Chemico-Biological Interactions 231 (2015) 27–34

Contents lists available at ScienceDirect

Chemico-Biological Interactions journal homepage: www.elsevier.com/locate/chembioint

Boldine, a natural aporphine alkaloid, inhibits telomerase at non-toxic concentrations Sakineh Kazemi Noureini ⇑, Fatemeh Tanavar Dept. of Biology, Faculty of Basic Sciences, Hakim Sabzevari University, P.O. Box 397, Sabzevar, Iran

a r t i c l e

i n f o

Article history: Received 27 September 2014 Received in revised form 31 January 2015 Accepted 23 February 2015 Available online 3 March 2015 Keywords: Cancer Boldine Telomerase hTERT Splicing variants

a b s t r a c t In a preliminary screening study of natural alkaloids, boldine, an aporphine alkaloid, showed an interesting dose and time dependent anti-proliferative effect in several cancer cell lines. Cytotoxicity of boldine in human fibroblasts was considerably lower than the telomerase positive embryonic kidney HEK293 and breast cancer MCF-7 and MDA-MB-231 cells. Whether boldine can inhibit telomerase was investigated here using a modified quantitative real-time telomere repeat amplification protocol (q-TRAP). This test showed that boldine inhibits telomerase in cells treated with sub-cytotoxic concentrations. Telomerase inhibition occurs via down-regulation of hTERT, the catalytic subunit of the enzyme. Boldine changed the splicing variants of hTERT towards shorter non-functional transcripts as well. A direct interaction of boldine with the enzyme may also be involved, though thermal FRET method did not detect any substantial interaction between boldine and synthetic telomere sequences. This study advocates boldine as a valuable candidate for telomerase-targeted cancer care. This study suggests that derivatives of boldine could be potent anti-cancer drugs. Ó 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Telomeres, the terminal DNA at chromosome ends, progressively shorten during each cell division and limit the replicative life span of dividing cells [1]. It is believed that continuous proliferation and avoidance of replicative senescence are caused by the action of telomerase, the cellular ribonucleoprotein reverse transcriptase that is responsible for telomere elongation [2,3]. Telomerase is active at extremely low levels in most human normal tissues, whereas it is strongly reactivated in more than 85% of human tumors [4–6]. Therefore, telomerase has been proposed as a critical anticancer target [7], the inhibition of which has the potential to serve as an anti-cancer strategy [8]. Natural compounds with extensive structural diversity are valuable resources for drug discovery. Several natural small molecules have been discovered so far that arrest proliferation of cancer cells by inhibiting telomerase [9–12]. This study has focused on evaluation of cytotoxicity and telomerase inhibitory effects of boldine (1,10-dimethoxy-2,9-dihydroxyaporphine), a natural aporphine alkaloid found abundantly in Peumus boldus [13], on cancer cells. This strong antioxidant has shown to have several pharmacological

activities, such as anti-inflammatory, antipyretic, antiatherogenic, antiplatelet, antitumor promoting, cytoprotective and tyrosinase inhibitory properties [14,15]. In addition, boldine has been suggested as a therapeutic agent in the treatment of diabetes mellitus [16]. It has also shown to attenuate catecholamine oxidation-induced brain mitochondrial dysfunction [17]. Aporphinoids in general exhibit a wide range of biological properties [18,19], for example isolated aporphines of plant Cassytha filiformis (Lauraceae) possess in vitro antiproliferative properties in a number of cancer and non-cancer cell lines [20,21]. Several of the aporphine alkaloids extracted from natural herbs and related derivatives of boldine are known as inhibitors of topoisomerase I or II [22]. In this study boldine showed strong suppression of proliferation in embryonic kidney cells HEK293 and two telomerase-positive breast cancer cells; consequently its ability to inhibit telomerase activity and probable involved mechanisms were investigated. By our knowledge, no other report exists on the ability of boldine to inhibit telomerase.

2. Materials and methods 2.1. Cell culture

⇑ Corresponding author. Tel.: +98 51 4401 3012, +98 915 572 5818 (cell); fax: +98 51 4401 3365. E-mail address: [email protected] (S. Kazemi Noureini). http://dx.doi.org/10.1016/j.cbi.2015.02.020 0009-2797/Ó 2015 Elsevier Ireland Ltd. All rights reserved.

