit, Fermentas, Espoo, Finland) according to manufacturer's instructions. Briefly, cell lysis and cytoplasmic protein extraction were performed by incubating cells with a cell lysis buffer supplemented with protease inhibitor. Cytoplasmic proteins were separated from nuclei by centrifugation at 1000ˆg for 10 min. Nuclear proteins were extracted from washed nuclei with nuclei lysis reagent by shaking samples at 1200 rpm for 15 min and separating proteins by centrifugation at 16,000ˆg for 15 min. All collection and extraction steps were performed on ice or at 4˝C. Protein samples were stored at´80˝C until protein amounts were determined by a colorimetric bicinchoninic acid (BCA) assay (Thermo Scientific, Waltham, MA, USA).

Western Blot Analysis of Cytoplasmic and Nuclear p53
Western blot analysis was performed as previously described by Lantto et al. [13]. Cytoplasmic proteins (15 µg) and nuclear proteins (30 µg) were first separated in 12% SDS-PAGE gel and transferred onto nitrocellulose membrane. Nonspecific binding of antibodies was blocked by 5% non-fat milk powder before exposing the membrane with a monoclonal p53 and β-actin primary antibodies (DO-7, Novocastra, Hämeenlinna, Finland; A1973, Sigma, Helsinki, Finland). To enable the visualization of proteins, the membranes were exposed to a horseradish peroxidase (HRP)-conjugated secondary antibody (HAF007, R&D Systems, Abingdon, UK) and a chemiluminescence reaction was induced by HRP-substrate (SuperSignal West Pico, Thermo Scientific, IL, USA). Proteins were detected and analyzed with GeneGnome system and GeneTools programme (Syngene, Frederick, MD, USA).

DNA Fragmentation
DNA fragmentation was determined by agarose gel electrophoresis as previously described [24,79], with minor modifications. In brief, SH-SY5Y cells were treated with 10 µg/mL of juniper berry extract. Both floating and attached cells were collected at different time points and centrifuged for 5 min at 5000ˆg at 37˝C. The pellets were resuspended in lysis buffer (0.2% Triton X-100; 10 mM Tris-HCl, pH 7.5, and 10 mM EDTA in water) for 20 min on ice. After a centrifugation of 20 min at 16,000ˆg at 4˝C, the supernatants were collected and incubated with RNAse enzyme (final concentration of 100 µg/mL) for 1 h at 37˝C. DNA was purified with phenol/chloroform and precipitated with 100% ethanol overnight at´20˝C. Precipitated DNA was collected by centrifugation for 30 min at 12,000ˆg at room temperature (RT) and washed with 70% ethanol. The pellets were dissolved in TE buffer (10 mM Tris-HCl, pH 8.0, and 10 mM EDTA in water) and mixed with DNA loading dye for electrophoresis. DNA fragments were separated in 2% agarose gel in TAE buffer (40 mM Tris acetate, 2 mM EDTA, pH 8,5) for 2 h 40 min using a voltage of 100 V. The image was photographed with a charge coupled device (CCD) camera (UltraLum, Claremont, CA, USA) under UV light.

cDNA Representational Difference Analysis (RDA)
Total RNA was isolated from juniper berry extract-treated and untreated cell samples by TRIzol-method (Gibco BRL, Gaithersburg, MD, USA). The total RNA samples were stored at 80˝C until the concentration, purity and integrity of total RNA were determined by 1.5% agarose gel and spectrophotometry. Double-stranded cDNA was synthesised from the total RNA by SuperScript™ Double-Stranded cDNA Synthesis kit (Invitrogen, Gaithersburg, MD, USA) according to manufacturer's instructions. Representational difference analysis was performed as previously described by Hubank et al. [80] with some modifications. The ds-cDNA was digested with DpnII-enzyme (R0543S, BioLabs, Ipswich, MA, USA) and desalted R-Bgl-12/24 linkers (TAG Copenhagen, Frederiksberg, Denmark) were ligated to enable the PCR amplification with R-Bgl-24 primer. PCR products were cut by DpnII-enzyme to remove R-linkers and to produce driver sample which represents gene expression of untreated control cells. Tester sample representing gene expression of juniper berry extract-treated cells was prepared from cut PCR products by purifying R-linkers from the samples and ligating J-Bgl-12/24 linkers (TAG Copenhagen). The first difference products (DP1) were generated by hybridizing a mixture of tester and driver in a ratio of 1:100 followed by an exponential PCR amplification of tester:tester hybrids. Driver:tester and driver:driver hybrids were eliminated. To remove false annealing products, the second difference products (DP2) were produced. The DP1 were cut, ligated to N-Bgl-12/24 linkers (TAG Copenhagen) and hybridized with the driver at the ratio of 1:2000. After amplification by PCR the DP2 were cut with DpnII-enzyme, separated and purified by gel electrophoresis and ligated with BamHI-digested pGEM vector (Fermentas). The DP2 were cloned by transferring vectors into E. coli DH10B (TOP10) (Invitrogen) by electroporation. Plasmids were purified and the plasmid DNA analyzed by cycle sequencing (PE Applied Biosystems, 3730 DNA Analyzer, Tartu, Estonia) using M13 forward primers (BigDye, Applied Biosystems, Carlsbad, CA, USA).

Statistical Analysis
The western blot results are shown as arithmetic means˘SEM from three independent experiments. Comparisons between untreated control cells and treated cells were evaluated using a one-tailed Student's t-test, p < 0.05 was defined as significant. Morphological changes of cells are shown as an arithmetic mean˘SD complemented with a median from three independent experiments.