Each of the breast cancer cell lines (MCF-7 and MDA-MB-231), human embryonic kidney 293 cells (HEK293) from DSMZ

S. Kazemi Noureini, F. Tanavar / Chemico-Biological Interactions 231 (2015) 27–34

2.2. Cytotoxicity assay Cell viability was evaluated using MTT (3-(4,5-dimethylthiazol2-yl)-2,5-diphenyl-tetrazolium bromide) (Sigma–Aldrich) assay [25]. Briefly, cells were seeded in 96-well plates at a density of 1  104 cells/well and treated with each compound at serially diluted concentrations. After various incubation times, MTT at 0.5 mg/ml final concentration was added and incubated 4 h to be reduced to blue formazan by viable cells. Absorbance of the dye, after dissolving in DMSO containing 10% SDS and 1% acetic acid, was measured at 570 nm using a plate reader (BioTek, USA) and cell viability was analyzed using Gen5 software version 1.06. The assay was carried out at least in three independent logical repeats each of which includes samples in triplicates. The concentration of boldine that caused cell growth decrease to 50% of untreated controls, IC50, was determined from the dose–response curves. The results are presented as means ± SD. 2.3. Telomerase assay (SYBR Green q-TRAP assay) Telomerase activity was measured using a SYBR-Green quantitative-telomere repeat amplification protocol (q-TRAP) based on the method of Hou et al. [26] with some small modifications [27]. Briefly the MCF-7 cells after 48 h incubating with various concentrations of boldine and/or berberine were washed with PBS, lysed in a buffer containing 10 mM Tris–HCl pH = 7.5, 1 mM MgCl2, 1 mM EGTA, 0.1 mM Phenylmethylsulfonylfluoride (PMSF), 5 mM beta-mercaptoethanol, 0.5% CHAPS and 10% glycerol and incubated for 30 min at 4 °C. Protein concentration of the cell lysates was measured using microBradford assay with plate reader (BioTek, USA) and analyzed with Gene5 software version 1.06. Using SYBR Green PCR Master Mix (GenetBio, South Korea), qTRAP assay was performed to compare telomerase activity in equal amounts of protein extracts from samples in a real-time thermal cycler Rotor-Gene 3000 (Corbett Research). The reaction mixtures including 1X SYBR Green master mix, 1 lg protein of cell extract, 10 pmol TS (50 -AATCCGTCGAGCAGATT-30 ) and 5 pmol ACX (50 -GCGCGGCTTACCCTTACCCTTACCCTAACC-30 ) primers were incubated 30 min at 25 °C. Then the amplification of telomerase products was started at 94 °C for 10 min to activate the hot-start Taq polymerase and the 40 cycles of 30 s at 94 °C, 30 s at 50 °C and 45 s at 72 °C with signal acquisition. The threshold cycle value (Ct) determined for each sample by using Rotor-Gene 6.01 software was compared with those of the standards generated from serially diluted cell lysate of untreated MCF-7 control. The

experiment has been repeated at least four times independently each including triplicates for each concentration of boldine. 2.4. q-TRAP-ligand assay In a q-TRAP assay two enzymes are involved; first telomerase elongates the synthetic substrate, TS, and then the products are amplified by a hot-start Taq polymerase. In q-TRAP-ligand experiments (also known as in vitro TRAP assay) the reactions were treated with boldine in two distinct steps to differentiate between its effects on the enzymes. Two different incubation times were applied; 0 (for adding boldine immediately) and 30 min before the correlated enzyme starts its activity. Briefly, a master mix of q-TRAP reaction including MCF-7 cell lysate was prepared and aliquoted to four sets on ice (A to D). Samples in each set were treated with different concentrations of boldine (10, 100 and 160 lM) at the specified step and for the defined incubation time. In two sets boldine was added before telomerase activity; one set treated immediately before enzyme activity (B) and another set were incubated 30 min in presence of boldine on ice (A); in these two sets, both telomerase and Taq polymerase are exposed to boldine. Then all samples were incubated 20 min at 25 °C for extending TS primer by telomerase. All four sets were put back on ice and

A

Boldine (Molecular Weight: 327.37)

** 100

*#

MCF7 MDA-MB-231 HEK293 HFF

**

90 80 % cell viability

(German Collection of Microorganisms and Cell Cultures) and normal human foreskin fibroblast cells (HFF) (Royan Institute, Iran) were cultivated in Dulbecco’s modified Eagle’s medium supplemented (DMEM High Glucose with stable Glutamine) with 10% fetal bovine serum (FBS gold), 100 U/ml penicillin, and 100 lg/ml streptomycin in a humidified atmosphere containing 5% CO2 at 37 °C. All the materials were purchased from PAA, Austria. All the cell types were sub-cultured routinely when reached to almost 80% confluence and the cell viability was briefly estimated using trypan blue exclusion method [23]. All the following experiments were performed with cells in the logarithmic growth phase. Berberine and boldine were purchased from Sigma–Aldrich and dissolved in absolute ethanol (Merck) at a concentration of 50 mM (stock solution) and stored at 20 °C until use. Each stock solution was serially diluted in medium before use and the maximum final concentration of ethanol in cell cultures did not exceed 0.1%. Berberine was used as a known compound with telomerase inhibitory effect [24] and the data collected for boldine was compared with that of MCF-7 treated with berberine as a control.

#

70

*

**

60 50 *

40 30

#

**

20 10 0 0

B

62.5

125 250 500 Boldine concentraon (μM)

1000

2000

100 90 80 70 % cell viability

28

60 50 40 30 20 10 0 0

31 62 125 250 Berberine concentraon (μM)

500

1000

Fig. 1. Viability of different cell cultures MCF-7, MDA-MB-231, HEK293, HFF incubated for 48 h with various concentrations of boldine using MTT assay (A). Cell viability of MCF-7 under treatment of berberine was shown in (B). The mean values ± SD have been represented; p values for ⁄, ⁄⁄ and # are 0.001, 0.05 and 0.02 respectively.

29

S. Kazemi Noureini, F. Tanavar / Chemico-Biological Interactions 231 (2015) 27–34

then boldine was added to set C and 30 min later to D. In these two sets only Taq polymerase are affected; samples of set C experienced 30 min incubation with boldine while set D had an immediate exposure to boldine (no incubation time). The amplification step and quantification were done exactly as mentioned in previous section. This experiment has been repeated at least three times, and each repeat included triplicate samples for each concentration of boldine.

Time from start point (min) Set

0

30

A B

Treatment with boldine –

C

–

Treatment with boldine –

D

–

–

Incubation on ice

50

Telomerase activity at 24 °C

Incubation on ice

2.5. Total RNA isolation, cDNA synthesis and real-time PCR Total RNA was isolated from control and boldine-treated MCF-7 cells using RNX-Plus solution (SinaClone BioScience, Iran) according to the manufacturer’s instructions and cDNA synthesis was performed on 2 lg of each sample using MMULV reverse transcriptase (Vivantis, Koera). Expression level of hTERT and b2 microglubolin genes was detected with the quantitative reverse transcription polymerase chain reaction (qRT-PCR) by using specific intron-spanning primers as explained above [27]. Briefly, 1 ll of each cDNA sample was subjected to PCR-amplification reaction including SYBR Green PCR Master Mix (GenetBio, Korea) and 5 pmol of each primer. Relative mRNA copy number of hTERT gene to that of the housekeeping gene, b2 microglubolin, was compared among the control and treated samples using the related standard curves calculated by Rotor-Gene 6.01 software.

*

90

*

**

24 h 48 h

#

72 h

% cell viability

80 70

**

*

60 50

# **

40 30

#

20

*

10 0 0

62.5 125 250 500 Boldine concentration (µM)

The hTERT transcript that encodes the catalytic subunit of telomerase has at least six alternate splicing sites including four insertions in human cancer cells [28,29], which cause premature translation terminations. To examine the probable effects of boldine on splicing pattern of hTERT, the full length functional and the defective transcripts were amplified using specific primers

30–50

The probable traces of RNase contamination that potentially can give rise to false positive results was checked by incubating total RNA with aliquots of boldine for 0 or 30 min at room temperature followed by electrophoresis in agarose gel.

100

2.6. Estimation of alternative hTERT splicing variants

1000

2000

Fig. 2. Time- and concentration-dependent reduction of MCF-7 cell viability under treatment with boldine.

80

–

–

–

–

Treatment with boldine –

–

80–180 Amplification of telomerase products (real-time PCR)

Treatment with boldine

(hTERT-2026F: GCCTGAGCTGTACTTTGTCAA and hTERT-2482R: CGCAAACAGCTTGTTCTCCATGTC) for its exon junction spanning the RT region of hTERT [30]. This region contains a and b splicing sites and products of four different sizes may potentially be seen. The hTERT mRNA with a full length transcript is identified as hTERT + a + b (457 bp). This transcript is translated to the functional hTERT. Three other products are non-functional and identified as hTERT-a (a variant; 421 bp), the major nonfunctional hTERT-b (b variant; 275 bp), and hTERT-a-b (-a-b variant; 239 bp).

2.7. Thermal FRET analysis Interaction of small molecules with telomerase substrate and interfering with the enzyme activity can be estimated using a simple thermal melting experiment based on Guédin et al. [31]. This method demonstrates the structural stabilization effect of the small molecule on a dual-labeled fluorescence synthetic oligonucleotide of human telomere sequence rich in guanine, F21T (FAM 30 -GGG(TTAGGG)3-50 TAMRA), which inherently folds to an intra-molecular quadruplex structure. Fluorescence intensity of F21T is measured while it is heated gradually; heat will unfold the quadruplex structure and fluorescence intensity will increase in an almost two-state denaturation pattern. The temperature at which the folded-to-unfolded transition occurs is defined as melting temperature (Tm). More stable structures are associated with higher melting temperatures. An increase in Tm indicates preferential ligand binding to the folded form of the molecule rather than the unfolded. Briefly, F21T at final concentration of 0.25 lM was heated at 95 °C for 10 min, quickly chilled on ice and then incubated at 37 °C for 2 h in presence of sodium cacodylate (10 mM), KCl (100 mM); a similar concentration of the major cation ions as inside the cells, and various concentrations of boldine from equimolar to F21T up to 2000 equimolar. Fluorescence intensity of F21T was measured using Rotor-Gene 3000 real-time thermal cycler (Corbett Research) using filters suited for FAM while temperature increased 1 °C/min. Differential fluorescence intensity over temperature was used to determine the melting temperature (Tm) using Rotor-Gene software version 1.06.

S. Kazemi Noureini, F. Tanavar / Chemico-Biological Interactions 231 (2015) 27–34

B 100 90 80 70 60 50 40 30 20 10 0

C

MDA-MB-231

0 10 100 160 boldine concentration (µM)

HEK293 100 90 80 70 60 50 40 30 20 10 0

% Telomerase activity

MCF-7 100 90 80 70 60 50 40 30 20 10 0

% Telomerase activity

A

% Telomerase activity

30

0 10 80 150 boldine concentration (µM)

0 10 60 110 boldine concentration (µM)

Fig. 3. The alterations of telomerase activity in MDA-MB-231, MCF-7, and HEK293 cell lines 48 h treated with various concentration of boldine.

2.8. Statistical analysis Statistical analysis was performed by using one-way ANOVA test and a p < 0.05 was considered as the cut off for significant difference. 3. Results 3.1. Boldine exhibits dose- and time-dependent cytotoxicity in immortal cell lines Cytotoxic effect of boldine after 48 h treatment in human foreskin fibroblasts (HFF) and telomerase positive immortal cells HEK293, MDA-MB-231, MCF-7 was estimated by MTT method. The results of at least four logical independent repetitions have been shown in dose response curves in Fig. 1A. Boldine was less toxic in fibroblasts than immortal cells (IC50 values after 48 h treatment are 110 ± 4.7, 150 ± 5.2, 160 ± 4.5 and 280 ± 6.1 lM in HEK293, MDA-MB-231, MCF-7 and HFF respectively). MTT method estimated 54 ± 2.3 lM as the 48 h IC50 value for berberine, the known telomerase inhibitor alkaloid, in MCF-7 cells (Fig. 1B). Altogether, exposure to boldine resulted in a dose- and timedependent decrease in cell survival of MCF-7 cells (Fig. 2).

Our data in Fig. 4 shows an obvious dose-dependent repression of transcripts of the hTERT gene after 48 h treatment in MCF-7 cells in a similar pattern with inhibition of telomerase activity. However, a stronger reduction is seen in the enzyme activity than in the hTERT mRNA level. To further elucidate the mechanism(s) of inhibition, q-TRAP in presence of boldine (q-TRAP-ligand assay), hTERT variants, and thermal FRET analysis were assessed. 3.2.3. Boldine inhibits telomerase activity in q-TRAP-ligand assay In q-TRAP-ligand assay, TRAP reactions were incubated with boldine at two different steps; before telomere extension by

A

100

% Telomerase activity

# *

90

% hTERT/β2MG expression *

80 70 60

*

50 #

40 30

#

20

3.2. Boldine effectively suppresses telomerase by multiple mechanisms

10 3.2.1. Boldine strongly inhibits telomerase activity in treated cells A real-time q-TRAP assay analysis based on Hou et al. was used to assess telomerase activity in MDA-MB-231, MCF-7 cancer and HEK293 non-cancerous immortal cell lines after 48 h treatment with boldine (Fig. 3). A clear decrease in telomerase activity was observed in all the telomerase-positive cell lines after treatment with sub-cytotoxic concentrations of boldine. Interestingly telomerase inhibition in MDA-MB-231, a more aggressive breast cancer cell line than MCF-7, is very strong even at 10 lM concentration of boldine (Fig. 3A and B). At this concentration boldine decreased telomerase activity of MDA-MB-231 to 61% of un-treated cells (p < 0.01). However, telomerase in HEK293 was also inhibited in a comparable level (Fig. 3C). 3.2.2. Boldine reduces hTERT mRNA level in the treated cells Telomerase inhibition may be due to down-regulation of the hTERT gene, the most important way in telomerase regulation [32], direct interaction of boldine with the enzyme, altered splicing patterns of the hTERT transcript that produce non-functional variant(s), and/or interference of the enzyme activity by limiting its access to the substrate.

0 0

10

100

Boldine concentration (µM)

160

B

Fig. 4. Dose-dependent alterations of telomerase activity and hTERT mRNA levels in MCF-7 cells treated 48 h with 0, 10, 100 and 160 lM boldine (A). The mean values ± SD are shown; ⁄p < 0.02 and #p < 0.05. TRAP products resolved in 15% polyacrylamide gel (B) show the 6 bp ladder of telomerase activity. The smallest band of the marker is 50 bp. N: negative control that contains heat-inactivated cell extract; 1, 1:5, 1:25 and 1:125 are the reactions containing serially diluted cell extract. The reactions related to extracts of the treated cells with 10, 100 and 160 lM boldine have been loaded after M, the DNA Size marker.

S. Kazemi Noureini, F. Tanavar / Chemico-Biological Interactions 231 (2015) 27–34

A

31

B

C

Boldine concentraon (μM)

treated aer telomere elongaon (in %)

∆ TA = TA treated before telomere elongaon - TA

0

10

100

160

0

D -20

* *

-40 # *

*

-60

-80

#

-100

Fig. 5. Telomerase activity (TA) in presence of different concentrations of boldine added before (A, p < 0.003) or after (B, p < 0.064) telomere elongation both without incubation time; resulted from 5 independent experiments. The difference of telomerase activity (TA) in presence of various concentrations of boldine without (gray) or with 30 min incubation time (black) (C). The columns show the difference between the q-TRAP estimated telomerase activities of the samples treated before and those treated after telomere extension step; gray: no incubation time, black: 30 min incubation. The mean values ± SD has been shown, ⁄p < 0.01 and #p < 0.05. To control the RNase contamination in boldine solution, total RNA without (left) and after 30 min incubation with stock solution of boldine (right) was resolved in 1% agarose gel (D).

telomerase (sets A and B) and after (sets C and D). The difference between the estimated telomerase activities of these two reaction sets represents telomerase inhibition. The difference between the reactions of A and C (A–C) in Fig. 5 shows a dose-dependent decrease in telomerase activity in treated samples. A greater difference between B and D (B–D) in Fig. 5C shows that this effect grows stronger when the enzyme is incubated for 30 min in presence of boldine. Altogether, results of this experiment suggest direct interaction of boldine with telomerase ribonucleoprotein. 3.2.4. Boldine stimulates non-functional splice variant(s) of hTERT Using specific primers to study splicing pattern of hTERT [30] all four bands of alternative spliced variants of hTERT were clearly visible in untreated control cells (Fig. 6). After 48 h treatment with increasing concentrations of boldine, a considerable decrease was clearly seen in the full-length variant while the three shorter transcripts were still visible. Although the full-length transcript exhibited a marked decrease by 10 lM treatment, all non-functional

variants remained unchanged. However, by increasing boldine concentration, the non-functional variants also decreased and the -b spliced form was the only detectable variant in the treated cells with 160 lM boldine. 3.2.5. Boldine does not interact with synthetic telomeric G-rich sequences Thermal FRET analysis showed only a small increase (